This document summarizes a distributed protocol for dynamic IP address assignment in mobile ad hoc networks (MANETs). In MANETs, nodes can enter and leave the network dynamically, so a static pre-assignment of IP addresses is not feasible. Existing solutions like DHCP require centralized servers that may not exist in MANETs. The proposed solution guarantees unique IP address assignment in MANETs under various conditions like message losses and network changes, with low latency and communication overhead. It provides a distributed alternative to address the dynamic nature of nodes in MANETs.
A Survey On IP Address Assignment In A Mobile Ad HocKristen Flores
This document summarizes research on IP address assignment protocols for mobile ad hoc networks. It begins with an abstract describing the challenges of address assignment in infrastructure-less mobile ad hoc networks. It then provides background on traditional address assignment schemes and why they cannot be directly applied to mobile ad hoc networks. The document reviews requirements for address assignment protocols in this context. It proceeds to describe several existing IP address assignment protocols for mobile ad hoc networks, categorizing them as best effort, leader-based, or decentralized approaches. It concludes that more research is needed to solve the challenging problem of dynamic address assignment in unpredictable and merging mobile ad hoc networks.
An Efficient FB Addressing Protocol for Auto configuration of Ad Hoc Networksijceronline
International Journal of Computational Engineering Research (IJCER) is dedicated to protecting personal information and will make every reasonable effort to handle collected information appropriately. All information collected, as well as related requests, will be handled as carefully and efficiently as possible in accordance with IJCER standards for integrity and objectivity.
IJRET : International Journal of Research in Engineering and Technology is an international peer reviewed, online journal published by eSAT Publishing House for the enhancement of research in various disciplines of Engineering and Technology. The aim and scope of the journal is to provide an academic medium and an important reference for the advancement and dissemination of research results that support high-level learning, teaching and research in the fields of Engineering and Technology. We bring together Scientists, Academician, Field Engineers, Scholars and Students of related fields of Engineering and Technology.
A Mobile Ad-hoc network is a set of mobile terminals moving in different directions at different
speed being wirelessly connected to each-other. In this paper we study the proactive Link State Routing
Protocol – OLSR, which uses hello and topology control (TC) messages to find then distribute link state
data throughout the mobile ad hoc network. Individual nodes use this topology data to calculate next hop
destinations for all nodes within the network using shortest hop forwarding paths. We then suggest ways
by which the existing algorithm can be optimized in terms of delay, throughput, power consumption, jitter
etc. Finally we summarize the applications of OLSR.
Mitigation of Selfish Node Attacks In Autoconfiguration of MANETsIJAAS Team
Mobile ad-hoc networks (MANETs) are composed of mobile nodes connected by wireless links without using any pre-existent infrastructure. Hence the assigning of unique IP address to the incoming node becomes difficult. There are various dynamic auto configuration protocols available to assign IP address to the incoming nodes including grid based protocol which assigns IP address with less delay and low protocol overhead. Such protocols get affected by presence of either selfish nodes or malicious nodes. Moreover there is no centralized approach to defend against these threats like in wired network such as firewall, intrusion detection system, proxy etc. The selfish nodes are the nodes which receive packet destined to it and drop packet destined to other nodes in order to save its energy and resources. This behavior of nodes affects normal functioning of auto configuration protocol. Many algorithms are available to isolate selfish nodes but they do not deal with presence of false alarm and protocol overhead. And also there are certain algorithms which use complex formulae and tedious mathematical calculations. The proposed algorithm in this paper helps to overcome the attack of selfish nodes effect in an efficient and scalable address auto configuration protocol that automatically configures a network by assigning unique IP addresses to all nodes with a very low protocol overhead, minimal address acquisition delay and computational overhead.
Detection of Duplicate Address in Mobile Adhoc Networks for On Demand Routing...IJMER
International Journal of Modern Engineering Research (IJMER) is Peer reviewed, online Journal. It serves as an international archival forum of scholarly research related to engineering and science education.
The document discusses several topics related to computer networking:
1. It explains how subnet masks are used to distinguish the network and host parts of IP addresses and how they allow controlling the number of addresses on a network.
2. It describes how checksums are used in internet protocols like IP, UDP and TCP to detect errors in transmitted data.
3. It discusses VLANs and how they provide logical network segmentation within a physical LAN, as well as inter-VLAN routing.
4. It covers load balancing techniques used to distribute network traffic across multiple servers for improved scalability and reliability.
ANALYSIS OF PROACTIVE AND REACTIVE MANET ROUTING PROTOCOLS UNDER SELECTED TCP...ijasuc
This document analyzes the performance of two reactive MANET routing protocols, DSR and DSDV, under TCP Vegas and TCP Newreno variants through simulations. The simulations measured packet delivery ratio, average end-to-end delay, and total packets dropped. The results showed that DSDV generally had a higher packet delivery ratio but also higher end-to-end delay and more packet drops compared to DSR. DSR performed better in terms of delay and drops due to its on-demand route discovery, while DSDV maintained more consistent routes leading to better packet delivery.
A Survey On IP Address Assignment In A Mobile Ad HocKristen Flores
This document summarizes research on IP address assignment protocols for mobile ad hoc networks. It begins with an abstract describing the challenges of address assignment in infrastructure-less mobile ad hoc networks. It then provides background on traditional address assignment schemes and why they cannot be directly applied to mobile ad hoc networks. The document reviews requirements for address assignment protocols in this context. It proceeds to describe several existing IP address assignment protocols for mobile ad hoc networks, categorizing them as best effort, leader-based, or decentralized approaches. It concludes that more research is needed to solve the challenging problem of dynamic address assignment in unpredictable and merging mobile ad hoc networks.
An Efficient FB Addressing Protocol for Auto configuration of Ad Hoc Networksijceronline
International Journal of Computational Engineering Research (IJCER) is dedicated to protecting personal information and will make every reasonable effort to handle collected information appropriately. All information collected, as well as related requests, will be handled as carefully and efficiently as possible in accordance with IJCER standards for integrity and objectivity.
IJRET : International Journal of Research in Engineering and Technology is an international peer reviewed, online journal published by eSAT Publishing House for the enhancement of research in various disciplines of Engineering and Technology. The aim and scope of the journal is to provide an academic medium and an important reference for the advancement and dissemination of research results that support high-level learning, teaching and research in the fields of Engineering and Technology. We bring together Scientists, Academician, Field Engineers, Scholars and Students of related fields of Engineering and Technology.
A Mobile Ad-hoc network is a set of mobile terminals moving in different directions at different
speed being wirelessly connected to each-other. In this paper we study the proactive Link State Routing
Protocol – OLSR, which uses hello and topology control (TC) messages to find then distribute link state
data throughout the mobile ad hoc network. Individual nodes use this topology data to calculate next hop
destinations for all nodes within the network using shortest hop forwarding paths. We then suggest ways
by which the existing algorithm can be optimized in terms of delay, throughput, power consumption, jitter
etc. Finally we summarize the applications of OLSR.
Mitigation of Selfish Node Attacks In Autoconfiguration of MANETsIJAAS Team
Mobile ad-hoc networks (MANETs) are composed of mobile nodes connected by wireless links without using any pre-existent infrastructure. Hence the assigning of unique IP address to the incoming node becomes difficult. There are various dynamic auto configuration protocols available to assign IP address to the incoming nodes including grid based protocol which assigns IP address with less delay and low protocol overhead. Such protocols get affected by presence of either selfish nodes or malicious nodes. Moreover there is no centralized approach to defend against these threats like in wired network such as firewall, intrusion detection system, proxy etc. The selfish nodes are the nodes which receive packet destined to it and drop packet destined to other nodes in order to save its energy and resources. This behavior of nodes affects normal functioning of auto configuration protocol. Many algorithms are available to isolate selfish nodes but they do not deal with presence of false alarm and protocol overhead. And also there are certain algorithms which use complex formulae and tedious mathematical calculations. The proposed algorithm in this paper helps to overcome the attack of selfish nodes effect in an efficient and scalable address auto configuration protocol that automatically configures a network by assigning unique IP addresses to all nodes with a very low protocol overhead, minimal address acquisition delay and computational overhead.
Detection of Duplicate Address in Mobile Adhoc Networks for On Demand Routing...IJMER
International Journal of Modern Engineering Research (IJMER) is Peer reviewed, online Journal. It serves as an international archival forum of scholarly research related to engineering and science education.
The document discusses several topics related to computer networking:
1. It explains how subnet masks are used to distinguish the network and host parts of IP addresses and how they allow controlling the number of addresses on a network.
2. It describes how checksums are used in internet protocols like IP, UDP and TCP to detect errors in transmitted data.
3. It discusses VLANs and how they provide logical network segmentation within a physical LAN, as well as inter-VLAN routing.
4. It covers load balancing techniques used to distribute network traffic across multiple servers for improved scalability and reliability.
ANALYSIS OF PROACTIVE AND REACTIVE MANET ROUTING PROTOCOLS UNDER SELECTED TCP...ijasuc
This document analyzes the performance of two reactive MANET routing protocols, DSR and DSDV, under TCP Vegas and TCP Newreno variants through simulations. The simulations measured packet delivery ratio, average end-to-end delay, and total packets dropped. The results showed that DSDV generally had a higher packet delivery ratio but also higher end-to-end delay and more packet drops compared to DSR. DSR performed better in terms of delay and drops due to its on-demand route discovery, while DSDV maintained more consistent routes leading to better packet delivery.
Address allocation for MANET merge and partition using cluster based routingsugandhasinghhooda
Network merges and partitions occur quite often in MANET wherein address auto-configuration is a critical
requirement. There are various approaches for address auto-configuration in MANETs which allocate address to
the nodes in a dynamic and distributed manner in which HOST ID and MANET ID are assigned on the basis of
their Base value. MANET merges and partitions employing Cluster Based Routing Protocol require a node to be
assigned as the Cluster Head (CH). This paper presents the Election Algorithm which assigns a node as the Cluster
Head on the basis of its weight. Through simulation using the NS-2, it has been shown that the Election Algorithm
improves the packet delivery ratio (PDR) significantly and decreases the packet delay to a great extent in comparison
to the existing AODV protocol.
MOBILE INTERNET PROTOCOL AND TRANSPORT LAYER
Overview of Mobile IP – Features of Mobile IP – Key Mechanism in Mobile IP – route Optimization. Overview of TCP/IP – Architecture of TCP/IP- Adaptation of TCP Window – Improvement in TCP Performance.
1. Serial0 is up but the line protocol is down, indicating a potential issue with the cable or configuration settings not matching on both ends.
2. Static IP addresses are manually assigned while dynamic IP addresses are assigned by a DHCP server from its IP range, and have a lease duration after which the address may change.
3. Protocols establish rules and standards for moving data across a network, such as routing protocols like RIP, IGRP, and EIGRP that determine the best path between networks.
This document discusses the Teredo protocol, which enables IPv6 connectivity for nodes located behind IPv4 NAT devices. It explains how Teredo works by tunneling IPv6 packets over UDP through NATs. While Teredo allows IPv6 connectivity, it also raises security concerns by bypassing security controls and allowing unsolicited traffic. The document analyzes attacks that could exploit vulnerabilities in Teredo tunnels, such as a denial of service attack against a Teredo server using a single packet. It investigates whether Teredo represents a security risk or is a worthwhile transition mechanism from IPv4 to IPv6.
AUTOCONFIGURATION ALGORITHM FOR A MULTIPLE INTERFACES ADHOC NETWORK RUNNING...IJCNC
Network configuration is the assignment of network parameters necessary for a device to integrate the
network, examples being: an IP address, netmask, the IP address of the gateway, etc ... In the case of
Mobile Ad hoc NETworks (MANETs), the connectivity of nodes is highly dynamic and a central
administration or configuration by the user is very difficult. This paper presents an autoconfiguration
solution for ad hoc networks running the widely implemented version of OLSR routing protocol, the 2003
RFC 3626 [1]. This solution is based on an efficient Duplicate Address Detection (DAD) algorithm,
which takes advantage of the genuine optimization of the OLSR routing protocol. The proposed
autoconfiguration algorithm is proved to operate correctly in a multiple interfaces OLSR network.
The document discusses topics related to the network layer, including:
1. It describes virtual circuits and datagrams, which are two methods for transferring data across networks.
2. It covers IPv4 addressing concepts such as address space, notations, classful and classless addressing, subnetting, and network address translation.
3. It provides an overview of additional network layer topics like IPv6 addressing, routing algorithms, internet control protocols, and routing protocols.
Routing protocols in Ad-hoc Networks- A Simulation StudyIOSR Journals
This document summarizes a study that simulated and compared several routing protocols for ad hoc networks using the Network Simulator 2 (NS-2). It began by describing ad hoc networks and reviewing related work comparing routing protocols. It then provided an overview of desirable routing protocol properties and described several prominent protocols in detail: DSDV, AODV, DSR, ZRP, TORA, IMEP, and CBRP. The document outlined the NS-2 simulator and the mobility extension used. It concluded by detailing the simulation methodology and modifications made to the AODV and DSR implementations in NS-2 to perform the comparative analysis.
UNIT IV MOBILE AD-HOC NETWORKS
Ad-Hoc Basic Concepts – Characteristics – Applications – Design Issues – Routing – Essential of Traditional Routing Protocols –Popular Routing Protocols – Vehicular Ad Hoc networks ( VANET) – MANET Vs VANET – Security
This document analyzes and compares the scalability of three mobile ad hoc network (MANET) routing protocols - AODV, TORA, and OLSR - by simulating them using varying numbers of nodes. The key findings are:
1) In terms of end-to-end delay, OLSR consistently performs better than AODV and TORA as the number of nodes increases from 25 to 75.
2) For throughput, OLSR outperforms AODV and TORA across all network sizes tested.
3) The study concludes that OLSR demonstrates better scalability than AODV and TORA based on its performance with increasing numbers of nodes in terms of
This document provides an overview of mobile ad hoc networks (MANETs). It defines MANETs as wireless networks without fixed infrastructure where nodes are free to move and dynamically self-organize. Key points:
- MANETs are formed spontaneously as nodes connect via wireless links without centralized administration.
- Each node acts as a router, forwarding data for other nodes. Network topology changes frequently as nodes move.
- Examples of MANET applications include military operations, emergency response, classroom environments, and home/office networks.
- Challenges for MANET routing protocols include dynamic topology, limited resources, and network security. Common routing approaches include proactive, reactive, and geographic routing.
IJERA (International journal of Engineering Research and Applications) is International online, ... peer reviewed journal. For more detail or submit your article, please visit www.ijera.com
Performance Analysis of Mobile Adhoc Network Routing Protocols Over Tcppijans
In order to reduce the communication cost and time we are looking forward for successful implementation
of an infrastructure less network like Mobile Ad Hoc Network in all arena of wireless mobile
communication. But still it is a challenge to decide a most appropriate routing protocol for MANET. In
MANET there is no fixed topology due to the mobility of nodes, interference, multipath propagation and
path loss. Since MANET does not use fixed infrastructure rather it always have to find suitable router and
routing path for each communication, existing established routing protocol is not suitable for MANET to
function properly. Different Routing protocols have been proposed to meet the challenges with MANETs.
This paper evaluates the performances of four MANET routing protocols which are DSDV, AODV, DSR
and TORA over TCP, a Transport Layer Protocol. The performance metrics which are considered in this
paper are packet delivery fractions, normalized routing overload, end to end delay and throughput. DSDV
is a proactive protocol and the others are reactive protocols.
Network Address Translation (NAT) allows private IP addresses to be used within a local area network (LAN) while providing access to the public internet. NAT maps private IP addresses to public IP addresses, allowing multiple devices to share public IP addresses. The main NAT traversal challenges are that NAT prevents outside systems from initiating connections to inside systems and communication between systems that are both behind NAT routers. Proposed solutions include using third-party servers to reverse connections or techniques like UDP and TCP hole punching that establish connections directly between systems.
MANET stands for mobile ad hoc network. It is a type of wireless network that can change locations and configure itself without a centralized administration. Nodes in a MANET can connect to each other to form a temporary network without any existing network infrastructure. Routing in MANETs is challenging due to the dynamic network topology, asymmetric links, and interference. Common routing algorithms for MANETs include distance vector, link state, and various protocols designed specifically for MANETs to handle mobility.
MANET stands for mobile ad hoc network. It is a type of wireless network that can change locations and configure itself without a centralized administration. Nodes in a MANET can connect to each other to form a temporary network without any existing network infrastructure. Routing in MANETs is challenging due to the dynamic network topology, asymmetric links, and interference. Common routing algorithms for MANETs include distance vector, link state, and various protocols designed specifically for MANETs to handle mobility.
Mobile ad-hoc networks (MANETs) are wireless networks formed without any preexisting infrastructure. Nodes in a MANET are mobile and communicate in a multi-hop fashion through wireless connections. Routing is challenging in MANETs due to their dynamic topology. There are several routing protocols for MANETs including proactive, reactive, and hybrid protocols. Common protocols discussed are AODV, DSDV, and DSR. Simulation tools like NS-2 are often used to study and compare MANET routing protocols. MANETs have applications in areas like disaster recovery, mesh networks, opportunistic networking, and vehicular networks.
Cisco discovery d homesb module 3 - v.4 in english.igede tirtanata
This document contains a quiz with 20 multiple choice questions about networking concepts such as encapsulation, MAC addresses, ARP protocol, routing tables, default gateways, and file sharing. It asks the learner to identify networking devices, standards, functions, and impacts of misconfigurations. The questions cover topics like Ethernet, TCP/IP, routing, and basic network security.
This presentation summarizes the Cisco Certified Network Associate (CCNA) certification and covers networking concepts relevant to the CCNA including networking devices, the OSI model, IP addressing, routing, access lists, network address translation, switches, virtual LANs, WAN connection types, wireless technology, and comparisons of 802.11 wireless standards.
Comparatively analysis of AODV and DSR in MAC layer for Ad Hoc Environmentijsrd.com
In Wireless Adhoc Network is a group of wireless mobile nodes is an autonomous system of mobile nodes connected by wireless links. Every node operates as an end system and as a router to forward packets. In this paper mainly focused on Mac layer because this layer is most important for the data communication using control the packet loss and we worked on the comparison based performance of wimax802.16 and wireless802.11 networks using Ad hoc on- demand Distance Vector Routing Protocol and Dynamic Source Routing Protocol. In this paper we used the different maximum speed for the network. And this comparison based on unicast On-demand routing procedure and our simulation for mobile ad hoc networks discover and maintain only needed the design and follows the idea that each node by sending routing packets whenever a communication is requested and compared various parameter packet delivery ratio, normalized routing load and e-e delay. These simulations are carried out using the Network simulator version-2. The results presented in this work illustrate the importance in carefully evaluating and implementing routing protocols in an ad hoc environment.
The document discusses ad hoc networks. It defines an ad hoc network as a temporary network connection between devices without fixed infrastructure. Key characteristics of ad hoc networks include dynamic topology, nodes that can freely join and leave, multi-hop routing, and limited bandwidth. The document compares ad hoc networks to wired and managed wireless networks. It also discusses different types of ad hoc networks and routing protocols like DSR and AODV. Applications of ad hoc networks include military operations, conferences, and emergency response situations.
10 Best Printable Primary Writing Paper TemplateChristina Bauer
The document provides instructions for requesting writing assistance from HelpWriting.net. It outlines a 5-step process: 1) Create an account with a password and email. 2) Complete a 10-minute order form providing instructions, sources, and deadline. 3) Review bids from writers and select one based on qualifications. 4) Review the completed paper and authorize payment or request revisions. 5) Request revisions to ensure satisfaction, and HelpWriting.net guarantees original, high-quality work or a full refund.
Essay Writing Format For Kids. Browse Printable EssChristina Bauer
The document discusses both the positive and negative aspects of the 1969 moon landing. Positively, it achieved an important goal of the Space Race with the Soviet Union during the Cold War. However, it also describes the human toll of the space program, including the deaths of astronauts in training accidents. It then provides details of President John F. Kennedy's visit to Fort Worth, Texas on November 21st, 1963, the day before his assassination in Dallas.
More Related Content
Similar to A Distributed Protocol For Dynamic Address Assignment In Mobile Ad Hoc Networks
Address allocation for MANET merge and partition using cluster based routingsugandhasinghhooda
Network merges and partitions occur quite often in MANET wherein address auto-configuration is a critical
requirement. There are various approaches for address auto-configuration in MANETs which allocate address to
the nodes in a dynamic and distributed manner in which HOST ID and MANET ID are assigned on the basis of
their Base value. MANET merges and partitions employing Cluster Based Routing Protocol require a node to be
assigned as the Cluster Head (CH). This paper presents the Election Algorithm which assigns a node as the Cluster
Head on the basis of its weight. Through simulation using the NS-2, it has been shown that the Election Algorithm
improves the packet delivery ratio (PDR) significantly and decreases the packet delay to a great extent in comparison
to the existing AODV protocol.
MOBILE INTERNET PROTOCOL AND TRANSPORT LAYER
Overview of Mobile IP – Features of Mobile IP – Key Mechanism in Mobile IP – route Optimization. Overview of TCP/IP – Architecture of TCP/IP- Adaptation of TCP Window – Improvement in TCP Performance.
1. Serial0 is up but the line protocol is down, indicating a potential issue with the cable or configuration settings not matching on both ends.
2. Static IP addresses are manually assigned while dynamic IP addresses are assigned by a DHCP server from its IP range, and have a lease duration after which the address may change.
3. Protocols establish rules and standards for moving data across a network, such as routing protocols like RIP, IGRP, and EIGRP that determine the best path between networks.
This document discusses the Teredo protocol, which enables IPv6 connectivity for nodes located behind IPv4 NAT devices. It explains how Teredo works by tunneling IPv6 packets over UDP through NATs. While Teredo allows IPv6 connectivity, it also raises security concerns by bypassing security controls and allowing unsolicited traffic. The document analyzes attacks that could exploit vulnerabilities in Teredo tunnels, such as a denial of service attack against a Teredo server using a single packet. It investigates whether Teredo represents a security risk or is a worthwhile transition mechanism from IPv4 to IPv6.
AUTOCONFIGURATION ALGORITHM FOR A MULTIPLE INTERFACES ADHOC NETWORK RUNNING...IJCNC
Network configuration is the assignment of network parameters necessary for a device to integrate the
network, examples being: an IP address, netmask, the IP address of the gateway, etc ... In the case of
Mobile Ad hoc NETworks (MANETs), the connectivity of nodes is highly dynamic and a central
administration or configuration by the user is very difficult. This paper presents an autoconfiguration
solution for ad hoc networks running the widely implemented version of OLSR routing protocol, the 2003
RFC 3626 [1]. This solution is based on an efficient Duplicate Address Detection (DAD) algorithm,
which takes advantage of the genuine optimization of the OLSR routing protocol. The proposed
autoconfiguration algorithm is proved to operate correctly in a multiple interfaces OLSR network.
The document discusses topics related to the network layer, including:
1. It describes virtual circuits and datagrams, which are two methods for transferring data across networks.
2. It covers IPv4 addressing concepts such as address space, notations, classful and classless addressing, subnetting, and network address translation.
3. It provides an overview of additional network layer topics like IPv6 addressing, routing algorithms, internet control protocols, and routing protocols.
Routing protocols in Ad-hoc Networks- A Simulation StudyIOSR Journals
This document summarizes a study that simulated and compared several routing protocols for ad hoc networks using the Network Simulator 2 (NS-2). It began by describing ad hoc networks and reviewing related work comparing routing protocols. It then provided an overview of desirable routing protocol properties and described several prominent protocols in detail: DSDV, AODV, DSR, ZRP, TORA, IMEP, and CBRP. The document outlined the NS-2 simulator and the mobility extension used. It concluded by detailing the simulation methodology and modifications made to the AODV and DSR implementations in NS-2 to perform the comparative analysis.
UNIT IV MOBILE AD-HOC NETWORKS
Ad-Hoc Basic Concepts – Characteristics – Applications – Design Issues – Routing – Essential of Traditional Routing Protocols –Popular Routing Protocols – Vehicular Ad Hoc networks ( VANET) – MANET Vs VANET – Security
This document analyzes and compares the scalability of three mobile ad hoc network (MANET) routing protocols - AODV, TORA, and OLSR - by simulating them using varying numbers of nodes. The key findings are:
1) In terms of end-to-end delay, OLSR consistently performs better than AODV and TORA as the number of nodes increases from 25 to 75.
2) For throughput, OLSR outperforms AODV and TORA across all network sizes tested.
3) The study concludes that OLSR demonstrates better scalability than AODV and TORA based on its performance with increasing numbers of nodes in terms of
This document provides an overview of mobile ad hoc networks (MANETs). It defines MANETs as wireless networks without fixed infrastructure where nodes are free to move and dynamically self-organize. Key points:
- MANETs are formed spontaneously as nodes connect via wireless links without centralized administration.
- Each node acts as a router, forwarding data for other nodes. Network topology changes frequently as nodes move.
- Examples of MANET applications include military operations, emergency response, classroom environments, and home/office networks.
- Challenges for MANET routing protocols include dynamic topology, limited resources, and network security. Common routing approaches include proactive, reactive, and geographic routing.
IJERA (International journal of Engineering Research and Applications) is International online, ... peer reviewed journal. For more detail or submit your article, please visit www.ijera.com
Performance Analysis of Mobile Adhoc Network Routing Protocols Over Tcppijans
In order to reduce the communication cost and time we are looking forward for successful implementation
of an infrastructure less network like Mobile Ad Hoc Network in all arena of wireless mobile
communication. But still it is a challenge to decide a most appropriate routing protocol for MANET. In
MANET there is no fixed topology due to the mobility of nodes, interference, multipath propagation and
path loss. Since MANET does not use fixed infrastructure rather it always have to find suitable router and
routing path for each communication, existing established routing protocol is not suitable for MANET to
function properly. Different Routing protocols have been proposed to meet the challenges with MANETs.
This paper evaluates the performances of four MANET routing protocols which are DSDV, AODV, DSR
and TORA over TCP, a Transport Layer Protocol. The performance metrics which are considered in this
paper are packet delivery fractions, normalized routing overload, end to end delay and throughput. DSDV
is a proactive protocol and the others are reactive protocols.
Network Address Translation (NAT) allows private IP addresses to be used within a local area network (LAN) while providing access to the public internet. NAT maps private IP addresses to public IP addresses, allowing multiple devices to share public IP addresses. The main NAT traversal challenges are that NAT prevents outside systems from initiating connections to inside systems and communication between systems that are both behind NAT routers. Proposed solutions include using third-party servers to reverse connections or techniques like UDP and TCP hole punching that establish connections directly between systems.
MANET stands for mobile ad hoc network. It is a type of wireless network that can change locations and configure itself without a centralized administration. Nodes in a MANET can connect to each other to form a temporary network without any existing network infrastructure. Routing in MANETs is challenging due to the dynamic network topology, asymmetric links, and interference. Common routing algorithms for MANETs include distance vector, link state, and various protocols designed specifically for MANETs to handle mobility.
MANET stands for mobile ad hoc network. It is a type of wireless network that can change locations and configure itself without a centralized administration. Nodes in a MANET can connect to each other to form a temporary network without any existing network infrastructure. Routing in MANETs is challenging due to the dynamic network topology, asymmetric links, and interference. Common routing algorithms for MANETs include distance vector, link state, and various protocols designed specifically for MANETs to handle mobility.
Mobile ad-hoc networks (MANETs) are wireless networks formed without any preexisting infrastructure. Nodes in a MANET are mobile and communicate in a multi-hop fashion through wireless connections. Routing is challenging in MANETs due to their dynamic topology. There are several routing protocols for MANETs including proactive, reactive, and hybrid protocols. Common protocols discussed are AODV, DSDV, and DSR. Simulation tools like NS-2 are often used to study and compare MANET routing protocols. MANETs have applications in areas like disaster recovery, mesh networks, opportunistic networking, and vehicular networks.
Cisco discovery d homesb module 3 - v.4 in english.igede tirtanata
This document contains a quiz with 20 multiple choice questions about networking concepts such as encapsulation, MAC addresses, ARP protocol, routing tables, default gateways, and file sharing. It asks the learner to identify networking devices, standards, functions, and impacts of misconfigurations. The questions cover topics like Ethernet, TCP/IP, routing, and basic network security.
This presentation summarizes the Cisco Certified Network Associate (CCNA) certification and covers networking concepts relevant to the CCNA including networking devices, the OSI model, IP addressing, routing, access lists, network address translation, switches, virtual LANs, WAN connection types, wireless technology, and comparisons of 802.11 wireless standards.
Comparatively analysis of AODV and DSR in MAC layer for Ad Hoc Environmentijsrd.com
In Wireless Adhoc Network is a group of wireless mobile nodes is an autonomous system of mobile nodes connected by wireless links. Every node operates as an end system and as a router to forward packets. In this paper mainly focused on Mac layer because this layer is most important for the data communication using control the packet loss and we worked on the comparison based performance of wimax802.16 and wireless802.11 networks using Ad hoc on- demand Distance Vector Routing Protocol and Dynamic Source Routing Protocol. In this paper we used the different maximum speed for the network. And this comparison based on unicast On-demand routing procedure and our simulation for mobile ad hoc networks discover and maintain only needed the design and follows the idea that each node by sending routing packets whenever a communication is requested and compared various parameter packet delivery ratio, normalized routing load and e-e delay. These simulations are carried out using the Network simulator version-2. The results presented in this work illustrate the importance in carefully evaluating and implementing routing protocols in an ad hoc environment.
The document discusses ad hoc networks. It defines an ad hoc network as a temporary network connection between devices without fixed infrastructure. Key characteristics of ad hoc networks include dynamic topology, nodes that can freely join and leave, multi-hop routing, and limited bandwidth. The document compares ad hoc networks to wired and managed wireless networks. It also discusses different types of ad hoc networks and routing protocols like DSR and AODV. Applications of ad hoc networks include military operations, conferences, and emergency response situations.
Similar to A Distributed Protocol For Dynamic Address Assignment In Mobile Ad Hoc Networks (20)
10 Best Printable Primary Writing Paper TemplateChristina Bauer
The document provides instructions for requesting writing assistance from HelpWriting.net. It outlines a 5-step process: 1) Create an account with a password and email. 2) Complete a 10-minute order form providing instructions, sources, and deadline. 3) Review bids from writers and select one based on qualifications. 4) Review the completed paper and authorize payment or request revisions. 5) Request revisions to ensure satisfaction, and HelpWriting.net guarantees original, high-quality work or a full refund.
Essay Writing Format For Kids. Browse Printable EssChristina Bauer
The document discusses both the positive and negative aspects of the 1969 moon landing. Positively, it achieved an important goal of the Space Race with the Soviet Union during the Cold War. However, it also describes the human toll of the space program, including the deaths of astronauts in training accidents. It then provides details of President John F. Kennedy's visit to Fort Worth, Texas on November 21st, 1963, the day before his assassination in Dallas.
24 Page Set Of Winter Themed Writing Paper ByChristina Bauer
I apologize, upon further review I do not feel comfortable providing a full summary of this document without having the full context and being able to verify the accuracy and appropriateness of the content. The document discusses technical topics related to physics and I do not have the expertise to fully comprehend or summarize it.
Summarize Paragraph In Short. Online assignment writing service.Christina Bauer
The document provides instructions for creating an account and submitting a paper writing request on the HelpWriting.net site. It outlines a 5-step process: 1) Create an account with an email and password. 2) Complete a form with paper details, sources, and deadline. 3) Writers will bid on the request and their qualifications can be reviewed. 4) Place a deposit to start the writing. 5) Review the completed paper and authorize final payment or request revisions. The summary highlights the key steps involved in obtaining writing help through the site.
How To Write About The Theme Of A Book CoverlChristina Bauer
The document provides instructions for requesting writing assistance from HelpWriting.net. It outlines a 5-step process: 1) Create an account with a password and email. 2) Complete an order form with instructions, sources, and deadline. 3) Review bids from writers and choose one. 4) Receive the paper and authorize payment if pleased. 5) Request revisions until satisfied.
Art Thesis Examples - What Is Art Essay Examples WhChristina Bauer
The document provides instructions for requesting writing assistance from HelpWriting.net. It outlines a 5-step process: 1) Create an account, 2) Complete an order form providing instructions and deadline, 3) Review bids from writers and select one, 4) Review the completed paper and authorize payment, 5) Request revisions until satisfied. It emphasizes that original, high-quality content is guaranteed or a full refund will be provided.
The document provides instructions for requesting and completing an assignment writing request through the HelpWriting.net website. It outlines a 5-step process: 1) Create an account with an email and password. 2) Complete a form with assignment details and attach samples. 3) Review bids from writers and select one. 4) Review the completed paper and authorize payment. 5) Request revisions until satisfied, with refunds offered for plagiarized work. The document promises original, high-quality content and full satisfaction of needs.
Thank You Writing Paper. Online assignment writing service.Christina Bauer
The critique analyzes a piece of music that opens quickly with the cello but is interrupted by the piano playing in a deep key, bringing anxiety. The piano and cello alternate taking the lead, with the deep vibratos of the cello associated with lament and the single notes of the piano with hope. Though the piano suggests hope, it always ends the movement in a deep, somber key, indicating an underlying melancholy.
Great Personal Narratives. How To Write A Personal NChristina Bauer
This document provides instructions for how to request and complete an assignment writing request on the HelpWriting.net website. It outlines a 5 step process: 1) Create an account with a password and email, 2) Complete an order form with instructions and deadline, 3) Review bids from writers and choose one, 4) Review the completed paper and authorize payment, 5) Request revisions until satisfied. It emphasizes that the site aims to provide original, high-quality content and offers refunds for plagiarized work.
How To Insert A Citation In Mla Style - LasgraceChristina Bauer
The document provides instructions for how to request an assignment writing service from HelpWriting.net in 5 steps:
1. Create an account with a password and email.
2. Complete a 10-minute order form providing instructions, sources, deadline, and attaching a sample work.
3. Review bids from writers and choose one based on qualifications, history, and feedback, then pay a deposit.
4. Review the completed paper and authorize final payment if pleased, or request revisions.
5. Request multiple revisions to ensure satisfaction, and HelpWriting.net promises original, high-quality work with refunds for plagiarism.
Autobiography Outline Template For Middle SchoolChristina Bauer
The document provides information about Operation Barbarossa, Germany's invasion of the Soviet Union in 1941. It begins by discussing the 1939 non-aggression pact between Germany and the Soviet Union, noting that while they agreed not to attack each other, Hitler planned to invade the Soviet Union to remove communists and exterminate Jews. The summary then explains that Germany launched Operation Barbarossa on June 22, 1941, attacking the Soviet Union with over 3 million soldiers, marking the start of the Great Patriotic War. Finally, it notes the Germans had the advantage over the Soviet air force at the beginning of the invasion.
In Herman Melville's novella Billy Budd, Sailor, the protagonist Billy Budd is portrayed as a naive but excellent sailor. While Billy is good-natured and diligent, he is emotionally underdeveloped and fails to recognize evil in others. These attributes, along with his impressment into the British Navy and the malicious intentions of others on his ship, ultimately lead to Billy's downfall. However, the essay concludes that while Billy faces wrongful acts, he is also a victim of his own naivety and inability to see the corruption in people.
Ghost Writing And Craft By Its MoNiques World TeacChristina Bauer
This document discusses social anxiety disorder and how it differs from just being shy. Social anxiety disorder is defined as a fear and anxiety of being negatively judged and evaluated by others. It is a chronic mental illness, unlike shyness which is a temporary feeling. While it is common for people to feel judged at times, social anxiety disorder causes excessive and disabling fear and anxiety in social situations. The document aims to raise awareness that social anxiety disorder is a real mental illness and not just an extreme form of shyness or something people pretend to have.
The document provides instructions for creating an account and submitting an assignment request on the HelpWriting.net website in 5 steps:
1. Create an account by providing a password and email.
2. Complete a 10-minute order form with instructions, sources, deadline, and attach a sample if wanting the writer to imitate your style.
3. Review bids from writers and choose one based on qualifications, history, and feedback, then pay a deposit to start the assignment.
4. Review the completed paper and authorize full payment if satisfied, or request free revisions.
5. You can request multiple revisions to ensure satisfaction, and HelpWriting guarantees original, high-quality work with a full refund
The Black Sox Scandal involved several members of the Chicago White Sox baseball team who were accused of intentionally losing the 1919 World Series against the Cincinnati Reds in exchange for money from gamblers; some key players had been offered bribes of $5,000 or more each to lose the series by Chicago gamblers allied with the notorious crime boss Arnold Rothstein; while the players were acquitted in court due to lack of evidence, they were banned from organized baseball for life due to the determination by the newly created Commissioner of Baseball that they had conspired to fix the series.
Paperback Writer Partitions The Beat. Online assignment writing service.Christina Bauer
1. Putnam and Campbell's book American Grace analyzes religion's role in both dividing and uniting American society over recent decades through a sociological lens.
2. The book traces America's social history to understand shifts in religious beliefs and behaviors. It finds religion has both divided Americans along denominational lines but also brought people together in new ways.
3. Putnam and Campbell take a nuanced approach, acknowledging both religion's positive and negative impacts on American social cohesion over time. Their sociological analysis provides insights into religion's complex role in the nation's social fabric.
Argument Analysis - Excelsior College OWL - EChristina Bauer
This document provides instructions for creating an account and requesting writing assistance from the HelpWriting.net website. It outlines a 5-step process: 1) Create an account with a password and email; 2) Complete a form with assignment details and attach samples; 3) Review bids from writers and select one; 4) Review the completed paper and authorize payment; 5) Request revisions until satisfied. It emphasizes the bidding system used to match clients with writers and promises original, high-quality work with refunds for plagiarism.
Short Essay For School Students O. Online assignment writing service.Christina Bauer
The document discusses how romance fiction focuses on women's triumphs over inequality, but this is primarily for Caucasian women. It notes that romance fiction seldom extends the same concern to other marginalized groups. While romance fiction aims to challenge justifications for women's subordination, it offers little insight into how sexism intersects with other forms of oppression like racism. The essay will examine how romance novels can incorporate intersectional perspectives by representing diverse protagonists and relationships.
My First Day At Secondary School. Online assignment writing service.Christina Bauer
The document provides instructions for requesting writing assistance from HelpWriting.net. It outlines a 5-step process: 1) Create an account with a password and email. 2) Complete a order form with instructions, sources, and deadline. 3) Review bids from writers and select one. 4) Review the completed paper and authorize payment. 5) Request revisions to ensure satisfaction, with a refund option for plagiarized work. The process aims to match clients with qualified writers to fully meet their needs for original, high-quality content.
Famous Quotes For Essays. 170 Writing Quotes By Famous AutChristina Bauer
The document provides instructions for using the HelpWriting.net service to request that writers complete assignments and papers. It outlines a 5-step process: 1) Create an account, 2) Submit a request with instructions and deadline, 3) Review writer bids and qualifications and select a writer, 4) Review the completed paper and authorize payment, 5) Request revisions if needed, knowing the service guarantees original work or a refund.
Exploiting Artificial Intelligence for Empowering Researchers and Faculty, In...Dr. Vinod Kumar Kanvaria
Exploiting Artificial Intelligence for Empowering Researchers and Faculty,
International FDP on Fundamentals of Research in Social Sciences
at Integral University, Lucknow, 06.06.2024
By Dr. Vinod Kumar Kanvaria
LAND USE LAND COVER AND NDVI OF MIRZAPUR DISTRICT, UPRAHUL
This Dissertation explores the particular circumstances of Mirzapur, a region located in the
core of India. Mirzapur, with its varied terrains and abundant biodiversity, offers an optimal
environment for investigating the changes in vegetation cover dynamics. Our study utilizes
advanced technologies such as GIS (Geographic Information Systems) and Remote sensing to
analyze the transformations that have taken place over the course of a decade.
The complex relationship between human activities and the environment has been the focus
of extensive research and worry. As the global community grapples with swift urbanization,
population expansion, and economic progress, the effects on natural ecosystems are becoming
more evident. A crucial element of this impact is the alteration of vegetation cover, which plays a
significant role in maintaining the ecological equilibrium of our planet.Land serves as the foundation for all human activities and provides the necessary materials for
these activities. As the most crucial natural resource, its utilization by humans results in different
'Land uses,' which are determined by both human activities and the physical characteristics of the
land.
The utilization of land is impacted by human needs and environmental factors. In countries
like India, rapid population growth and the emphasis on extensive resource exploitation can lead
to significant land degradation, adversely affecting the region's land cover.
Therefore, human intervention has significantly influenced land use patterns over many
centuries, evolving its structure over time and space. In the present era, these changes have
accelerated due to factors such as agriculture and urbanization. Information regarding land use and
cover is essential for various planning and management tasks related to the Earth's surface,
providing crucial environmental data for scientific, resource management, policy purposes, and
diverse human activities.
Accurate understanding of land use and cover is imperative for the development planning
of any area. Consequently, a wide range of professionals, including earth system scientists, land
and water managers, and urban planners, are interested in obtaining data on land use and cover
changes, conversion trends, and other related patterns. The spatial dimensions of land use and
cover support policymakers and scientists in making well-informed decisions, as alterations in
these patterns indicate shifts in economic and social conditions. Monitoring such changes with the
help of Advanced technologies like Remote Sensing and Geographic Information Systems is
crucial for coordinated efforts across different administrative levels. Advanced technologies like
Remote Sensing and Geographic Information Systems
9
Changes in vegetation cover refer to variations in the distribution, composition, and overall
structure of plant communities across different temporal and spatial scales. These changes can
occur natural.
How to Build a Module in Odoo 17 Using the Scaffold MethodCeline George
Odoo provides an option for creating a module by using a single line command. By using this command the user can make a whole structure of a module. It is very easy for a beginner to make a module. There is no need to make each file manually. This slide will show how to create a module using the scaffold method.
A review of the growth of the Israel Genealogy Research Association Database Collection for the last 12 months. Our collection is now passed the 3 million mark and still growing. See which archives have contributed the most. See the different types of records we have, and which years have had records added. You can also see what we have for the future.
Executive Directors Chat Leveraging AI for Diversity, Equity, and InclusionTechSoup
Let’s explore the intersection of technology and equity in the final session of our DEI series. Discover how AI tools, like ChatGPT, can be used to support and enhance your nonprofit's DEI initiatives. Participants will gain insights into practical AI applications and get tips for leveraging technology to advance their DEI goals.
Main Java[All of the Base Concepts}.docxadhitya5119
This is part 1 of my Java Learning Journey. This Contains Custom methods, classes, constructors, packages, multithreading , try- catch block, finally block and more.
A Distributed Protocol For Dynamic Address Assignment In Mobile Ad Hoc Networks
1. 1
A Distributed Protocol for Dynamic Address
Assignment in Mobile Ad Hoc Networks
Mansi Thoppian and Ravi Prakash
Abstract
A Mobile Ad hoc NETwork (MANET) is a group of mobile nodes that form a multi-hop wireless
network. The topology of the network can change randomly due to unpredictable mobility of nodes and
propagation characteristics. Previously, it was assumed that the nodes in the network were assigned IP
addresses a priori. This may not be feasible as nodes can enter and leave the network dynamically. A
dynamic IP address assignment protocol like DHCP requires centralized servers that may not be present
in MANETs. Hence, we propose a distributed protocol for dynamic IP address assignment to nodes in
MANETs. The proposed solution guarantees unique IP address assignment under a variety of network
conditions including message losses, network partitioning and merging. Simulation results show that
the protocol incurs low latency and communication overhead for an IP address assignment.
Index Terms
MANET, address allocation, IP-networks.
I. INTRODUCTION
M
OBILE ad hoc networks (MANETs) are multi-hop wireless networks of mobile nodes
without any fixed, or pre-existing infrastructure. Nodes within the wireless range of each
other can communicate directly. Nodes outside each other’s wireless range must communicate
indirectly, using a multi-hop route through other nodes in the network. This multi-hop route
may change if the network topology changes. Several routing protocols like DSR [5], CBR [6],
TORA [7], ZRP [8], DSDV [9], AODV [10], etc. have been proposed for MANETs.
This work is supported in part by NSF CAREER Grant # CCR-9796331
The authors are with the Department of Computer Science at The University of Texas at Dallas, Richardson, TX 75083.
Email: mansi, ravip@utdallas.edu
2. MANETs may operate in a stand-alone mode, or may have gateways to interconnect to a fixed
network. In the stand-alone mode, the network is spontaneously formed by nodes gathering at
a remote location with no network infrastructure. Such a network can also be formed when the
gateways to the external world fail or when all the existing network infrastructure goes down
due to natural/man-made disasters. In the presence of a gateway, a MANET is envisioned to
operate as a stub network connected to a fixed internetwork.
In most networks, including MANETs, each node needs a unique identifier to communicate.
It may be argued that the MAC address or the home IP address of the node should be sufficient
for this purpose. However, use of the MAC address as a unique identifier has the following
limitations:
1) MANET nodes are not restricted to using network interface cards (NICs) with a 48-bit
IEEE-assigned unique MAC address. In fact, the TCP/IP protocol stack should work on
a variety of data-link layer implementations. So, if this approach were to be employed,
specific implementations would be required for each type of hardware.
2) The uniqueness of a MAC address cannot always be guaranteed, as it is possible to change
the MAC address using commands like ifconfig.
3) There are known instances of multiple NIC cards from the same vendor having the same
MAC address [11] [12].
The home IP address of the mobile node may not be usable as a unique identifier at all times.
The home IP address may not be permanent, for example when the node acquires an IP address
during boot up through DHCP and releases it when it leaves the network. It is possible that two
nodes belonging to the same home network, at different times, may join the MANET with the
same home IP address. Moreover, even if a node owns a unique home IP address, it needs a
unique care-of IP address in the MANET if it is to be addressable from the Internet.
Static IP address assignment for MANET nodes is difficult as it needs to be done manually
with prior knowledge about the MANET’s current network configuration. Dynamic configura-
tion protocols like Dynamic Host Configuration Protocol (DHCP) [1] require the presence of
centralized servers. MANETs may not have such dedicated servers. Hence, centralized protocols
cannot be used to configure nodes in MANETs. In this paper we present a distributed protocol for
dynamic IP address assignment. The proposed solution is targeted towards the stand-alone mode
2
3. of operation. It may also be used in situations where the gateway only provides connectivity to
external network(s) with no support for IP address assignment. While we limit our examples to
IPv4, the proposed protocol is applicable to both IPv4 and IPv6 networks.
The remainder of the paper is organized as follows. Section II presents the related work.
Section III provides the basic idea of the protocol. The protocol messages and timers are discussed
in Sections IV and V, respectively. In Section VI, a detailed description of the protocol is provided
followed by a discussion of its correctness in Section VII. Simulation experiments and results
are presented in Section VIII, followed by conclusion in Section IX.
II. RELATED WORK
Cheshire et al. [13] describe a method for auto-configuration of the host by randomly choosing
a link-local address within the range 169.254.1.0 to 169.254.254.255. After selecting an address,
the host tests to see if the address is already in use by any other node. This approach focuses on
wired networks and ensures link-local uniqueness of the address. It is required that every node
in the network be within the communication range of every other node in the network, which
is not always possible in the case of MANETs. To extend the solution to MANETs, conflict
detection message(s) will have to be flooded throughout the network.
Perkins et al. [14] propose a solution for address auto-configuration in ad hoc networks.
An address is randomly chosen within the range 2048 to 65534 from the 169.254/16 address
block. A node floods Route Requests (RREQs) for the selected IP address. If no Route Reply
(RREP) is received within a timeout period, the node retries for RREQ RETRIES times. At
the end of all the retries, if no response is received, the chosen IP address is assumed to be
free. The node assigns itself that IP address. Here the latency is the timeout value multiplied
by RREQ RETRIES. This approach requires the routing protocol to have a “route discovery”
phase. It does not address the network partitioning issues.
IPv6 Stateless Autoconfiguration [15] specifies the steps a node takes to configure its interfaces
in IPv6. The steps include construction of link-local address, Duplicate Address Detection, and
construction of a site-local address. During Duplicate Address Detection for MANETs flooding
is required, thus making the approach unscalable. To overcome this scalability issue, an extension
is proposed in [16] by building a hierarchical structure. But the cost incurred in maintaining such
a hierarchical structure may be high.
3
4. In MANETconf [12], a new node entering the MANET requests for configuration information
from its neighbors. One of these neighbors initiates the IP address allocation process for the new
node. For each IP address assignment, this approach requires a network-wide broadcast leading
to scalability problems. However, this approach handles network partitioning and subsequent
merging.
Weak Duplicate Address Detection (DAD) protocol [17] requires each node in the network to
have a unique key. Weak DAD requires that packets “meant for” one node must not be routed
to another node, even if the two nodes have chosen the same address. This is achieved by using
the key information for duplicate address detection. In this approach, the routing protocol related
control packets need to be modified to carry the key information. In the weak DAD scheme, the
packet can still be misrouted in the interval between the occurrence of duplicate IP addresses
in the network and their actual detection. Enhanced Weak DAD [17] was proposed to eliminate
the above shortcoming by using sequence numbers and some bookkeeping.
Passive Duplicate Address Detection (PDAD) presented in [23] tries to detect duplicate ad-
dresses without disseminating additional control information. Based on classic link state routing,
the following three schemes are proposed:
PDAD based on sequence numbers (PDAD-SN): Detects duplicate addresses by using
sequence numbers and link state information.
PDAD based on locality principle (PDAD-LP): Exploits the fact that nodes move with
limited speed.
PDAD based on neighborhood (PDAD-NH): Exploits the property that a node knows its
own neighborhood and the neighborhood of the originator of a link state packet.
The proposed PDAD schemes use random source IDs to detect duplicate addresses within two
hop neighborhood. This approach requires the use of a link state routing protocol.
The stateless address autoconfiguration scheme presented in [24] consists of three phases: (1)
selection of random address, (2) verification of the uniqueness of the address and, (3) assignment
of the address to the network interface. The verification of uniqueness is done by a hybrid DAD
scheme consisting of two phases: (a) strong DAD phase and, (b) weak DAD phase. Within a
connected ad hoc network, a node configures itself with an IP address using strong DAD. During
strong DAD, the node chooses a tentative address and checks for any address duplication by
4
5. broadcasting AREQ message with the chosen tentative address for a fixed number of times. If
no response is received for the AREQ message, the node configures itself with the tentative
address. Weak DAD uses a “key” in addition to the IP address to detect duplicate addresses
during ad hoc routing. This scheme ensures that during resolution of an address conflict, the
sessions using conflicting addresses are maintained until the sessions are closed.
Dynamic Registration and Configuration Protocol (DRCP) [18] [19] extends DHCP for wire-
less networks. In this protocol, each node acts as both server and client and owns an ad-
dress pool. The address pool distribution is done using Dynamic Address Allocation Protocol
(DAAP) [18] [19]. Each node obtains the address pool by requesting half of the addresses from
the address pool of a neighboring node. The protocol does not discuss the network partitioning
issues and the impact of message loss.
Dynamic Address Allocation Protocol proposed in [20] requires that new nodes approach the
leader of the network to obtain an IP address. The leader is the node with the highest IP address
in the network. Network partitioning and merger are considered. The unique identifier used for
identifying the network is the MAC address of the initiator which is the first node that comes
up in the network. This could lead to a problem of multiple networks having the same identifier
when the initiator itself moves out of the network and forms another network. The impact of
message losses is not considered.
Zhou et al. [21] propose a solution which is derived from a sequence generation scheme
whereby a sequence consisting of numbers within a range R is generated using a function f(n).
The function f(n) is chosen such that in the sequences generated by f(n), the interval between two
occurrences of the same number is very large and the probability of more than one occurrence of
the same number in a limited number of different sequences initiated by different seeds during
some interval is extremely low. The protocol requires the first node to choose a random number
from the range R as its IP address and use a random state value as the seed for function f(n).
When a new node joins the network, the configured node generates another integer and a new
state value using f(n). The new node obtains these values and configures itself. In this approach,
the needed block of IP addresses may be significantly bigger than the number of nodes in the
network. Even if the minimal interval between two occurrences of the same address in the
sequence generated is extremely large, it is still possible for two nodes to have the same IP
5
6. address if the nodes keep moving in and out of the network at a high rate.
The solutions mentioned above have made significant contributions to our understanding of the
problem. However, we believe that all of these approaches handle only a subset of the network
conditions listed below:
1) Topology changes: Nodes in the network can move arbitrarily and can enter and leave the
network dynamically.
2) Message losses and node crashes: Message loss can be quite prevalent and can lead to
duplicate IP address assignment if not handled effectively. Nodes can leave the network
abruptly either due to link failure or crash.
3) Partitioning and subsequent merging: During the course of MANET operation, the network
can split into multiple networks and subsequently merge into one. During network merging,
it is possible to have duplicate IP addresses in the merged network.
4) Concurrent address requests: Multiple nodes can join the network simultaneously.
5) Limited energy and bandwidth: Nodes in MANET are energy limited and links are band-
width limited. Hence, the communication overhead incurred should be low.
In this paper, we propose a solution similar to DAAP [18], [19] that guarantees unique IP
address assignment under the above network conditions. In our approach, most of the address
assignments require local communication leading to low communication overhead and latency
(see Section VIII for details).
III. BASIC IDEA
The objective of the proposed protocol is to assign a unique IP address to a new node joining
the MANET. The new node joining the network is called a requester. The configured node
responsible for assigning an IP address to the requester is called an allocator. We assume
that the MANET starts with a single node. We call this node the Initiator of the network
and the configuration of this first node as MANET initialization. For simplicity, it is assumed
that at least the first node in the MANET knows the IP address block from which the IP
addresses are to be assigned to the nodes in the MANET. For example, this block can be the IPv4
private address block: 10.0.0.1 - 10.255.255.255, 172.16.0.0 - 172.31.255.255 or 192.168.0.0 -
192.168.255.255. The proposed protocol is equally applicable for IPv6 address space. The address
6
7. block information can be propagated to other nodes joining the network during the assignment
process. The term “broadcast” in this paper stands for local broadcast, unless specified otherwise.
When the Initiator starts operation in MANET mode, it broadcasts a message requesting an
IP address. As there are no other MANET nodes in the neighborhood, the Initiator will not
receive any response. The Initiator re-broadcasts its request message for a constant number of
times after which it assigns itself the first IP address from the IP address block and forms its
free ip set from the remaining addresses. The free ip set is an ordered set containing addresses
that are not in use by any node in the network.
After MANET initialization, every time a new node (requester) requests an IP address, one
of the existing MANET nodes (allocator) within communication range of the requester initiates
address allocation process for the requester. If the allocator has a non-empty free ip set, it allots
the second half of the addresses from its free ip set to the requester (this approach is similar
to the buddy system for memory management [2], [3]). Otherwise, it performs an expanding
ring search whereby it propagates the request through the network. If the allocator finds a node
(say node A) with a non-empty free ip set during the expanding ring search, it allots half of the
addresses from node A’s free ip set to the requester. The requester configures itself with the first
address from the alloted address block and forms its free ip set with the remaining addresses in
the block.
During the expanding ring search, the allocator might fail to find a free IP address either be-
cause: (i) all IP addresses have been assigned to nodes currently in the MANET or (ii) some nodes
have left the MANET without releasing their IP address and/or free ip set (leaked-addresses). In
the first scenario, no new node can be admitted (without expanding the address block range) as
the MANET has reached its maximum size. In the second case, the leaked-addresses have to be
reclaimed and assigned to new nodes joining the network. Address reclamation is done as follows:
the allocator determines the addresses of the nodes that failed to respond during the expanding
ring search (call this set the missing addresses set). The allocator performs a network-wide
broadcast targeted towards these addresses. The nodes in the network, on hearing this broadcast
message, respond with a message indicating a conflict if their IP address is among the broadcast
addresses. If the allocator receives messages indicating conflict, it removes the corresponding
IP address from the missing addresses set. The addresses in the resultant missing addresses set
7
8. are then declared as free IP addresses. One of these addresses is assigned to the requester. The
rest of the free IP addresses are re-distributed among the existing nodes in the network to form
their free ip sets. Concurrent reclamation operations are serialized based on the priorities of the
allocators (see Section VI-E for details).
During the course of MANET operation, nodes can split from the network and form/join
different networks. These networks can later merge into a single network. Each network is as-
signed a unique identifier (network id). To detect network-merging, nodes periodically broadcast
“Hello” messages to their neighbors. These “Hello” messages contain the IP address and the
network id of the sending node. Whenever a node (say A) receives a “Hello” message from
its neighboring node (say B) containing a different network id than its own, node A detects
merging of networks and replies. When node B receives a reply from A, it detects network-
merging as well. Furthermore, if node A has higher IP address than node B (ties are broken
using network ids), then node B is responsible for re-configuring the merged network in terms
of re-distributing free IP addresses and removing duplicate address assignments. We call node
B the merge agent. Node B initiates a network-wide flood to collect the network configuration
information, i.e., the IP addresses assigned to the nodes and their corresponding free ip sets within
its network. Node B asks node A to initiate similar network-wide flood in node A’s network1
. We
call node A as co-merge agent. Node A sends the collected network-wide information to node B.
After collecting the IP address and free ip set information from all the responses, node B detects
duplicate IP address assignments among nodes in the merged network. Node B invalidates the
IP address of the nodes with conflicting IP addresses and assigns them with new IP address.
After invalidation, node B performs re-distribution of free addresses among nodes in the merged
network.
The protocol is described in detail in Section VI.
IV. PROTOCOL MESSAGES
The following messages are exchanged during IP address assignment:
IPAddressRequest: A local broadcast message from the requester to neighboring nodes in
the network requesting an IP address.
1
By A’s and B’s network we mean the networks they belonged to prior to merging.
8
9. IPAddressAvail: Unicast2
responses to IPAddressRequest from the neighboring nodes of the
requester. This message contains the replying node’s IP address and proposed block of IP
addresses, if any.
AllocatorChosen: Unicast messages from the requester to all its neighbors on electing
an allocator from among the neighbors that sent IPAddressAvail messages. This message
contains the IP address of the chosen allocator.
IPAddressAssign: Unicast message from the allocator to the requester assigning an IP
address. This message contains an IP address block for the requester, IP address range
from which the addresses can be assigned to nodes, and the network id.
IPAddressUpdate: Network-wide message exchanged between nodes to update the network
configuration information. This message is sent to all nodes in the network: (i) by the
allocator responsible for reclamation of IP addresses during IP address allocation process,
and (ii) by the merge agent during network-merging process.
WaitPeriod: Unicast message from the allocator informing the requester to extend its
assign ip timer (explained in Section V).
The following four messages are exchanged during the expanding ring search:
IPAddressInfoRequest: Unicast messages sent from the allocator, during expanding ring
search, to nodes in the network requesting network configuration information namely the
IP address and free ip set.
IPAddressInfoReply: Unicast message in response to IPAddressInfoRequest from a node to
the allocator containing its IP address, free ip set and the neighbor list information.
IPAddressChosen: Unicast message sent by the allocator to a node in the network informing
the node that addresses from its free ip set are chosen for assignment.
IPAddressConfirm: Unicast response to IPAddressChosen message confirming the receipt of
IPAddressChosen message. This message contains the IP address block for the requester.
The following five messages are exchanged during the reclamation operation :
TentativeFreeAddresses: Network-wide flood message sent by the allocator during the recla-
mation of IP addresses. This message contains the missing addresses set.
2
Unicast messages directed to requester or originating from requester use MAC address for communication.
9
10. ConflictNotification: Unicast message sent by a node in response to a TentativeFreeAd-
dresses message indicating that the node’s IP address is in conflict with an address in the
TentativeFreeAddresses message.
NoConflict: Unicast message sent by a node in response to a TentativeFreeAddresses message
indicating that the node’s IP address is not in conflict with the addresses in the Tentative-
FreeAddresses message.
Defer: Unicast messages sent by the nodes/co-merge agents on receiving multiple Tenta-
tiveFreeAddresses/PartitionMergeQuery messages from different allocators/ merge agents.
This is sent to all the allocators/merge agents except the allocator/merge agent with the
lowest IP address. On receipt of this message, the receiver (allocator/merge agent) suspends
the address allocation/network-merge process.
Resume: Unicast message sent by a node that had previously sent a Defer message. A node
on receiving this message resumes the suspended address allocation/network-merge process.
The following message is exchanged during migration of a requester:
IPAddressForward: When a requester moves away from its allocator before acquiring an
IP address, the requester chooses a new allocator. The new allocator sends this message
to the old allocator.
The following four messages are used during partition handling:
Hello: Local broadcast messages exchanged periodically between neighbors to detect neigh-
borhood changes and merging of networks. Each node sends this message containing its
network id to its neighbors.
PartitionMergeQuery: Network-wide flood message sent by the merge agent/co-merge agent
to collect the network-wide IP address information. This message contains the network ids
of the merging networks and the IP address of merge agent .
PartitionMergeResponse: Unicast message sent in response to the PartitionMergeQuery
message by a node to a merge agent/co-merge agent. This message contains the responding
node’s IP address, free ip set and pending ip set (explained in SectionVI).
IPAddressInvalidate: Network-wide broadcast message sent from the merge agent to inval-
idate conflicting IP addresses.
The following message is for graceful departure of nodes:
10
11. HandOver: A node departing the network sends this unicast message, containing its IP
address, free ip set and pending ip set, to one of its neighbors.
V. TIMERS
Timers are used to ensure that the protocol is deadlock-free and works correctly in the event
of message losses or node crashes. Let t indicate the time it takes for a message to reach a
node one hop away from the sender, and let p indicate the upper bound on the time required
to process a message (including the queuing delay) at a node. The following are the various
timers3
used in the protocol.
offer ip timer: The requester sets this timer after sending IPAddressRequest message. If the
requester does not receive any reply before the timer expires, it retries for requester retry
times, where requester retry is the maximum number of attempts by a requester to acquire
an IP address. The timer value should at least be (2t + p).
allocation pending timer: A node in the network sets this timer after sending an IPAddres-
sAvail message and creating an entry in the pending ip set. If no AllocatorChosen message
is received before the timer expires, the proposed address block is withdrawn from the
pending ip set and appended to the free ip set of the node. The timer value should at least
be (2t + p).
assign ip timer: This timer is set by the requester after choosing an allocator. If no address is
assigned before the timer expires, the requester retries assign retry times, where assign retry
is the maximum number of times a requester tries to choose a new allocator. The timer
value should at least be (d + 1) (2t + p), where the value of the hop count, d, is provided by
the allocator through a WaitPeriod message (refer to Section VI for details on hop count).
confirm ip timer: The allocator sets this timer after sending an IPAddressChosen message to
a node with largest free ip set. If the timer expires before the receipt of an IPAddressConfirm
message, the allocator chooses the next node with largest free ip set. The timer value should
at least be (d (2t + p)).
state collection timer: The allocator sets this timer before sending an IPAddressInfoRe-
quest/TentativeFreeAddresses message. The timer should at least be (d (2t + p)).
3
Here we have provided the minimum values for the timers. It is possible to choose higher values for the timers to make the
protocol less aggressive while increasing its latency, without compromising its correctness.
11
12. hello timer: This timer is used to detect network-merging. It is set after sending the Hello
messages. The timer value corresponds to the interval between consecutive Hello messages.
partition timer: This timer is set by the merge agent. It is set after sending the network-wide
flood of PartitionMergeQuery message. If
✂✁ and
☎✄ are the average diameters of the two
networks, the timer value should be at least (2 (max(
✂✁✝✆✞✟✄ )(t + p)))).
VI. DETAILED DESCRIPTION OF THE ALGORITHM
In MANETs, the following scenarios could occur due to dynamic topology changes and the
dynamic arrival/departure of nodes in the network: (i) MANET initialization, (ii) new nodes join-
ing the network, (iii) graceful departure of nodes, (iv) abrupt departure of nodes, (v) concurrent
address requests, (vi) migration of the requester, (vii) message losses, and (viii) partitioning of
network and subsequent merging. We address each of these scenarios in this section:
A. MANET Initialization
The Initiator of a MANET broadcasts IPAddressRequest message and waits for an IPAddres-
sAvail message until the offer ip timer expires. If it does not receive any IPAddressAvail message,
it re-broadcasts the IPAddressRequest message. This continues for requester retry times. This
is to ensure that, if there are other configured nodes in the network, in the event of message
losses the node does not assume itself to be the Initiator (details on message losses given in
Section VII-A). After all the failed retries, the node concludes that it is the only node in the
network. It then assigns itself the first IP address from the IP address block, initializes the
free ip set to the rest of the IP address block, and sets its network id to the following 4 tuple
✠ Initiator’s MAC address, Initiator’s IP address, timestamp, random number✡ .
B. New node joining the network
After MANET initialization, any new node (requester) appearing in the neighborhood of exist-
ing node(s) broadcasts an IPAddressRequest message and starts the offer ip timer. Each address
allocation is assigned a unique transaction id. The transaction id consists of the requester’s
MAC address and timestamp.
Let ☛✌☞✎✍✏✍ ✑✓✒✕✔✗✖✙✘ set at a configured node X in the neighborhood be ✚✜✛ ✁✝✆ ✛ ✄✢✆✢✣✤✣✓✣✤✆ ✛✦✥★✧ . On
receiving the IPAddressRequest message, node X computes k = ✩✫✪✭✬✯✮✜✰ .
12
13. If k ✡ 0, node X sends IPAddressAvail message to the requester with the addresses ✚✜✛✁ ,...,✛✦✥ ✧
as the proposed address block (proposed block)4
. It updates its free ip set to : free ip✔ ✖ ✘✄✂
✚✜✛
✁ ✆
✛
✄✢✆✢✣✤✣✓✣✤✆
✛☎✝✆ ✁
✧ . Node X also creates a ✚ transaction id, proposed block ✧ entry in its
pending ip set and starts the allocation pending timer.
Otherwise, node X sends a NULL IPAddressAvail message to the requester.
When the offer ip timer expires5
, requester sorts all the received non-NULL IPAddressAvail
replies based on the size of the proposed block. The requester selects the neighbor with the
largest proposed block6
as the allocator. The requester sends an AllocatorChosen message with
the selected allocator’s IP address and corresponding transaction id to all its neighbors. If all
the received IPAddressAvail messages contain NULL responses, the requester randomly chooses
one of the nodes as its allocator.
On receiving an AllocatorChosen message, each neighboring node that is not chosen as the al-
locator, removes the proposed block from its pending ip set for the corresponding transaction id
and appends it to its free ip set. Thus, it is ensured that nodes not chosen as the allocator get
back their offered address blocks.
On receiving the AllocatorChosen message, the node chosen as the allocator creates an entry
in the transaction info set for the transaction id. The transaction info is a set of ordered pairs.
The first field of the ordered pair is the transaction id and the second field is the allocator’s
IP address. The allocator, then retrieves the proposed block from its pending ip set for the
corresponding transaction id.
1) If proposed block is not NULL: The allocator cancels the allocation pending timer and
sends IPAddressAssign message with the proposed block to the requester. It then removes the
entry for the transaction id from its pending ip set.
2) If proposed block is NULL: The allocator initiates an expanding ring search. The hop
count d is initialized to 1. Each node in the network maintains a neighbor list. The neighbor
list is updated by the exchange of Hello messages used for partition handling. The allocator
performs the following sequence of steps during the expanding ring search until it finds either
4
If the proposed block is contiguous only the fi rst and the last addresses are sent. A fi eld can be added to the protocol header
to indicate whether the message contains the range or individual addresses of the block.
5
The fi rst IPAddressAvail message received need not contain the largest proposed block, hence we wait for a time-out to
collect more IPAddressAvail messages.
6
In case of a tie, one of the neighbors is chosen randomly.
13
14. a free IP address or all the nodes in the network are visited:
1) The allocator sends an IPAddressInfoRequest message containing its IP address, transaction id,
and hop count (d) to every node in its d-hop neighborhood and to all the nodes that did
not respond during previous iterations. It also starts the statecollection timer and sends
a WaitPeriod message to the requester with an estimated delay of the allocation process.
The requester on receiving the WaitPeriod message extends its assign ip timer.
2) On receiving an IPAddressInfoReply response, if the received IPAddressInfoReply message
has a non-empty free ip set, the allocator goes to step 3. Otherwise, it takes a union of all
the received neighbor lists to check if the resultant union set is same as the set obtained
during the previous iteration. If so, there are no nodes left to be visited. Thus, if the
IPAddressInfoReply message responses do not have any free IP addresses and if all nodes
in the network are visited, the allocator performs reclamation of IP address (explained in
next subsection). Otherwise it goes to step 5.
3) The allocator sorts all the received non-empty free ip sets based on their cardinality. It
sends an IPAddressChosen message to the node with highest cardinality free ip set. It also
sends a WaitPeriod message to the requester with current hop count. It then starts the
confirm ip timer and waits for IPAddressConfirm message. If the timer expires before it
receives an IPAddressConfirm message, it chooses the next node in the sorted list and
repeats this step until either an IPAddressConfirm response is received or all the nodes in
the list are visited. If all nodes in the list are visited and no free IP address is found, it
goes to step 5.
4) If the allocator receives IPAddressConfirm message before confirm ip timer expires, it
cancels the confirm ip timer and sends an IPAddressAssign message to the requester with
the proposed block received in the IPAddressConfirm message. This ends the IP address
allocation process.
5) The allocator increments hop count d by an integer constant and goes to step 1.
During the expanding ring search, an intermediate node does the following:
On receiving an IPAddressInfoRequest message, it responds with an IPAddressInfoReply
message containing its free ip set and neighbor list information.
On receiving IPAddressChosen message, it performs actions similar to the ones performed
14
15. by the neighbors of the requester when they receive an IPAddressRequest message. It
sends an IPAddressConfirm message with half of the addresses from its free ip set as the
proposed block to the allocator. In case the IPAddressConfirm message is lost, the allocator
eventually times out as explained in step 3 of the expanding ring search. The addresses of
the proposed block would be unavailable for allocation in future as they are assumed to
be allocated and are removed from the free ip set of the node. These addresses would be
reclaimed during subsequent reclamation process (Section VI-B.3).
3) Reclamation of IP addresses: Suppose, no free IP addresses are found during the ex-
panding ring search, it does not necessarily mean that there are no free IP addresses in the
network. It is possible that some nodes may have left the network abruptly. Hence, reclamation
of addresses needs to be done whereby the IP addresses of nodes that may have left abruptly
are reclaimed. Reclamation of addresses is done by the allocator as follows:
It finds the addresses that are not in use and forms the missing addresses set where:
missing addresses = ✚ x:x ✚ IP address block - alive addresses ✧ ✧
alive addresses = ✚ IP addresses of nodes that responded during expanding ring search ✧
It floods a TentativeFreeAddresses message in the network. This message contains the
missing addresses set. Nodes that receive this message respond either with a ConflictNotifi-
cation message (if the node’s IP address is in the missing addresses set) or with a NoConflict
message. If the allocator receives a ConflictNotification message from any node, it removes
the corresponding IP address from the missing addresses set. If addresses in a node’s free ip
set are in the missing addresses set, then the node removes the corresponding addresses from
its free ip set. The resulting missing addresses set contains addresses that are free.
It then sends an IPAddressAssign message with the first address from these free IP ad-
dresses and a new network id (
✠
allocator’s MAC address, its IP address, timestamp, ran-
dom number✡ ) to the requester. It then re-distributes the remaining free IP addresses
among the nodes in the network by sending a network-wide IPAddressUpdate message.
The addresses could be re-distributed randomly or evenly among the nodes. Instead of
distributing reclaimed IP addresses randomly, they can also be allocated such that nodes
finally have contiguous block of addresses. The latter approach would reduce the problem of
address block fragmentation. The IPAddressUpdate message also carries the new network id.
15
16. The nodes receiving the IPAddressUpdate message update their network id and free ip set. If
nodes that had left the network later rejoin the network, their network id would be different
from the new network id. Thus, by changing the network id during reclamation it is ensured
that rejoining of nodes that left the network is detected by the partition handling process
(Section VI-G).
The requester, on receiving the IPAddressAssign message, cancels the assign ip timer and
configures itself with the first address from the proposed block. It also sets its free ip set to the
remaining addresses in the proposed block . If it does not receive an IPAddressAssign message
before assign ip timer expires, it retries assign retry times. The likelihood of all the assign retry
attempts being unsuccessful due to message losses is very low (refer to Section VII-A for details).
C. Migration of Requester
Let a requester (say node X) move away from the allocator (say node Y) before Y could
assign an IP address to X. Node X sends node Y’s IP address to the new allocator (say node
Z). Node Z sends IPAddressForward message to Y. Node Y, on receiving the IPAddressForward
message, updates the entry for the corresponding transaction id in the transaction info set with
the IP address of Z. Node Y then sends the IPAddressAssign message to node X via node
Z. Alternately, the old allocator (Y) could abort the IP address allocation process when the
requester migrates and the new allocator (Z) could start the allocation process afresh. In the
former approach, the communication overhead is reduced by maintaining state information at
the nodes. The latter approach does not maintain such state information but requires the address
allocation process to be started all over again. Our simulations use the first approach.
D. Departure of a node:
The nodes in the network can either depart abruptly or gracefully from the network. The IP
addresses and the free ip sets of nodes that abruptly depart the network are reclaimed during
subsequent IP address allocation processes. A node that wishes to gracefully depart the network
sends a HandOver message with its IP address, free ip set, and pending ip set to one of its
neighbors before leaving the network. The neighbor on receiving the HandOver message, appends
the received IP address, free ip set, and pending ip set to its own free ip set.
16
17. E. Concurrent Address Requests
If a node receives concurrent IP address requests, and if it has a non-empty free ip set, then
it allots disjoint IP address blocks from its free ip set to the requesting nodes. If two allocators
perform IP address reclamation concurrently, one of them suspends the allocation process as fol-
lows. Let the two allocators be P and Q with P’s IP address higher than Q’s. Both nodes flood the
network with TentativeFreeAddresses message. A node that receives the TentativeFreeAddresses
message from Q before receiving that from P, sends a NoConflict/ConflictNotification message
(based on whether there is conflict between its address and addresses in the broadcast message)
to Q and a Defer message to P. A node that receives the TentativeFreeAddresses message from
P before receiving that from Q, responds with a NoConflict/ConflictNotification message to both
the allocators. P suspends the IP address allocation process on receiving a Defer message. While
Q continues the IP address allocation process. After completing the address allocation process,
Q sends the IPAddressUpdate message to all nodes in the network. Nodes that had sent the Defer
messages to P, now send Resume messages to P. Node P, on receiving the Resume message, re-
broadcasts TentativeFreeAddresses message and continues the address reclamation process. After
the reclamation process at Q, the free ip sets at nodes in the network might have changed. Hence,
it is possible that the addresses that were in missing addresses set of P are either in use or are re-
distributed among nodes to form their free ip set. By re-broadcasting TentativeFreeAddresses after
suspension, it is ensured that P has the updated network information. The proposed approach is
similar to the one described in [12] and Ricart-Agrawala algorithm [25] for mutual exclusion.
Unlike the Ricart-Agrawala algorithm, in our proposed scheme, explicit Defer messages are
used to suspend the allocation process. In the absence of such Defer messages, the timer at
the allocator would have expired and the allocation process at multiple allocators would have
continued leading to duplicate IP address assignment. The approach described above avoids such
duplication.
F. Message Losses
We propose to use UDP for communication. Hence, it is important to handle message losses
as they can lead to duplicate IP address assignments. In this protocol, message losses are handled
using appropriate timers and confirmation messages (refer to Section VII-A).
17
18. G. Partition Handling
During the course of MANET operation, nodes can split from a network and form/join other
networks. These networks can later merge into one. To detect merging of networks, each network
needs a unique identifier (network id). The network id of the network is initially the following
4-tuple:
✠
Initiator’s MAC address, Initiator’s IP address, timestamp, random number✡ , where
Initiator is the first node that forms the network. The probability of two nodes having the
same MAC address is low. Furthermore the probability of nodes having the same MAC address
getting assigned identical IP addresses with equal timestamps is even lower. Adding the random
number field to the 4-tuple, makes the probability of duplicate network ids negligible. Thus, for
all practical purposes, network ids can be considered to be unique.
The network id of a network is changed every time an address reclamation is performed. By
changing the network id it is ensured that networks have unique network ids. The inclusion of
timestamp as part of the network id ensures that even if the same allocator performs IP address
reclamation more than once, the resulting networks would still have different network ids. If a
MANET splits into multiple MANETs at time
✁
and the networks merge at a later time
✄
such
that no reclamation is performed between time
✁
and
✄
, then the network ids will not change.
This does not violate the correctness of the protocol because there are no conflicts among the
networks merging.
Figure 1 describes a sample network partitioning and merging process. Suppose network id of
network X is
✠✂✁ ✛☎✄✝✆ , ✞✠✟✡✆ , ☛☞✆ , ✌☎✆ ✡ . Let free ip sets of all the nodes except node A be empty
(Figure 1(a)). Nodes P, Q, and R split from this network and form a new network Y. The nodes
in networks X and Y have unique IP addresses, disjoint free ip sets, and same network ids after
partitioning (Figure 1(b)). New nodes joining either networks are assigned unique IP addresses
as long as the two networks do not initiate an IP address reclamation process. Now, suppose new
node F joins the network X and let node E be the allocator (Figure 1(c)). Since free ip set of node
E is empty, it initiates an expanding ring search and finds that none of the active nodes in network
X have a non-empty free ip set. At this point, node E performs address reclamation whereby it
finds that nodes P, Q, and R have left the network (the protocol does not distinguish between
nodes leaving the network and abrupt crashing of nodes). Node E reclaims the IP addresses and
the free ip sets (in this example, they are empty) of P, Q, and R. Node E then propagates new
18
19. A A A
(MAC , IP , T , R )
A
Q
R
P
A
B
C
D
10.0.0.1
10.0.0.3
10.0.0.4
10.0.0.5
10.0.0.6
10.0.0.7
10.0.0.8
free_ip = {10.0.0.2}
Network X
(a) Network X before Partitioning
A A A
(MAC , IP , T , R )
A
A A A
(MAC , IP , T , R )
A
Q
R
P
A
B
C
D
E
10.0.0.1
10.0.0.3
10.0.0.4
10.0.0.5
10.0.0.2
10.0.0.6
10.0.0.7
10.0.0.8
Network X Network Y
(b) Network after Partitioning. Node E joins the net-
work X after partitioning
E E E
(MAC , IP , T , R )
E
A A A
(MAC , IP , T , R )
A
Network Y
Q
R
P
10.0.0.6
10.0.0.7
10.0.0.8
free_ip = {10.0.0.8}
A
B
C
D
F
E
10.0.0.1
10.0.0.3
10.0.0.4
10.0.0.2
10.0.0.6
Network X
10.0.0.5
free_ip ={10.0.0.7}
(c) Network X performs reclamation of IP address
E E E
(MAC , IP , T , R )
E
A A A
(MAC , IP , T , R )
A
free_ip = {10.0.0.8}
A
B
C
F
E
10.0.0.1
10.0.0.3
10.0.0.2
10.0.0.6
10.0.0.5
free_ip ={10.0.0.7}
R
P
10.0.0.6
10.0.0.7
Q
10.0.0.4
D
10.0.0.8
Network X
Network Y
(d) Network X and Network Y merge
Fig. 1. Network Partitioning and Merging
network id,
✠
Node E’s MAC address, Node E’s IP address, timestamp, random number✡ , to all
the nodes in network X. Thus, at the end of the reclamation process, the network ids of X and
Y are different (Figure 1(c)).
Detection of network merging is done as follows. Each node in the network locally broadcasts
a periodic Hello message with its network id and IP address to its neighbors. Whenever a
node (say A) receives Hello message from a neighboring node (say B) containing a network id
different from its own, A detects merging of networks and responds to B. When B receives such
a response from A, B detects merging of networks as well. Furthermore, if A has higher IP
address than B (ties are broken using network ids), then B (merge agent) initiates a network-
wide flood of PartitionMergeQuery message to collect the network configuration information in
terms of IP addresses assigned to the nodes and their respective free ip sets within its network.
Node B asks node A (co-merge agent) to do the same within A’s network7
. On receiving a
7
By A’s and B’s network we mean the networks they belonged to prior to merging.
19
20. PartitionMergeQuery message, a node responds with a PartitionMergeResponse message, if its
network id matches with sender’s network id. The PartitionMergeResponse message contains
IP address, free ip set, and network id of the node sending the message. Nodes receiving a
PartitionMergeQuery message store the network ids of networks being merged. Node A sends
all the collected IP address and free ip set information to node B. The assigned IP addresses and
free IP addresses that are present in both the networks being merged are called as conflicting
addresses and conflicting free ip sets, respectively. By aggregating all the information obtained
from the PartitionMergeResponse messages and information collected from A, node B learns
of all the non-conflicting IP addresses and the number of nodes with conflicting IP addresses.
For each conflicting IP address, node B further knows those nodes that own the IP address and
the corresponding network id. Based on all these information, node B performs a network-wide
broadcast of an IPAddressInvalidate message containing three tuples of the following form:
✠
x,
y, z✡ . The first element, x, is a conflicting IP address and y is a network id. While multiple
nodes may have the same IP address, x, they can be differentiated by their network ids. The first
two elements (x and y) can uniquely identify the nodes whose address needs to be invalidated
and the third element z8
is the new IP address assigned to that particular node. Nodes receiving
the IPAddressInvalidate message with their IP address and network id contained in one of the
3-tuples, change their IP address to the new IP address9
given in that three tuple. After the
invalidation process, node B computes all the addresses not in use. It re-distributes these free IP
addresses among nodes and propagates the new network id to all the nodes in the merged network
by sending an IPAddressUpdate message. The free IP addresses can be re-distributed such that
the nodes finally have contiguous block of addresses. A node on receiving the IPAddressUpdate
message updates its network id and its free ip set. Any node that misses the IPAddressUpdate
message remains as part of the old network. Later, when the nodes in the network exchange
Hello messages, this node gets merged into the network.
When more than one node initiates the network-wide flood of PartitionMergeQuery message
simultaneously on detecting the same merger between networks, the merge agent with lower
IP address has precedence over the merge agent with higher IP address. If a node receives
PartitionMergeQuery messages from multiple nodes that detected the same merger concurrently,
8
If there are no free IP addresses, then z is set to null.
9
If the new IP address fi eld is null, then the node invalidates its IP address and retries to acquire a new IP address later.
20
21. then: (i) If the node has already sent a response to one flood request, it ignores the Parti-
tionMergeQuery message from the other nodes if the merge agent’s address in the received
PartitionMergeQuery message is higher than the merge agent’s address in the previous flood
message or if the network id of the sender is different from its own network id, else it responds;
(ii) If the node itself is a merge agent and receives a flood message from another merge agent
with lower IP address, the node aborts the merging process that it had initiated and responds
to the received flood message. Thus, the merge agent with the lowest IP address among all
the merge agents succeeds. Figure 1(d) illustrates this situation. Nodes C-P and D-Q detect the
merging of networks X and Y simultaneously. Node C has a lower IP address than node P.
So, C initiates network-wide flood within X and P initiates flood within Y on behalf of C.
Both the flood messages contain node C’s address as merge agent’s address. Similarly, node D
(merge agent) initiates the network-wide flood in X and Q initiates the network-wide flood on
D’s behalf in network Y. Later, when D learns about node C’s merge process, node D aborts its
merge process.
When more than two networks merge simultaneously, the merge requests are serialized based
on the priorities of the merge agents. If a node receives PartitionMergeQuery messages from mul-
tiple nodes that detected different mergers simultaneously, then: (i) If the node is a merge agent
and receives a flood message from another merge agent with lower IP address, then the node
suspends the merging process that it had initiated and responds to the flood message, (ii) If
the node is a co-merge agent and receives a flood message from a merge agent with lower IP
address, then the co-merge agent sends a Defer message to the merge agent with higher IP
address. The suspended merge process is resumed on completion of ongoing merge process.
This approach is quite similar to the one described in Section VI-E.
VII. DISCUSSION
In this section we discuss the correctness of the address assignment process and the partition
handling process. Theoretically, it is impossible to guarantee correctness in the event of message
losses and node failures [4]. As message delays are non-deterministic and timers are best guesses,
it is possible that timers may always expire prematurely and all the retry attempts fail all the
time. In this section we prove the correctness based on the assumption that not all the retry
attempts fail.
21
22. A. Correctness of IP address assignment
Given: Prior to an address assignment all the nodes have unique IP addresses and disjoint free ip
sets.
Assertion: Following the IP address assignment all the nodes have unique IP addresses and
disjoint free ip sets.
Proof: We prove the assertion in two parts. First, we prove the correctness of address assignment
when there are no message losses. Then we relax our assumption and prove the correctness of
address assignment in the event of message losses.
Part 1: Correctness under reliable communication: When an IP address request arrives, one of
the following possibilities exist.
Possibility 1: The allocator has a non-empty free ip set: The allocator allots the second half of
its free ip set to the requester and removes the corresponding addresses from its own free ip set.
The requester configures itself with the first address from the alloted set and forms its free ip
set with the remaining addresses in the alloted set (refer to Section VI-B.1 for details). Thus,
following the IP address assignment, the IP address at both the allocator and the requester would
be different and their free ip sets disjoint. The IP addresses and free ip sets at the other nodes
in the network remain unaffected.
Possibility 2: The allocator has an empty free ip set and finds a non-empty free ip set in the
network during expanding ring search: During expanding ring search, the allocator chooses one
of the nodes (say X) with non-empty free ip set and allots the second half of the addresses from
X’s free ip set to the requester. X deletes the alloted addresses from its free ip set (see Section
VI-B.2). Thus, all the nodes in the the network have unique IP addresses and disjoint free ip
sets following the IP address assignment.
Possibility 3: The allocator does not find any free IP address during the expanding ring search: It
does reclamation of addresses. As described in Section VI-B.3, after reclamation, the addresses
in the missing addresses set are free. One of the addresses from the resultant missing addresses
set is assigned to the requester. The remaining addresses are re-distributed among other nodes
in the network.
Possibility 4: During concurrent IP address requests, a node X receives multiple IP address
requests and has a non-empty free ip set: X assigns different IP addresses and disjoint free ip
22
23. sets to the requesters (Section VI-E).
Possibility 5: Multiple allocators perform reclamations concurrently: The allocator(s) with lower
priority receives Defer message(s) and suspends the allocation process while the allocator with
highest priority continues its allocation process. After the highest priority allocator completes
the allocation process, the next highest priority allocator resumes its allocation process. Thus,
concurrent IP address requests are serialized based on the priorities of the allocators.
Therefore, when there are no message losses, all the five possibilities result in unique IP address
assignment and disjoint free ip set allotment.
Part 2: Correctness of address assignment in the event of message losses: Suppose the IP address
assignment results in duplicate addresses in the network in the event of message losses. The
following messages are exchanged during the address allocation process.
If IPAddressRequest and/or IPAddressAvail messages are lost, when offer ip timer expires,
the requester retries by sending IPAddressRequest message for a maximum of requester retry
times. If it does not receive any response, it assigns itself an IP address, a free ip set and
a network id. If requester is the only node in the network, then there are no duplicate
addresses. If that is not the case, the requester assigns itself an IP address which may
already be in use in the network. This amounts to a single node MANET (composed of
the requester) co-located with another MANET consisting of all the other nodes. However,
the network id assigned to the requester would be different from the network id of other
nodes in the network. Eventually, when the MANETs are able to communicate with each
other by network-merging process the duplicate address assignments would be detected and
removed (Section VI-G). Thus, there would be no duplicate addresses in the network.
If AllocatorChosen message is lost and the assign ip timer at the requester expires, the
requester retries by sending IPAddressRequest message and this IP address allocation pro-
cess is started all over. This is done for a maximum of assign retry times after which the
requester is either successful in obtaining a unique IP address or it fails to acquire one.
If IPAddressInfoRequest and/or IPAddressInfoReply messages from nodes get lost during
expanding ring search, then during reclamation of addresses, these nodes would respond
with ConflictNotification messages. If ConflictNotification message is also lost then either:
(i) Node receives IPAddressUpdate message, invalidates its IP address and tries to acquire
23
24. a new IP address, or (ii) Node does not receive the IPAddressUpdate message, hence it
would not receive the new network id. This node remains as a part of the old network and
eventually when it communicates with any node in the current network, it gets merged into
the current network (Section VI-G).
If IPAddressChosen and/or IPAddressConfirm messages are lost, the allocator on confirm ip
timer expiry sends IPAddressChosen message to another node with non-empty free ip set
and continues the allocation process. If the IPAddressConfirm message is lost, the proposed
free block of IP addresses would no longer be available for allocation because they are
assumed to be already allocated. Thus, it is possible that some addresses are unavailable
temporarily. This does not violate the correctness requirement and those addresses would
be available after subsequent address reclamation.
Thus, none of the message losses lead to duplicate address assignment. Therefore, we prove that
IP address assignment process works correctly in the event of message losses.
B. Correctness of Partition Handling process
Given: Prior to network-merging all the nodes in the network have unique IP addresses and
disjoint free ip sets.
Assertion: Following network-merging, all the nodes in the merged network have unique IP
addresses and disjoint free ip sets.
Proof: The network id assigned to each network is unique (see Section VI-G for details on
uniqueness of network id). The Hello messages are exchanged periodically to detect merging of
networks. Thus, in finite time, the merging of networks would be detected. When two networks
merge, there could be address conflicts. In the proposed scheme, the node responsible for
configuring the merged network (merge agent) collects the network-wide information in terms
of IP addresses, free ip sets, and pending ip sets. If there are address conflicts, the merge agent
would send address invalidation message with new IP addresses for nodes with conflicting IP
addresses. The merge agent would also send a flood message to re-distribute the free IP addresses
among the nodes in the merged network and update the network id of the merged network. As
all the conflicting addresses are removed and disjoint free ip sets are distributed, all nodes in
the merged network would have unique IP addresses and disjoint free ip sets.
24
25. VIII. SIMULATION EXPERIMENTS
Simulation experiments were performed using the network simulator ns-2 (ver 2.1b9a) [18]
with CMU extensions to support MANETs to evaluate the performance of the protocol in terms
of message complexity and latency.
A. Simulation scenario and parameters
The random waypoint mobility model was used. The speed of the nodes in the network was 5
meters/second and the pause time was 10 seconds. The simulation duration was 4500 seconds.
The routing protocol used was AODV, although any routing protocol can be used. The hello
timer period was 3 seconds, and the requester retry threshold and assign retry threshold were
set to 3 and 2, respectively. The hop count d was incremented by 1 during the expanding ring
search. Following terms are used in subsequent discussions:
1) Network density: the number of nodes per unit area.
2) Degree of a node: the number of nodes in the neighborhood of the node.
Two kinds of simulation experiments were carried out:
1) In the first kind, performance of the protocol was evaluated in terms of latency and
communication overhead. The following three sets of simulation were performed:
a) Varying Node Population: Networks with n nodes were simulated where n varied from
50 nodes to 300 nodes. The area of the network was 408m 408m, 578m 578m, 816m
816m and 1001m 1001m for the 50, 100, 200 and 300 node networks, respectively (ensuring
the network density of around 300 nodes/✂✁
✄
). The size of the IP address block was 2 n.
Simulation started with zero configured nodes. The network was initialized with a single node.
The inter-arrival time of new nodes in the MANET was exponentially distributed with mean of
0.2 node arrivals/second. The arriving node could appear anywhere within the area served by
the network. Once the population reached n, departure of nodes began. The inter-departure time
of nodes was exponentially distributed with a mean of 0.24 node departures/second. The arrivals
and departures continued independently until the number of IP address allocations reached (n +
500) requests. Simulation results were collected for the last 500 address allocations. From these
experiments the behavior of protocol for different node populations was analyzed.
25
26. b) Varying Address block range: Networks with 50 and 300 nodes, respectively with varying
address block size were simulated. The area of network was set to 578m 578m and 1415m
1415m for the 50 node and 300 node systems, respectively. The address block size was varied
from twice the node population to 10 times the node population. The arrival/departure pattern
was the same as in the previous case. These experiments were used to analyze the effect of
varying address block size on address allocation latency and communication overhead.
c) Varying Network Density: Networks with 50 and 300 nodes, respectively with varying
network density were simulated. The address block size was three times the node population.
The arrival/departure pattern was the same as in the previous case. The goal of these experiments
was to study the address allocation latency and communication overhead with varying network
density (which also affects the degree of each node).
2) In the second kind, the effect of the protocol on application traffic was analyzed: We used
50 and 300 node networks with 25% of node population generating CBR traffic. For every source
there was a corresponding destination. The source and destination pairs were randomly selected.
Each source injected packets with rate varying from 4 packets/second to 20 packets/second. The
packet size was 512 bytes. The effect of the control traffic induced by our protocol on the end-
to-end latency of application traffic was studied. This required us to run two sets of experiments
for a given network scenario: (i) Using the proposed protocol for IP address assignment with its
associated messages competing with application traffic for network bandwidth, and (ii) Using
static IP address assignment with no resultant address assignment traffic. The area of the network
was 578m 578m and 1415m 1415m for the 50 and 300 node systems, respectively.
Following statistics were gathered:
Number of unicast messages: This includes the various unicast messages required namely
IPAddressAvail, AllocatorChosen, IPAddressInfoRequest, IPAddressInfoReply, IPAddressCho-
sen, IPAddressConfirm, NoConflict, ConflictNotification, Defer, Resume and WaitPeriod mes-
sage.
Latency: The total time taken by the protocol, from the instant the node enters the network
and requests for an IP address until the node is assigned an IP address.
Number of address reclamations: Number of times address reclamations were needed during
the entire simulation. Two network-wide floods are required during each address reclama-
26
27. tion: (i) TentativeFreeAddresses sent by the allocator when none of the active nodes in the
network have a non-empty free ip set, and (ii) IPAddressUpdate message sent to update the
network id at nodes in the network.
Free ip set distribution: This gives the mean number of addresses in the free ip set of the
nodes in the network and the mean size of largest contiguous block of addresses in the
free ip set of the nodes in the network.
Partition Handling overhead: This indicates the communication overhead incurred during
network-merging, including the number of network-wide flood messages and number of
unicast messages. The network-wide flood messages include the following: (i) Partition-
MergeQuery flooded by the node detecting the network-merging, (ii) network-wide flood
message with IP addresses to be invalidated, sent by the node responsible for configuring
the merged network, and (iii) network-wide IPAddressUpdate message sent to all the nodes
in the network. The unicast messages include the PartitionMergeResponse messages sent
in response by the nodes in network for the PartitionMergeQuery message.
Vulnerability Period: This is the total time taken by the protocol, from the instant network-
merging is detected to the instant when the merged network is re-configured.
End-to-end latency for application data: The time taken for application data to reach the
destination node from the source node.
B. Simulation Results
1) Message overhead:
a) Impact of varying node population on number of unicast messages: The 95% confidence
interval (CI) for mean number of unicast messages for varying node population over 500 address
allocations is plotted in Figure 2(a). Figure 2(b) gives 95% confidence interval for mean degree
of requester in the network. It was observed that, the mean degree of requester was about 20
and the mean number of unicast messages varied from around 50 to 60 messages. Every address
allocation needs at least ((2 degree of requester) + 1) unicast messages. Hence, the number
of unicast messages is between 2.5 to 3 times the degree of requester. This is because most
of the addresses were allocated locally. Only for few allocations, expanding ring search and
reclamations were needed as can be seen from Figure 2(c). Thus, for the same network density,
27
28. 50 100 150 200 250 300
0
10
20
30
40
50
60
70
Node Population (Number of Nodes)
Unicast
Message
(Number
of
Messages)
95% CI Lower
Mean
95% CI Upper
(a) Node Population vs Number of
Unicast Messages
50 100 150 200 250 300
0
5
10
15
20
25
Node Population (Number of Nodes)
Degree
of
Requester
(Number
of
Neighbors) 95% CI Lower
Mean
95% CI Upper
(b) Node Population vs Degree of
requester
50 100 200 300
0
10
20
30
40
50
60
70
80
90
100
110
Node Population (Number of Nodes)
Percentage
(%)
of
address
allocations
Locally
Expanding ring searches
Address Reclamations
(c) Node Population vs Percent-
age of Expanding ring searches and
Reclamations
Fig. 2. Communication overhead for varying Node population ( address block size = 2 x node population, network density =
300 nodes/✂✁☎✄ )
the number of unicast messages required increases sub-linearly with increase in network size.
0 50 100 150 200 250 300
0
10
20
30
40
50
60
70
Network Density (Nodes/Km2
)
Number
of
Unicast
Messages
95% CI Lower for 300 node
Mean for 300 node
95% CI Upper for 300 node
95% CI Lower for 50 node
Mean for 50 node
95% CI Upper for 50 node
(a) Network Density (nodes/✂✁✆✄ )
vs Number of Unicast Messages
0 50 100 150 200 250 300
0
5
10
15
20
25
Network Density (Nodes/Km2
)
Degree
of
requester
(Number
of
Neighbors)
95% CI Lower for 300 node
Mean for 300 node
95% CI Upper for 300 node
95% CI Lower for 50 node
Mean for 50 node
95% CI Upper for 50 node
(b) Network Density (nodes/✂✁✆✄ )
vs Degree of requester
300 150 75 38 19
0
10
20
30
40
50
60
Network Density (Nodes/Km2
)
One Hop searches
Two Hop searches
Three Hop Searches
Number of Reclamations
50
node
300
node
50
node
300
node
50
node
300
node
50
node
300
node
50
node
300
node
(c) Network Density (nodes/✂✁✆✄ )
vs Expanding ring searches and
reclamations
Fig. 3. Communication overhead for varying network density ( node population = 50 and 300 nodes, Address Block Size = 3
x node population )
b) Impact of varying network density on number of unicast messages: Figure 3(a) shows
the 95% confidence interval for mean number of unicast messages for 50 node and 300 node
systems with varying network density. It was observed that the mean number of unicast messages
exchanged increases with the increase in network density. In networks with high density, there
are more nodes in the neighborhood of the requester, leading to a higher number of message
28
29. exchanges. Thus, we can say that the number of message exchanges required is a function of
degree of nodes in the network. This can further be confirmed by Figure 3(b) which shows
the degree of the requester. The number of unicast messages exchanged is 2 to 3 times the
degree of requester. The same reason given in previous section applies. Figure 3(c) shows the
number of expanding ring searches required along with their radius and the number of address
reclamations done during the entire simulation duration. From Figure 3(c), we can infer that
with the increase in area there was need to search farther in the network, with expanding ring
searches upto a distance of 3 hops from the allocator. For networks with network density less
than 75 nodes/✂✁
✄
, we observed partitioning of network into smaller networks, which in turn
lead to lower number of expanding ring searches and reclamations.
2 x n 3 x n 4 x n 5 x n 10 x n
0
10
20
30
40
50
60
70
Address Block Size (Number of Addresses)
Number
of
Unicast
Messages
95% CI Lower for 300 node
Mean for 300 node
95% CI Upper for 300 node
95% CI Lower for 50 node
Mean for 50 node
95% CI Upper for 50 node
(a) Address Block Size vs Number
of Unicast Messages
2 x n 3 x n 4 x n 5 x n 10 x n
0
20
40
60
80
100
120
Address Block Size (Number of Addresses)
Percentage
(%)
of
addresses
assigned
locally
For 50 node n/w
For 300 node n/w
(b) Address Block Size vs Percent-
age of address allocations done lo-
cally
2 x n 3 x n 4 x n 5 x n 10 x n
0
5
10
15
20
25
30
35
40
45
50
Address Block Size (Number of Addresses)
One Hop searches
Two Hop searches
Three Hop Searches
Number of Reclamations
50
node
300
node
50
node
300
node
50
node
300
node
50
node
300
node
50
node
300
node
(c) Address Block Size vs Expand-
ing ring searches and reclamations
Fig. 4. Communication overhead for varying address block size for 50 and 300 node network ( network density = 150
nodes/ ✁ ✄ )
c) Impact of varying address block size on number of unicast messages: The 95% confi-
dence interval for mean number of unicast messages over 500 address allocations for 50 node
network and 300 node network is plotted in figure 4(a). From the plot, it can be seen that the
mean number of unicast messages exchanged decreases as the size of address block increases.
We observed around 17% decrease in number of unicast messages for a 300 node system when
the address block size was increased from 2n to 10n. When the IP address block size is small,
the probability of requester finding a free IP address within its neighborhood is small. Hence,
the request needs to be propagated farther through the network, leading to a higher number
29
30. of message exchanges. Figure 4(b) shows the fraction of address requests that were completed
locally by the allocator. It was observed that around 95% of addresses were allocated locally
without requiring expanding ring search or reclamation of addresses. Figure 4(c) shows the
number of address allocations that required expanding ring search and address reclamation to
obtain an IP address. In our simulation experiments we observed that the number of expanding
ring searches and the number of address reclamations decreased with increase in the address
block size. Most of them were one or two hop searches.
d) Free ip set distribution: Free ip set distribution for 50 and 300 node systems is plotted
in Figure 5. We collected snapshots at different stages of the simulation. In our simulations,
the addresses are re-distributed randomly among nodes during reclamation and network-merging
processes. During graceful departure of a node, the addresses given by the departing node to
one of its neighbors need not be contiguous with the addresses in the neighbor’s free ip set. To
find a node with free ip set contiguous to addresses in free ip set of the departing node would
require significant communication overhead. We observed that the difference between mean size
of free ip sets and the mean size of contiguous block in free ip sets is not much (around 1 to
2 addresses). As the simulation progresses, the mean size of contiguous block decreases, but
again the decrease is not significant. Hence, there is little fragmentation of the free ip set and
not much reason for re-distribution of addresses to yield contiguous free ip set at nodes.
1000 1500 2000 2500 3000 3500 4000 4500
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
5
Time (sec)
Size
of
free_ip
set
Mean Contiguous free_ip set size
Mean free_ip set size
(a) Time vs Size of free ip set for
50 node network
1000 1500 2000 2500 3000 3500 4000 4500
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
5
Time (sec)
Size
of
free_ip
set
Mean Contiguous free_ip set size
Mean free_ip set size
(b) Time vs Size of free ip set for
300 node network
Fig. 5. Distribution of IP address blocks at different time instances for 50 and 300 node network (network density = 150
nodes/ ✁
✄
, address block size = 5 x node population)
2) Latency:
30
31. a) Impact of varying node population on latency: Figure 6(a) shows the 95% confidence
interval for mean latency for 50, 100, 200 and 300 node network scenarios. It was observed that
around 95% of address allocations were completed within 0.19 seconds as the requester had
neighbors with non-empty free ip sets. A small fraction of the allocations did require as much
as 4 seconds in a 300 node network. This was when the allocator had to perform a network-
wide search for an IP address. Figure 6(a) shows an increase in latency with increase in node
population. This was because as node population increased, for the same network density, the
network diameter also increased. But again increase in latency is sub-linear with respect to the
increase in node population. This is because of the same reasons provided in Section VIII-B.1.a.
50 100 150 200 250 300
0
0.1
0.2
0.3
0.4
0.5
0.6
Node Population (Number of Nodes)
Latency
(Seconds)
95% CI Lower
Mean
95% CI Upper
(a) Node Population vs Latency
(seconds)
2 x n 3 x n 4 x n 5 x n 10 x n
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
Address Block Size (Number of Addresses)
Latency
(secs)
(Network Density = 150 Nodes/Km.2
)
95% CI Lower for 300 node
Mean for 300 node
95% CI Upper for 300 node
95% CI Lower for 50 node
Mean for 50 node
95% CI Upper for 50 node
(b) Address Block Size vs Latency
(seconds)
0 50 100 150 200 250 300
0
0.2
0.4
0.6
0.8
1
Network Density (Nodes/Km2
)
Latency
(Seconds)
(Address Block size = 2 x node population)
95% CI Lower for 300 node
Mean for 300 node
95% CI Upper for 300 node
95% CI Lower for 50 node
Mean for 50 node
95% CI Upper for 50 node
(c) Network Density (nodes/ ✁
✄
)
vs Latency (seconds)
Fig. 6. Latency for 50 and 300 node network with varying node population, varying address block size and varying network
density
b) Impact of varying address block size on latency: The 95% confidence interval for mean
latency for 50 and 300 node network scenarios is plotted in Figure 6(b) with varying address
block size. It was observed that mean latency decreased with an increase in IP address block
size. We observed a 33% decrease in latency when the address block size was increased from
2n to 10n in the 300 node system. Larger the address block size, greater is the probability
of the requester finding a free IP address from its immediate neighbors. Hence, the number of
expanding ring searches/address reclamations required decreases leading to lower latency (Figure
4(c)). Again the decrease in latency is sub-linear with increase in address block size.
c) Impact of varying network density on latency: The 95% confidence interval for mean
latency for 50 node and 300 node network scenarios with varying network density is plotted
31
32. in Figure 6(c). It was observed that as the network density increased, mean latency decreased.
At higher network density, there are more nodes in the neighborhood of the requester. Hence,
the probability of a requester finding a free IP address from its immediate neighbors increases.
Thus, the number of expanding ring searches/reclamations required is low. Even if there is need
for expanding ring search, a free IP address is found within a few number of hops. At lower
network density, more expanding ring searches are required leading to higher latency. In a 300
node network the latency increases upto a point, after which the networks have partitions and
each partition in itself is a smaller network. Hence, we see a decrease in latency.
300 150 75 38 19
0
100
200
300
400
500
600
700
800
Network Density (Nodes/Kms2
)
Partition
Handling
Overhead
(Number
of
Messages)
Broadcasts
Unicasts
50 node
300 node
300 node
50 node
300 node
50 node
300 node
50 node
(a) Network Density (nodes/✂✁☎✄ )
vs Partition Messages (Number of
Messages)
0 50 100 150
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
5
Network Density (Nodes/Km2
)
Mean
Vulnerability
Period
(Seconds)
For 300 node
For 50 node
(b) Network Density (nodes/✂✁☎✄ )
vs Vulnerability Period (seconds)
Fig. 7. Partition Handling overhead for varying network density
3) Overhead due to Partitioning:
a) Impact of varying network density on partition handling: Figure 7(a) shows the com-
munication overhead incurred due to partition merging in 50 and 300 node networks with
varying network density. The lower the network density, the greater the probability of network-
partitioning. In our simulations we do not consider sparse networks which could experience
disconnections and merger very often. In a 50 node system, we observed that the network had
partitions but by the end of the simulation all the partitions had merged to form one single
network. For a 300 node system, we observed that the network had partitions until the end of
the simulation when the network density was low. Figure 7(b) shows the mean vulnerability
period for 50 and 300 node networks with varying network density. During the vulnerability
period, there could be duplicate addresses and it is possible that the traffic destined for one node
32
33. might get mis-routed to some other node. However, we observed that this period was always
less than 4 seconds.
4 6 8 10 12 14 16 18 20
0
0.5
1
1.5
2
2.5
3
CBR Traffic (packets/second)
End−to−End
Latency
(seconds)
Static IP addressing
Using proposed protocol
(a) CBR traffi c (packets/second) vs
End-to-End Latency (seconds) for
50 node network
4 6 8 10 12 14 16 18 20
0
1
2
3
4
5
6
CBR Traffic (packets/second)
End−to−End
Latency
(seconds)
Static IP addressing
Using proposed protocol
(b) CBR traffi c (packets/second) vs
End-to-End Latency (seconds) for
300 node network
Fig. 8. Effect of protocol on application data ( Address Block Size = 2 x node population )
4) Effect of protocol on application data: The mean end-to-end latency for CBR flows with
and without the traffic generated by the proposed protocol is shown in Figure 8. We observed
that there is a slight increase in end-to-end latency in the presence of control traffic generated
by address assignment protocol. This is due to contention between the CBR traffic and protocol
messages for communication medium. From Figure 8 we see that the application traffic is not
adversely affected by the control traffic induced by our IP address assignment protocol.
From our simulation experiments, we can see that the communication overhead and latency
is low when compared to the solution presented in [12]. This is because regardless of where a
new node appears in the network, most IP address requests are satisfied locally incurring low
latency and communication overhead. All the simulations were conducted with 2Mbps wireless
channels. With current IEEE 802.11b and IEEE 802.11a hardware support for 11Mbps and
54Mbps channels respectively, latency would be less than what is shown in Figures 6 and 8.
IX. CONCLUSION
We presented a distributed protocol for dynamic configuration of nodes in MANETs. We have
addressed the issue of unique IP address assignment to nodes in MANETs in the absence of any
static configuration or centralized servers. The proposed protocol is based on the buddy system
33
34. [2], [3] used for memory management. The basic idea is to dynamically distribute the IP address
block among the nodes in the network. Our approach guarantees unique IP address assignment
under all network conditions including message losses, node crashes, network partitioning and
merging.
We have presented the results obtained from simulation experiments. Simulation results show
that most of the address allocations were done locally and required around 0.19 seconds. We
observed that the latency and communication overhead increased in a sub-linear manner with
increase in node population. We also observed that decrease in latency and communication
overhead was sub-linear with increase in the address block size. Thus, the proposed protocol
incurs low latency and communication overhead.
X. ACKNOWLEDGEMENTS
The authors wish to thank the anonymous reviewers for their useful suggestions that helped in
improving the quality of this paper. The authors also wish to thank Mansoor Mohsin for helpful
suggestions and discussions.
REFERENCES
[1] R. Droms, “Dynamic Host Confi guration Protocol,” Network Working Group, RFC 2131, Mar 1997.
[2] K. Knowlton, “Fast Storage Allocator,” Communications of the ACM, Oct 1965.
[3] J. L. Peterson and T. A. Norman, “Buddy Systems,” Communications of the ACM, Jun 1977.
[4] N. A. Lynch, “
A hundred impossibility proofs for distributed computing,” Proceedings of 8th Annual Symposium on
Principles of Distributed Computing, pp. 1-28, Aug 1989.
[5] D. B. Johnson, D. A. Maltz, Y. Hu and J. Jetcheva, “DSR: The Dynamic Source Routing for Multi-hop wireless Ad-hoc
Networks,”Ad Hoc Networking, pp.139 - 172, Addison-Wesley, 2001.
[6] M. Jiang, J. Li and Y.C. Tay, “Cluster Based Routing Protocol (CBRP) Function Specifi cation,” Internet Draft, Aug 1999,
work in progress, http://www.math.nus.edu.sg/ mattyc/cbrp.txt.
[7] V. D. Park and M. S. Corson, “
A Highly Adaptive Distributed Routing Algorithm for Mobile Wireless Networks,” IEEE
Conference on Computer Communications (Infocom ’97), 1997.
[8] M. R. Pearlman and Z. J. Haas, “Determining the Optimal Confi guration for the Zone Routing Protocol,” IEEE Journal
on Selected Areas in Communications, vol. 17, issue 8, pp. 1395 - 1414, Aug 1999.
[9] C. Perkins and P. Bhagwat, “Routing over Multihop Wireless Network in Mobile Computers,”SIGCOMM’94 Computer
Communications Review, Oct 1994.
[10] C. Perkins and E. Royer, “
Ad Hoc On-Demand Distance Vector Routing,” 2nd IEEE Workshop on Selected Area in
Communications, pp. 90-100, Feb 1999.
[11] Cisco, “Duplicate MAC addresses on Cisco 3600 series,” May 1997, http://www.cisco.com/warp/public/770/7.html.
34
35. [12] S. Nesargi and R. Prakash, “MANETconf: Confi guration of Hosts in a Mobile Ad Hoc Network,”Proceedings of INFOCOM
2002, 2002.
[13] S. Cheshire, B. Aboba and E. Guttman, “Dynamic Confi guration of IPv4 Link-Local Addresses,” Internet Draft, Jul 2004,
work in progress, “http://fi les.zeroconf.org/draft-ietf-zeroconf-ipv4-linklocal.txt”.
[14] C. Perkins, J. Malinen, R. Wakikawa, E. Royer and Y. Sun, “IP Address Autoconfi guration for Ad Hoc Networks,”Internet
Draft, Nov 2001, work in progress, http://people.nokia.net/ charliep/txt/aodvid/autoconf.txt.
[15] S. Thomson and T. Narten, “IPv6 Stateless Address Autoconfi guration,” RFC 2462, Dec 1998.
[16] K. Weniger and M. Zitterbart, “IPv6 Autoconfi guration in Large Scale Mobile Ad-Hoc Networks,”Proceedings of European
Wireless 2002, Florence, Italy, Feb 2002.
[17] N. Vaidya, “Weak Duplicate Address Detection in Mobile Ad Hoc Networks,” ACM International Symposium on Mobile
Ad Hoc Networking and Computing (MobiHoc), Jun 2002.
[18] A. J. McAuley and K. Manousakis, “Self-Confi guring Networks,” 21st Century Military Communications Conference
Proceedings, Oct 2000, vol.1, pp. 315 - 319.
[19] A. Misra, S. Das, A. McAuley and S.K. Das, “
Autoconfi guration, registration, and mobility management for pervasive
computing,” IEEE Personal Communications, pp. 24 - 31, Aug 2001.
[20] P. Patchipulusu, “Dynamic Address Allocation Protocols For Mobile Ad Hoc Networks,” Master’s Thesis, Texas A&M
University, Aug 2001.
[21] H. Zhou, L. Ni, and M. Mutka, “Prophet Address Allocation for Large Scale MANETs,” Proceedings of IEEE INFOCOM
2003, Mar 2003.
[22] K. Fall and K. Varadhan, “The ns Manual,”, http://www.isi.edu/nsnam/ns/ns-documentation.html.
[23] K. Weniger, “Passive Duplicate Address Detection in Mobile Ad Hoc Networks,” Proceedings of IEEE WCNC 2003, Mar
2003.
[24] J. Jeong, J. Park, H. Kim and D. Kim, “
Ad Hoc IP Address Autoconfi guration,” Internet Draft, draft-jeong-adhoc-ip-addr-
autoconf-03.txt, work in progress, Jul 2004.
[25] G. Ricart and A. K. Agrawala, “
An Optimal Algorithm for Mutual Exclusion in Computer Networks,” Communications
of the ACM, vol. 24, no.1, pp.9-17, Jan 1981.
35