Ad hoc networks are multi-hop networks consisting of wireless autonomous hosts, where each host may serve as a router to assists traffic from other nodes. Wireless ad hoc networks cover a wide range of network scenarios, including sensor, mobile ad hoc, personal area, and rooftop/mesh networks

1. 1
Ad Hoc and Sensor Networks
Dr. Parveen Kakkar
Department of Computer Science & Engineering
DAVIET, Jalandhar

2. Outline
 Introduction
 Characteristics of MANETs
 Challenges & Issues
 Applications
 Routing
 Table-driven Routing Protocols
 Source-initiated On-demand Routing
 Hybrid (Routing) Protocols
 Wireless Sensor Networks
 General Wireless Sensor Networks
 Fixed Wireless Sensor Networks

3. Introduction
 An ad hoc network is a collection of mobile nodes
forming a temporary network without the aid of any
centralized administration or standard support services.
 A mobile Ad hoc network (MANET) is formed by a
group of autonomous mobile nodes connected by wireless
links, in which there is no backbone infrastructure.
 In general, Ad hoc networks are self-creating, self-
organizing, and self-administrating networks. Hence, they
offer unique benefits and flexibility for a variety of
situations and applications.
10/24/20
23
3

4. Introduction(Contd..)
10/24/20
23
4
Ad hoc network with three participating nodes
Figure shows an example of a simple Ad hoc network of
three nodes.
• The outermost nodes are not within the transmission range
of each other.
• However the middle nodes can be used to forward the
packets between the outermost nodes.
• The middle node is acting as a router and three nodes have
formed an ad hoc network.

5. Characteristics of MANET
 Dynamic Topology
 Lack of Secure Boundaries
 Wireless Radio Medium
 Limited Resources
 No Predefined Infrastructure
 Lack of Centralized Administration
 Rapidly Deployable
10/24/20
23
5

6. Challenges & Issues of
MANETs
 Power Constraints
 Radio Interference
 Routing Protocols
 Limited Security
 Mobility Management
 Service Discovery
 Bandwidth Constraints
 Quality of Services (QoS)
10/24/20
23
6

7. Introduction(Contd..)
 A fundamental problem in ad hoc networking is
how to deliver data packets among Mobile Nodes
efficiently without predetermined topology or
centralized control, which is the main objective of
ad hoc routing protocols.
 Since mobile ad hoc networks change their
topology frequently, routing in such networks is a
challenging task.
 Moreover bandwidth, energy and physical
security are limited.
 With the increasing popularity of mobile devices
and wireless networks over the past few years,
wireless ad-hoc networks has now become one of
the most vibrant and active fields of communication
and networking research.
10/24/20
23
7

8. 13.3. Applications (Examples)
 “Wearable” computing
 Defense applications
 Crisis-management applications
 Natural disasters, where the entire communication infrastructure is in disarray
 Telemedicine
 E.g., assistance by a surgeon for an emergency
 Tele-geo processing applications
 Queries dependent on location of the users
 Integrating geographical info systems (GIS) & GPS
 Virtual navigation
 Data from a remote database transmitted to navigation device in car or in hand
 May contain graphical representation of streets, buildings, and the latest traffic
information
 May assist driver in selecting a route
 Education and Internet access
 K-12, continued education, etc., for people in remote areas
 E.g., email-by-bus in remote villages

9. 13.4. Routing in MANETS
 Many factors affecting routing in MANETs:
 Models of topology
 Selection of routers
 Initiation of route requests
 Specific underlying characteristics
 E.g. application-based characteristics
 Major goals in selecting routing protocols:
 Provide the maximum possible reliability - use alternative
routes if an intermediate node fails
 Choose a route with the least cost
 E.g., minimal # of hops from source to destination
 Give the nodes the best possible response time and
throughput
 Each node in MANETs expected to serve as a router
 All execute the same routing protocol
 Protocol calculates a route

10. 13.4.1. Need for Routing in MANETS
 Goals for routing in MANETs:
1) Route computation must be distributed
 Centralized routing in a dynamic network is usually too expensive
2) Routing computation should not involve the maintenance of
global state
 Would require xmitting a lot of information in dynamic topology
3) As few nodes as possible involved in route computation and
state propagation
 But every node must have quick access to routes on demand
 Ensure infrequent topology updates for MANET portions that have
no traffic
4) Each node must be only concerned about the routes to its
destinations
5) Stale routes must be avoided or detected & eliminated
6) Broadcasts should be avoided
 Broadcasts can be highly unreliable in MANETs
7) If topology stabilizes, the routes must converge to the
optimal routes
8) It is desirable to have a backup route
 Used when the primary route becomes stale

11. 13.4.1. Need for Routing in MANETS – cont.
 A major challenge in designing routing protocol:
 Must know at least how to route a packet via its
neighbors
 Network topology can change frequently
 Large # of network nodes (MSs)
=> Could have a lot of info to update
 Unless a clever protocol design

12. 13.4.2. Routing Classification
 Types of routing protocols:
1) Proactive protocols
 Keep routes ready at all times
 When a packet needs to be forwarded, the route is already known
 Example: distance vector routing protocols (more below)
2) Reactive (= on-demand) protocols
 Route determination on demand
 Determine a route only when there is a packet to send
 Examples: flooding routing algorithms, ad hoc on-demand
distance vector (AODV), temporarily ordered routing
algorithm (TORA) (more below)
 Proactive vs. reactive
 In proactive: a route available immediately
 In reactive: (1) a significant delay
(2) significant traffic of control msgs
(searching for a route)

13. 13.4.2. Routing Classification – cont.
 Is proactive or reactive better for MANETs?
 Pure proactive use too much bandwidth for control -
updating routing information
 Too keep it current all the time
 But topology changes are so quick that most routes never used
– a waste!
 Pure reactive too slow for real-time applications
 Types of routing protocols:
1) Table-driven protocols
2) Source-initiated on-demand protocols
3) Hybrid protocols
More on all of them in the following subsections

14. 13.5. Table-driven Routing Protocols
 “Table-driven” because:
Each node maintains table(s) with routing information for
every other nodes in the network
 Table-driven is proactive
 When the topology changes, updates are propagated throughout
the network.
 Examples of table-driven routing protocols are:
 Destination Sequenced Distance Vector routing (DSDV)
 Cluster-head Gateway Switch routing (CGSR)
 Wireless Routing Protocol (WRP)

15. 13.5.1. Destination-Sequenced
Distance-Vector Routing (DSDV)
 Based on the Bellman-Ford algorithm
 Each mobile node maintains a routing table
 In terms of number of hops to each destination
 Routing table updates are periodically transmitted
 Each entry in the table is marked by a sequence number
 Helps to distinguish stale routes from new ones, and thereby
avoiding loops
 To minimize the routing updates, variable sized update
packets are used
 Depending on the number of topological changes

16. 13.5.2. Cluster-head Gateway Switch Routing
(CGSR)
 Cluster-head gateway switch routing (CGSR) –
a clustered multi-hop routing, heuristic
 MANET divided into clusters
 For each cluster, a cluster-head (CH) elected
 By a distributed CH selection algorithm
 It modifies DSDV by using a hierarchical CH to route traffic
 Gateway nodes are bridges connecting cluster heads
 Frequent changes of CHs affect the performance
7 1
0
8 9
Cluster Head
Internal (non-
cluster) Node
Gateway Node

17. 1
3
2
4 7
10
5
6
8
9
11
12
Cluster Head
Internal (non-
cluster) Node
Gateway Node
13.5.2. Cluster Head Gateway Switch Routing (CGSR)
 CGSR scenario - routing from Node 1 to Node 12:
 Packet sent by Node 1 is first routed to its CH (2)
 The packet is routed from CH (2) to gateway (4) connecting to CH of
another cluster (5)
 Packet routed from gateway (4) to the next CH (5)
 …
 Packet routed from gateway (10) to CH of destination cluster (11)
 Packet routed to destination node (12)

18. 13.5.3. The Wireless Routing Protocol (WRP)
 Each node maintains 4 tables:
 Distance table
 Routing table
 Link cost table
 Message Retransmission List table (MRL)
MRL contains the sequence number of the update
message, a retransmission counter and a list of
updates sent in the update message

19. 13.5.3. The Wireless Routing Protocol (WRP) – cont.
 Nodes inform each other of link changes using update
messages
 Nodes send update messages after processing updates
from their neighbors or after detecting a change in the
link
 If a node is not sending messages, it must send a HELLO
message within a specified time to ensure connectivity
 If the node receives a HELLO message from a new node,
that node is added to the table
 It avoids the “count to infinity” problem

20. 13.6. Source-Initiated On-Demand Routing (the
second category of routing protocols)
 Source-initiated on-demand routing - the 2nd
category (“family”) of routing protocols
 This family of protocols includes:
1) Ad hoc On-Demand Distance Vector (AODV)
2) Dynamic Source Routing (DSR)
3) Temporary Ordered Routing Algorithm (TORA)
4) Associativity Based Routing (ABR)
5) Signal Stability Routing (SSR)

21. 13.6.1. Ad hoc On-Demand Distance
Vector Routing
 AODV is an improvement over DSDV, which
minimizes the number of required broadcasts
by creating routes on demand
 Nodes that are not in a selected path do not
maintain routing information or participate in
routing table exchanges
 A source node initiates a path discovery
process to locate the other intermediate
nodes (and the destination), by broadcasting
a Route Request (RREQ) packet to its
neighbors

22. Route Discovery in AODV Protocol
Source
Destinatio
n
1
3
2
5
7
4
6
8
(a) Propagation of Route Request (RREQ) Packet
Source
Destinatio
n
(b) Path Taken by the Route Reply (RREP) Packet
1
3
2
5
7
4
6
8
Hop1 Hop2 Hop3

23. 13.6.2. Dynamic Source Routing
 The protocol consists of two major phases: Route
Discovery, Route Maintenance
 When a mobile node has a packet to send to some
destination, it first consults its route cache to check
whether it has a route to that destination
 If it is an un-expired route, it will use this route
 If the node does not have a route, it initiates route
discovery by broadcasting a Route Request packet
 This Route Request contains the address of the
destination, along with the source address

24. 13.6.2. Dynamic Source Routing – cont. 1
 Each node receiving the packet checks to see
whether it has a route to the destination. If it does
not, it adds its own address to the route record of
the packet and forwards it.
 A route reply is generated when the request
reaches either the destination itself or an
intermediate node that contains in its route cache
an un-expired route to that destination.
 If the node generating the route reply is the
destination, it places the the route record
contained in the route request into the route reply.

25. Hop1 Hop2 Hop4
Hop3
1
4
7
5
3
Source
Destination
<1>
<1>
<1,2>
<1>
<1,3>
<1,3,5>
<1,3,5,7>
<1,4>
<1,4,6>
(a) Building Record Route During Route Discovery
4
1
4
6
8
7
2
3
Source
Destination
<1,4,6>
<1,4,6>
<1,4,6>
(b) Propagation of Route Reply with the Route Record
1
2
3
5
6
7
8
Creation of Route Record in DSR
13.6.2. Dynamic Source Routing – cont. 2

26. 13.6.3. Temporarily Ordered Routing
Algorithm (TORA)
 TORA is a highly adaptive loop-free distributed
routing algorithm based on the concept of link
reversal
 TORA decouples the generation of potentially far-
reaching control messages from the rate of
topological changes
 The height metric is used to model the routing state
of the network

27. Source
Destinatio
n
H = 0
H = 1
H = 2
H = 3
Illustration of Tora height metric
13.6.3. TORA – cont. 1

28.  The protocol performs three basic functions: route
creation, route maintenance, route erasure
 During the route creation and maintenance phases
nodes use a height metric to establish a Directed
Acyclic Graph (DAG) rooted at the destination
 Thereafter links are assigned a direction based on
the relative heights
13.6.3. TORA – cont. 2

29. 13.6.4. Associativity Based Routing (ABR)
 The three phases of ABR are: route discovery, route
reconstruction, route deletion
 In ABR a route is selected based on the degree of stability
associated with mobile node
 Association stability is defined by connection stability of one
node with respect to another node over time and space
 Each node generates a beacon to signify its existence
 When received by neighboring nodes, the beacon causes their
associativity tables to be updated
 The route discovery is accomplished by a Broadcast Query- Reply
(BQ-REPLY) cycle
 When a discovered route is no longer desired, the source node
initiates a Route Delete broadcast so that all the nodes along the
route update their routing tables

30. 13.6.5. Signal-Stability-Based Routing (SSR)
 SSR selects a route based on the signal strength
between nodes and a node’s location stability
 This route selection criteria has the effect of
choosing routes that have a better link connectivity

31. 13.7. Hybrid routing protocols
 Many of hybrid routing protocols
 7 example protocols follow

32. 13.7. Hybrid routing protocols – cont. 1
1) Zone Routing Protocol (ZRP): A node proactively maintains
routes to destinations within a local neighborhood
 The construction of a routing zone requires a node to first
know who its neighbor, which is implemented through a MAC
layer Neighbor Discovery Protocol
2) Fisheye State Routing (FSR): There are multi-level fisheye
scopes to reduce routing update overhead in large networks
 It helps to make a routing protocol scalable by gathering
data on the topology, which may be needed soon
3) Landmark Routing (LANMAR): Uses a landmark to keep track of
a logical subnet
 The LANMAR routing table includes only those nodes within
the scope and the landmark nodes themselves

33. 13.7. Hybrid routing protocols – cont. 2
4) Location-Aided Routing (LAR): Exploits location information to
limit the scope of routing
 LAR limits the search based on the expected location of the
destination node and thereby restricts and controls the
flood of Route Request packets
5) Distance Routing Effect Algorithm for Mobility (DREAM) : Based
on the distance effect and a node’s mobility rate
 Each node can optimize the frequency at which it sends
updates to the networks and correspondingly reduce the
bandwidth and energy used
6) Relative Distance Micro-discovery Ad Hoc Routing (RDMAR):
Based on the calculated relative distance between two
terminals
 The query flood is localized to a limited region centered at
the source node

34. 13.7. Hybrid routing protocols – cont. 3
7) Power Aware Routing: Power-aware metrics are used for
determining routes
 It reduces the cost, ensures that the mean time to
node failure is increased, without any further delay in
packet delivery

35. 13.7.7B. Summary of Some Characteristics of
Selected Protocols
Pro-
tocol
Route
Acquisi-
tion
Flood for
Route
Discovery
Delay for
Route
Discovery
Multipath
Capability
Effect of Route
Failure
DSDV Computed
a priori
No No No Updates the routing
tables of all nodes
WRP Computed
a priori
No No No Ultimately, updates
the routing tables of
all nodes by
exchanging MRL
between neighbors
DSR On-
demand,
only when
needed
Yes.
Aggressive
use of
caching may
reduce flood
Yes Not explicitly.
The
technique of
salvaging
may quickly
restore a
route
Route error
propagated up to the
source to erase
invalid path

36. Summary of Some Characteristics of Selected Protocols – cont.
Pro-
tocol
Route
Acquisi-
tion
Flood for
Route
Discovery
Delay for
Route
Discovery
Multipath
Capability
Effect of Route
Failure
AODV On-
demand,
only when
needed
Yes.
Controlled
use of cache
to reduce
flood
Yes No, although
recent
research
indicate
viability
Route error
propagated up to
the source to erase
invalid path
TORA On-
demand,
only when
needed
Basically one
for initial
route
discovery
Yes. Once the
DAG is
constructed,
multiple paths
are found
Yes Error is recovered
locally
LAR On-
demand,
only when
needed
Reduced by
using
location
information
Yes No Route error
propagated up to
the source
ZRP Hybrid Only outside
a source's
zone
Only if the
destination is
outside the
source's zone
No Hybrid of updating
nodes' tables within
a zone and
propagating route
error to the source

37. 13.8. Wireless Sensor Networks
 Wireless sensor networks (WSNs)
 A class of ad hoc networks
 A collection of hundreds or thousands of tiny, disposable
(bec. low-cost) & low-power (bec. battery-operated) sensor nodes
 Communicating together to achieve an assigned task:
monitoring & analysis of an area
 Sensor node in WSN
 Converts a sensed physical attribute (e.g., temperature)
into data
 Includes :
 Sensing module
 Communications module
 Computing module
 Memory module
 Power source (usually a battery)

38. 13.8. Wireless Sensor Networks – cont. 1
 Wired sensor networks
 Example: wired sensor net within a plane
 Known for years in many applications
 Monitor critical physical parameters
 Alert when anomalies perceived
 Sensor locations carefully predesigned
 Sensors distributed in strategic locations
 # of sensor nodes can be huge
 As many as needed to cover the monitored area
 Connected via wires
 Fault tolerance requirements
 Avoid single points of failure
 If work unattended => can not repair if failed
 E.g., in inhospitable

39. 13.8. Wireless Sensor Networks – cont. 2
 Advances in technology enabled wireless sensornets
(WSNs)
 Esp. advances in miniaturization, low-cost sensors (&
multisensors), wireless communications, batteries
 Low costs (due to advances in technology) enabled massive
deployment of WSN nodes
 No need for predesigned locations
 Can drop them from a plane, a speeding car, etc.
=> no installation costs (saves more)

40. 13.8. Wireless Sensor Networks – cont. 3
 Advantages of WSNs over wired sensornets
1) Ease of deployment
 Deployed without careful design
 E.g., dropped from a plane
2) Extended range
 At the same cost can cover a larger area
 No need for infrastructure
 One large wired sensornet replaced by many smaller wireless
sensornets at the same cost
 Can easily move from area to area
3) Fault tolerance
 A design requirement!
 Must tolerate node failures (maybe reduces monitoring accuracy)
 Some fault tolerance is “natural”
 Bec. failure of a sensor node is “masked” by other nodes collecting similar data in the same area
4) Mobility
 No wires inhibiting mobility – not even for power (batteries)
 Still WSNs less mobile than ad hoc networks

41. 13.8. Wireless Sensor Networks – cont. 4
 Inherent limitations of WSNs
1) Limited energy (batteries, etc.)
2) Low-bandwidth transmission
3) Error-prone transmission
 Bec. xmitted at low power (Item 1 above) & over limited bandwidth (Item 2
above)

42. 13.8. Wireless Sensor Networks – cont. 5
 Design requirements for WSNs & their protocols
1) Maximize WSN lifetime
 Discussed below
2) Accommodate dynamic, fast-changing physical para-
meters affecting WSNs, such as:
(1) Power availability for nodes
(2) Positions of nodes
(3) Reachability
 For a given node, which other nodes can it reach
(4) Types of tasks executed by nodes
 I.e. what attributes (such as temperature) the node monitors & reports

43. 13.8. Wireless Sensor Networks – cont. 6
 Dealing with limited energy:
 Design WSN & its protocols carefully to maximize WSN’s
lifetime
 One approach: balance energy use in such a way that all nodes die at approximately the
same time
 Once this happens, can replace the whole sensornet
 Better than replacing individual nodes (impossible or inconvenient)

44. 13.8. Wireless Sensor Networks – cont. 7
 Data exchange in WSN is fundamentally different than in
other wireless networks
 WSNs are data-centric networks
 The interest is in “what is the data?” rather than “where is the data?”
 E.g., WSNs focus on attributes (e.g., temperature, velocity)
 WSNs must efficiently respond to application/user queries asking for data
 => WSNs require different routing protocols then MANETs
 Routing protocols for WSNs must be application-data-
specific

45. 13.8. Wireless Sensor Networks – cont. 8
 Challenges in design of (application-data-specific) routing
protocols for WSNs
1) No unique node ID to be used for routing
 Which is typical in traditional wired/wireless networks
 Bec. (#1) routing to/from a specific node is not required in data-
centric WSNs
 Recall: It does not matter “where is the data?”
 Bec. (#2) with the large # of nodes in WSNs, ID would be large
 ID might be larger than amount of actual data being xmitted
2) Nodes often send aggregated data, not raw data
 Adjacent nodes may have similar data, so aggregation cuts traffic
3) Routing protocols must be application-specific and data-
centric
 Bec. WSNs are application-specific and data-centric
 E.g., WSN may require protocol customized to very efficient
delivery of data on a single attribute (e.g., temperature)
 Could be very inefficient for delivery of data on other single
attribute, or delivery of multi-attribute data
4) Minimizing energy consumption

46. 13.8. Wireless Sensor Networks – cont. 9
 Features of an ideal WSN
1) Attribute-based addresses
 Composed of a series of attribute-value pairs
 Specify physical parameters to be sensed
 Example attribute address: “temperature>100C, location=?”
 All nodes that sense temperature>100C must report their locations
2) Location-awareness of nodes
 A node often (as in example in (1) above) needs to know its location
 Otherwise cannot provide sensed data (cf. example above)
3) Must react immediately to drastic environment changes
 Necessary for time-critical monitoring applications
 Can react slowly to non-critical changes/events
 Saving bandwidth & energy at the cost of increased latency
4) Efficient handling of queries
 Efficient xmission of queries from usesr/applications to
appropriate nodes (=> need efficient routing!)
 Efficient xmission of answers to queries from nodes to
users/applications (=> need efficient routing again!)
 Can reply with larger latency for noncritical changes/events
 E.g., can increase interval for reporting periodic data

47. Routing Attacks in MANET
1. Denial-of-Service (DoS) Attacks:-
 DoS is an intentional attempt by malicious users/
attackers to completely disrupt or degrade
availability of service/resource to legitimate/ authorized
users.
Main targets:
 CPU Cycles
 Memory
 Bandwidth
 DoS attacks exploit the vulnerabilities of the network
protocol architecture.
 They do not need complicated technology, and they are very
easy for attackers to launch, but very hard for victims to
prevent and trace back.
10/24/20
23
47

48. Flooding Attack
 In flooding attack attacker will flood the network with useless
packets to exploit the bandwidth and useful resources of the
network.
It can be broadly classified into two types:
 Data Flooding
 RREQ Flooding
10/24/20
23
48

49. Data Flooding
 In data flooding attacker will send unwanted data items to
congest the network.
 Once the paths are established between all the nodes attacker
will flood unwanted data packets.
 Which will ultimately causes to Denial of Service.
10/24/20
23
49

50. RREQ Flooding
S
D
Route Request
The RREQ sent for the destination which is
not present will traverse in whole of the
network.
10/24/20
23
50

51. RREQ Flooding
 Attacker will send large amount of RREQ requests to waste the
bandwidth and resources of the network.
 Usually the destination IP chosen for RREQ will not exists in
the network, so no node knows the location of the required IP.
 As a result the RREQ will pass through whole the network.
10/24/20
23
51

52. Routing Attacks in MANET
(continue..)
Black hole attack
 In a black hole attack, a malicious node sends fake routing
information, claiming that it has an optimum route and
causes other good nodes to route data packets through the
malicious one.
 For example, in AODV, the attacker can send a fake RREP
(including a fake destination sequence number that is
fabricated to be equal or higher than the one contained in
the RREQ) to the source node, claiming that it has a
sufficiently fresh route to the destination node.
 This causes the source node to select the route that passes
through the attacker.
 Therefore, all traffic will be routed through the attacker, and
therefore, the attacker can misuse or discard the traffic.
10/24/20
23
52

53. Routing Attacks in MANET
(continue..) Black hole attack
10/24/20
23
53
Black Hole Attack
Figure shows an example of a black hole attack, where
attacker A sends a fake RREP to the source node S,
claiming that it has a sufficiently fresher route than other
nodes.
Since the attacker’s advertised sequence number is higher
than other nodes’ sequence numbers, the source node S will
choose the route that passes through node A.

54. Routing Attacks in MANET
(continue..)
Other DoS Attacks
I. “Byzantine” attackers respond to the RREQ with wrong route
information to disrupt or degrade the routing services, such as
creating routing loops, forwarding packets through non-
optimal paths, or selectively dropping packets.
II. “Location disclosure” attackers disclose the security-sensitive
location information of nodes or the topology of the network.
III. “Resource consumption” or so-called “Sleep deprivation”
attackers try to waste the power of the legitimate nodes by
requesting excessive route discovery, forwarding useless
packets to the victim node, or endlessly “dangling” useless
packets between two distant attackers.
10/24/20
23
54

55. Routing Attacks in MANET
(continue..)
10/24/20
23
55
IV. “Rushing” attackers have more power and quicker links
than legitimate nodes. They may forward the RREQ and
RREP faster. By this way, they are always involved in the
routes.
Rushing Attack
V. “Selfish” nodes use the network but do not cooperate. They
save the battery life, CPU cycles, and other resources for
their own packets. Though they do not intend to directly
damage other nodes, the result is less damaging.

56. Routing Attacks in MANET
(continue..)
VI. “Spoofing” attackers impersonate a legitimate
node to misrepresent the network topology to
cause network loops or partitions.
VII. The Wormhole attack is a kind of tunnelling
attack which is extremely dangerous and
damaging to defend against even though the
routing information is confidential, authenticated
or encrypted.
 Routing may be disrupted by tunneled routing
control messages.
 Wormhole attacks are severe threats to MANET
on-demand routing protocols.
10/24/20
23
56