SOFROP: Self-Organizing and Fair Routing Protocol for Wireless Networks with MobileSensors and Stationary Actors

1. SOFROP: Self-Organizing and Fair Routing
Protocol for Wireless Networks with Mobile
Sensors and Stationary Actors
Mustafa I. Akbas1, Matthias R. Brust2 and Damla Turgut1
1University of Central Florida
2Technological Institute of Aeronautics, Brazil
LCN 2010
October XX, 2010

2. Environment Monitoring: Amazon
Scenario
 Sensor nodes
 Thrown in the river
 Local communication
 No geographical
information
 Actors
 At rare accessible points
 Full communication range
used only in
 Backbone formation
 Actor-actor data
exchange

3. The problem definition
 Challenges
 Rapid changes in neighborhood and actor associations
 Fast and reliable communication
1. Adaptive network organization
2. Fair usage of bandwidth/data transmission
 Objective
 Develop a fair and self-organizing routing protocol for
Amazon scenario

4. Network organization
 Clustering to build an
overlay network
 Multi-hop clusters
 Actors
 Pre-assigned
clusterheads
 Node weight
 Actor weight: 0
 Weight for each
node: 1 to k
 Weight is “min in
neighborhood + 1”
 Local neighbor
information only

5. Network organization: Illustrations

6. Data Transmission
 Sensor nodes:
 Interest table
 “on” and “off”
interests
 Number of “on”
interests N i
 Max. packet transmission
rate (C o )
 No global state
 Packets encoded with
the rate and interest

7. Data Transmission – cont’d
 For each sensor node: Fair rate: f Co / N i
 If output capacity is not exceeded
 No drop
 Cs : Capacity shared by flows with rates > f
 Probability to drop a packet: Pd 1 C s /( N s p )
Cs i
p Ns
 The packets are encoded with rate:
j
j 0

8. Data transmission – Illustration 1

9. Data Transmission – Illustration 2

10. Simulation Study
 OPNET modeler
 Network
 Interest area: 200x300 m
 Number of sensor nodes: 60
 Sensor transmission range = 40 m
 Traffic type: CBR
 Mobile nodes:
 Starting, destination, pausing points
 Pausing time: 0-10 sec
 Speed: 0-3 m/sec
 MAC attributes: OPNET 802.11

11. Number of Received Packets - 1
50% type-1, 30% type-2 , 20% type-3

12. Number of Received Packets - 2
50 % type-1, 45 % type-2 and 5% type-3

13. Utilization

14. Delay

15. Number of sensor nodes – Hop values

16. Conclusions
 Focus is Amazon scenario
 SOFROP builds a structured network topology that
permanently adapts according to the river dynamics
 SOFROP provides fairness among different types of
applications and it is efficient in utilization of the
bandwidth
 Future work:
 Adaptation to animal monitoring
 Data aggregation
 Other performance metrics (e.g. power conservation)