![Comparative Evaluation Of Swarm
Intelligence Based Routing Protocols
The swarm intelligence based routing protocols considered in our comparative
study are –
IABR[3], EEABR[3], IEEABR[4], SC[5]
ACO based Energy Aware Routing[6], Sensor Ant[7], MACS[8], SEB[9]
ACA for Data Aggregation[10], Ant-Aggregation[11]
Basic-DAACA[12], ES-DAACA[12]
ACS-DAACA[12], SIBER XLP[2]
SIBER-DELTA[13].](https://image.slidesharecdn.com/designconsiderationandcomparativeevaluationofswarmintelligence-181119183157/85/Design-consideration-and-comparative-evaluation-of-swarm-intelligence-10-320.jpg)
Design consideration and comparative evaluation of swarm intelligence
- 1. Design Consideration and Comparative Evaluation of Swarm Intelligence based Approaches for Efficient Routing in Wireless Sensor Networks PRESENTED BY : PALAKURTHI SHASHANK
- 2. Presentation Outlines Introduction Routing Challenges and Design Issues in WSNs Comparative Evaluation Of Swarm Intelligence Based Routing Protocols Conclusion References
- 3. SWARM Swarm : Swarm is a collection of agents interacting locally with one another and with their environment. Swarm Intelligence : “Any attempt to design algorithms or distributed problem-solving devices inspired by the collective behaviour of social insect colonies and other animal societies “ Computer scientists are increasing interested in swarm intelligence since it can be used to solve many optimization problems.
- 4. What is Wireless Sensor Network ? A Wireless Sensor Network (WSN) is a collection of compact size low power computational nodes capable of detecting local environmental conditions and forward such information to a central base station (sink) for appropriate processing . WSN FEATURES : Large Network Size Self-organised All nodes acts as routers No wired infrastructure
- 5. The Difference between WSN &Ad-hoc The number of nodes The topology of a sensor network changes very frequently WSN broadcast but ad hoc point-to point Sensor node are limited in power computation capacities and memory Sensor nodes may not have global identification
- 6. Routing Challenges and Design Issues in WSNs The design of routing protocols in WSNs is influenced by many challenging factors. These factors must be overcome before efficient communication can be achieved in WSNs. Node deployment Data delivery model Node/link heterogeneity Scalability Transmission media WSN RESOURCE CONSTRAINTS : Most of the wireless sensor networks are composed of sensor motes equipped with a 8-bit microcontroller running at 8/16MHz and having memory comprised of 4KB of RAM, 4KB of EEPROM and 128 KB of flash.
- 7. Large Scale Network & Scalability The routing protocol must be able to handle large and dense networks It is to be noted that the expansion of the network scale results in drastic increase in the number of possible paths from source to sink Routing protocols should be designed to exhibit scalable performance.
- 8. Design & Implementation Issues in WSN Routing Data Aggregation Solves implosion and overlap problem Energy efficient Traffic Patterns – Data Delivery Models The case of MANETs traffic patterns strictly depend on the application and are address centric. Common data delivery models present in WSNs include - event- driven, query driven, continuous monitoring . Self-Organization A routing protocol must be resilient to such dynamic and generally unpredictable variations The failure of sensor nodes such as blockage due to lack of power
- 9. Need for Bioinspired Approaches The network protocols and communication techniques must be designed to provide robust, reliable and highly efficient wireless sensor networks It is evident that the adaptiveness, flexibility, robustness characteristics exhibited in coordination of their behaviors have made them capable of solving real world problems. Self-organization may help in distributed control and management tasks.
- 10. Comparative Evaluation Of Swarm Intelligence Based Routing Protocols The swarm intelligence based routing protocols considered in our comparative study are – IABR[3], EEABR[3], IEEABR[4], SC[5] ACO based Energy Aware Routing[6], Sensor Ant[7], MACS[8], SEB[9] ACA for Data Aggregation[10], Ant-Aggregation[11] Basic-DAACA[12], ES-DAACA[12] ACS-DAACA[12], SIBER XLP[2] SIBER-DELTA[13].
- 11. PARAMETERS USED IN ROUTING DECISION & PERFORMANCE EVALUATION METRICS USED Routing Protocol Parameters used in the Routing Decision Performance Metrics BABR D ARE,ME,SD,EE IABR E ARE,ME,SD,EE EEABR D,E ARE,ME,SD,EE IEEABR D,E EC,EE,SD,L,SR SC D L,SR,EC,EE ACO based Energy Aware Routing E EC, NL SensorAnt E, QR EC,EE MACS D TEC,ATD SEB D,E NL, RSD ACA for Data Aggregation D TEC, AEC, NL,TO,TTAD Ant-Aggregation D OC,EE Basic-DAACA D,E AET,EDNL,SSR ES-DAACA E AET,EDNL,SSR MM-DAACA E AET,EDNL,SSR ACS-DAACA E AET,EDNL,SSR SIBER-XLP D,E,LQ PDR,EE, L, ARE, ME, SD SIBER-DELTA D,E,LQ,TR PDR,EE, L, ARE, ME, SD Parameters Used Legend : D-Distance, E- Energy, QR- Quality of Route, LQ- Link Quality, TR-Trust Rating Performance Metrics Legend: ARE- Average Minimum Energy, ME- Minimum Energy, SD- Standard Deviation, EE- Energy Efficiency, L- Latency, EC-Energy Consumption, SR-Success Rate/Ratio, NL-Network Lifetime, TEC-Total Energy Consumption, ATD-Average Transmission Delay, RSD-Relative Standard Deviation, ED-Energy difference, AET-Average Energy cost, PDR-Packet delivery Ratio
- 12. Comparison of simulation area, initial energy and number of nodes used in the simulation Number of nodes deployed in the network of given simulation area with specified initial energy on the nodes used by different protocols Deploying network of node size of 10 to 2000 nodes varies from 40x50 to 2000x1000 sq mts . Initial energy on the nodes used by different protocols vary from minimum value of 0.25J to maximum value of 1000J.
- 13. Initial energy and number of nodes used in the simulation Ant- Aggregatio n A12,A13, A14,A15,A16 Not Specified 50 Basic-DAACA A1,A2,A3,A4,A5,A6, A7,A8,A9,A10,A11 10J 200-2000 ES-DAACA A1,A2,A3,A4,A5,A6, A7,A8,A9,A10,A11 10J 200-2000 MM-DAACA A1,A2,A3,A4,A5,A6, A7,A8,A9,A10,A11 10J 200-2000 ACS-DAACA A1,A2,A3,A4,A5,A6, A7,A8,A9,A10,A11 10J 200-2000 SIBER-XLP A14,A16,A17,A18 30J 25,49,64, 100 SIBER-DELTA A16, A18 30J 50,100 Routing Protocol Simulation Area sq mts Initial Energy Number of BABR A12,A13, A14,A15,A16 50J(S),30J(D), 20J(M) 10-100 IABR A12,A13, A14,A15,A16 50J(S),30J(D), 20J(M) 10-100 EEABR A12,A13, A14,A15,A16 50J(S),30J(D), 20J(M) 10-100 IEEABR A11 30J(S),60J(D) 49 SC A11 3N 49 ACO based Energy Aware Routing A12 40J 50,100,200, 300,500 SensorAnt A12,A13, A14,A15,A16 1250mJoule 10-100 MACS A12 1000J 50,100,200, 300,500 SEB A19,A20,A21,A22 80-100units 50,100,150, 200 ACA for Data Aggregation A6 0.25J 300-500 Simulation Area Legend : A1- 40 × 50, A2-60 × 60, A3-70 × 80, A5-80 × 90, A6-100 × 100, A7- 100 × 120, A8- 100 × 140, A9- 120 × 130, A10-130 × 120, A11-140 × 140, A12-200 X 200, A13- 300 X 300, A14-400 X 400, A15- 500X500, A16- 600X600, A17- 700X700, A18-900X900, A19-1000X500, A20-1000X1000, A21-500X1000, A22-2000X1000 Initial Energy Legend : S-static, D-Dynamic, M- Mesh Networks, N- number of nodes
- 14. Simulation Tools and Control parameters used Tools like NS-2, QualNet , Prowler , Matlab , J-SIM are used by researchers to evaluate different routing protocols. Routing Protocol Simulation Tool Simulation Control Parameters BABR NS-2 α, β, ƿ IABR NS-2 α, β, ƿ EEABR NS-2 α, β, ƿ , IEEABR Prowler α, β, ƿ SC Prowler ACO based Energy Aware routing MATLAB SensorAnt QualNet V5 β , 𝜔 MACS NS-2 α=0.9, β=2, q0=0.45, ƿ=0.95 ,m =30 SEB J-SIM α, β, γ ACA for Data Aggregation NS-2 β=20, ƿ=0.3, ξ=1 Ant-Aggreagation MATLAB Q=8, ƿ=0.8, α=3, β=6, T=0.5, H=0.4, K=0.1 Basic-DAACA NS-2 α =2,β=2,initial pheromone=0.8, ηmin=0.5, ηmax=0.9, ƿ=0.2, roundToUpdate=100,200,ξ=0.9, q0=0.5, rho in ACA=0.3 ES- DAACA NS-2 α=2,β α α =2,initial pheromone=0.8, ηmin=0.5, ηmax=0.9, ƿ=0.2, roundToUpdate=100,200 ξ=0.9, q0=0.5, rho in ACA=0.3 MM-DAACA NS-2 α=2,β=2,initial pheromone=0.8, ηmin=0.5, ηmax=0.9, ƿ=0.2, roundToUpdate=100,200 ξ=0.9, q0=0.5, rho in ACA=0.3 ACS-DAACA NS-2 α=2,β=2,initial pheromone=0.8, ηmin=0.5, ηmax=0.9, ƿ=0.2, roundToUpdate=100,200 ξ=0.9, q0=0.5, rho in ACA=0.3 SIBER-XLP NS-2 α=1, β =1, γ =1, ƿ=0.2 SIBER-DELTA NS-2 α=2, β =2, γ =1, δ=1, ƿ=0.2
- 15. Probability and Pheromone Update Functions One or combinations of more parameters such as distance, energy, link probability and trust ratings are considered by different routing protocols to find the optimal route to the destination from source node. Protocol uses its own pheromone update scheme to increase the pheromone on the path so as to select the best path to the destination.
- 16. PROBILITY FUNCTIONS USED TO SELECT NEXT NODE
- 17. PHEROMONE UPDATE FUNCTIONS USED
- 18. Conclusion This paper provides the comparative study carried out to evaluate various Swarm Intelligence based approaches reported in literature for efficient routing in wireless sensor network. Main objective of this work is to perform detailed analysis of these algorithms to determine their advantages and limitations to provide guidelines to design a robust Design and implementation issues in the context of WSN deployed environment and networking conditions based routing protocols for wireless sensor networks.
- 19. THANK YOU !