Smart data packet ad hoc routing protocol

Abstract

This paper introduces a smart data packet routing protocol (SMART) based on swarm technology for mobile ad hoc networks. The main challenge facing a routing protocol is to cope with the dynamic environment of mobile ad hoc networks. The problem of finding best route between communication end points in such networks is an NP problem. Swarm algorithm is one of the methods used solve such a problem. However, copping with the dynamic environment will demand the use of a lot of training iterations. We present a new infrastructure where data packets are smart enough to guide themselves through best available route in the network. This approach uses distributed swarm learning approach which will minimize convergence time by using smart data packets. This will decrease the number of control packets in the network as well as it provides continues learning which in turn provides better reaction to changes in the network environment. The learning information is distributed throughout the nodes of the network. This information can be used and updated by successive packets in order to maintain and find better routes. This protocol is a hybrid Ant Colony Optimization (ACO) and river formation dynamics (RFD) swarm algorithms protocol. ACO is used to set up multi-path routes to destination at the initialization, while RFD mainly used as a base algorithm for the routing protocol. RFD offers many advantages toward implementing this approach. The main two reasons of using RFD are the small amount of information that required to be added to the packets (12 bytes in our approach) and the main idea of the RFD algorithm which is based on one kind of agent called drop that moves from source to destination only. This will eliminate the need of feedback packets to update the network and offers a suitable solution to change data packet into smart packets. Simulation results shows improvement in the throughput and reduction in end to end delay and jitter compared to AODV and AntHocNet protocols. © 2013 Elsevier B.V. All rights reserved.