New NSGA-II-based OLSR self-organized routing protocol for mobile ad hoc networks

The problem of routing in ad hoc networks, in particular proactive routing, has attracted the attention of many researchers. Although the protocols proposed in the literature present some relevant characteristics, they also have their limitations, especially in terms of high number of mobile nodes or in terms of high number of load-dependent parameters of the ad hoc network which are often chosen intuitively by a seasoned expert. This paper describes our work to solve this difficult task using a multi-objective genetic algorithm to automate the selection process of the routing protocol parameters. The realized experiments showed the effectiveness of the proposed NGSA-II-OLSR compared to the original OLSR. In case of low node mobility, the proposed NSGA-II-OLSR improves the PLR between 8.59 and 33.17%; the E2ED between 18.17 and 27.56%; and the NRL between 35.18 and 36.60%. While in case of high mobility node, it improves the PLR between 3.47 and 9.94%; the E2ED between 1.47 and 9.40%; and the NRL between 0.14 and 2.34%. In addition, the algorithm can adapt the ad hoc network to each topology change which makes it adaptive to any environment changing.