Optimizing the Parameters of Relay Selection Model in OLSRv2 D2D Network

The Fifth generation (5G) cellular network's traffic load is certain to expand significantly in the near future as a result of its flexibility, high speed, increased bandwidth, better connectivity and low latency. Consequently, it is necessary to investigate the application of technologies that may be able to reduce the amount of the traffic on the core network. One such technique is D2D communication in 5G cellular network, which allows two or more devices to communicate with each other directly without transversing the cellular network. However, deployment of D2D communication required an effective relay and routing technique to off load the traffic of the network. On the other hand, Multipoint Relay (MPR) nodes in OLSRv2 experiences quick power failure and frequent disconnection due to selection of unsuitable MPR node. To solve this problem, this research employed Linear Programming (LP) technique to determine the optimum weight combination of four parameters in OLSRv2 relay selection model in D2D communication scenario. The four parameters are battery level, mobility speed, node degree and connection to base station. These parameters have been modelled based on their influences in maintaining D2D connectivity. However, to ensure each parameter contributes fairly to the OLSRv2 relay selection, the proposed model works out the weights of the parameters in the selection process. In addition, the proposed model is a general one and can handle
$n$
number of parameters. Meanwhile, to determine the values of Wi for each parameter i, we formulated the problem as an LP model to obtain the optimum contribution for each of the parameters considered in our system for the selection of the optimum node. The results showed that application of LP has led to optimum weight allocation to the four parameters thereby leading to efficient OLSRv2 relay selection in D2D communication.