Efficient Spectrum Sensing in Cognitive Radio Network Using Multi-Objective Improved Salp Swarm Algorithm

The Cognitive Radio (CR) has been established as an emerging technique which enables underused or unused spectrum for dynamically enhance the spectral effectively. To enhance the CR performance which is extensively significant to reconfigure or adapt the model parameters. The Multi-Objective Improved Salp Swarm Algorithm (MOISSA) is proposed for spectrum sensing in CR networks. The objective functions such as throughput, energy efficiency and interference are formulated to define the spectrum sensing performance in CR networks. In this paper, the parametric alteration of CR is accomplished through SLSSA which exploits the areas near to individual location to attain its global optimum. The SSA is used to avoid local optima and define the suitable optimal solution evaluation in whole optimization process. In traditional SSA, the Self Learning (SL) rule is applied to provides every learner opportunity to enhance its individual experience through search. The performance of MOISSA is estimated by various metrics like delay, Packet Delivery Ratio (PDR), energy consumption and network lifetime are used with 20, 40, 60, 80 and 100 nodes. The MOISSA attains delay of 13.4s, PDR of 59.7%, 8.55J of energy consumption and 27.5 of network lifetime for 100 nodes which is better when compared to existing method such as Improved Whale Optimization Algorithm with Convolutional Neural Network (IWOA-CNN).