GA with SVM to Optimize the Dynamic Channel Assignment for Enhancing SIR in Cellular Networks

There is a reduction in the signal-to-noise ratio of cellular networks due to interference caused by assigning the channels to the cell. As the demand for connectivity is on rise with limited spectrum availability, the interference may increase, so channels are required to be assigned optimally. This work presents applying Genetic algorithm (GA) along with Support Vector Machine (SVM) to assigning the channels dynamically for reducing co-channel and co-site interference with constraints. In this paper, we propose to adopt the GA to solve the minimum interference channel assignment problem (MICAP) and the nonlinear dataset are best classified using SVM. The fitness function is designed using SVM and the optimization is done with GA with a focus on MICAP. The performance of the GA-SVM is enhanced SIR, reduces interference, and requires less computation time than the work reported by GA.