A novel method for the optimization of Spectral -Energy efficiency tradeoff in 5 G heterogeneous Cognitive Radio Network

Abstract

Cognitve Radio (CR) technology has recently become popular for future communication networks since it allows co-existence of Secondary User (SU) and Primary User (PU) networks. Compared to interweave and overlay, underlay mode of operation is best suited for practical scenarios since it facilitates PU-SU co-existence with minimum overhead. Aiming to enhance Spectral Efficiency (SE) alone using CR may lead to huge power consumption and thus increase the overall cost of network operation. Hence, effort must be taken to maintain a tradeoff between SE and Energy Efficiency (EE). In this work, a novel technique, called Superior Population Generation Algorithm (SPGA), is developed to solve the crucial problem of SE-EE tradeoff for a Fifth generation (5 G) heterogeneous network which uses key future technologies such as CR and Carrier Aggregation (CA). The proposed technique performs optimum resource allocation, by modifying regular Genetic Algorithm (GA) to incorporate the capabilities of conventional techniques. SPGA follows a multi objective optimization approach towards obtaining SE-EE tradeoff, rather than focusing on SE alone. Simulation results show that SPGA outperforms conventional methods and regular GA in terms of SE-EE tradeoff with about 5% and 10% better EE for the same SE respectively. It is also observed that SPGA converges to the Pareto optimal solutions, much faster using only 25% of iterations required for regular GA. All simulations in this work are performed according to 3GPP specifications for 5 G communication networks. Also, power consumption of various base station components is also considered in addition to the transmit power to make the EE calculations further close to reality.