دانلود رایگان مقاله تخصیص انرژی در شبکه رادیو شناختی با بهینه سازی ذرات ازدحام بی نظم هم تکاملی

Abstract

In this paper, the trade-off between utility and energy consumption in orthogonal frequency division multiplexing (OFDM)-based cognitive radio (CR) network is investigated. Energy efficiency problem is very important in the field of CR network, where the utility is maximized and the energy consumption is minimized in such a CR network. Since the trade-off between them has been paying more attentions in literature, this study summarizes the power allocation as an optimization problem that maximizes the energy efficiency via a new energy efficiency metric defined by this paper. The formulated problem is a large-scale nonconvex problem, which is very difficult to solve. In this paper, we present an improved particle swarm optimization (PSO) algorithm to solve the difficult large-scale optimization problem directly. Given the weak convergence of the original PSO around local optima, an improved version that combines the chaos theory is proposed in this study, where chaos theory can help PSO search for solutions around the personal and global bests. In addition, for the purpose of accelerating the convergence process when facing with such a large-scale optimization, the original problem is decomposed into a number of small ones by employing the coevolutionary methodology, and then divide-and-conquer strategy is used to avoid producing infeasible solutions. Simulations demonstrate that the proposed coevolution chaotic PSO needs a smaller number of iterations and can achieve more energy efficiency than the other algorithms.

6. Conclusion

In this paper, a new metric that reflects the trade-off between the utility and energy consumption is defined in CR networks. Since the optimization problem is a large scale and nonconvex one, our proposed algorithm exploits the coevolutionary and chaotic ideas for the dynamic power allocation problem in CR networks. Strict assump- tions such as continuity, differentiability, and convexity of the objective function are not necessary. The formu-lated optimization problem is solved by using max–min approach, where two populations of PSO are included. The performance of the proposed algorithm is compared with those of related methods in the literature. It is observed that the proposed algorithm is indeed capable of quickly achieving energy-efficient solutions. Future research topics may include dynamic PSO algorithms for cooperative CR networks in which one SU may help relay other SUs’ signal to the secondary BS such that cooperative diversity can be achieved. Then each SU may need to distribute its power budget in transmitting its own signal and in relaying other SUs’ signals. The problem is much more complex, and deserves further investigation.