CAJ | 학술논문

针对现有的跨层资源分配算法计算量大、复杂度较高的问题，提出了一种基于遗传算法的OFDMA系统的跨层资源分配算法。利用遗传算法隐形并行处理、较好的全局搜索性能、易收敛到最优解的特点，在系统性能一定，且满足各个用户业务要求的条件下，对资源进行优化分配，较好地解决了跨层资源的分配问题。仿真结果表明，在满足用户间公平性的前提下，算法有效提高了系统的频谱利用率及吞吐量，减小了用户的平均等待时延，提高了服务质量，并且随着子载波数和用户数的增加，算法在复杂度方面优势更突出。
침대현유적과층자원분배산법계산량대、복잡도교고적문제，제출료일충기우유전산법적OFDMA계통적과층자원분배산법。이용유전산법은형병행처리、교호적전국수색성능、역수렴도최우해적특점，재계통성능일정，차만족각개용호업무요구적조건하，대자원진행우화분배，교호지해결료과층자원적분배문제。방진결과표명，재만족용호간공평성적전제하，산법유효제고료계통적빈보이용솔급탄토량，감소료용호적평균등대시연，제고료복무질량，병차수착자재파수화용호수적증가，산법재복잡도방면우세경돌출。
According to the amount and complexity of computation for cross-layer resource allocation algorithm, an adap- tive cross-layer resource allocation scheme employing Genetic algorithm (GA) in muhiuser OFDMA system is proposed. A better solution to the cross-layer resource allocation satisfying system performances and users' requirements is presented based on the GA characteristics such as parallel processing, comprehensive ability of global search and easy convergence to the optimum. Under the prerequisite of fairness between users met, an improved algorithm is proposed to enhance the availability of frequency spectrum and also shorter the user's average wait delay. The experiment simulation result shows that the method proposed effectively improves cross-layer resource allocation for OFDMA System, and with the increase of Carrier number and the number of users, it has more advantage in complexity of computation.