中文摘要：

分析电力线通信系统在各种约束下,多用户多业务在多子载波上自适应资源分配的多层多目标模型.基于快速的非支配分类遗传算法Ⅱ,提出改进的功率或速率自适应的资源分配多目标和单目标优化遗传算法。在使用多目标遗传算法获得实时用户资源分配的所有Pareto非劣解后,由每个Pareto解计算系统的剩余资源,再采用单目标遗传算法把剩余资源分给非实时用户,最后从所有的资源分配方案中寻找全局最优方案.在典型电力线信道环境下仿真结果表明,本文算法其性能更好且能更好地满足多用户资源分配的多目标要求.

英文摘要：

The adaptive resource allocation＇s multilayer multiobjective model for multiuser multiserver power-line communication systems is analyzed with the various restrictions.Based on the fast nondominated sorting genetic algorithm II,an improved resource allocation genetic algorithm with rate or power adaptation for single and multiple objective optimization is proposed,which firstly assigns resource for RT user by using the improved multiple objective genetic algorithm so as to obtain all Pareto nondominated solutions,secondly computes the system＇s remainder resource for every Pareto solution and assigns them for NRT user by using the improved single objective genetic algorithm,lastly searches the global optimal scheme from all resource allocation schemes.The algorithm is tested in the typical power-line channel environment and the simulation results indicate that it has better performances and satisfies betterly the multiple aims of multiuser resource allocation.