混沌免疫多目标算法求解认知引擎参数优化问题

合理的认知引擎参数设置可以提高频谱的使用性能. 通过分析认知无线网络中的认知引擎参数配置, 给出了其数学模型, 并将其转化为一个多目标优化问题, 进而提出一种基于混沌免疫多目标优化的求解方法. 算法使用Logistic混沌映射初始化种群, 并在每一代将混沌特性用于最优解集的搜索; 设计了适合此问题的免疫克隆算子和抗体群更新算子, 保证了Pateto最优解集分布的多样性和均匀性. 最后, 在多载波环境下对算法进行了仿真实验. 结果表明, 算法可以根据信道条件和用户服务的动态变化, 自适应调整各个子载波的发射功率和调制方式, 可以求出更多满足偏好需求的解, 满足认知引擎参数优化要求.

Parameter optimization of cognitive engine based on chaos multi-objective immune algorithm

Reasonable setting of engine parameters can improve the performance of the spectrum use. By analyzing engine parameter adjustment of cognitive wireless network, the mathematical model is given, and then it is converted into a multi-objective optimization problem. A chaos multi-objective immune algorithm is proposed to solve the problem. Logistic mapping is used to initialize population and search for the best solutions in every generation. The operators of cloning and antibodies updating are designed, which ensures that the distribution of Pateto optimal solutions is more diverse and uniform. Finally, the simulation experiments are done to test the algorithm under multi-carrier system. The results show that the algorithm can adjust transmission power and modulation mode according to the change of channel and user demands. More solutions with preferences are obtained, so it meets the demands for parameter optimization of cognitive engine.