基于神经网络的感知无线电学习与评估研究*

为了得到感知无线电最优决策，对感知循环、人工智能技术应用、基于遗传算法的感知引擎进行了分析。感知无线电决策机应该既考虑带宽、信号速率、功率等可变因素，又考虑费用等不可变因素。基于神经网络，提出了感知无线电决策机学习与评估模型，并研究、讨论了感知无线电知识库信息与决策机设计思路。

Research on learning and evaluation of CR based on neural network

This paper introduced the research of cognitive engine and application of artificial intelligence techniques in cognitive radio.The limitation of CR engine based on GA was analyzed,propose for improvement was proposed.The decision maker of CR engine should consider both the changeable and the unchangeable factors such as cost,bandwidth,signal rate and ARQ.Based on Neural Network,the method of evaluating and learning best decision was proposed.Several key architectural issues for cognitive radio engine based on Neural Network were discussed,including knowledge base information model and learning model Neural Network design.