【摘要】 针对感知节点能量和汇报信道带宽受限的认知无线电网络中协作频谱感知问题,提出了一种基于距离准则的优化量化器设计方法。首先,计算融合中心接收的量化数据的巴氏距离(BD, Bhattacharyya distance)为性能准则,构建量化器的优化数学模型,采用粒子群优化算法求解得出最优量化阈值。根据融合中心接收的各感知节点的量化数据,构造对数似然比检测器,对是否存在主用户信号做出决策,最后推导了未量化条件下能量检测器的性能上界。仿真实验结果与已有方法对比,所提出的3 bit量化方法的性能接近能量检测器的性能上界,在获得类似检测性能的前提下降低了对通信带宽的需求。更多还原

【Abstract】 In terms of sensing node’s energy and reporting channel’s bandwidth constrains problem for cooperative spectrum sensing in cognitive radio networks,an optimal quantizer design method based on distance criterion was proposed.First of all,the Bhattacharyya distance of received quantized data at the fusion center(FC) was calculated as performance criteria,the optimization mathematical model of the quantizer was constructed,and the optimum quantization thresholds were obtained by using particle swarm optimization algorithm.According to received sensing nodes’ quantized data at the FC,a log-likelihood ratio detector was constructed to decide the presence or absence of primary user signal,the upper bound to sensing performance of energy detector that without quantization was derived.Compared with the existing methods in literatures,the performance of proposed 3-bit quantization method approaches to the upper bound performance of energy detector,under the premise of obtaining comparable detection performance,the requirement of communication bandwidth is reduced.更多还原