一种基于似然比准则的频谱感知算法

Spectrum detection algorithm based on likelihood ratio decision criterion

为了改善低信噪比环境下无线认知网络频谱检测性能,提出了一种基于似然比准则的频谱感知算法。该算法以主用户信号的循环谱峰值群为检验统计量,以虚警概率和检测概率加权组合为频谱检测性能的主要指标,构建最佳频谱检测性能下的似然比准则。由于该算法不需要主用户信号及其传输信道的任何先验知识,克服了噪声功率不确定性对频谱检测性能的影响,解决了无线认知网络在低信噪比以及噪声波动环境下的频谱检测问题。该算法可通过调节虚警概率和检测概率的加权因子以便获得频谱检测最佳性能。与基于似然比拟合优度的频谱检测算法相比,频谱检测性能改善了5 dB左右。

To improve the performance of spectrum detection in wireless cognitive networks with low signal to noise ratio(SNR),a spectrum sensing algorithm based on likelihood ratio criterion is proposed.Taking the cyclic spectral peak group of the main user signal as the test statistic and the weighted combination of the false alarm probability and the detection probability as the main index of the spectrum detection performance,the likelihood ratio criterion on the optimal spectrum detection performance is constructed by the algorithm.Since the algorithm does not need any prior knowledge of the main user signal and its transmission channel,thus overcoming the influence of noise power uncertainty on the spectrum detection performance and solving the spectrum detection problem of wireless cognitive network in the environment of low SNR and noise fluctuation.The algorithm can achieve the best performance of spectrum sensing by adjusting the weighted factors of the false alarm probability and the detection probability.Compared with the spectrum detection algorithm based on likelihood match fitness,the spectrum detection performance of the algorithm is improved by about 5 dB.