一种基于分数低阶矩的α稳定分布噪声中频谱感知方案

An FLOM-based Spectrum Sensing Scheme in α-Stable Distributed Noise

针对在非高斯背景、主用户信息未知的条件下,传统的基于二阶统计量的频谱感知方法性能将出现退化或失效的问题,建立了以α稳定分布为背景噪声的频谱感知模型,给出了一种基于分数低阶矩的感知方法,较好地解决了非高斯背景中主用户先验信息未知条件下的频谱感知问题。同时,根据中心极限定理推导了感知门限与虚警概率的关系式,通过蒙特卡洛仿真分析了该算法在不同广义信噪比、特征指数α以及协作用户数条件下的感知性能,并与传统的感知方法进行比较。仿真结果表明,基于分数低阶矩的感知方法在α稳定分布背景噪声中的感知性能明显优于基于二阶统计量的能量检测,且采用多用户协作可以进一步提高感知性能。

The traditional spectrum sensing methods based on second order statistic are in general not applicable to detecting the primary user in non-Gaussian noise with unknown parameters.This paper presents a fractional lower order moment （FLOM）-based spectrum sensing scheme for the background noise with α-stable distribution.The new method does not need any priori knowledges about the primary user signal.The relationship between the sensing threshold and the probability of false alarm is derived by using the central limit theorem.The detection performances of the proposed method versus the generalized signal-to-noise ratio,dispersion α and the number of cooperative users are analyzed by Monte Carlo simulations compared with the traditional methods.Simulation results show that the proposed method has a much better performance than the energy detection method in the α-stable distributed noise environment.It is also shown that using multi-user cooperation can provide a significantly higher probability of detection than that of the single user version.