基于变分稀疏贝叶斯学习的频谱检测方法

为了降低对宽带信号进行压缩频谱感知的复杂度,提出了一种基于变分稀疏贝叶斯学习的频谱检测方法.该算法直接利用压缩测量值对授权用户的位置、个数以及功率传播图进行了估计,在先验知识未知的情况下,利用变分稀疏贝叶斯求解稀疏权值.而且用简单函数因子逼近的方法降低了边缘似然函数的计算难度.实验结果表明:该方法在感知精度和速度上有显著提高.

Spectrum Detection Based on Variational Sparse Bayesian Learning

A novel spectrum detection algorithm based on variational sparse bayesian learning is proposed for reducing the complexity of the compressed spectrum sensing for wideband.The algorithm directly uses the compressed measurement to estimate the location and number of the primary users and adopts variational sparse learning to obtain the sparse weights although the priori knowledge is unknown. It reduces the computational difficulty of marginal likelihood function by adopting the approximation of simple factorial function.The experimental results show that the algorithm significantly improves the accuracy and speed of sensing.