quotation:		[Copy]
		[Copy]

This Paper:Browse 3941 Download 151
卫星稀疏信道中频谱空穴的新型检测算法
陈鹏,徐烽,邱乐德,王宇
0 Fontlarge +\|Default\|Small
(中国空间技术研究院西安分院，西安 710100)

摘要:

面对宽带移动卫星通信中信道占用呈现的稀疏状态和时而存在的稀疏干扰，以低速率采样和运算是简化设备的必要途径。针对宽带频谱范围内稀疏信道信号和干扰信号的检测，提出了一种用于直接估计频谱空穴的压缩采样算法，以低速率对信号进行采样，通过相关向量机进行迭代优化，给出检测估计值。该算法基于最大似然贝叶斯估计，具有良好的抗噪性能，同时能够降低设备复杂度。仿真结果表明,其抗噪性能明显好于CVX方法而略逊于理想情况，而前端随机序列的相关性是影响检测性能的重要因素。

关键词: 卫星通信认知无线电频谱空穴压缩采样贝叶斯检测相关向量机

DOI：

基金项目:

A new scheme for detecting spectrum holes of sparse satellite channel

CHEN Peng,XU Feng,QIU Le-de,WANG Yu

()

Abstract:

Broadband satellite systems encounter occasionally sparse channel occu pations and sparse interferences,so lowrate sampling and oper a tion is required. For detection of sparse signals and interferences among a wide spectrum range,a compressive sampling(CS) scheme is proposed that can detect sp ectr um holes directly. Iterative optimization by a Relevance Vector Machine(RVM) is appli ed on subNyquist rate samples to produce the estimate. This scheme is based on Maximum Likelihood(ML) Bayesian estimation, and features good antinoise perfor mance and effective d evice complexity reduction. Simulation results show a better antinoise perform a nce over CVX method. And the relevance of the frontend pseudorandom sequences is shown to have a large impact on the detection performance.

Key words: satellite communication cognitive radio spectrum hole compressive sampling Ba ye sian detection relevance vector machine