This paper presents the implementation of a modified version of Bayesian relevance vector machine (RVM)‐based compressive sensing method on cognitive radio network with wavelet transform for spectrum hole detection. Bayesian compressive sensing is used in this work to deal with the complexity and uncertainty of the process. The dependency of the Bayesian compressive sensing on the knowledge of noise levels in the measurement has been relaxed through the proposed Bayesian RVM‐based compressive sensing algorithm. This technique recovers the wideband signals even with fewer measurements maintaining considerably good accuracy and speed. Wavelet transform is used in this paper to enable the detection of primary user (PU) even in the low regulated transmission from unlicensed user. The advantage of this approach lies in the fact that it enables the evaluation of all possible hypotheses simultaneously in the global optimization framework. Simulation study is performed to evaluate the efficacy of the proposed technique over the cognitive radio environment. The performance of the proposed technique is compared with the conventional Bayesian approach on the basis of recovery error, recovery time and covariance to verify its superiority.

中文翻译：

---

基于小波变换的基于改进贝叶斯算法的压缩采样用于认知无线电网络中的宽带频谱感知

本文提出了一种基于小波变换的认知无线电网络上基于贝叶斯相关向量机（RVM）的压缩感知的改进版本的实现，该小波变换用于频谱空洞检测。在这项工作中使用贝叶斯压缩感测来处理过程的复杂性和不确定性。通过提出的基于贝叶斯RVM的压缩感测算法，放宽了贝叶斯压缩感测对测量中的噪声水平的依赖性。即使使用较少的测量值，该技术也可以恢复宽带信号，从而保持相当高的准确性和速度。本文使用小波变换来检测主要用户（PU），即使在来自未授权用户的低管制传输中也是如此。这种方法的优点在于，它可以在全局优化框架中同时评估所有可能的假设。进行仿真研究以评估所提出技术在认知无线电环境中的功效。在恢复误差，恢复时间和协方差的基础上，将所提技术的性能与常规贝叶斯方法进行了比较，以验证其优越性。