压缩传感技术在无线电监测中的应用探讨

Abstract

无线电监测是指探测、搜索、截获无线电管理地域内的无线电信号，并对该无线电信号进行分析、识别、监视并获取其技术参数、工作特征和辐射位置等技术信息的活动。在日常的监测工作中将产生大量的监测数据，并且随着监测站数量的增加，这些数据的传输将占用大量的网络资源。由于无线电频谱数据本身具有稀疏性，因此可以通过压缩来减少传输的数据量。本文介绍了几种压缩传感算法及其Matlab工具箱实现方式，采用不同的算法对20~3000 MHz频段的无线电频谱数据进行压缩处理，并比较了算法的时间以及数据的恢复重构误差。结果表明在5%的重构误差下，采用压缩传感可减少2/3的数据流量，有利于降低网络传输成本，这在无线电监测中有潜在应用。 Radio monitoring refers to the action of detection, search, interception of the radio signal within a certain radio management area, as well as analyzing and recognizing its technique parameters, working characteristics and radiation location. The daily monitoring work will generate massive amounts of data and with the increase of the number of monitoring stations, the transmission of the datum will take up a lot of network resources. Since the radio spectrum itself has sparse fea-ture, it is possible to be compressed to reduce the amount of data. This paper introduces several compression sensing algorithms and their Matlab toolbox implementation method. Different compression algorithms are implemented to compress the radio spectrum data in the whole fre-quency range of 20~3000 MHz. The running time and data recovery reconstruction error of the algorithms are analyzed and compared. The results show that if 5% of the reconstruction error is allowed, the use of compression sensing can reduce the data flow of 2/3, which is beneficial to reduce the network transmission cost, and has potential application in radio monitoring.