基于多尺度小波融合决策的宽带频谱感知  

Wideband Spectrum Sensing Based on Multi-scale Wavelet Fusion Decision

作　　者：李佳强 王方刚 贺勃翔 刘钰 王雪刚 LI Jiaqiang;WANG Fanggang;HE Boxiang;LIU Yu;WANG Xuegang(School of Electronic and Information Engineering,Beijing Jiaotong University,Beijing 100044,China)

机构地区：[1]北京交通大学电子信息工程学院,北京100044

出　　处：《无线电工程》2022年第10期1741-1746,共6页Radio Engineering

基　　金：国家自然基金高铁联合基金(U1834210);轨道交通控制与安全国家重点实验室自主课题(RCS2022ZT011);中央高校基本科研业务费(2020JBM081)。

摘　　要：针对宽带无线频谱范围内同时存在多个带宽悬殊信号的场景,提出了基于多尺度小波融合决策的宽带频谱感知算法。对接收信号进行滑动平均滤波,根据采样率和频率分辨率划分宽中窄带信号,并设置相应的小波变换尺度候选集,对宽中窄带信号进行串行多尺度小波融合决策获得频谱感知结果,解决了传统算法无法有效感知不同带宽信号的问题。仿真结果表明,提出的算法能有效检测宽带范围内信号的个数、带宽和载频等参数,且优于传统的多尺度感知算法。To deal with the situation that there are multiple signals with different bandwidth over a wide frequency range at the same time,a novel wideband spectrum sensing algorithm based on multi-scale wavelet fusion decision is proposed.First,the received signals are preprocessed by moving average filtering.Then the signals are divided into three levels:wide,medium and narrow signals,according to the sampling rate and frequency resolution.Meanwhile,the candidate set of wavelet transform scale is set.The final spectrum sensing results can be gained by make serial multi-scale wavelet fusion decision on wide,medium and narrow signals.The proposed algorithm solves the problems of traditional algorithms that signals with different bandwidth cannot be sensed the same time.The simulation results show that the algorithm can effectively detect the number,carrier frequency and bandwidth of signals in wideband,which performs better than traditional algorithms.

关 键 词：频谱感知 多尺度小波 边缘检测 融合决策 

分 类 号：TN92[电子电信—通信与信息系统]