基于稀疏傅里叶变换的低采样率宽带频谱感知

Wideband spectrum sensing at low sampling rate based on the sparse Fourier transform

针对传统频谱感知方法应用于宽带频谱感知时存在采样率过高的现象,提出一种基于稀疏傅里叶变换的采用奈奎斯特采样进行宽带信号频谱感知的方法。该算法在频谱分布稀疏时具有极低的误判率,并在频谱占用率增加时,提出了改进的算法,最后利用MATLAB仿真验证了稀疏傅里叶变换用于宽带频谱感知方案的可行性。相比传统方法,本文的频谱感知方案需要的采样率仅为奈奎斯特采样率的20%。

Because the sampling rate is too high in the traditional spectrum sensing method, this paper propose a wideband signal spectrum sensing method based on the sparse Fourier transform using sub-Nyquist sampling rate. This algorithm has a very low rate of false positives when spectrum is sparse, and a modified solution is proposed even if the spectrum occupancy increases. MATLAB simulation verified the feasibility of wideband spectrum sensing scheme. Compared to the traditional method, the sampling rate that spectrum sensing based on the sparse Fourier transform required is 20% of the Nyquist sampling rate.