Digital Instantaneous Frequency Measurement of a Real Sinusoid Based on Three Sub-Nyquist Sampling Channels

Abstract

Multichannel sub-Nyquist sampling is an efficient technique to break through the limitation of the Nyquist sampling theorem for the wideband digital instantaneous frequency measurement (DIFM) receiver. The significant challenge is calculating the folding frequency and solving the ambiguity quickly and accurately. Usually, the researchers adopt a fast Fourier transform (FFT) to calculate the folding frequency and a Chinese remainder theorem (CRT) to solve the ambiguity caused by undersampling. However, these algorithms have the drawback of long response time. In this paper, we use a frequency deduction algorithm based on the total least-squares estimation to measure the folding frequency accurately within a few clocks. We propose a frequency-band division algorithm by dividing the measurable frequency range into a few sub-bands to solve the frequency ambiguity. This deblurring algorithm is more efficient and faster than CRT. We analyse the performance of our algorithms by numerical simulations and functional hardware simulations and compare them with FFT and CRT. We also conduct experiments on a hardware platform to confirm the effectiveness of our algorithms. The simulation results show that our algorithms have better performance than FFT and CRT in accuracy and response time. Board-level measurement results verify that the proposed DIFM technique can capture the frequency from 0 to 5,000 MHz with a maximum of 0.8 MHz root-mean-squared error in less than 200 ns.