2 October 2023 in IEEE Sensors Journal
Frequency Modulation Continuous Wave radar will produce large ranging error when the Signal-to-Noise Ratio (SNR) is low. Both traditional frequency thinning technology and band selection technology have complex modulation and filtering processes, and the processing steps are complex. In order to quickly fit more accurate signal frequency under the condition of low SNR, this paper proposes an improved sine Curve Fitting (CF) method for signal frequency. This algorithm can not only fit the exact signal frequency, but also save the process of complex modulation and filtering, so it can estimate the signal frequency quickly and accurately. In frequency fitting, in order to avoid falling into local optimum when finding the solution coefficient in the fitting process, a new particle optimal fitting algorithm (PSO-CF) is combined by Particle Swarm Optimization (PSO) algorithm and CF algorithm. The algorithm can quickly find the optimal solution coefficient and fit the curve waveform closer to the original signal. Because the PSO algorithm is influenced by the initial value in the iterative sine curve parameters, Fast Fourier Transform (FFT) is used to roughly estimate the carrier signal frequency, and then the signal frequency roughly estimated by FFT is used as the initial value of the PSO algorithm iteration. Simulation and radar tests show that the improved sine CF algorithm can quickly and accurately fit the signal frequency to be estimated.