Performance analysis of frequency estimation methods for electronic support systems

In electronic warfare systems, accurate estimation of the threat radar frequency is so important for radar identification and electronic attack. Digital instantaneous frequency estimation (DIFM) and fast Fourier transform (FFT) are the most widely used frequency estimation methods. If the intermediate frequency (IF) value of the threat radar is not an exact multiple of the FFT bins, frequency estimation with the FFT method cannot be obtained with the desired accuracy. Interpolation techniques are often applied to the output of the FFT to improve accuracy. In the DIFM method, the digitized IF signal is delayed for a certain time and its conjugate is taken and phase is calculated by multiplying the conjugate with the original I/Q signal. In this study, the frequency estimation performances of the FFT technique that Jacobsen, Improved Quinn and Ligges interpolation techniques were applied and DIFM method with appropriate delay time extensively analyzed by changing the signal to noise ratio (SNR) under Gauss noise. Also, since fast frequency estimation is so important for real time systems, FFT, FFT and Jacobsen, FFT and improved Quinn, FFT and Ligges and DIFM frequency estimation methods were also compared in terms of computation time. For each method, one hundred Monte Carlo trials were applied and the error in the frequency estimation is presented in terms of the root mean square error (RMSE). According to the simulation results performed in the MATLAB environment, it has been observed that the FFT and the improved Quinn method generally provide better frequency estimation according to other two interpolation methods. In addition, as the SNR level increased, it was observed that the DIFM method had better performance (lower RMSE value) compared with Jacobsen, Ligges and improved Quinn interpolation methods.