A comparison of FFT-based techniques for Doppler velocity estimation in SuperDARN radars

In radar applications, the target velocity is commonly determined using the Doppler effect. By comparing the transmit-receive differential frequency, the Doppler frequency shift can be measured, and as a result, the target velocity can be determined. The Tasman International Geospace Environment Radars (TIGER) form part of an international network of similar HF radars called Super Dual Auroral Radar Network (SuperDARN) which explores the impact of solar disturbances on the Earth's upper atmosphere. These radars utilise an Auto Correlation Function (ACF) to measure the changing phase of the ACF between lag times to determine the Doppler frequency and the target velocity. Measured velocity results can show large, and sometimes unrealistic errors. As part of the development of the third TIGER radar at Buckland Park, Adelaide, South Australia, a Spectrum Difference Function (SDF) technique for measuring velocity has been proposed as a means for cross-checking results. The SDF technique uses the Fast Fourier Transform (FFT) to calculate the transmit and receive signal magnitude spectrums which are then compared to find the Doppler frequency. In this paper the developed technique is compared to existing interpolation techniques using SuperDARN radar parameters. Simulation results show that the accuracy and computational complexity of the SDF technique are comparable to those of other techniques using FFT.