Comparison of two overlapping averaging schemes for HF ocean radar

The spectral amplitudes of High-frequency (HF) radar signals backscattered from the ocean are known to be a Rayleigh distributed variable. The mean radar Doppler spectrum forms the basis for the extraction of sea-state parameters. Conventionaly, the mean spectrum is obtained by averaging the power of several Doppler spectra, a process which utilizes only the amplitude information of the sea echoes. The phase information is assumed to be meaningless. An assessment of a complex-value averaging scheme for HF radar systems is attempted through a simulation-based study and a field-data comparison with measurements from a moored current meter for a year of time. Results show that, although the power-averaging method outperforms the complex-value averaging for every simulated case, when field data are used, the RMS difference of the power-averaging method is significantly higher than that of the complex-value averaging, when the time series is long and extra smoothing is not ulized.