Proceedings with Extended Abstracts (single PDF file) - Radio ...

RANGE IMAGING OBSERVATIONS OF PMSEUSING THE EISCAT VHF RADAR: PHASECALIBRATION AND FIRST RESULTSJ. R. Fernandez a , R. D. Palmer a , P. B. Chilson b , I. Häggström c , M. T. Rietveld da Department of Electrical Engineering, University of Nebraska, Lincoln, USAb CIRES–University of Colorado and NOAA Environmental Technology Lab., USAc EISCAT Scientific Association, Box 164, S-98123 Kiruna, Swedend Max–Planck-Institut fur Aeronomie, 37191 Katlenburg–Lindau, Germany1 IntroductionIn this work, a novel phase calibration technique for use with the multiple-frequencyRange IMaging (RIM) technique is introduced based on genetic algorithms. The methodis used on data collected with the European Incoherent SCAtter (EISCAT) VHF radarduring a 2002 experiment with the goal of characterizing the vertical structure of PolarMesosphere Summer Echoes (PMSE) over northern Norway. For typical Doppler measurements,the initial phases of the transmitter and receiver are not required to be thesame. The EISCAT receiver systems exploit this fact, allowing a multi-static configuration.However, the RIM method relies on the small phase differences between closelyspaced frequencies. As a result, the high-resolution images produced by the RIM methodcan be significantly degraded if not properly calibrated. The novel method is applied topreliminary data from the EISCAT radar providing first results of RIM images of PMSE.The EISCAT VHF radar has a nominal operating frequency of 224 MHz, but it iscapable of operating at frequencies ranging from 222.4–225.4 MHz in steps of 200 kHz.The 16 frequencies are denoted by F0–F15, respectively. Although more frequencies areavailable, only five receiver boards were available at the time of the experiment. For thepreliminary results presented here, the following frequency combination, F7, F9, F11,F13, and F15 were used, providing a 400 kHz frequency separation.2 Effect of phase errors on range imagingFor notation purposes, the initial phase values for the transmitter and receiver usingfrequency n will be denoted by δT n and δn R , respectively. Assuming a single layer locatedat range z I , the returned, coherently detected signal x n can be modeled as the followingx n = Ãne j[wn d t+δn−2kn˜z] + ɛ n (1)where Ãn is the returned amplitude and wdn represents the Doppler frequency for frequencyn. The term δ n = δT n − δn R is the difference in the phase of the transmitter andreceiver, k n is the corresponding wavenumber for frequency n, and ˜z = z I + z 0 is thesum of the layer range (z I ) and the range shift due to the system delay (z 0 ). Finally, ɛ nrepresents the additive white Gaussian noise in each signal with zero mean and varianceσ 2 . It should be noted that the phase difference δ n should remain constant throughoutthe experiment.Using the cross-covariance definition between two signals l and m and the FourierRIM power as defined in Palmer et al. [1999], and assuming one layer located at z I , itcan be shown that the Fourier RIM power is given by138P F (r) =N∑(Ã2 i + σ 2 ) + 2i=1N−1∑N∑n=1 m=n+1à n à m cos [2(k n − k m )(r − z I ) + ˆδ nm ] (2)