Wavelet Decomposition for Discrete Probability Maps
Emily Brown, Samuel Picton Drake, Anthony Finn · Defense Technical Information Center (DTIC) · 2007
Modern day electronic warfare often contains a heterogeneous mix of distributed sensors. This mix of sensors provides information about the probability of emitters being located at certain points. This discrete probability map (DPM) must be reported to the commander or some other decision maker in a timely fashion. This report shows that with respect to current methods the most effective way to transmit the DPM is through wavelet decomposition. Following an introduction into wavelet theory we go on to discuss the specifics of the Haar wavelet. Using a sample image, we show how to decompose data by wavelets, specify a compression ratio, transmit a specific region of interest only and reconstruct the data from the wavelets. Having established these techniques we give a specific example of a DPM generated by noisy sensors trying to locate a radar from time difference of arrival, bearings and scan-rate measurements. We conclude the report with a discussion of wavelet basis functions other than the Haar wavelets.