Multipath Signal Processing Using a Vertical Array.
DUKE UNIV DURHAM N C ADAPTIVE SIGNAL DETECTION LAB
Pagination or Media Count:
Performance of a vertical array for estimating multipath signal parameters and for detecting multipath Gaussian signals is investigated. For known multipath signal direction cases, the parameters of the signal from each individual path may be estimated so that the multipath signal cancellation can be eliminated. The performance of the Maximum Likelihood Estimates MLE for the signal parameters of individual paths are shown to be dependent on the angular separation of signal paths and array length. The maximum A Posteriori Probability Estimate MAP is considered when the signal statistics are known. The linear relationship between the MLE and MAP estimates is also shown. The Bayesian optimum array processor for detecting multipath Gaussian signals with known arrival angles not only has multiple beamformers pointing on multipath signal directions, but also has an additional structure which cross-correlates multiple beams. Optimum detection performance of a vertical array can be evaluated by an analytical method if the array has no more than four elements. With the aid of a digital computer the optimum performance can be calculated for an array of any size. The trade-off between the processor structure complexity and the detection performance is shown in terms of receiver operating characteristic ROC curves and processor block diagrams. A two-path example is shown for the performance comparison between optimum detector, single beamformer and double beamformer arrays. Author