Threshold Detection and Estimation in Correlated Interference
David Middleton · 1991
For electromagnetic compatibility (EMC) applications, as well as for telecommunications in the large, signal detection and estimation methods are essential elements for improved performance in such interference environments. Accordingly, the Bayes theory of optimum threshold detection of signals and the estimation of their parameters in generalized noise is extended to the often critically important cases where the noise sampled data are unavoidably correlated. This is particularly relevant for array sampling of noise fields in the overall reception operation, where "matched field" (in space) as well as "matched-signal" (in time) processing is desired. Earlier canonical theory depends for its optimal results on the condition of independent noise samples, which cannot always be safely assumed.