The detection of a deterministic signal in additive noise is considered. The log-likelihood ratio is studied for three cases: stationary white noise process, stationary Markov noise process, and noise process with independent increments. Procedures leading to realizable optimum detection schemes and related examples are given. An approximation to the log-likelihood ratio, which leads to the locally optimum detector, is also presented for the white noise case.
ASJC Scopus subject areas
- Control and Systems Engineering
- Signal Processing
- Computer Networks and Communications
- Applied Mathematics