Recently a nonlinear optical processor has been introduced that can perform real-time pattern recognition of a scene containing multiple objects with multiple reference patterns in parallel and it can optically and/or electrically update both the input scene and the reference signals in real-time. Theoretical and numerical studies have shown that the performance of this nonlinear correlator is substantially superior to the existing conventional optical correlators in the areas of light efficiency, autocorrelation peak to sidelobe ration, autocorrelation bandwidth, and discrimination sensitivity. The nonlinear optical processor produces delta function like correlation signals with significantly higher autocorrelation peak intensity, lower auto correlation sidelobes, and lower cross correlation values. It is proposed that the results of those theoretical and numerical investigators be confirmed by measurements to be made on an experimental model of the real-time nonlinear optical correlator
