Rapid detection of weak signals using adaptive recursive filter weights

The adaptive weights of a frequency domain adaptive recursive filter in the line enhance mode are investigated for their ability to rapidly detect and identify a weak tonal in the presence of additive broadband noise. The investigation utilizes both Monte Carlo computer simulations and first order theory to obtain an approximate analytic solution. In the simulation, rapid identification meant within one minute. For the case considered, frequency was determined for dB and amplitude was determined for dB. The analytic results show the limiting factors. Some of the results apply to the LMS transversal filter.