Kernel adaptive interference suppression system
AssigneeThe United States of America as represented by the Secretary of the Navy
InventorsJames W. Bond, David J. Marchette, Carey E. Priebe, Thomas W. Schlosser
Life of this patent
Estimating the term…
Abstract
An interference suppression system utilizes an adaptive locally optimum detection algorithm derived from kernel estimation. By utilizing this algorithm in the processor of the invention, fewer receive samples are required to obtain useful estimates of required probability density functions. The new adaptive locally optimum detection algorithm is useable in applications involving very high sampling rates and is thus suitable for rapidly and radically changing interference environments encompassing a wide range of frequencies.
Claims9
A method comprising the steps of:
- receiving signals;
- taking samples of said signals;
- finding a signal magnitude value of each of said signal samples;
- storing said signal magnitude values;
- calculating the variance of said stored magnitude values;
- calculating approximations of probability density functions for each of said stored magnitude values using said variance, so that a number of probability density functions are calculated;
- averaging the probability density functions calculated;
- finding the derivative of the average of the probability density functions;
- forming a gain factor from said derivative of the average of the probability density functions, said average of the probability density functions and one of said signal magnitude values;
- normalizing each baseband signal sample taken;
- multiplying said gain factor times each normalized signal sample to obtain gain factor adjusted signal samples;
- and
- inserting said gain factor adjusted signal samples into a demodulator.
A method for interference suppression comprising the steps of:
- receiving radio frequency signals;
- taking baseband signal samples of said radio frequency signals;
- finding a signal amplitude value of each of said baseband signal samples;
- storing said signal amplitude values;
- calculating the variance of said stored amplitude values;
- calculating approximations of probability density functions using said variance for each of said stored amplitude values, so that a number of probability density functions are calculated;
- averaging the probability density functions calculated;
- finding the derivative of the average of the probability density functions;
- forming a gain factor from said derivative of the average of the probability density functions, said average of the probability density functions and one of said signal amplitude values;
- normalizing each baseband signal sample taken;
- multiplying said gain factor times each normalized baseband signal sample to obtain gain factor adjusted signal samples;
- and
- inserting said gain factor adjusted signal samples into a radio demodulator to output interference adjusted signals.
Dependencies are inferred from the claim text.