Peak detection method using blind source separation
AssigneesKing Fahd University of Petroleum and Minerals, King Abdulaziz City for Science and Technology
InventorsZahid Saleem, Samir Al-Ghadhban, Tareq Y. Al-Naffouri
Life of this patent
Estimating the term…
Abstract
The peak detection method using blind source separation extracts true peaks from noisy peaks in a more robust way that does not require any a priori information. Information regarding true peak location is obtained by thresholding the output of a wavelet transform. The value of the threshold is dependent on noise variance. While noise variance is normally unknown, the present method implements a blind source separation technique to calculate the noise variance. The blind source separation technique does not require information of the incoming signal or the channel noise, and hence is suitable for CR (cognitive radio) peak detection.
Claims10
A peak detection method using blind source separation, comprising the steps of:
- receiving mixed signals in a radio frequency (RF) receiver, the mixed signals including a wideband RF signal;
- calculating a sample mean of the received wideband RF signal;
- edge detecting the received wideband RF signal to provide a peak signal containing information regarding frequency band boundaries;
- performing a blind source separation procedure on the peak signal to calculate an un-mixing matrix W;
- estimating the source signals based on the un-mixing matrix, assuming a difference between the original received signal and an estimated source signal is a noise signal;
- calculating a variance of the noise signal using a statistical combination of the blind source separation un-mixing matrix W and an output of the edge detecting step;
- calculating a threshold value T for the mixed signals, the threshold calculation being based on the noise variance calculation;
- applying the threshold value T to the resultant peak signal of the edge detecting step;
- iteratively computing locations of the frequency band boundaries based on the step of applying the threshold;
- and
- computing Power Spectral Density within the frequency band boundaries.
A computer software product, comprising a non-transitory medium readable by a processor, the non-transitory medium having stored thereon a set of instructions for performing a peak detection method using blind source separation, the set of instructions including:
- (a) a first sequence of instructions which, when executed by the processor, causes said processor to receive mixed signals including a wideband RF signal;
- (b) a second sequence of instructions which, when executed by the processor, causes said processor to calculate a sample mean of the received wideband RF signal;
- (c) a third sequence of instructions which, when executed by the processor, causes said processor to perform edge detection on the received wideband RF signal to provide a peak signal containing information regarding frequency band boundaries;
- (d) a fourth sequence of instructions which, when executed by the processor, causes said processor to perform a blind source separation procedure on the peak signal to calculate an un-mixing matrix W;
- (e) a fifth sequence of instructions which, when executed by the processor, causes said processor to estimate the source signals based on the un-mixing matrix, assuming a difference between the original received signal and the estimated signal is a noise signal;
- (f) a sixth sequence of instructions which, when executed by the processor, causes said processor to calculate a variance of the noise signal using a statistical combination of the blind source separation un-mixing matrix W and an output of the edge detection;
- (g) a seventh sequence of instructions which, when executed by the processor, causes said processor to calculate threshold value T for the mixed signals, the threshold calculation being based on the noise variance calculation;
- (h) an eighth sequence of instructions which, when executed by the processor, causes said processor to calculate a normalized noise variance;
- (i) a ninth sequence of instructions which, when executed by the processor, causes said processor to apply the threshold value T to the edge detection resultant peak signal;
- (j) a tenth sequence of instructions which, when executed by the processor, causes said processor to iteratively compute locations of the frequency band boundaries based on the threshold value application;
- and
- (k) an eleventh sequence of instructions which, when executed by the processor, causes said processor to compute Power Spectral Density within the frequency band boundaries.
Dependencies are inferred from the claim text.