Real-time signal detection over very wide band in low SNR environments

Patent metadata

Field	Value
Publication number	US-9398587-B1
Application number	US-201414184609-A
Country	US
Kind code	B1
Filing date	Feb 19, 2014
Priority date	Apr 20, 2011
Publication date	Jul 19, 2016
Grant date	Jul 19, 2016

Abstract

A method of and apparatus removing of a plurality of relatively narrow banded signals in a relatively wide banded input signal. The method involves and the apparatus provides for compressively sensing one relatively narrow banded signal in the relatively wide banded input signal and removing one relatively narrow banded signal from the relatively wide banded input signal before detecting and removing another relatively narrow banded signal in the relatively wide banded input signal, the step of and apparatus for compressing sensing occurring with respect to both (i) the input signal with the previously detected narrow banded signals removed therefrom and (ii) a frequency shifted version of (i).

First claim

What is claimed is:

1. A method for detecting of a plurality of relatively narrowbanded signals in a relatively widebanded input signal, the method comprising:
(i) compressively sensing a strongest narrowbanded signal in said input signal and determining a center frequency of said first narrowbanded signal;
(ii) utilizing the center frequency and the bandwidth of said strongest narrowbanded signal to remove the strongest narrowbanded signal from said input signal and to produce a version of the input signal which version corresponds to said input signal without said strongest narrowbanded signal;
(iii) repeatedly compressively sensing additional narrowbanded signals in the input signal and repeatedly utilizing a center frequency and a bandwidth of each additional narrowbanded signal to remove the each additional narrowbanded signal from said input signal and thereby to produce said version of the input signal which version corresponds to said input signal less said strongest narrowbanded signal and less each additional narrowbanded signal as the additional narrowbanded signals are repeatedly compressively sensed and removed from said input signal;
(iv) assigning each compressively sensed narrowbanded signal which is removed from the input signal to a separate narrowband receiver for signal analysis; and
wherein the compressive sensing which occurs for said strongest narrowbanded signal and with respect to each additional narrowbanded occurs with respect to both said input signal less all previously detected narrowbanded signals and a frequency shifted version of said input signal less all previously detected narrowbanded signals.

2. The method of claim 1 wherein step (iii) is repeated at least until either (1) all of the plurality of relatively narrowbanded signals have been removed from said input signal or (2) hardware resources to accomplish the removal of the relatively narrowbanded signals are exhausted.

3. The method of claim 2 wherein the step of compressively sensing includes:
(a) mixing the input signal less all previously detected narrowbanded signals with a random sequence; and
(b) sampling the mixed signal from step (a) at a subnyquist rate.

4. The method of claim 3 further including the step of:
(c) projecting the sampled signal from step (b) to a space spanned by a set of sinusoid functions whereby a maximum projection coefficient is located and the frequency of the corresponding sinusoid function is determined, the frequency of the corresponding sinusoid function being the center frequency is utilized to subtract the narrowbanded signals from the input signal and to produce said cleaner version of the input signal.

5. The method of claim 3 wherein the step of compressively sensing further includes:
(a) mixing said frequency shifted version of the input signal less all previously detected narrowbanded signals with a random sequence; and
(b) sampling the mixed signal from step (a) at a subnyquist rate.

6. The method of claim 5 further including the step of:
(c) projecting the sampled signal from step (b) to a space spanned by a set of sinusoid functions whereby a maximum projection coefficient is located and the frequency of the corresponding sinusoid function is determined, the frequency of the corresponding sinusoid function being the center frequency is utilized to subtract the narrowbanded signals from the input signal and to produce said cleaner version of the input signal.

7. The method of claim 1 further including testing for a continued presence of previously detected narrowbanded signals, and if one or more previously detected narrowbanded signals disappears, then hardware resource associated with subtracting those one or more previously detected narrowbanded signals are released and those signals are no longer removed from input signal to form said cleaner version thereof.

8. The method of claim 7 wherein the released hardware resources are then allocated to other detected narrowbanded signals.

9. The method of claim 1 further including, while repeatedly detecting additional narrowbanded signals, testing whether the additional narrowbanded signals exceed a predetermined energy threshold, and if they do not exceed said predetermined energy threshold, then designating them to be noise and ignoring them rather than removing them from the input signal x(t) in forming said version of the input signal.

10. A computer program product for processing a plurality of relatively narrowbanded signals in a relatively widebanded input signal data, the computer program product comprising computer technology readable instruction means stored on a non-transitory computer technology readable medium that are executable by a processor for causing the processor to perform operations of:
(i) compressively sensing a first narrowbanded signal in said input signal data and determining a center frequency of said first narrowbanded signal;
(ii) utilizing a center frequency and bandwidth of said first narrowbanded signal to remove the first narrowbanded signal from said input signal data and to produce a cleaner version of the input signal data which version initially corresponds to said input signal data without said first narrowbanded signal;
(iii) repeatedly compressively sensing additional narrowbanded signals in the cleaner version of the input signal and repeatedly utilizing a center frequency and bandwidth of each additional narrowbanded signal to remove the each additional narrowbanded signal from said input signal and thereby to produce said cleaner version of the input signal data which cleaner version corresponds to said input signal data less first narrowbanded signal and less each additional narrowbanded signal as the additional narrowbanded signals are repeatedly detected and removed from said input signal data;
wherein the compressive sensing which occurs for said first narrowbanded signal and with respect to each additional narrowbanded occurs with respect to both said input signal with all previously detected narrowbanded signals removed therefrom and a frequency shifted version of said input signal with all previously detected narrowbanded signals removed therefrom.

11. The computer program product of claim 10 wherein step (iii) is repeated according to the computer technology readable instruction means at least until either (1) all of the plurality of relatively narrowbanded signals have been removed from said input signal or (2) hardware resources to accomplish the removal of the relatively narrowbanded signals are exhausted.

12. The computer program product of claim 10 wherein the step of detecting the first and additional narrowbanded signals is performed compressively in accordance with the computer technology readable instruction means.

13. The computer program product of claim 12 wherein the step of compressively sensing includes:
(a) mixing the input signal with all previously sensed narrowbanded signals removed therefrom with a random sequence; and
(b) sampling the mixed signal from step (a) at a subnyquist rate.

14. The computer program product of claim 10 wherein the computer technology readable instruction means include instructions that are executable by the processor for causing the processor to perform operations of testing for a continued presence of previously detected narrowbanded signals, and if one or more previously detected narrowbanded signals disappears, then hardware resource associated with subtracting those one or more previously detected narrowbanded signals are released and those signals are no longer removed from input signal x(t) to form said cleaner version thereof.

Assignees

• Hrl Lab Llc

Inventors

• Kong Cathy Xiangming
• Ni Kang-Yu

Classifications

• H04W72/044 Primary

  based on the type of the allocated resource · CPC title

• H04L5/0064

  Rate requirement of the data, e.g. scalable bandwidth, data priority · CPC title

• H04J11/004 Primary

  using regenerative subtractive interference cancellation · CPC title

• H04J11/0066

  of narrowband interference (narrowband interference reduction H04B1/1036) · CPC title

• H04W72/00

  Local resource management · CPC title