Methods of blind source separation filter resource management
원문보기
IPC분류정보
국가/구분
United States(US) Patent
등록
국제특허분류(IPC7판)
H04L-027/26
H04L-012/26
출원번호
US-0296233
(2016-10-18)
등록번호
US-9866422
(2018-01-09)
발명자
/ 주소
Ray, Gary A.
출원인 / 주소
The Boeing Company
대리인 / 주소
Ostrager Chong Flaherty & Broitman P.C.
인용정보
피인용 횟수 :
1인용 특허 :
6
초록▼
Systems and methods that solve the problem of scheduling the fixed filter resources of a blind source separation subsystem by choosing in real time the center frequency and bandwidth of each filter in such a way as to allow new and existing signals to be separated out in consistent channels, with as
Systems and methods that solve the problem of scheduling the fixed filter resources of a blind source separation subsystem by choosing in real time the center frequency and bandwidth of each filter in such a way as to allow new and existing signals to be separated out in consistent channels, with as few missed signals as possible given the filter resources available. The proposed method comprises an algorithm that uses a periodic time/frequency covering map to set the center frequency and bandwidth of each filter over all time by acquiring energy measurements from the filtering subsystem using existing filter settings and continuously adaptively updating those settings while maintaining optimal coverage in time and frequency.
대표청구항▼
1. A method for performing blind source separation of signals, comprising: (a) receiving signals;(b) denoising the signals received using a denoising module;(c) separating the denoised signals as a function of frequency using respective filters of a multiplicity of channels;(d) generating first info
1. A method for performing blind source separation of signals, comprising: (a) receiving signals;(b) denoising the signals received using a denoising module;(c) separating the denoised signals as a function of frequency using respective filters of a multiplicity of channels;(d) generating first information vectors comprising respective data sets of parameter values of separated signals, wherein each data set of parameter values making up an information vector of the first information vectors comprises a frequency, an amplitude and a pulse width of a respective separated signal separated by the filter of a respective channel;(e) collecting the data sets of parameter values of the first information vectors; and(f) directing a filter of one channel of the multiplicity of channels to separate a particular frequency based on the collected data sets of parameter values of the first information vectors. 2. The method as recited in claim 1, further comprising: (g) finding an event in the collected data sets of parameter values of the first information vectors, which found event is associated with the particular frequency;(h) associating a signal-of-interest priority value with the found event;(i) determining a current minimum signal-of-interest priority value associated with one channel of the multiplicity of channels; and(j) comparing the signal-of-interest priority value associated with the found event to the current minimum signal-of-interest priority value,wherein the one channel is directed to tune its filter with the particular frequency in response to the signal-of-interest priority value associated with the found event being greater than the current minimum signal-of-interest priority value. 3. The method as recited in claim 2, further comprising tuning the filter of the one channel to process the particular frequency. 4. The method as recited in claim 2, further comprising: (k) determining whether any other channel of the multiplicity of channels is currently processing the particular frequency or not; and(l) performing step (f) in response to a determination in step (k) that none of the multiplicity of channels are currently processing the particular frequency. 5. The method as recited in claim 2, further comprising: generating second information vectors comprising respective data sets of parameter values of denoised signals, wherein each data set of parameter values making up an information vector of the second information vectors comprises a frequency range and an energy of a respective denoiser state;collecting the data sets of parameter values of the second information vectors;computing respective energies corresponding to the collected first information vectors; andsubtracting the computed energies corresponding to the collected first information vectors from the total energies of the collected second information vectors for each frequency range. 6. The method as recited in claim 5, wherein the step of associating a signal-of-interest priority value with the found event comprises setting the associated signal-of-interest priority value equal to a maximum of a first a priori signal-of-interest priority value corresponding to the particular frequency and a second a priori signal-of-interest priority value corresponding to a denoiser energy of a denoiser state having the particular frequency. 7. A method for performing blind source separation of signals, comprising: receiving signals;denoising the signals received using a denoising module;separating the denoised signals as a function of frequency using respective filters of a multiplicity of channels;generating first information vectors comprising respective data sets of parameter values of separated signals, wherein each data set of parameter values making up an information vector of the first information vectors comprises a frequency, an amplitude and a pulse width of a respective separated signal separated by the filter of a respective channel;adding the data sets of parameter values of the first information vectors to a first histogram;generating second information vectors comprising respective data sets of parameter values of denoised signals, wherein each data set of parameter values making up an information vector of the second information vectors comprises a frequency range and an energy of a respective denoiser state;adding the data sets of parameter values of the second information vectors to a second histogram;computing respective energies of the first information vectors;subtracting the computed energies from the energies of associated denoiser states in the second histogram;finding an event having an associated frequency in the first histogram;associating a signal-of-interest priority value with the found event;determining a current minimum signal-of-interest priority value associated with one channel of the multiplicity of channels; andcomparing the signal-of-interest priority value associated with the found event to the current minimum signal-of-interest priority value. 8. The method as recited in claim 7, further comprising sending a request to the one channel to tune its filter with the frequency associated with the found event in response to the signal-of-interest priority value associated with the found event being greater than the current minimum signal-of-interest priority value. 9. The method as recited in claim 8, further comprising tuning the filter of the one channel with the frequency associated with the found event. 10. The method as recited in claim 7, wherein the step of associating a signal-of-interest priority value with the found event comprises setting the associated signal-of-interest priority value equal to a maximum of a first a priori signal-of-interest priority value corresponding to the frequency associated with the found event and a second a priori signal-of-interest priority value corresponding to the denoiser energy of a denoiser state having the frequency associated with the found event. 11. The method as recited in claim 7, further comprising determining whether the frequency associated with the event is currently being processed by any one of the multiplicity of channels or not. 12. The method as recited in claim 7, wherein the event finding uses a technique selected from the following group: a mean above mean technique and a barcode tracking technique. 13. The method as recited in claim 7, wherein the first histogram is a data array comprising a respective set of frequency bins for each frequency of a plurality of frequencies, the frequency bins in each set of frequency bins containing energy information for signals having respective times of arrival to form a respective diffogram. 14. A method for performing blind source separation of signals, comprising: (a) receiving signals;(b) denoising the signals received using a denoising module;(c) separating the denoised signals as a function of frequency using respective filters of a multiplicity of channels;(d) generating information vectors comprising respective data sets of parameter values of denoised signals, wherein each data set of parameter values making up an information vector comprises a frequency range and an energy of a respective denoiser state;(e) collecting the data sets of parameter values of the information vectors; and(f) directing a filter of one channel of the multiplicity of channels to separate a particular frequency based on the collected data sets of parameter values of the information vectors. 15. The method as recited in claim 14, further comprising: (g) finding a maximum signal-of-interest priority value corresponding to one denoiser state of the respective denoiser states associated with the particular frequency;(h) determining a current minimum signal-of-interest priority value associated with one channel of the multiplicity of channels; and(i) comparing the maximum signal-of-interest priority value to the current minimum signal-of-interest priority value,wherein the one channel is directed to tune its filter with the particular frequency in response to the maximum signal-of-interest priority value associated with the found event being greater than the current minimum signal-of-interest priority value. 16. The method as recited in claim 15, further comprising tuning the filter of the one channel to process the particular frequency. 17. A system for processing received signals, comprising: a signal denoising module;a multiplicity of channels communicatively coupled to the signal denoising module, each channel comprising a respective filter and respective channel state machines for configuring the filters to separate denoised signals as a function of frequency;a control module communicatively coupled to the channel state machines of the multiplicity of channels; anda pulse descriptor words (PDW) generation module communicatively coupled to the multiplicity of channels and to the control module, wherein the pulse descriptor words (PDW) generation module is configured to generate first information vectors comprising respective data sets of parameter values of separated signals, wherein each data set of parameter values of the first information vectors comprises a frequency, an amplitude and a pulse width of a respective separated signal separated by the filter of a respective channel,wherein the control module is configured to:collect the data sets of parameter values of the first information vectors; anddirect a filter of one channel of the multiplicity of channels to separate a particular frequency based on the collected data sets of parameter values of the information vectors. 18. The system as recited in claim 17, wherein the control module is further configured to: find an event in the collected data sets of parameter values of the first information vectors, which found event is associated with the particular frequency;associate a signal-of-interest priority value with the found event;determine a current minimum signal-of-interest priority value associated with one channel of the multiplicity of channels; andcompare the signal-of-interest priority value associated with the found event to the current minimum signal-of-interest priority value,wherein the one channel is directed to tune its filter with the particular frequency in response to the signal-of-interest priority value associated with the found event being greater than the current minimum signal-of-interest priority value. 19. The system as recited in claim 18, wherein the control module is further configured to determine whether any other channel of the multiplicity of channels is currently processing the particular frequency or not, wherein the control module determines the current minimum signal-of-interest priority value in response to a determination that none of the multiplicity of channels are currently processing the particular frequency. 20. The system as recited in claim 18, wherein the signal denoising module is further configured to generate second information vectors comprising respective data sets of parameter values of denoised signals, wherein each data set of parameter values of the second information vectors comprises a frequency range and an energy of a respective denoiser state, and the control module is further configured to: collect the data sets of parameter values of the second information vectors;compute respective energies corresponding to the collected first information vectors; andsubtract the computed energies corresponding to the collected first information vectors from the total energies of the collected second information vectors for each frequency range. 21. The system as recited in claim 20, wherein the operation of associating a signal-of-interest priority value with the found event comprises setting the associated signal-of-interest priority value equal to a maximum of a first a priori signal-of-interest priority value corresponding to the particular frequency and a second a priori signal-of-interest priority value corresponding to a denoiser energy of a denoiser state having the particular frequency.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.