Signal pulse detection scheme for use in real-time spectrum analysis

• US 7,292,656 B2
• Filed: 08/02/2004
• Issued: 11/06/2007
• Est. Priority Date: 04/22/2002
• Status: Active Grant

• Current Assignee: Cisco Technology, Inc. (Cisco Systems, Inc.)
• Original Assignee: Cognio, Inc. (Cisco Systems, Inc.)
• Inventors: Kane, Brian F., Sugar, Cary L., Kloper, David S.
• Primary Examiner(s): PHU, PHUONG M
• Application Number:
  US10/909,450
• Publication Number:
  US 20050002473A1
• Time in Patent Office: 1,191 Days
• Field of Search: 375/340, 375/259, 375/224, 375/228, 375/316, 708/5, 455/226.1, 455/226.2, 702 75- 77, 324/76.19, 324 7621- 7625, 324/76.43, 324/76.47, 324 7652- 7657, 370/241
• US Class Current: 375/340
• CPC Class Codes: G01R 23/16 Spectrum analysis; Fourier analysis
• CPC Class Codes: H04L 1/1664 the supervisory signal being transmitted together with payload signals; piggybacking
• CPC Class Codes: H04L 1/1867 Arrangements specially adapted for the transmitter end
• CPC Class Codes: H04W 16/14 Spectrum sharing arrangements between different networks
• CPC Class Codes: H04W 24/00 Supervisory, monitoring or testing arrangements

Abstract

A spectrum analysis engine (SAGE) that comprises a spectrum analyzer component, a signal detector component, a universal signal synchronizer component and a snapshot buffer component. The spectrum analyzer component generates data representing a real-time spectrogram of a bandwidth of radio frequency (RF) spectrum. The signal detector detects signal pulses in the frequency band and outputs pulse event information entries output, which include the start time, duration, power, center frequency and bandwidth of each detected pulse. The signal detector also provides pulse trigger outputs which may be used to enable/disable the collection of information by the spectrum analyzer and the snapshot buffer components. An alternative pulse detection module is provided that tracks signal pulses by comparing peak data from successive FFT cycles with existing signal pulse data that is derived from comparing peak data for prior FFT cycles. Peaks for new FFT cycles are matched to data associated with signal pulses determined to be occurring over many FFT intervals.

Claims

1. A method for detecting signal pulses from received radio frequency (RF) energy, comprising:

a. storing data for one or more spectral peaks of RF energy;

b. analyzing the data over multiple time intervals to identify one or more signal pulses, a signal pulse comprised of a sequence of spectral peaks having similar spectral characteristics over multiple time intervals;

c. storing signal pulse data describing one or more identified signal pulses based on said analyzing; and

d. updating the stored signal pulse data with data for a spectral peak that has spectral characteristics that matches data in the stored signal pulse data to thereby extend a duration of a signal pulse, add data for a new signal pulse to the stored signal pulse data for a spectral peak associated with a time interval, or terminate a signal pulse in the stored signal pulse data.

2. The method of claim 1, and further comprising comparing a bandwidth of each of the spectral peaks with a minimum bandwidth threshold and discarding data for spectral peaks that do not exceed the minimum bandwidth threshold.

3. The method of claim 1, and further comprising detecting spectral peaks of RE energy as power values above a power threshold at a set of frequencies.

4. The method of claim 3, wherein detecting spectral peaks comprises detecting spectral peaks of RE energy as power values above the power threshold at a set of contiguous frequencies.

5. The method of claim 3, wherein detecting spectral peaks comprises comparing power values at a plurality of frequencies with the power threshold, wherein the power threshold for the comparison at a particular frequency may vary with a power level at a frequency near the particular frequency.

6. The method of claim 3, wherein detecting a peak comprises detecting power values that exceed the power threshold at Fast Fourier Transform (FFT) frequency bins.

7. The method of claim 1, wherein storing data comprises storing data describing a lefthand frequency of each of the one or more spectral peaks.

8. The method of claim 1, wherein storing data comprises storing data describing a center frequency of each of the one or more spectral peaks.

9. The method of claim 1, wherein storing data comprises storing data describing a righthand frequency of each of the one or more spectral peaks.

10. The method of claim 1, wherein storing data comprises storing an incoming peak list comprising data for one or more spectral peaks detected for a time interval.

11. The method of claim 10, wherein storing signal pulse data comprises storing an existing pulse list containing pulse data for one or more signal pulses that have resulted over one or more time intervals, wherein the signal pulse data describes at least frequency and bandwidth for each signal pulse.

12. The method of claim 11, wherein analyzing comprises comparing data in the incoming peak list with signal pulse data in the existing pulse list, and depending on the comparison of the data for spectral peaks in the incoming peak list with the signal pulse data of signal pulses in the existing pulse list (a) storing an updated pulse list by adding data for a new signal pulse to the existing pulse list;

(b) modifying data in the existing pulse list to extend a time duration of a signal pulse in the existing pulse list with data for a spectral peak in the incoming pulse list;

or (b) removing data in the existing pulse list for a signal pulse that is no longer occurring.

13. The method of claim 12, and further comprising storing the contents of the updated pulse list into the existing pulse list for comparison against data in the incoming pulse list for a new time interval.

14. The method of claim 12, wherein storing an incoming peak list comprises storing the data for one or more spectral peaks detected for a time interval in order of increasing frequency, and the step of storing an existing pulse list comprises storing pulse data for one or more pulses in order of increasing frequency.

15. The method of claim 1, and wherein analyzing comprises comparing data with the stored signal pulse data subject to configurable values for at least one of a bandwidth deviation, a frequency deviation and a power deviation.

16. The method of claim 1, wherein analyzing comprises comparing frequency, bandwidth and power for one or more spectral peaks associated with a time interval with frequency, bandwidth and power of signal pulses in the stored signal pulse data.

17. The method of claim 1, and further comprising outputting signal pulse data describing characteristics of signal pulses detected over a period of time.

18. The method of claim 17, wherein outputting signal pulse data comprises outputting data describing at least one of time of occurrence, power, frequency, bandwidth and duration for each signal pulse.

19. The method of claim 17, wherein outputting signal pulse data comprises outputting ranges for at least one of power, frequency, bandwidth and duration for each signal pulse.

20. The method of claim 17, wherein outputting signal pulse data comprises outputting averages for at least one of power, frequency, bandwidth and duration for signal pulses detected over time.

21. The method of claim 17, and further comprising joining signal pulse data for similar pulses that may have been broken up due to signal fading and other conditions.

22. The method of claim 17, and further comprising filtering out signal pulse data for pulses having a duration less than a minimum duration threshold.

23. The method of claim 1, and further comprising storing digital signals representing samples of received RF energy when a signal pulse of a particular type is detected.

28. The signal pulse detector of claim 23, wherein the peak calculation module, pulse comparison module and first memory are implemented on an integrated circuit.

46. The method of claim 1, wherein analyzing comprises comparing frequency location and bandwidth of one or more spectral peaks for a time interval with frequency location and bandwidth of signal pulses in the stored signal pulse data so as to update the stored signal pulse data with data for a spectral peak that matches data in the stored signal pulse data to thereby (a) extend a duration of a signal pulse;

(b) add data for a new signal pulse to the stored signal pulse data;

or (c) terminate a signal pulse in the stored signal pulse data.

47. The method of claim 1, wherein storing data for one or more spectral peaks comprises storing data describing bandwidth and frequency location for each spectral peak for a time interval.

24. A signal pulse detector that detects signal pulses from received radio frequency (RF) energy, comprising:

a. a peak calculation module that computes data for one or more spectral peaks of RF energy that exceed a power threshold and a minimum bandwidth threshold during each of a plurality of time intervals;

b. a first memory that stores an incoming peak list comprising data for one or more spectral peaks detected for a current time interval; and

c. a pulse comparison module coupled to the first memory, and which compares the data in the incoming peak list with existing pulse data to determine whether to (1) extend an existing pulse;

(2) terminate an existing pulse;

or (3) add data for a new pulse to the existing pulse data.

25. The signal pulse detector of claim 24, and further comprising a second memory that stores the existing pulse data in an existing pulse list for signal pulses determined to occur based on data in the incoming peak list, and that stores an updated pulse list containing changes to the existing pulse list based on comparison of data in the incoming peak list for the current sampling cycle with the data in the existing pulse list.

26. The signal pulse detector of claim 24, wherein the pulse comparison module (a) writes data to the updated pulse list to add data for a new signal pulse;

(b) extends a time duration of a signal pulse in the existing pulse list with data associated with a spectral peak in the incoming peak list;

or (c) removes data for a signal pulse in the existing pulse list that has terminated.

27. The signal pulse detector of claim 24, wherein the peak calculation module and pulse comparison module are implemented by digital logic gates.

29. The signal pulse detector of claim 24, and further comprising a memory interface module coupled to the second memory and to the pulse comparison module, wherein the pulse comparison module triggers the memory interface module to read the contents of the second memory in order to output signal pulse data.

30. A system comprising the signal pulse detector of claim 24, and further comprising a Fast Fourier Transform (FFT) block that receives as input a digital signal representing received RF energy and computes FFT values spanning a plurality of frequency bins for a time interval;

and a power calculation block coupled to the FFT block that computes the power at each frequency bin at each time interval, wherein the power is supplied as input to the peak calculation module.

31. The system of claim 30, and further comprising a memory that stores one or more of:

a. a running sum of the power at each frequency bin over time intervals;

b. a duty count comprising a running sum at each time interval of the number of times the power at each frequency bin exceeds the power threshold;

c. a maximum power for each frequency bin for the current and prior time intervals; and

d. a running count of the number of time intervals in which a certain number of spectral peaks have been detected.

32. A system comprising the signal pulse detector of claim 24, and further comprising a memory controller coupled to an output of the signal pulse detector, wherein the signal pulse detector outputs a trigger signal in response to detecting a certain type of signal pulse that causes the memory controller to store digital samples of received RF energy for a time interval in a memory.

33. The system of claim 32, wherein the memory controller writes a timestamp identifying a time associated with an occurrence of the trigger signal.

34. The system of claim 32, wherein the memory controller continuously stores the digital samples in the memory and stops storing in response to the trigger signal.

35. A system comprising the signal pulse detector of claim 24, and further comprising a memory controller coupled to an output of the signal pulse detector, wherein the signal pulse detector outputs a trigger signal in response to detecting a certain type of pulse that causes the memory controller to write to a memory the output of the power calculation block for one or more time intervals.

36. The signal pulse detector of claim 24, and further comprising a configuration register that stores configuration data for the power threshold and the minimum bandwidth threshold.

37. The signal pulse detector of claim 36, wherein the configuration register further stores configuration data for a bandwidth deviation, a frequency deviation and a power deviation for use by the pulse comparison module when comparing peak data in the incoming peak list with existing pulse data for determining whether or not to extend an existing pulse.

38. The signal pulse detector of claim 24, and further comprising a memory interface circuit that outputs signal pulse data for existing pulses to a memory.

39. A system comprising the signal pulse detector of claim 38, and further comprising a processor coupled to the memory that processes the signal pulse data stored in the memory.

40. The system of claim 39, wherein the processor joins pulse data for similar pulses that may have been broken up due to signal fading and other conditions.

41. A spectrum analysis system comprising:

a. a Fast Fourier Transform (FFT) block that receives as input a digital signal representing received RF energy and computes FFT values spanning a plurality of frequency bins for a time interval;

b. a power calculation block coupled to the FFT block that computes the power at each frequency bin for a each time interval;

c. a peak detector that receives as input an output of the power calculation block for successive time intervals, detects one or more peaks in the output of the power calculation block and outputs data describing spectral peaks of RF energy for each time interval;

d. a signal pulse detector coupled to receive as input the output of the peak detector;

i. a peak calculation module that computes data for one or more spectral peaks that exceed a power threshold and a minimum bandwidth threshold;

ii. a first memory that stores an incoming peak list comprising data for one or more spectral peaks detected for a current time interval; and

iii. a pulse comparison module coupled to the first memory, and which compares the data in the incoming peak list with existing pulse data to determine whether to extend or terminate an existing pulse, or to add a new pulse to the existing pulse data; and

e. a statistics computation module coupled to the power calculation module that computes one or more of a running sum of the power at each frequency bin over time intervals;

a duty count comprising a running sum at each time interval of the number of times the power at each frequency bin exceeds the power threshold;

a maximum power for each frequency bin for the current and prior time intervals; and

a running count of the number of time intervals in which a certain number of spectral peaks have been detected.

42. The spectrum analysis system of claim 41, and further comprising a second memory that stores the existing pulse data in an existing pulse list for signal pulses determined to occur based on data in the incoming peak list, and that stores an updated pulse list containing changes to the existing pulse list based on comparison of data in the incoming peak list for the current time interval with the pulse data in the existing pulse list.

43. The spectrum analysis system of claim 41, wherein the pulse comparison module writes data to the updated pulse list to (a) add data for a new signal pulse, (b) extend a time duration of a signal pulse in the existing pulse list with data associated with a peak in the incoming peak list and (c) remove data for a signal pulse in the existing pulse list that has terminated.

44. The spectrum analysis system of claim 41, wherein the peak calculation module and pulse comparison module are implemented by digital logic gates.

45. The spectrum analysis system of claim 43, and further comprising a memory interface module coupled to the second memory and to the pulse comparison module, wherein the pulse comparison module triggers the memory interface module to read the contents of the second memory in order to output signal pulse data.

48. A processor readable medium storing instructions, that when executed by a processor, cause the processor to perform functions comprising:

a. storing data for one or more spectral peaks detected from received RF energy;

b. analyzing the data over multiple time intervals to identify one or more signal pulses, a signal pulse comprised of a sequence of spectral peaks having similar frequency location and bandwidth characteristics over multiple time intervals; and

c. storing signal pulse data describing one or more identified signal pulses based on said analyzing;

d. wherein the instructions for said analyzing comprise instructions for comparing frequency location and bandwidth of one or more spectral peaks for a time interval with frequency location and bandwidth of signal pulses in the stored signal pulse data so as to update the stored signal pulse data with data for a spectral peak that matches data in the stored signal pulse data to thereby extend a duration of a signal pulse, add data for a new signal pulse to the stored signal pulse data, or (c) terminate a signal pulse in the stored signal pulse data.

49. The processor readable medium of claim 48, wherein the instructions for said storing of data for one or more peaks comprise instructions, for storing data describing bandwidth and frequency location for each spectral peak for a time interval.