Periodogram-based wireless signal detection method

• US 9,462,400 B2
• Filed: 05/22/2012
• Issued: 10/04/2016
• Est. Priority Date: 12/31/2011
• Status: Active Grant

• Current Assignee: University of Science and Technology of China (Chinese Academy of Sciences)
• Original Assignee: University of Science and Technology of China (Chinese Academy of Sciences)
• Inventors: Zhang, Taotao, Zhang, Wenyi, Sun, Huanhuan
• Primary Examiner(s): GAY, SONIA L
• Application Number:
  US14/363,995
• Publication Number:
  US 20150049875A1
• Time in Patent Office: 1,596 Days
• US Class Current: 1/1
• CPC Class Codes: H04B 17/336 Signal-to-interference ratio [SIR] or carrier-to-interference ratio [CIR]
• CPC Class Codes: H04B 17/345 Interference values signal-to-interference ratio [SIR] or carrier-to-interference ratio [CIR] H04B17/336
• CPC Class Codes: H04R 2420/07 Applications of wireless loudspeakers or wireless microphones
• CPC Class Codes: H04R 29/004 for microphones H04R29/007 takes precedence

Abstract

A periodogram-based wireless microphone signal detection method, which solves the problem of incapability of distinguishing wireless microphone signal from narrow-band interference, includes: acquiring a time-domain digital signal for detection through antenna module, RF front-end module, ADC module and time-domain signal preprocessing module; and performing detection of microphone signal in frequency domain, that is, calculating an average value of periodogram of M segments of the time-domain digital signal for detection, shifting the average value to obtain a shifted vector, estimating a mean and a covariance matrix of the shifted vector, locating a maximum point of the vector, and taking several points on each side of the point as a center to form an information vector; calculating a decision statistical quantity of the information vector using the decision theory, simulating or calculating a threshold in a predefined method, and deciding the signal is narrow-band interference if the decision statistical quantity is less than the threshold, otherwise the signal is wireless microphone signal.

Claims

1. A periodogram-based wireless microphone signal detection method implemented in a computer, comprising:

acquiring M segments of frequency-domain signal, obtaining a shifted vector by preprocessing and shifting the frequency-domain signal, estimating a mean and a covariance matrix of the shifted vector, locating a maximum point or a nearby large-value point of the vector, and taking several points on each side of the point as a center to form an information vector μ

;

calculating a decision statistical quantity T_p of the information vector μ

, simulating or calculating a threshold γ

in a predefined method, and deciding whether the acquired M segments of frequency-domain signal are wireless microphone signal based on a comparison between the decision statistical quantity T_p and the threshold γ

to reduce high false alarm rates in distinguishing between wireless microphone signals and noise in a TV band white space.

2. The method of claim 1, wherein said acquiring M segments of frequency signal comprises converting non-overlapped or partially-overlapped M segments of time-domain signal into frequency-domain signal for detection by using Discrete Fourier Transform or Fast Fourier Transform algorithm.

3. The method of claim 1, wherein said preprocessing the frequency-domain signal comprises calculating a periodogram of the M segments of frequency-domain signal for detection, averaging the M segments of periodogram, and preprocessing an average value to remove frequency deviation;

said preprocessing the average value to remove frequency deviation comprises;

step 1, estimating a noise variance using the average value of the M segments of periodogram, and locating two greatest elements in the average value;

step 2, substituting the two greatest elements obtained in step 1 into a theoretical mean expression to obtain a frequency deviation of the signals for detection;

step 3, substituting the estimated frequency deviation obtained in step 2 into the theoretical mean expression to estimate a signal-to-noise ratio of the signals for detection.

4. The method of claim 3, wherein said estimating a noise variance σ

² in the preprocessing of frequency deviation removal comprises;

selecting a segment of band including only noise, and measuring a power of the noise;

or

selecting a segment of band including only noise based on the obtained periodogram, and estimating a power of the noise in a method such as a method of taking a median or mean.

5. The method of claim 3, wherein said estimating a signal-to-noise ratio in the preprocessing of frequency deviation removal comprises:

measuring an overall power of signal and noise for a certain bandwidth with an equipment, and measuring a power of the noise to estimate the signal-to-noise ratio;

or

selecting a segment of band including only noise based on the obtained periodogram, and estimating a power of the noise in a method such as a method of taking a median or mean; and

then estimating an overall power of the signal for a certain bandwidth using a similar method to estimate the signal-to-noise ratio.

6. The method of claim 1, wherein said shifting comprises shifting an average value of a preprocessed periodogram and forming a shifted vector;

calculating a mean and a covariance matrix of the shifted vector based on estimation of noise variance and signal-to-noise ratio, M, and a length of frequency-domain signal samples; and

locating a maximum point or a nearby large-value point of the shifted vector, and taking several points on each side of the point as a center to form the information vector.

7. The method of claim 1, wherein said calculating a decision statistical quantity comprises calculating the decision statistical quantity T_p with a decision theory, simulating or calculating the threshold γ

in a predefined method, and deciding, based on a comparison between T_p and γ

, that the frequency-domain signals are narrow-band interference if T_p≦

γ

, otherwise they are wireless microphone signals.

10. The method of claim 5, wherein the equipment comprises a spectrum analyzer.

8. A periodogram-based wireless microphone signal detection device comprising an antenna module, a low-noise amplifier, a bandpass filter, a downconverter for down-converting to intermediate frequency or baseband, a lowpass filter, a gain controller, an analog-to-digital converter, and a time-domain signal preprocessing module, wherein a signal output of the antenna module is coupled to a signal input of the low-noise amplifier, a signal output of the low-noise amplifier is coupled to a signal input of the bandpass filter, a signal output of the bandpass filter is coupled to a signal input of the downconverter, a signal output of the downconverter is coupled to a signal input of the lowpass filter, a signal output of the lowpass filter is coupled to a signal input of the gain controller, a signal output of the gain controller is coupled to the signal input of the analog-to-digital converter, and a signal output of the analog-to-digital converter is coupled to a signal input of the time-domain signal preprocessing module:

wherein said lowpass filter has a bandwidth equal to or wider than a half of a bandwidth of one TV band;

said gain controller is configured to output a voltage or current value that is adjustable manually or automatically according to a voltage or current required by the analog-to-digital converter; and

said time-domain signal preprocessing module is configured to implement, with a programmable chip or software, two functions of

     1) performing processing on two branches of I and Q signals, and

     2) providing control signals required by another module, and

said performing processing on two branches of I and Q signals comprises receiving, down-sampling, down-converting, integrating and storing the I and Q signals.

9. The device of claim 8, wherein said antenna module comprises Very-High Frequency and Ultra-High Frequency receive antennas, and the downconverter is configured to set a local oscillator frequency based on a TV band for detection, and down-converting the signal to intermediate frequency or baseband using an orthogonal downconverter.

11. The device of claim 8, wherein the other module comprises one or more of a local oscillator, the analog-to-digital converter, and a USB chip.