CN101577564A - Self-adaptive signal frequency spectrum sensing and detection method based on decision threshold - Google Patents

Abstract

The invention relates to self-adaptive signal frequency spectrum sensing and detection technology based on decision threshold. Two assumptions are given: no wireless frequency spectrum signal H0: y(t)=n(t) exists and a wireless frequency spectrum signal H1: y(t)=x(t) +n(t) exists. Whether a signal x(t) exists in a received signal y(t) is detected, which comprises the following steps of: 1) signal separation, in which the received wide frequency signal y(t) is subjected to orthogonal basis space transformation, and a useful signal s1(t) and a channel noise signal s2(t) are effectively separated from the detected signal y(t); 2) threshold formation, in which the separated channel noise signal s2(t) is used to estimate channel noise energy or power, and decision threshold needed by a decider is set according to the channel noise energy or power, and is provided to the decider for decision; 3) frequency spectrum detection, in which the separated useful signal s1(t) is subjected to frequency spectrum detection, and a frequency spectrum detection result is taken as an evidence for the decision; and 4) decision, in which the decider judges the frequency spectrum detection result according to the set decision threshold to determine whether the signal x(t) exists. The technology has the advantages that the decision threshold can adapt to the dynamic change of channel noise, and the technology has high accuracy rate of signal sensing and detection, and short detection time.

Description

Based on decision threshold self-adaptive signal frequency spectrum sensing and detection method

Technical field

The present invention relates to the frequency spectrum perception and the detection technique of random signal, more specifically to a kind of under wireless channel environment based on decision threshold self-adaptive signal frequency spectrum sensing and detection technique.

Background technology

Current, show growing spectrum requirement and the contradiction day between the limited frequency spectrum resources outstanding, seriously restricted the development of radio communication service.But from actual wireless frequency spectrum operation situation, distribute the wireless frequency spectrum of (mandate) to exist the idle of certain degree on time and space, according to the measurement data report to wireless frequency spectrum, the frequency spectrum utilization rate of most of radio band is only about 10%.How effectively to solve the contradiction of the rare and frequency spectrum utilization rate of frequency spectrum resource between low and become key technology in the radio communication.The radio (CR) of giving cognitive function is acknowledged as the effective technology means of efficiently utilizing wireless frequency spectrum.

The core of CR technology then is to survey " frequency spectrum cavity-pocket " by the dynamic spectrum perception, rationally take interim available frequency band, and according to the perception information self adaptation, dynamically change transmission parameters such as self signal transmission power, tranmitting frequency, modulation system to evade the main user (authorized user) who is communicating by letter.It requires time user (CR user) existing main user not to be produced any interference by wireless environment, change self transmission parameter around the perception with assurance.This just feature that can change the self transmission parameter rapidly makes the CR technology be considered to " revolution next time " of future communications.In today that spectral compatibility and interoperability become more and more difficult, the CR technology that has physical layer (PHY) and network layer (MAC) perceptional function concurrently has been expressed great expectations.

Wireless environment is the prerequisite of CR work around correct perception and the detection.The method of cognition wireless signal sensing and detection can be divided into three major types at present: energy measuring, matched filter detect and cyclostationary characteristic detects.No matter adopt that a kind of detection method, all need a suitable decision threshold.The setting of decision threshold is a difficult problem of besieging signal sensing and detection always, and it directly has influence on the accuracy rate of signal sensing and detection.It mainly is to set according to signal power that receives and channel priori that present decision threshold is set, and the noise in can not adaptive channel changes, and has had a strong impact on the accuracy rate of signal sensing and detection.

Summary of the invention

The objective of the invention is to overcome above-mentioned the deficiencies in the prior art, solve the difficult problem that decision threshold is difficult to the variation of adaptive channel noise in the frequency spectrum detecting method, realize quick, accurate and effective signal frequency spectrum sensing and detection under the wireless environment, and design is a kind of based on decision threshold self-adaptive signal frequency spectrum sensing and detection method.

Above-mentioned purpose is achieved by following technical proposals:

The wireless frequency spectrum signal that surrounding environment is existed carries out two kinds of hypothesis: do not have wireless frequency spectrum signal H ₀There is wireless frequency spectrum signal H ₁

H ₀：y(t)＝n(t)

H ₁：y(t)＝x(t)+n(t)

N (t) represents additive Gaussian noise, the wireless frequency spectrum signal that x (t) expression has been authorized, and the signal that receives is y (t), 0≤t≤T to whether there being signal x (t) among the signal y (t) detects, and comprises the following steps:

1) Signal Separation is carried out the orthogonal basis spatial alternation with the broadband signal y (t) that receives, and effectively isolates useful signal s from tested signal y (t) ₁(t) and interchannel noise signal s ₂(t);

2) thresholding forms, by isolated interchannel noise signal s ₂(t) estimate interchannel noise energy or power, set the required decision threshold of decision device thus, and offer decision device and adjudicate;

3) to frequency spectrum detection, the useful signal s that separates ₁(t) carry out frequency spectrum detection, frequency spectrum detecting result is as the foundation of judgement;

4) judgement, decision device are judged described frequency spectrum detecting result according to the described decision threshold of setting judge whether there is signal x (t).

Further design is that the orthogonal basis spatial alternation in the described step 1) is:

1) y (t) carries out wavelet transformation by wavelet transformer to the received signal, and wavelet transformation is

${\mathrm{WT}}_y(a,b)=\frac{1}{\sqrt{a}}{\∫}_{-\∞}^{\∞}y\left(t\right){\mathrm{\ψ}}^{*}\left(\frac{t-b}{a}\right)\mathrm{dt}$

Wherein, the basic small echo of ψ (t) for choosing, the complex conjugate of " * " representative function, a is a scale parameter, b is a translation parameters;

2) y (t) carries out time-frequency conversion by fast Fourier transformer to the received signal

$Y\left(f\right)={\&Integral;}_0^{T}y\left(t\right)e^{-\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="A20091003287300121.png"\; state="\{\{state\}\}">j</figure-callout>}2\mathrm{\&pi;ft}}\mathrm{dt};$

And, obtain the pairing frequency f of maximum value in view of the above to the Y behind the time-frequency conversion (f) delivery (absolute value) ₀, system bandwidth B and three dB bandwidth B ₁

3) according to the frequency f of above-mentioned gained ₀, system bandwidth B and three dB bandwidth B ₁Obtain frequency n, scale parameter a and translation parameters b that wavelet transformation decomposes;

4) wavelet transformation that will decompose for the last time is divided in half into two signals in frequency domain, and according to frequency f ₀Determine useful signal wavelet transformation WT ₁(a is b) with noise signal wavelet transformation WT ₂(a, b);

5) under corresponding scaling function, above-mentioned wavelet transformation is reconstructed, produces useful signal s respectively ₁(t) interchannel noise signal s ₂(t)

$s_1\left(t\right)=\frac{1}{C_{\mathrm{\&psi;}}}{\&Integral;}_0^{+\&infin;}\frac{\mathrm{da}}{a^2}{\&Integral;}_{-\&infin;}^{+\&infin;}{\mathrm{WT}}_1(a,b)\frac{1}{\sqrt{a}}\mathrm{\&psi;}\left(\frac{t-b}{a}\right)\mathrm{db}$

$s_2\left(t\right)=\frac{1}{C_{\mathrm{\&psi;}}}{\&Integral;}_0^{+\&infin;}\frac{\mathrm{da}}{a^2}{\&Integral;}_{-\&infin;}^{+\&infin;}{\mathrm{WT}}_2(a,b)\frac{1}{\sqrt{a}}\mathrm{\&psi;}\left(\frac{t-b}{a}\right)\mathrm{db}$

6) described orthogonal basis spatial alternation comprises fractal and wavelet transformation.

Further design is described step 2) in, described isolated interchannel noise signal s ₂(t) carry out following Energy Estimation or power estimation through the noise power device

$E_n={\&Integral;}_0^{\mathrm{<figure-callout\; id="2"\; label="T\; s"\; filenames="A20091003287300121.png"\; state="\{\{state\}\}">T</figure-callout>}}{s}_{}^{}{}_2^2\left(t\right)\mathrm{dt}$

P _n＝E _n/T

And according to the ENERGY E of having estimated _nOr power P _nAs priori, set decision threshold.If frequency spectrum detector adopts energy measuring and matching detection, then decision threshold is set at

η＝αE _n

If frequency spectrum detector adopts feature detection, then decision threshold is

η＝αP _n

Wherein factor alpha will be set according to the false alarm rate and the false dismissed rate of system requirements.

Further design is, the decision device in the described step 4) carries out hypothesis testing according to above-mentioned decision threshold η to the result of frequency spectrum detector perception, if H ₀Set up, received signal y (t)=n (t) does not wherein contain useful signal x (t), can use y (t) corresponding frequency band to communicate; If H ₁Set up, contain signal x (t) in y (t) corresponding frequency band, therefore cannot use the corresponding frequency range of y (t) to communicate.

The inventive method is applied to fractal technology and wavelet analysis in the frequency spectrum detection, by fractal or small echo tested signal is carried out Fast transforms, and the useful signal in the tested signal is effectively separated with noise, realizes quick, the accurately perception and the detection of signal.Be exactly that Signal Separation combines with frequency spectrum detection specifically, noise power in the Signal Separation Fast estimation channel, and estimating corresponding decision threshold, the accurate perception useful signal of frequency spectrum detection frequency spectrum has solved the difficult problem that decision threshold is difficult to the variation of adaptive channel noise in the frequency spectrum detecting method.The interchannel noise signal that utilizes Signal Separation to extract, the self adaptation of realization decision threshold reaches quick, the accurately perception and the detection of signal.Can produce such beneficial effect thus:

(1) by the orthogonal basis spatial alternation, realizes that the useful signal of tested signal and the effective of interchannel noise separate;

(2) set decision threshold according to interchannel noise, and give adaptation function, the dynamic change of energy adaptive channel noise improves signal sensing and detects accuracy rate;

(3) system configuration is simple, computational complexity is low, detection time is short, applied range.

Description of drawings

Fig. 1 is the FB(flow block) of the inventive method.

Fig. 2 is the FB(flow block) of the signal processing control operation of system of the present invention.

Embodiment

The present invention will be further described below in conjunction with the drawings and specific embodiments.

Two kinds of hypothesis: H of wireless frequency spectrum signal feeding that surrounding environment is existed ₀Only there is additive Gaussian noise n (t) in-surrounding environment; H ₁There are the wireless frequency spectrum signal x (t) and additive Gaussian noise n (t) that distribute (mandate) in-the surrounding environment, promptly

H ₀：y(t)＝n(t)

H ₁：y(t)＝x(t)+n(t)

If the signal that receives is y (t), 0≤t≤T.

Now whether exist signal x (t) to estimate and detection among the signal y (t) to acceptance, the key step of enforcement FB(flow block) as shown in Figure 1, the processing controls flow process of concrete signal sees also Fig. 2, and detailed process is as follows:

Y (t) carries out the orthogonal basis spatial alternation at first to the received signal

${\mathrm{WT}}_y(a,b)=\frac{1}{\sqrt{a}}{\&Integral;}_{-\&infin;}^{\&infin;}y\left(t\right){\mathrm{\&psi;}}^{*}\left(\frac{t-b}{a}\right)\mathrm{dt}$

Wherein, the basic small echo of ψ (t) for choosing, the complex conjugate of " * " representative function, a is a scale parameter, b is a translation parameters;

Y (t) carries out time-frequency conversion by fast Fourier transformer to the received signal simultaneously

$Y\left(f\right)={\&Integral;}_0^{T}y\left(t\right)e^{-\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="A20091003287300121.png"\; state="\{\{state\}\}">j</figure-callout>}2\mathrm{\&pi;ft}}\mathrm{dt};$

And, obtain the pairing frequency f of maximum value thus to the Y behind the time-frequency conversion (f) delivery (promptly taking absolute value) ₀, system bandwidth B and three dB bandwidth B ₁, and then draw frequency n, scale parameter a and the translation parameters b that wavelet transformation decomposes.

Secondly the wavelet transformation that will decompose for the last time is divided in half into two parts in frequency domain, and according to frequency f ₀Determine useful signal wavelet transformation WT ₁(a is b) with noise signal wavelet transformation WT ₂(a, b).Under corresponding scaling function, above-mentioned wavelet transformation is reconstructed, produces useful signal s respectively ₁(t) interchannel noise signal and detail signal s ₂(t)

$s_1\left(t\right)=\frac{1}{C_{\mathrm{\&psi;}}}{\&Integral;}_0^{+\&infin;}\frac{\mathrm{da}}{a^2}{\&Integral;}_{-\&infin;}^{+\&infin;}{\mathrm{WT}}_1(a,b)\frac{1}{\sqrt{a}}\mathrm{\&psi;}\left(\frac{t-b}{a}\right)\mathrm{db}$

$s_2\left(t\right)=\frac{1}{C_{\mathrm{\&psi;}}}{\&Integral;}_0^{+\&infin;}\frac{\mathrm{da}}{a^2}{\&Integral;}_{-\&infin;}^{+\&infin;}{\mathrm{WT}}_2(a,b)\frac{1}{\sqrt{a}}\mathrm{\&psi;}\left(\frac{t-b}{a}\right)\mathrm{db}$

Once more, to isolated interchannel noise signal s from signal y (t) ₂(t) carry out following Energy Estimation and power estimation by the noise power device

$E_n={\&Integral;}_0^{\mathrm{<figure-callout\; id="2"\; label="T\; s"\; filenames="A20091003287300121.png"\; state="\{\{state\}\}">T</figure-callout>}}{s}_{}^{}{}_2^2\left(t\right)\mathrm{dt}$

P _n＝E _n/T

And according to the ENERGY E of having estimated _nAnd power P _nAs priori, set decision threshold.If frequency spectrum detector adopts energy measuring and matching detection, then decision threshold is set at

η＝αE _n

If frequency spectrum detector adopts feature detection, then decision threshold is

η＝αP _n

Wherein factor alpha will be set according to the false alarm rate and the false dismissed rate of system requirements.

At last, the approximate signal s of detection method that adopts Energy Estimation or power to estimate to from signal y (t), separating ₁(t) carry out frequency spectrum detection, and frequency spectrum detecting result is offered decision device, as the foundation of judgement, so that make further judgement.The method of frequency spectrum detection can select for use energy measuring, matched filter to detect and the cyclostationary characteristic detection according to the actual detected needs.

Decision device carries out hypothesis testing according to above-mentioned decision threshold η to the result of frequency spectrum detector perception, if H ₀Set up, received signal y (t)=n (t) does not wherein contain signal x (t), can use y (t) corresponding frequency band to communicate; If H ₁Set up, contain signal x (t) in y (t) corresponding frequency band, therefore cannot use the corresponding frequency range of y (t) to communicate.

Claims (4)

1. based on decision threshold self-adaptive signal frequency spectrum sensing and detection method, the wireless frequency spectrum signal that surrounding environment is existed carries out two kinds of hypothesis: do not have wireless frequency spectrum signal H ₀There is wireless frequency spectrum signal H ₁

H ₀：y(t)＝n(t)

H ₁：y(t)＝x(t)+n(t)

N (t) represents additive Gaussian noise, the wireless frequency spectrum signal that x (t) expression has been authorized, and the signal that receives is y (t), 0≤t≤T to whether there being signal x (t) among the signal y (t) detects, and it is characterized in that comprising the following steps:

1) Signal Separation is carried out the orthogonal basis spatial alternation with the broadband signal y (t) that receives, and effectively isolates useful signal s from tested signal y (t) ₁(t) and interchannel noise signal s ₂(t);

2) thresholding forms, by isolated interchannel noise signal s ₂(t) estimate interchannel noise energy or power, set the required decision threshold of decision device thus, and offer decision device and adjudicate;

3) frequency spectrum detection is to the useful signal s that separates ₁(t) carry out frequency spectrum detection, frequency spectrum detecting result is as the foundation of judgement;

4) judgement, decision device are judged described frequency spectrum detecting result according to the described decision threshold of setting judge whether there is signal x (t).

2. according to claim 1 based on decision threshold self-adaptive signal frequency spectrum sensing and detection method, it is characterized in that the orthogonal basis spatial alternation process in the described step 1) is:

1) y (t) carries out wavelet transformation by wavelet transformer to the received signal, and wavelet transformation is

${\mathrm{WT}}_y(a,b)=\frac{1}{\sqrt{a}}{\&Integral;}_{-\&infin;}^{\&infin;}y\left(t\right){\mathrm{\&psi;}}^{*}\left(\frac{t-b}{a}\right)\mathrm{dt}$

Wherein, the basic small echo of ψ (t) for choosing, the complex conjugate of " * " representative function, a is a scale parameter, b is a translation parameters;

2) y (t) carries out time-frequency conversion by fast Fourier transformer to the received signal

$Y\left(f\right)={\&Integral;}_0^{T}y\left(t\right)e^{-j2\mathrm{\&pi;ft}}\mathrm{dt};$

And, obtain the pairing frequency f of maximum in view of the above to the Y behind the time-frequency conversion (f) delivery ₀, system bandwidth B and three dB bandwidth B ₁

3) according to the frequency f of above-mentioned gained ₀, system bandwidth B and three dB bandwidth B ₁Obtain frequency n, scale parameter a and translation parameters b that wavelet transformation decomposes;

4) wavelet transformation that will decompose for the last time is divided in half into two signals in frequency domain, and according to frequency f ₀Determine useful signal wavelet transformation WT ₁(a is b) with noise signal wavelet transformation WT ₂(a, b);

5) under corresponding scaling function, above-mentioned wavelet transformation is reconstructed, produces useful signal s respectively ₁(t) and interchannel noise signal s ₂(t)

$s_1\left(t\right)=\frac{1}{C_{\mathrm{\&psi;}}}{\&Integral;}_0^{+\&infin;}\frac{\mathrm{da}}{a^2}{\&Integral;}_{-\&infin;}^{+\&infin;}{\mathrm{WT}}_1(a,b)\frac{1}{\sqrt{a}}\mathrm{\&psi;}\left(\frac{t-b}{a}\right)\mathrm{db}$

$s_2\left(t\right)=\frac{1}{C_{\mathrm{\&psi;}}}{\&Integral;}_0^{+\&infin;}\frac{\mathrm{da}}{a^2}{\&Integral;}_{-\&infin;}^{+\&infin;}{\mathrm{WT}}_2(a,b)\frac{1}{\sqrt{a}}\mathrm{\&psi;}\left(\frac{t-b}{a}\right)\mathrm{db}$

6) described orthogonal basis spatial alternation comprises fractal and wavelet transformation.

3. according to claim 1 based on decision threshold self-adaptive signal frequency spectrum sensing and detection method, it is characterized in that described step 2) in, described isolated interchannel noise signal s ₂(t) carry out following Energy Estimation or power estimation through the noise power device

$E_n={\&Integral;}_0^T{s}_2^2\left(t\right)\mathrm{dt}$

P _n＝E _n/T

And according to the ENERGY E of having estimated _nOr power P _nAs priori, set decision threshold:

If frequency spectrum detector adopts energy measuring and matching detection, then decision threshold is set at

η＝αE _n

If frequency spectrum detector adopts feature detection, then decision threshold is

η＝αP _n

Wherein factor alpha will be set according to the false alarm rate and the false dismissed rate of system requirements.

4. according to claim 1 based on decision threshold self-adaptive signal frequency spectrum sensing and detection method, it is characterized in that the decision device in the described step 4) carries out hypothesis testing according to above-mentioned decision threshold η to the result of frequency spectrum detector perception, if H ₀Set up, received signal y (t)=n (t) does not wherein contain signal x (t), can use y (t) corresponding frequency band to communicate; If H ₁Set up, contain signal x (t) in y (t) corresponding frequency band, therefore cannot use the corresponding frequency range of y (t) to communicate.

Images