CN102364885A - A Spectrum Sensing Method Based on Signal Spectrum Envelope - Google Patents

Abstract

The invention discloses a frequency spectrum sensing method based on signal frequency spectrum envelope. The frequency spectrum sensing method comprises the following steps of: firstly, receiving a time domain continuous radio frequency signal by using two receiving antennae; secondly, carrying out down conversion and time domain sampling processing on the time domain continuous radio frequency signal to obtain a time domain discrete base band signal; carrying out discrete Fourier transform on the time domain discrete base band signal to obtain a signal frequency spectrum; calculating a signal frequency spectrum enveloped crosscorrelation coefficient; and finally, judging whether a frequency band is occupied by other wireless communication services by comparing the signal frequency spectrum enveloped crosscorrelation coefficient with a decision threshold and realizing the frequency spectrum cognition. According to the frequency spectrum sensing method disclosed by the invention, the defects that the frequency spectrum cognition is failed when the time-domain correlation among a plurality of antenna receiving signals is lower or uncorrelated in a traditional covariance matrix detection method and a characteristic value detection method are overcome.

Description

A Spectrum Sensing Method Based on Signal Spectrum Envelope

technical field

The invention relates to a spectrum sensing technology in a cognitive radio system, in particular to a spectrum sensing method based on a signal spectrum envelope.

Background technique

With the rapid growth of wireless communication services, people's demand for spectrum resources continues to increase. However, the available physical spectrum resources that can be used for wireless communication services are limited, and the existing fixed spectrum resource allocation strategy makes the utilization rate of spectrum resources low, which results in a serious shortage of spectrum resources. The cognitive radio (CognitiveRadio, CR) technology can effectively improve the utilization rate of spectrum resources, thereby alleviating the problem of lack of spectrum resources. Spectrum sensing is an important part of cognitive radio technology. It can effectively prevent wireless communication services using cognitive radio technology from interfering with other wireless communication services in the same frequency band. The performance of spectrum sensing is directly related to the performance of wireless communication services. quality.

The existing spectrum sensing methods mainly include energy detection method, cyclic feature detection method, covariance matrix detection method, eigenvalue detection method and so on. Among them, the energy detection method requires the noise power to be accurately known, but in practice the noise power cannot be accurately obtained, and the performance of the energy detection method will drop sharply at this time; the cycle feature detection method requires prior knowledge of the signal cycle characteristic frequency, and in practice In the case of multiple antennas, the covariance matrix detection method and the eigenvalue detection method can use the time-domain correlation between the signals received by multiple antennas to realize spectrum sensing, but in practical applications , in order to obtain diversity gain, the time-domain correlation between signals received by multiple antennas is low or even irrelevant, at this time the covariance matrix detection method and eigenvalue detection method will fail.

Contents of the invention

The technical problem to be solved by the present invention is to provide a signal spectrum envelope based signal detection system with good spectrum sensing performance, which can effectively overcome the disadvantage of low time-domain correlation or poor spectrum sensing performance between multi-antenna received signals. Spectrum Sensing Methods.

The technical solution adopted by the present invention to solve the above technical problems is: a spectrum sensing method based on signal spectrum envelope, which is characterized in that it comprises the following steps:

① Assume that the cognitive radio system uses two receiving antennas to receive time-continuous radio frequency signals, and express the time-domain continuous radio frequency signals received by the two receiving antennas as a function of time t. The time domain continuous radio frequency signal is denoted as x ₁ (t), and the time domain continuous radio frequency signal received by the second receiving antenna is denoted as x ₂ (t);

②Respectively perform down-conversion processing on the time domain continuous radio frequency signal x ₁ (t) received by the first receiving antenna and the time domain continuous radio frequency signal x ₂ (t) received by the second receiving antenna, and then respectively The time-domain continuous radio frequency signals obtained after x ₁ (t) and x ₂ (t) are subjected to down-conversion processing are subjected to K times of time-domain sampling to obtain the time-domain discrete baseband signals on the first receiving antenna and the second receiving antenna For the time-domain discrete baseband signal on the antenna, the time-domain discrete baseband signal on the first receiving antenna and the time-domain discrete baseband signal on the second receiving antenna are expressed as functions of the time-domain sampling sequence number k, respectively Denoted as y ₁ (k) and y ₂ (k), where k=1, 2, ..., K, K represents the number of time-domain sampling;

③Respectively perform discrete Fourier transform on the time-domain discrete baseband signal y ₁ (k) on the first receiving antenna and the time-domain discrete baseband signal y ₂ (k) on the second receiving antenna to obtain the first receiving antenna The signal spectrum on the antenna and the signal spectrum on the second receiving antenna, the signal spectrum on the first receiving antenna and the signal spectrum on the second receiving antenna are expressed as a function of the frequency domain sampling sequence number k′, denoted respectively are w ₁ (k') and w ₂ (k'), where k'=1, 2, ..., K', K' represents the frequency domain sampling times;

符号“||”为绝对值符号，|w₁(k′)|表示第一根接收天线上的信号频谱w₁(k′)的信号频谱包络，|w₂(k′)|表示第二根接收天线上的信号频谱w₂(k′)的信号频谱包络；④ According to the signal spectrum y ₁ (k′) on the first receiving antenna and the signal spectrum w ₂ (k′) on the second receiving antenna, calculate the cross-correlation coefficient of the signal spectrum envelope, denoted as r, $r=\frac{\frac{1}{K^{\′}}\underset{k^{\′}=1}{\overset{K^{\′}}{\mathrm{\Σ}}}\left|w_1\left(k^{\′}\right)w_2\left(k^{\′}\right)\right|-{\mathrm{\μ}}_1{\mathrm{\μ}}_2}{\sqrt{\left(\frac{1}{K_{\′}}\underset{k^{\′}=1}{\overset{K^{\′}}{\mathrm{\Σ}}}{\left(\right|w_1\left(k^{\′}\right)|-{\mathrm{\μ}}_1)}^2\right)\left(\frac{1}{K^{\′}}\underset{k^{\′}=1}{\overset{K^{\′}}{\mathrm{\Σ}}}{\left(\right|w_2\left(k^{\′}\right)|-{\mathrm{\μ}}_2)}^2\right)}},$ in, ${\mathrm{\μ}}_1=\frac{1}{K^{\′}}\underset{k^{\′}=1}{\overset{K^{\′}}{\mathrm{\Σ}}}\left|w_1\left(k^{\′}\right)\right|,$

The symbol "||" is an absolute value symbol, |w ₁ (k′)| represents the signal spectrum envelope of the signal spectrum w ₁ (k′) on the first receiving antenna, and |w ₂ (k′)| represents the The signal spectrum envelope of the signal spectrum w ₂ (k') on the two receiving antennas;

⑤ Calculate the decision threshold according to the number of sampling times K′ in the frequency domain, which is denoted as λ, λ=Q ^-1 (1-P _f ), where P _f represents the false alarm probability, and Q ^-1 (1-P _f ) is Q The inverse function of (1-P _f ), u is the integral variable;

⑥Comparing the cross-correlation coefficient r of the signal spectrum envelope and the size of the decision threshold λ, if the cross-correlation coefficient r of the signal spectrum envelope is greater than or equal to the judgment threshold λ, it is determined that other wireless communication services are occupying the frequency band; if the signal spectrum envelope If the cross-correlation coefficient r is smaller than the decision threshold λ, it is determined that other wireless communication services do not occupy the frequency band.

Compared with the prior art, the advantage of the present invention is that two receiving antennas are used to receive time domain continuous radio frequency signals, and then the time domain continuous radio frequency signals are down-converted and time domain sampled to obtain time domain discrete baseband signals. Discrete baseband signals in the time domain are subjected to discrete Fourier transform to obtain the signal spectrum, and then calculate the cross-correlation coefficient of the signal spectrum envelope, and finally determine whether there are other wireless communication services by comparing the cross-correlation coefficient of the signal spectrum envelope with the judgment threshold The frequency band realizes spectrum sensing, and the method of the present invention overcomes the shortcomings of the existing covariance matrix detection method and eigenvalue detection method when the time domain correlation between signals received by multiple antennas is low or irrelevant.

Description of drawings

Fig. 1 is the block flow diagram of spectrum sensing method of the present invention;

Fig. 2 is a schematic diagram of spectrum sensing performance comparison between the method of the present invention and the existing covariance matrix detection method under different signal-to-noise ratios.

Detailed ways

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

A kind of spectrum sensing method based on the signal spectrum envelope proposed by the present invention, its flow chart is as shown in Figure 1, and it mainly comprises the following steps:

① Assume that the cognitive radio system uses two receiving antennas to receive time-continuous radio frequency signals, and express the time-domain continuous radio frequency signals received by the two receiving antennas as a function of time t. The time domain continuous radio frequency signal is denoted as x ₁ (t), and the time domain continuous radio frequency signal received by the second receiving antenna is denoted as x ₂ (t).

②Respectively perform down-conversion processing on the time domain continuous radio frequency signal x ₁ (t) received by the first receiving antenna and the time domain continuous radio frequency signal x ₂ (t) received by the second receiving antenna, and then respectively The time-domain continuous radio frequency signals obtained after x ₁ (t) and x ₂ (t) are subjected to down-conversion processing are subjected to K times of time-domain sampling to obtain the time-domain discrete baseband signals on the first receiving antenna and the second receiving antenna For the time-domain discrete baseband signal on the antenna, the time-domain discrete baseband signal on the first receiving antenna and the time-domain discrete baseband signal on the second receiving antenna are expressed as functions of the time-domain sampling sequence number k, respectively Denote as y ₁ (k) and y ₂ (k), where k=1, 2, . . . , K, K represents the number of time-domain sampling.

③Respectively perform discrete Fourier transform on the time-domain discrete baseband signal y ₁ (k) on the first receiving antenna and the time-domain discrete baseband signal y ₂ (k) on the second receiving antenna to obtain the first receiving antenna The signal spectrum on the antenna and the signal spectrum on the second receiving antenna, the signal spectrum on the first receiving antenna and the signal spectrum on the second receiving antenna are expressed as a function of the frequency domain sampling sequence number k′, denoted respectively are w ₁ (k′) and w ₂ (k′), where k′=1, 2, . . . , K′, K′ represents the frequency domain sampling times. Here, performing discrete Fourier transform on the time domain discrete baseband signal is essentially performing discrete frequency domain sampling on the time domain discrete baseband signal.

符号“||”为绝对值符号，|w₁(k′)|表示第一根接收天线上的信号频谱w₁(k′)的信号频谱包络，|w₂(k′)|表示第二根接收天线上的信号频谱w₂(k′)的信号频谱包络。④ According to the signal spectrum w ₁ (k′) on the first receiving antenna and the signal spectrum w ₂ (k′) on the second receiving antenna, calculate the cross-correlation coefficient of the signal spectrum envelope, denoted as r, $r=\frac{\frac{1}{K^{\′}}\underset{k^{\′}=1}{\overset{K^{\′}}{\mathrm{\Σ}}}\left|w_1\left(k^{\′}\right)w_2\left(k^{\′}\right)\right|-{\mathrm{\μ}}_1{\mathrm{\μ}}_2}{\sqrt{\left(\frac{1}{K_{\′}}\underset{k^{\′}=1}{\overset{K^{\′}}{\mathrm{\Σ}}}{\left(\right|w_1\left(k^{\′}\right)|-{\mathrm{\μ}}_1)}^2\right)\left(\frac{1}{K^{\′}}\underset{k^{\′}=1}{\overset{K^{\′}}{\mathrm{\Σ}}}{\left(\right|w_2\left(k^{\′}\right)|-{\mathrm{\μ}}_2)}^2\right)}},$ in, ${\mathrm{\μ}}_1=\frac{1}{K^{\′}}\underset{k^{\′}=1}{\overset{K^{\′}}{\mathrm{\Σ}}}\left|w_1\left(k^{\′}\right)\right|,$

The symbol "||" is an absolute value symbol, |w ₁ (k′)| represents the signal spectrum envelope of the signal spectrum w ₁ (k′) on the first receiving antenna, and |w ₂ (k′)| represents the The signal spectrum envelope of the signal spectrum w ₂ (k′) on the two receiving antennas.

u为积分变量。⑤ Calculate the decision threshold according to the number of sampling times K′ in the frequency domain, which is denoted as λ, λ=Q ^-1 (1-P _f ), where P _f represents the false alarm probability, and Q ^-1 (1-P _f ) is Q The inverse function of (1-P _f ),

u is the integration variable.

⑥Comparing the cross-correlation coefficient r of the signal spectrum envelope and the size of the decision threshold λ, if the cross-correlation coefficient r of the signal spectrum envelope is greater than or equal to the judgment threshold λ, it is determined that other wireless communication services are occupying the frequency band; if the signal spectrum envelope If the cross-correlation coefficient r is smaller than the decision threshold λ, it is determined that other wireless communication services do not occupy the frequency band.

In the following, the feasibility and effectiveness of the spectrum sensing method of the present invention will be further illustrated through the computer simulation of the measured digital TV signal.

Assume that the time-domain sampling frequency is 6 MHz, and the time-domain sampling time length is 1.3 ms, that is, the number of time-domain sampling K=7800, and the false alarm probability P _f =0.1. Fig. 2 has provided the detection probability that the measured digital television signal utilizes the existing covariance matrix detection method and the inventive method under different signal-to-noise ratio situations, and the measured digital television signal is to utilize two antennas to collect digital data in different places here. TV signal is obtained. Analyzing Figure 2, it can be seen that due to the long distance between the two antennas, their receiving channels are independent of each other, which causes the received signals on the two antennas to be uncorrelated in the time domain, so the existing covariance matrix The detection method fails. As shown in Figure 2, no matter what the signal-to-noise ratio is, the detection probability of the existing covariance matrix detection method is around 0.1, because the false alarm probability set in the simulation is 0.1. Since the signals received on the two antennas come from the same transmission source, the spectrum envelopes of the received signals on the two antennas are correlated, so when the signal-to-noise ratio is high enough, the method of the present invention has a higher Detection performance. In this simulation, the detection probability can be higher than 0.95 when the signal-to-noise ratio is greater than -4dB.

Claims (1)

1. A spectrum sensing method based on signal spectrum envelope is characterized by comprising the following steps:

firstly, supposing that the cognitive radio system adopts two receiving antennas to receive radio frequency signals with continuous time domains, the radio frequency signals with continuous time domains received by the two receiving antennas are both expressed as a function of time t, and the radio frequency signals with continuous time domains received by the first receiving antenna are recorded as x₁(t), recording the time-domain continuous radio frequency signal received by the second receiving antenna as x₂(t)；

Receiving to the first root respectivelyTime-domain continuous radio frequency signal x received by antenna₁(t) and a second receiving antenna receiving a continuous radio frequency signal x in the time domain₂(t) performing down-conversion treatment on the obtained product, and then respectively comparing the obtained product with x₁(t) and x₂(t) performing time domain sampling for K times on the continuous time domain radio frequency signal obtained after the down-conversion treatment to obtain a time domain discrete baseband signal on the first receiving antenna and a time domain discrete baseband signal on the second receiving antenna, and representing the time domain discrete baseband signal on the first receiving antenna and the time domain discrete baseband signal on the second receiving antenna as functions of time domain sampling sequence numbers K which are respectively recorded as y₁(k) And y₂(k) Where K is 1, 2, …, K denotes the number of time-domain samples;

thirdly, respectively carrying out time domain discrete baseband signals y on the first receiving antenna₁(k) And a time-domain discrete baseband signal y on a second receiving antenna₂(k) Performing discrete Fourier transform to obtain signal spectrum on the first receiving antenna and signal spectrum on the second receiving antenna, representing the signal spectrum on the first receiving antenna and the signal spectrum on the second receiving antenna as functions of frequency domain sampling sequence numbers k', respectively denoted as w₁(k') and w₂(K '), wherein K ' is 1, 2, …, K ' represents the number of frequency domain samples;

fourthly, according to the signal frequency spectrum w on the first receiving antenna₁(k') and the signal spectrum w on the second receiving antenna₂(k') calculating a cross-correlation coefficient, denoted as r, <math> <mrow> <mi>r</mi> <mo>=</mo> <mfrac> <mrow> <mfrac> <mn>1</mn> <msup> <mi>K</mi> <mo>&prime;</mo> </msup> </mfrac> <munderover> <mi>&Sigma;</mi> <mrow> <msup> <mi>k</mi> <mo>&prime;</mo> </msup> <mo>=</mo> <mn>1</mn> </mrow> <msup> <mi>K</mi> <mo>&prime;</mo> </msup> </munderover> <mo>|</mo> <msub> <mi>w</mi> <mn>1</mn> </msub> <mrow> <mo>(</mo> <msup> <mi>k</mi> <mo>&prime;</mo> </msup> <mo>)</mo> </mrow> <msub> <mi>w</mi> <mn>2</mn> </msub> <mrow> <mo>(</mo> <msup> <mi>k</mi> <mo>&prime;</mo> </msup> <mo>)</mo> </mrow> <mo>|</mo> <mo>-</mo> <msub> <mi>&mu;</mi> <mn>1</mn> </msub> <msub> <mi>&mu;</mi> <mn>2</mn> </msub> </mrow> <msqrt> <mrow> <mo>(</mo> <mfrac> <mn>1</mn> <msub> <mi>K</mi> <mo>&prime;</mo> </msub> </mfrac> <munderover> <mi>&Sigma;</mi> <mrow> <msup> <mi>k</mi> <mo>&prime;</mo> </msup> <mo>=</mo> <mn>1</mn> </mrow> <msup> <mi>K</mi> <mo>&prime;</mo> </msup> </munderover> <msup> <mrow> <mo>(</mo> <mo>|</mo> <msub> <mi>w</mi> <mn>1</mn> </msub> <mrow> <mo>(</mo> <msup> <mi>k</mi> <mo>&prime;</mo> </msup> <mo>)</mo> </mrow> <mo>|</mo> <mo>-</mo> <msub> <mi>&mu;</mi> <mn>1</mn> </msub> <mo>)</mo> </mrow> <mn>2</mn> </msup> <mo>)</mo> </mrow> <mrow> <mo>(</mo> <mfrac> <mn>1</mn> <msup> <mi>K</mi> <mo>&prime;</mo> </msup> </mfrac> <munderover> <mi>&Sigma;</mi> <mrow> <msup> <mi>k</mi> <mo>&prime;</mo> </msup> <mo>=</mo> <mn>1</mn> </mrow> <msup> <mi>K</mi> <mo>&prime;</mo> </msup> </munderover> <msup> <mrow> <mo>(</mo> <mo>|</mo> <msub> <mi>w</mi> <mn>2</mn> </msub> <mrow> <mo>(</mo> <msup> <mi>k</mi> <mo>&prime;</mo> </msup> <mo>)</mo> </mrow> <mo>|</mo> <mo>-</mo> <msub> <mi>&mu;</mi> <mn>2</mn> </msub> <mo>)</mo> </mrow> <mn>2</mn> </msup> <mo>)</mo> </mrow> </msqrt> </mfrac> <mo>,</mo> </mrow> </math> wherein, <math> <mrow> <msub> <mi>&mu;</mi> <mn>1</mn> </msub> <mo>=</mo> <mfrac> <mn>1</mn> <msup> <mi>K</mi> <mo>&prime;</mo> </msup> </mfrac> <munderover> <mi>&Sigma;</mi> <mrow> <msup> <mi>k</mi> <mo>&prime;</mo> </msup> <mo>=</mo> <mn>1</mn> </mrow> <msup> <mi>K</mi> <mo>&prime;</mo> </msup> </munderover> <mo>|</mo> <msub> <mi>w</mi> <mn>1</mn> </msub> <mrow> <mo>(</mo> <msup> <mi>k</mi> <mo>&prime;</mo> </msup> <mo>)</mo> </mrow> <mo>|</mo> <mo>,</mo> </mrow> </math>

the symbol "|" is an absolute value symbol, | w₁(k') l represents the signal spectrum w on the first receive antenna₁(k') signal spectral envelope, | w₂(k') l represents the signal spectrum w on the second receiving antenna₂(k') a signal spectral envelope;

calculating judgment threshold according to frequency domain sampling times K', and recording as lambda, lambda being Q^-1(1-P_f) Wherein P is_fIndicating false alarm probability, Q^-1(1-P_f) Is Q (1-P)_f) The inverse function of (a) is,

u is an integral variable;

comparing the cross correlation coefficient r of the signal spectrum envelope with the size of a judgment threshold lambda, and if the cross correlation coefficient r of the signal spectrum envelope is larger than or equal to the judgment threshold lambda, judging that other wireless communication services are occupying the frequency band; and if the cross correlation coefficient r of the signal spectrum envelope is smaller than the judgment threshold lambda, judging that other wireless communication services do not occupy the frequency band.

Images