Non-circular signal spectrum sensing method and system

Abstract

The invention discloses a non-circular signal spectrum sensing method and system, and the method comprises the steps: calculating a standard sample covariance matrix and a compensation sample covariance matrix based on a received signal of a cognitive user, and constructing an augmented sample covariance matrix; calculating a Floribenius norm and a trace of the augmented sample covariance matrix to obtain a local optimal invariant test statistic; setting a false alarm probability, and calculating a corresponding sensing threshold value; and comparing the local optimal invariant test statistic with the sensing threshold, and determining a spectrum sensing result. By adopting the technical scheme of the invention, under the framework of the local optimal invariant test, the statistical magnitude is designed by fully utilizing the characteristics of the non-zero compensation covariance of the non-circular signal, and the detection performance is better.

Description

Technical Field

[0001] The present invention belongs to the technical field of wireless communications, and in particular relates to a method and system for sensing a non-circular signal spectrum. Background Art

[0002] With the growing demand for wireless services and mobile devices, mobile data traffic has increased dramatically, and the scarcity of spectrum resources has become a key issue for future communications. With the help of spectrum sensing technology, the spectrum status of the authorized frequency band can be detected without interfering with the communication of the primary user, and idle spectrum holes can be found, so that cognitive users can access the authorized frequency band for communication by dynamic spectrum access, thereby realizing dynamic management of spectrum resources. The current spectrum sensing algorithms are all based on circular signals, that is, only the standard covariance matrix of the received signal is considered. However, in actual application scenarios, the signal usually exhibits non-circular characteristics, and its compensation covariance is not zero, which greatly limits the perception performance of the existing algorithms. To this end, the researchers took advantage of the feature of non-zero compensation covariance and proposed a Hadamard detector and a local mean variance detector based on non-circular signals. However, the perception performance of these two algorithms still needs to be further improved. Summary of the invention

[0003] The technical problem to be solved by the present invention is to provide a method and system for non-circular signal spectrum perception based on local optimal invariant test.

[0004] To achieve the above object, the present invention adopts the following technical solution:

[0005] A non-circular signal spectrum sensing method, comprising:

[0006] Based on the received signal of the cognitive user, the standard sample covariance matrix and the compensated sample covariance matrix are calculated to construct the augmented sample covariance matrix;

[0007] Calculating the Frobenius norm and trace of the augmented sample covariance matrix to obtain a local optimal invariant test statistic;

[0008] Set the false alarm probability and calculate the corresponding perception threshold;

[0009] The local optimal invariant test statistic and the sensing threshold are compared to determine the result of spectrum sensing.

[0010] As a preference, the augmented sample covariance matrix is ​​constructed as:

[0011]

[0012] in, is the received signal matrix, W is the standard sample covariance matrix, To compensate the sample covariance matrix,W is the augmented sample covariance matrix, with superscript (·) * Indicates the conjugation symbol; superscript (·) H represents the conjugate transposed symbol, M is the number of receiving antennas, and N is the sample length.

[0013] As a preferred method, the formula for obtaining the local optimal invariant test statistic is:

[0014]

[0015] in, W is the augmented sample covariance matrix, || W || F is the Frobenius norm of the augmented sample covariance matrix, tr( W ) is the trace of the augmented sample covariance matrix, T is the local optimal invariant test statistic, and M is the number of receiving antennas.

[0016] Preferably, the corresponding perception threshold is calculated as:

[0017]

[0018]

[0019] Among them, P f is the set false alarm probability, η is the perception threshold, M is the number of receiving antennas, I is the cumulative distribution function of the Beta distribution, and I -1 represents its inverse function, α0 and β0 represent the two parameters of Beta distribution, κ1 and κ2 are the first and second moments of the local optimal invariant test statistic, respectively.

[0020] The present invention also provides a non-circular signal spectrum sensing system, comprising:

[0021] An augmented sample covariance matrix construction module is used to calculate a standard sample covariance matrix and a compensation sample covariance matrix according to a received signal of a cognitive user, and to construct an augmented sample covariance matrix;

[0022] A statistics calculation module, used to calculate the Frobenius norm and trace of the augmented sample covariance matrix to obtain a local optimal invariant test statistic;

[0023] A perception threshold calculation module is used to set the false alarm probability and calculate the corresponding perception threshold;

[0024] The spectrum sensing module is used to compare the local optimal invariant test statistic and the sensing threshold to determine the result of spectrum sensing.

[0025] As a preference, the augmented sample covariance matrix is ​​constructed as:

[0026]

[0027] in, is the received signal matrix, W is the standard sample covariance matrix, To compensate the sample covariance matrix, W is the augmented sample covariance matrix, with superscript (·) * Indicates the conjugation symbol; superscript (·) H represents the conjugate transposed symbol, M is the number of receiving antennas, and N is the sample length.

[0028] As a preferred method, the formula for obtaining the local optimal invariant test statistic is:

[0029]

[0030] in, W is the augmented sample covariance matrix, || W || F is the Frobenius norm of the augmented sample covariance matrix, tr( W ) is the trace of the augmented sample covariance matrix, T is the local optimal invariant test statistic, and M is the number of receiving antennas.

[0031] Preferably, the corresponding perception threshold is calculated as:

[0032]

[0033] Among them, P f is the set false alarm probability, η is the perception threshold, M is the number of receiving antennas, I is the cumulative distribution function of the Beta distribution, and I -1 represents its inverse function, α0 and β0 represent the two parameters of Beta distribution, κ1 and κ2 are the first and second moments of the local optimal invariant test statistic, respectively.

[0034] The present invention calculates the standard sample covariance matrix and the compensation sample covariance matrix based on the received signal of the cognitive user, constructs the augmented sample covariance matrix; calculates the Frobenius norm and trace of the augmented sample covariance matrix, constructs the local optimal invariant test statistic; sets the false alarm probability, calculates the corresponding perception threshold; compares the size of the local optimal invariant test statistic and the perception threshold, and determines the result of spectrum perception. The technical solution of the present invention makes full use of the characteristic design statistic of the non-zero compensation covariance of the non-circular signal under the framework of the local optimal invariant test, and has better detection performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0036] Figure 1 This is a schematic diagram of a flow chart of a non-circular signal spectrum sensing method according to an embodiment of the present invention;

[0037] Figure 2 It is a comparison diagram of detection probabilities of the local optimal invariant test statistic detector, the Hadamard detector, and the local mean variance detector according to an embodiment of the present invention. DETAILED DESCRIPTION

[0038] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0039] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0040] Embodiment 1:

[0041] like Figure 1 As shown, an embodiment of the present invention provides a non-circular signal spectrum sensing method, including:

[0042] Step S1, calculating a standard sample covariance matrix and a compensated sample covariance matrix based on the received signal of the cognitive user, and constructing an augmented sample covariance matrix;

[0043] Step S2, calculating the Frobenius norm and trace of the augmented sample covariance matrix to obtain a local optimal invariant test statistic;

[0044] Step S3, setting the false alarm probability and calculating the corresponding perception threshold;

[0045] Step S4: compare the local optimal invariant test statistic and the sensing threshold to determine the result of spectrum sensing.

[0046] The embodiment of the present invention fully utilizes the characteristic that the compensation covariance of the signal is not 0, and has a higher detection effect in non-circular signal detection.

[0047] As a preferred implementation of the embodiment of the present invention, the received signal of the cognitive user is specifically:

[0048]

[0049] In the formula, H0 indicates that the cognitive user does not exist, that is, the spectrum is in an idle state, H1 indicates that the cognitive user exists, that is, the spectrum is in an occupied state; x(n) represents the received signal of the cognitive user at time n, u(n) represents the source signal sent by the primary user at time n, w(n) represents the noise vector at time n, and H represents the channel gain matrix between the primary user and the cognitive user. In addition,

[0050]

[0051] Where M and P represent the number of antennas configured at the cognitive user receiving end and the primary user transmitting end, respectively. T represents the transpose symbol; represents the channel gain coefficient matrix; x m (n) represents the received signal of the mth receiving antenna at time n, w m (n) represents the noise vector of the mth receiving antenna at time n, which obeys the standard complex Gaussian distribution; u i (n) represents the i-th primary user signal, which obeys the non-circular complex Gaussian distribution, that is, The signal power γ i =E(|u i | 2 ).in addition, τ i represents the signal non-circularity, φ i ∈[-π,π) represents the non-circular phase of the signal.

[0052] As a preferred implementation of the embodiment of the present invention, the calculation of the standard sample covariance matrix and the compensation sample covariance matrix to construct the augmented sample covariance matrix is ​​specifically as follows:

[0053]

[0054] In the formula, is the received signal matrix, W is the standard sample covariance matrix, To compensate the sample covariance matrix, W is the augmented sample covariance matrix, with superscript (·) * Indicates the conjugation symbol; superscript (·) H represents the conjugate transposed symbol, M is the number of receiving antennas, and N is the sample length.

[0055] As a preferred implementation of the embodiment of the present invention, the calculation of the Frobenius norm and trace of the augmented sample covariance matrix and the construction of the local optimal invariant test statistic are specifically:

[0056]

[0057] In the formula, W is the augmented sample covariance matrix, || W || F is the Frobenius norm of the augmented sample covariance matrix, tr( W ) is the trace of the augmented sample covariance matrix, T is the local optimal invariant test statistic, and M is the number of receiving antennas.

[0058] As a preferred implementation of the embodiment of the present invention, the false alarm probability is set and the corresponding perception threshold is calculated using the formula:

[0059]

[0060] Where P f is the set false alarm probability, η is the perception threshold, M is the number of receiving antennas, I is the cumulative distribution function of the Beta distribution, and I -1 represents its inverse function, α0 and β0 represent the two parameters of the Beta distribution, κ1 and κ2 are the first and second moments of the local optimal invariant test statistic, respectively.

[0061]

[0062]

[0063] As a preferred implementation of the embodiment of the present invention, the process of comparing the local optimal invariant test statistic and the sensing threshold to determine the result of spectrum sensing is specifically as follows:

[0064] When the local optimal invariant test statistic is greater than the perception threshold, the spectrum is in an occupied state; when the local optimal invariant test statistic is less than the perception threshold, the spectrum is idle.

[0065] The embodiments of the present invention are described in detail below through simulation experiments.

[0066] Simulation parameter setting: The simulation experiment assumes a Rayleigh fading channel, that is, the column components of the channel gain matrix H obey an independent and identically distributed standard complex Gaussian distribution. Once generated in the current Monte Carlo experiment, they remain unchanged, and will be randomly regenerated in the next Monte Carlo experiment. Under the H0 assumption, the primary user signal does not exist; under the H1 assumption, the primary user signal is set as follows: signal non-circularity τ1 = ··· = τ P =1, non-circular phase φi The uniform distribution is defined in the interval [-π,π). All simulation results are obtained by 10 4 Monte Carlo experiment.

[0067] like Figure 2 As shown in the figure, the detection probability of various sensing algorithms under different signal-to-noise ratio environments is compared. The sample length N = 200, the number of receiving antennas M = 5, the number of primary users P = 3, and the false alarm probability P f =0.1. It can be seen that the proposed local optimal invariant test statistic detector is superior to the Hadamard detector and the local mean variance detector based on non-circular signals, showing better detection performance.

[0068] Embodiment 2:

[0069] The embodiment of the present invention further provides a non-circular signal spectrum sensing system, including:

[0070] An augmented sample covariance matrix construction module is used to calculate a standard sample covariance matrix and a compensated sample covariance matrix according to a received signal of a cognitive user, and to construct an augmented sample covariance matrix;

[0071] A statistics calculation module, used to calculate the Frobenius norm and trace of the augmented sample covariance matrix to obtain a local optimal invariant test statistic;

[0072] A perception threshold calculation module is used to set the false alarm probability and calculate the corresponding perception threshold;

[0073] The spectrum sensing module is used to compare the local optimal invariant test statistic and the sensing threshold to determine the result of spectrum sensing.

[0074] As an implementation method of an embodiment of the present invention, the augmented sample covariance matrix is ​​constructed as:

[0075]

[0076] in, is the received signal matrix, W is the standard sample covariance matrix, To compensate the sample covariance matrix, W is the augmented sample covariance matrix, with superscript (·) * Indicates the conjugation symbol; superscript (·) H represents the conjugate transposed symbol, M is the number of receiving antennas, and N is the sample length.

[0077] As an implementation method of an embodiment of the present invention, the formula for obtaining the local optimal invariant test statistic is:

[0078]

[0079] in, W is the augmented sample covariance matrix, || W || F is the Frobenius norm of the augmented sample covariance matrix, tr( W ) is the trace of the augmented sample covariance matrix, T is the local optimal invariant test statistic, and M is the number of receiving antennas.

[0080] As an implementation of an embodiment of the present invention, the corresponding perception threshold is calculated as:

[0081]

[0082]

[0083] Among them, P f is the set false alarm probability, η is the perception threshold, M is the number of receiving antennas, I is the cumulative distribution function of the Beta distribution, and I -1 represents its inverse function, α0 and β0 represent the two parameters of Beta distribution, κ1 and κ2 are the first and second moments of the local optimal invariant test statistic, respectively.

[0084] The embodiments described above are only descriptions of the preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope determined by the claims of the present invention.

Claims

1. A non-circular signal spectrum sensing method, characterized in that: include: Based on the received signal of the cognitive user, the standard sample covariance matrix and the compensated sample covariance matrix are calculated to construct the augmented sample covariance matrix; Calculating the Frobenius norm and trace of the augmented sample covariance matrix to obtain a local optimal invariant test statistic; Set the false alarm probability and calculate the corresponding perception threshold; The local optimal invariant test statistic and the sensing threshold are compared to determine the result of spectrum sensing.

2. The non-circular signal spectrum sensing method according to claim 1, characterized in that: The augmented sample covariance matrix is ​​constructed as: in, is the received signal matrix, W is the standard sample covariance matrix, To compensate the sample covariance matrix, W is the augmented sample covariance matrix, with superscript (·) * Indicates the conjugation symbol; superscript (·) H represents the conjugate transposed symbol, M is the number of receiving antennas, and N is the sample length.

3. The non-circular signal spectrum sensing method according to claim 2, characterized in that: The formula for obtaining the locally optimal invariant test statistic is: in, W is the augmented sample covariance matrix, || W || F is the Frobenius norm of the augmented sample covariance matrix, tr( W ) is the trace of the augmented sample covariance matrix, T is the local optimal invariant test statistic, and M is the number of receiving antennas.

4. The non-circular signal spectrum sensing method according to claim 3, characterized in that: The corresponding perception threshold is calculated as: Among them, P f is the set false alarm probability, η is the perception threshold, M is the number of receiving antennas, I is the cumulative distribution function of the Beta distribution, and I -1 represents its inverse function, α0 and β0 represent the two parameters of Beta distribution, κ1 and κ2 are the first and second moments of the local optimal invariant test statistic, respectively.

5. A non-circular signal spectrum sensing system, characterized in that: include: An augmented sample covariance matrix construction module is used to calculate a standard sample covariance matrix and a compensation sample covariance matrix according to a received signal of a cognitive user, and to construct an augmented sample covariance matrix; A statistics calculation module, used to calculate the Frobenius norm and trace of the augmented sample covariance matrix to obtain a local optimal invariant test statistic; A perception threshold calculation module is used to set the false alarm probability and calculate the corresponding perception threshold; The spectrum sensing module is used to compare the local optimal invariant test statistic and the sensing threshold to determine the result of spectrum sensing.

6. The non-circular signal spectrum sensing system according to claim 5, characterized in that: The augmented sample covariance matrix is ​​constructed as: in, is the received signal matrix, W is the standard sample covariance matrix, To compensate the sample covariance matrix, W is the augmented sample covariance matrix, with superscript (·) * Indicates the conjugation symbol; superscript (·) H represents the conjugate transposed symbol, M is the number of receiving antennas, and N is the sample length.

7. The non-circular signal spectrum sensing system according to claim 6, characterized in that: The formula for obtaining the locally optimal invariant test statistic is: in, W is the augmented sample covariance matrix, || W || F is the Frobenius norm of the augmented sample covariance matrix, tr( W ) is the trace of the augmented sample covariance matrix, T is the local optimal invariant test statistic, and M is the number of receiving antennas.

8. The non-circular signal spectrum sensing system according to claim 7, characterized in that: The corresponding perception threshold is calculated as: Among them, P f is the set false alarm probability, η is the perception threshold, M is the number of receiving antennas, I is the cumulative distribution function of the Beta distribution, and I -1 represents its inverse function, α0 and β0 represent the two parameters of Beta distribution, κ1 and κ2 are the first and second moments of the local optimal invariant test statistic, respectively.

Images