Broadband signal detection method and system

Abstract

The invention provides a broadband signal detection method and system, and the method comprises the steps: carrying out the multi-path Nyquist uniform sampling of a broadband signal, obtaining a superposition spectrum, detecting a superposition sub-frequency band containing an active sub-channel in the superposition spectrum, and carrying out the power value estimation of the active sub-channel in the superposition sub-frequency band. According to the method, multi-path Nyquist uniform sampling is carried out on the broadband signal, coarse grain detection and fine grain detection are carried out on the sampled superposed spectrum, and only part of power spectrum is calculated, so that the calculation complexity is reduced, and the calculation resources are reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of signal detection, and in particular to a broadband signal detection method; in addition, the present invention also relates to a broadband signal detection system. Background Art

[0002] The working frequency band used by current radar systems has become wider and wider. In radar reconnaissance, in order to ensure high probability of interception of non-cooperative radar signals on ultra-wide frequency bands, radar reconnaissance receivers need to have ultra-wide instantaneous monitoring bands. Currently, NYFR is conventionally used for sampling of broadband signal acquisition and detection. However, the computational complexity of NYFR sampling and detection of broadband signals is high, which consumes a lot of computing resources. Summary of the invention

[0003] In order to solve the problems existing in the prior art, at least one embodiment of the present invention provides a broadband signal detection method, which greatly reduces the computational complexity and alleviates the computational resources. To this end, at least one embodiment of the present invention also provides a broadband signal detection system.

[0004] In a first aspect, an embodiment of the present invention provides a broadband signal detection method, comprising:

[0005] Perform multi-channel Nyquist uniform sampling on the broadband signal to obtain the superposition spectrum;

[0006] Detecting the superimposed sub-frequency bands in the superimposed spectrum that contain active sub-channels;

[0007] Estimate the power values ​​of active sub-channels in the superimposed sub-bands.

[0008] In some embodiments, the present invention provides a broadband signal detection method, which performs multi-path Nyquist uniform sampling on the signal to obtain a superposition spectrum, including:

[0009] The broadband signal is uniformly sampled at a sub-Nyquist rate through multiple sets of parallel analog-to-digital converters.

[0010] In some embodiments, the present invention provides a broadband signal detection method, and the continuous time sampling signal of the broadband signal is expressed by the following formula 1:

[0011] y i (t) = x(t)∑ k∈Z δ(t-(kN+c i )T) Formula 1; Among them, δ(t) is the impulse function, T is the sampling interval, and i is the sampling end sequence number; y i (t) is Fourier transformed and expressed by the following equation 2:

[0012]

[0013] Where i is the sampling end serial number.

[0014] In some embodiments, the present invention provides a broadband signal detection method, and the discrete time signal obtained by the sampling end is expressed by the following formula 3:

[0015]

[0016] Among them, each Nyquist sampling value in the discrete time signal sequence of each sampling end contains N-1 0 values; suppose is a sub-Nyquist sampling sequence with zero value removed, The y value of point F can be obtained by fast Fourier transform i The K-point fast Fourier transform of (n), K=FN, is expressed by the following equation 4:

[0017]

[0018] Among them, y i (n) is a discrete time signal, and i is the sampling end sequence number.

[0019] In some embodiments, the present invention provides a broadband signal detection method, wherein detecting a superimposed sub-frequency band containing an active sub-channel in a superimposed spectrum includes:

[0020] The coarse-grained detection is performed on the superimposed spectrum, and the test is set for each superimposed sub-band by the following formula 5:

[0021]

[0022] Assuming that the signals transmitted on the broadband spectrum are all zero-mean and independent of each other, and the noise is Gaussian white noise, it can be obtained by the following formula 6:

[0023]

[0024] Wherein, a is the number of the occupied subchannel, and 0≤a≤N-1.

[0025] In some embodiments, a broadband signal detection method provided by the present invention estimates the power value of an active sub-channel in a superimposed sub-band, including:

[0026] Fine-grained detection is performed on the active sub-channels in the superimposed sub-bands, and each sub-channel to be detected is represented by the following formula 7:

[0027]

[0028] in, is the power value of the occupied subchannel, for The variance of .

[0029] In a second aspect, an embodiment of the present invention further provides a broadband signal detection system, including:

[0030] A sampling module, used for performing multi-channel Nyquist uniform sampling on broadband signals to obtain a superposition spectrum;

[0031] A detection module, used to detect superimposed sub-frequency bands containing active sub-channels in the superimposed spectrum;

[0032] The power value estimation module is used to estimate the power value of the active sub-channel in the superposition sub-band.

[0033] In a third aspect, an embodiment of the present invention further provides a broadband signal detection device, comprising at least one processor; a memory coupled to the at least one processor, the memory storing executable instructions, which when executed by the at least one processor implement the steps of any one of the methods of the first aspect above.

[0034] In a fourth aspect, an embodiment of the present invention further provides a chip for executing the steps of the method in the first aspect. Specifically, the chip includes: a processor for calling and running a computer program from a memory, so that a device equipped with the chip is used to execute the steps of the method in the first aspect.

[0035] In a fifth aspect, an embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of any one of the methods in the first aspect above are implemented.

[0036] It can be seen that a broadband signal detection method and system in an embodiment of the present invention performs multi-way Nyquist uniform sampling on the broadband signal, and then performs coarse-grained detection and fine-grained detection on the superimposed spectrum after sampling, and only calculates part of the power spectrum, thereby reducing the computational complexity and alleviating computing resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, 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 some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0038] Figure 1 Shown is a schematic diagram of a NERT receiving structure in an embodiment of the present invention;

[0039] Figure 2 Shown is a schematic diagram of the NERT receiving principle in an embodiment of the present invention;

[0040] Figure 3 Shown is a flow chart of a broadband signal detection method according to an embodiment of the present invention;

[0041] Figure 4 Shown is a schematic diagram of an infinite dimensional matrix in an embodiment of the present invention;

[0042] Figure 5 Shown is a schematic diagram of the framework of a broadband signal detection system in an embodiment of the present invention. Specific implementation plan

[0043] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are 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.

[0044] It should be noted that, in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. In this article, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.

[0045] [Example 1]

[0046] The inventor of the present invention has found that in the prior art, NYFR is generally used for sampling broadband signal collection and detection. However, the sampling and detection of broadband signals by NYFR has high computational complexity, thus consuming a large amount of computing resources.

[0047] NYFR receiving technology is an analog information receiving technology based on CS theory. It can theoretically realize the structure of single-channel full-band signal reception under low-speed sampling. Its essence is to use RF pulse train to sample the received signal once in the analog domain, so as to modulate the phase information in the RF LO into each NZ of the received signal, and to distinguish different NZ signals by using the characteristics that the modulation bandwidth and amplitude in different NZ are not consistent. Its secondary sampling is the analog-to-digital signal conversion completed by low-speed ADC, which lays the foundation for subsequent digital signal processing.

[0048] The NYFR receiving structure is a special secondary sampling structure, such as Figure 1 To filter out noise and other signals outside the monitoring bandwidth, the RF input signal is first filtered by a preselected broadband filter and then subjected to low-noise amplification. Subsequently, the signal x(t) is directly converted into a quantization-free RF pulse train p δ (t) performs a sampling, and the sampling pulse train is composed of the template pulse p MODEL (t) trigger is generated, and the trigger time tk is the ZCR point of the LO clock function. Among them, the LO clock function is a phase modulation function. The signal y'(t) after the RF sampling is filtered by a low-pass interpolation filter to obtain the output signal y(t), and finally the low-speed ADC device performs secondary sampling and quantization to obtain the discrete output signal y[n].

[0049] Assume that the RF input signal is processed by two orthogonal paths I and Q to form a complex analog signal x(t). A sampling pulse train is represented by the following equation 8:

[0050]

[0051] Among them, P MODEL (t) is the template pulse, It is the convolution operation.

[0052] Assume that the phase modulated RF LO clock function is a sinusoidal frequency modulation (SFM) function, and the ZCR time t k It can be expressed by the following formula 9:

[0053] ω s t+θ(t)=2πk Equation 9; Among them, ω s is the average sampling angular frequency, and θ(t) is the phase modulation function.

[0054] According to the scaling property of the Dirac sequence, a sampling pulse train can be re-expressed by the following equation 10:

[0055]

[0056] Among them, the requirement for the approximate equality to hold is ωs >>ω θ Therefore, the Fourier spectrum of a sampled pulse train can be expressed by the following equation 11:

[0057] P δ (ω)≈∑ k T k (ω-kω s ) Formula 11;

[0058] The Fourier spectrum of a sampled pulse train is given by ω s Move, that is, the size of NZ is ω s The spectrum of a sampled pulse train in each NZ zone is different, which is related to the index position of the NZ zone and the phase modulation function θ(t). In theory, spectrum monitoring of the entire frequency band can be achieved. From the perspective of signal modulation theory, there is no need to compare with the template pulse p MODEL (t) Convolution operation can also be used to modulate the input signal.

[0059] Taking the sine function as an example of the phase modulation function, a sampling pulse train can be re-expressed by the following equation 12:

[0060]

[0061] The approximate partial sum can be expressed by the following formula 13:

[0062]

[0063] When k = 0, that is, the index value of NZ is 0, there is only one spectrum line at the zero frequency position. s As the center, with the change of NZ index value k; at the same time, its sideband spectrum lines are all ω θ is the interval, and about kω s Symmetrical, and the number of sideband spectral lines is proportional to the NZ index value k.

[0064] The basic principle of NYFR receiving technology is as follows Figure 2 As shown, the total number of NZs covered by the monitoring bandwidth is 5. If the NZ index value of the single-frequency analog signal x(t) is 3, the signal y'(t) is obtained after one RF pulse sampling, completing the spectrum shift and bandwidth expansion of the received signal, and the amplitude change of its extended bandwidth satisfies the Bessel function law. Subsequently, the output signal y(t) with NZ zone modulation information is obtained through low-pass filtering. Compared with the uniformly undersampled output signal, the modulation component is significantly increased in the time-frequency of the NYFR output signal. The type of this modulation component is the same as the type of the clock modulation function, and the size is related to the index value of the NZ zone where the signal is located.

[0065] Let the complex-valued input signal X N×1The vector is within the K NZ zones covered by the monitoring bandwidth, and the frequency domain representation signal vector X based on Nyquist sampling N×1 It has sparse characteristics. Consider Figure 1 The NYFR receiving structure shown in the figure has a discrete output signal vector y. M×1 The measurement matrix and the complex-valued input signal vector x N×1 Obtain, which can be expressed by the following formula 14:

[0066] y M×1 =φ M×N x N×1 Formula 14; Or by the perception matrix Φ M×N and the discrete Fourier spectrum vector X N×1 To obtain, it can be expressed by the following formula 15:

[0067]

[0068] in,

[0069]

[0070] y M×1 =[y1 y2 ... y n ] T

[0071] X N×1 =[X1 X2 ... X M |X M+1 X M+2 ... X 2M |...X N ] T

[0072]

[0073] Square Piece N is the unit FT basis matrix, and its rotation factor is ψ=e -j2π / N . Block diagonal matrix Each block matrix is the unit IFT basis matrix of size M×M. The block diagonal matrix S N×N Each block matrix in the figure shows the modulation effect on the signal in different NZ zones. M×N It is composed of K unit matrices IM to map the NZ area within the monitoring bandwidth to the low-pass filter band. However, the signal recovery process of the NYFR compressed sensing direct measurement matrix model based on the RD structure requires the estimation of the NZ area where the input signal is located first.

[0074] Embodiment 1 of the present invention provides the following solution:

[0075] As Figure 3 shown, this embodiment provides a broadband signal detection method, and the method includes the following steps:

[0076] Step 101: Perform multi-channel sub-Nyquist uniform sampling on the broadband signal to obtain a superimposed spectrum.

[0077] In some embodiments, the broadband spectrum to be detected is regarded as being composed of a set of continuous non-overlapping narrow sub-channels superimposed. Let the spectrum range occupied by the entire broadband be in [0, f max , and let b represent the bandwidth of a sub-channel. Suppose there are p occupied sub-channels in the broadband to be detected, and the signals in each occupied sub-channel are all zero-mean and independent of each other. The signal on the i-th sub-channel is expressed by Equation 16 below:

[0078]

[0079] where a i (n) is a modulation symbol, g i (t) is a pulse shaping equation, f i is the carrier frequency of the signal s i (t).

[0080] In some embodiments, the broadband signal is uniformly sampled at the sub-Nyquist rate by multiple groups of parallel analog-to-digital converters.

[0081] It should be noted that multi-channel sampling uses M groups of parallel analog-to-digital converters. Each group of analog-to-digital converters, that is, each sampling end, uniformly samples the signal at the sub-Nyquist rate of 1 / NT, where T represents the Nyquist sampling interval. Here, r represents the Nyquist sampling interval. When M < N, multi-channel sampling satisfies sub-Nyquist sampling. Set the time offset of the i-th sampling end to c i T, and there is 0 ≤ c0 < c1 <... < c M-1 < N - 1.

[0082] In some embodiments, after the broadband signal undergoes multi-channel uniform sampling, the continuous-time sampling signal of the i-th sampling end is expressed by Equation 1 below:

[0083] y i (T) = x(T) ∑ k∈Z δ(t - (kN + c i )T) Equation 1; where δ(T) is an impulse function, T is the sampling interval, and i is the serial number of the sampling end; perform Fourier transform on y i (t) and express it by Equation 2 below:

[0084]

[0085] Where i is the sampling end serial number.

[0086] The 1 / NT Nyquist sampling rate causes spectrum overlap, and the entire broadband g is evenly divided into N sub-channels. i=0,1,…,N-1,sub-band g1,g2,…g N-1 All are superimposed on g0 to form a superposition spectrum.

[0087] Under the multi-channel uniform sampling mechanism, the discrete time signal obtained at the i-th sampling end is expressed by the following equation 3:

[0088]

[0089] Among them, each Nyquist sampling value in the discrete time signal sequence of each sampling end contains N-1 0 values; suppose is a sub-Nyquist sampling sequence with zero value removed, The y value of point F can be obtained by fast Fourier transform i The K-point fast Fourier transform of (n), K=FN, is expressed by the following equation 4:

[0090]

[0091] Among them, y i (n) is a discrete time signal, and i is the sampling end sequence number.

[0092] Step 102: Detect the superimposed sub-frequency bands containing active sub-channels in the superimposed spectrum.

[0093] In some embodiments, the superimposed spectrum is subjected to coarse-grained detection, and for each superimposed sub-band, the following

[0094] Formula 5 sets the test:

[0095]

[0096] Assuming that the signals transmitted on the broadband spectrum are all zero-mean and independent of each other, and the noise is Gaussian white noise, it can be obtained by the following formula 6:

[0097]

[0098] Where a is the number of the occupied subchannel, and 0≤a≤N-1.

[0099] in for The variance of . Let the test statistic in the test problem represent the sum of the energies of each sub-channel superimposed on the same sub-band. Therefore, the essence of coarse-grained detection is energy detection, and energy detection requires only a small amount of computational consumption.

[0100] When the false alarm rate is determined, the detection threshold is calculated by the following formula 18:

[0101]

[0102] in, Each superimposed sub-band is tested. When , it is determined that the superposition sub-band k is occupied, and the number k is stored in the set G.

[0103] Step 103: Estimate the power value of the active sub-channel in the superposition sub-band.

[0104] In some embodiments, fine-grained detection is performed on active sub-channels in the superimposed sub-bands, and each sub-channel to be detected is represented by the following formula 7:

[0105]

[0106] in, is the power value of the occupied subchannel, for The threshold for fine particle detection is calculated by the following formula 19:

[0107]

[0108] in, The occupancy of each subchannel is shown as follows: Then subchannel (nF+1) is occupied, and other subchannels in the broadband are considered to be occupied.

[0109] In some other embodiments, it is assumed that spectrum sensing is performed on x(t) under a wideband signal. The bandwidth of the wideband signal spectrum is 1 / T, that is, 1 / T also corresponds to the Nyquist rate of the wideband signal, and the wideband signal is composed of U continuous non-overlapping narrowband sub-channels.

[0110] The signal x(t) is sampled using a multi-channel uniform sampling mechanism. The multi-channel uniform sampling uses parallel analog-to-digital converters to uniformly sample the signal. The sampling rate of each group of samplers is lower than the Nyquist rate, and the multiple (factor) of the sampling rate reduction is set to N. When the number of sampling terminals satisfies M<N, the multi-channel sampling is sub-Nyquist sampling.

[0111] The discrete signal obtained from the sampling terminal i is expressed by the following equation 20:

[0112]

[0113] y i The U-point FFT of (n) can be obtained by i The FFT of point F of (k) is obtained and is expressed by the following equation 21:

[0114]

[0115] The special structure of the superposition spectrum under multi-channel uniform sampling, the special ADFS structure of the superposition spectrum obtained after multi-channel uniform sub-Nyquist sampling. ADFS is defined as almost disjoint sub-channels.

[0116] When each occupied superimposed sub-band in the superimposed spectrum X contains only one occupied sub-channel, the superimposed spectrum is said to have an ADFS structure, which is expressed by the following equation 22:

[0117]

[0118] Among them, X n (k) is the occupied subchannel.

[0119] Assume that the number of occupied subchannels, that is, the sparsity, is D. When the total number of subchannels in the broadband to be detected U>D satisfies and the number of superimposed sub-bands in the superimposed spectrum satisfies The probability of occupied subchannels being superimposed on the same sub-band approaches 0. Here, It means that F is better than and at least approximately

[0120] Since the superposition spectrum must satisfy the condition U>D if there is an ADFS structure, In general broadband spectrum sensing applications, it is necessary to determine whether the ADFS structure of the superimposed spectrum exists, that is, the condition U>D. Is it easy to satisfy?

[0121] It should be noted that broadband spectrum is usually sparse, that is, U>D, which is also a prerequisite for most sub-Nyquist sampling broadband perception. It means that F is better than and at least Approximately Since U=FU, it is equivalent to That is, the sampling rate reduction factor N is not greater than In fact, in the sub-Nyquist sampling broadband sensing algorithm, the sampling rate reduction factor N will not be too high. Because the larger N is, the larger the interval between the sampling points is, which weakens the correlation between the sampling points, which will affect the performance of the algorithm. It is understandable that when the parameters of broadband spectrum sensing are set reasonably and in line with the actual situation, the ADFS structure usually exists.

[0122] Under the ADFS structure, the superimposed sub-bands in the superimposed spectrum obtained at each sampling end have similar amplitudes, that is, |Y o (k)|≈|Y1(k)|≈···≈|Y M-1 (k)|. In the superposition sub-band detection, the superposition test statistic is set to Assume that the signals in the occupied subchannels are all zero mean and independent of each other, and that there is Gaussian white noise in the broadband signal, and the noise power spectral density is σ 2 For each free (unoccupied) overlay sub-band, there is a distribution:

[0123] For each superimposed sub-band, the distribution of its test statistic is expressed by the following equation 23:

[0124]

[0125] Among them, H 0,k There is no occupied sub-channel in the superimposed sub-band k, H 1,k is the superposition sub-band k is occupied, and γ is the signal-to-noise ratio.

[0126] Under the ADFS structure, when a superimposed sub-band is detected as occupied, it is necessary to identify which of the sub-channels superimposed on the sub-band is occupied, which is a classification task.

[0127] Under the Gaussian white noise assumption, the optimal classification result can be obtained by finding the maximum absolute value of the inner product between Y(k) and the phase vector, and as long as the N phase vectors p(n) = 0, 1, ..., N-1 are different from each other, the method can still be used even if the number of samplers is as low as M = 2. Under the ADFS structure, the smaller the correlation coefficient between the phase vectors, the more accurate the classification method result: If you want to reduce the correlation coefficient A, you can do so by increasing the number of samplers.

[0128] In other embodiments, since the sampling frequencies of the channels satisfy a coprime relationship, the information redundancy between the sampling channels is reduced, thereby further reducing the average data rate of the system. The continuous baseband spectrum of each channel is estimated using super-resolution theory to obtain a gridless spectrum model, which can accurately estimate multiple targets in the scene with an accuracy that exceeds that of discrete grid units, and can overcome the spectrum leakage effect caused by finite-length data by solving a simple semidefinite programming problem to achieve the purpose of breaking through the Rayleigh limit. The baseband spectrum of each channel is then periodically extended at its own sampling frequency to extract frequency points where signals may exist, namely, the candidate frequency support set.

[0129] Specifically, suppose there are K narrowband signals in space, and the carrier frequencies of the signals are {f1, f2, ..., f k}, these signals fall within the frequency band of interest [0, F MAX ), can be written as a complex exponential superposition and expressed by the following formula 24:

[0130]

[0131] Among them, a m (t) is the baseband component of the mth signal. Since the signal is assumed to be narrowband, its baseband component can be approximately considered to contain only the DC component, that is, a m (t) has no significant change in multiple carrier frequency cycles, so a m (t) can be directly used as constant a m In addition, X(f) is used to represent the Fourier transform of x(t). Let x(t) pass through P sampling channels, and the sampling frequency of each sampling channel is F si =M nyq (i=1, 2, ..., P) i (i=1,2,…,P) are integers that are mutually prime, and let F si =M nyq Δf,M nyq M i The sampling signal of the i-th channel is expressed by the following equation 25:

[0132]

[0133] Performing Fourier transform on both sides of the equal sign to obtain the spectrum of the i-th channel sampling signal is expressed by the following equation 26:

[0134]

[0135] Since the original signal is undersampled in the time domain, the spectrum after sampling will produce aliasing, and the frequency spectrum is F si For the extension of the cycle, it is obvious that X i All information of (f) is stored in [0,F si ), called the baseband of the i-th channel.

[0136] The sampling frequency is selected correctly, then the sampling rate of the i-th channel is F si =M i Δf, the highest frequency of the frequency band of interest is F max =M nyq Δf. Assuming it is a fixed value here, the system sampling rate can be expressed as

[0137] When the highest frequency F in the frequency band of interest max =Mnyq When Δf is fixed, if and only if the number of sampling channels For When the nearest coprime numbers are reached, the system sampling frequency reaches the minimum.

[0138] Baseband super-resolution frequency estimation. After undersampling, all frequency components in the i-th channel will be aliased to the baseband [0,F si ], the baseband ambiguity frequency is Assume that the time span of the sampled signal is [0, T). Here, [0, T) is divided into L sub-time periods, namely [0, T / L), ..., [(L-1)T / L, T). Although such a division will reduce the final processing performance, it will greatly reduce the amount of calculation. L is a compromise factor between the two. Ideally, within the sub-time period, the baseband spectrum of the signal can be written as a superposition of multiple line spectra, which is expressed by the following equation 27:

[0139]

[0140] In the i-th channel, let the number of time domain samples in each sub-time period be 2N i +1=odd(TF si / L) time domain samples, where odd(·) returns the largest odd number that is not greater than the input positive number. Write the data of the L sub-time periods into vector form respectively.

[0141] It should be noted that the traditional method of solving the full variance norm can process each sub-time period one by one, but this does not fully utilize the relationship between the sub-time periods. In order to reduce the amount of calculation while taking into account the algorithm estimation performance, the idea of ​​group full variance norm is adopted to process the data of each sub-time period in parallel. It is precisely because of the introduction of this norm that the discrete grid mismatch problem existing in the traditional discrete norm optimization is overcome, so that this method can estimate the point target in the scene with an accuracy beyond the grid unit. The optimization of the direct solution is an uncertain polynomial problem.

[0142] In other embodiments, the signal amplitude at the fuzzy frequency point corresponding to each time period has little effect on the final frequency result, so the spectrum line amplitude of each signal in each channel in each time period is averaged, which is expressed by the following formula 28:

[0143]

[0144] The relationship between and x is expressed by the following equation 29:

[0145]

[0146] Where Qi is an ∞(F si)×∞(F max ) infinite - dimensional matrix, whose structure is as Figure 4 shown, where the elements on the diagonal of each sub - matrix are 1 and the other elements are 0. Since the sampling frequencies of each channel satisfy the co - prime relationship, the size of the alternative support set can be further reduced.

[0147] It should be noted that the essence of coarse - grained detection is energy detection that utilizes the channel overlap characteristic. Coarse - grained detection is used to further compress the sub - Nyquist sampling values with active user information, thereby further reducing the scale of the power spectrum to be reconstructed before fine - grained detection, thus greatly reducing the computational complexity.

[0148]

Embodiment 2

[0149] As Figure 5 shown, this embodiment provides a broadband signal detection system, including:

[0150] A sampling module 201, configured to perform multi - channel sub - Nyquist uniform sampling on the broadband signal to obtain a superimposed spectrum.

[0151] In some embodiments, the broadband spectrum to be detected is regarded as being composed of a set of continuous non - overlapping narrow - band sub - channels superimposed. Let the entire broadband occupy a spectrum range of [0, f max , and let b represent the bandwidth of a sub - channel. Assume that there are p occupied sub - channels in the broadband to be detected, and the signals in each occupied sub - channel are zero - mean and independent of each other. The signal on the i - th sub - channel is represented by the following equation (16):

[0152]

[0153] where a i (n) is a modulation symbol, g i (t) is a pulse - shaping equation, and f i is the carrier frequency of the signal s i (t).

[0154] In some embodiments, the broadband signal is uniformly sampled at a sub - Nyquist rate through multiple groups of parallel analog - to - digital converters.

[0155] It should be noted that multi - channel sampling uses M groups of parallel analog - to - digital converters. Each group of analog - to - digital converters, that is, each sampling end, uniformly samples the signal at a sub - Nyquist rate of 1 / NT, where T represents the Nyquist sampling interval. Here, r represents the Nyquist sampling interval. When M < N, multi - channel sampling satisfies sub - Nyquist sampling. Let the time offset of the i - th sampling end be c i T, and there is 0 ≤ c0 < c1 <... < c M-1 < N - 1.

[0156] In some embodiments, the broadband signal is uniformly sampled in multiple channels, and the continuous time sampling signal of the i-th sampling end is expressed by the following formula 1:

[0157] y i (t) = x(t)∑ k∈Z δ(t-(kN+c i )T) Formula 1;

[0158] Among them, δ(t) is the impulse function, T is the sampling interval, and i is the sampling end sequence number; y i (t) is Fourier transformed and expressed by the following equation 2:

[0159]

[0160] Where i is the sampling end serial number.

[0161] The 1 / NT Nyquist sampling rate causes spectrum overlap, and the entire broadband g is evenly divided into N sub-channels. i=0,1,…,N-1,sub-band g1,g2,…g N-1 All are superimposed on g0 to form a superposition spectrum.

[0162] Under the multi-channel uniform sampling mechanism, the discrete time signal obtained at the i-th sampling end is expressed by the following equation 3:

[0163]

[0164] Among them, each Nyquist sampling value in the discrete time signal sequence of each sampling end contains N-1 0 values; suppose is a sub-Nyquist sampling sequence with zero value removed, The y value of point F can be obtained by fast Fourier transform i The K-point fast Fourier transform of (n), K=FN, is expressed by the following equation 4:

[0165]

[0166] Among them, y i (n) is a discrete time signal, and i is the sampling end sequence number.

[0167] The detection module 202 is configured to detect the superimposed sub-frequency bands in the superimposed spectrum that contain active sub-channels.

[0168] In some embodiments, the superimposed spectrum is subjected to coarse-grained detection, and for each superimposed sub-band, the following

[0169] Formula 5 sets the test:

[0170]

[0171] Assuming that the signals transmitted on the broadband spectrum are all zero-mean and independent of each other, and the noise is Gaussian white noise, it can be obtained by the following formula 6:

[0172]

[0173] Where a is the number of the occupied subchannel, and 0≤a≤N-1.

[0174] Formula 5 can be transformed into the following formula 17:

[0175]

[0176] in for The variance of . Let the test statistic in the test problem represent the sum of the energies of each sub-channel superimposed on the same sub-band. Therefore, the essence of coarse-grained detection is energy detection, and energy detection requires only a small amount of computational consumption.

[0177] When the false alarm rate is determined, the detection threshold is calculated by the following formula 18:

[0178]

[0179] in, Each superimposed sub-band is tested. When , it is determined that the superposition sub-band k is occupied, and the number k is stored in the set G.

[0180] The power value estimation module 203 is used to estimate the power value of the active sub-channel in the superposition sub-frequency band.

[0181] In some embodiments, fine-grained detection is performed on active sub-channels in the superimposed sub-bands, and each sub-channel to be detected is represented by the following formula 7:

[0182]

[0183] in, is the power value of the occupied subchannel, for The threshold for fine particle detection is calculated by the following formula 19:

[0184]

[0185] in, The occupancy of each subchannel is shown as follows: Then subchannel (nF+1) is occupied, and other subchannels in the broadband are considered to be occupied.

[0186] It should be noted that the essence of coarse-grained detection is energy detection that utilizes the channel overlap characteristics. Coarse-grained detection is used to further compress the sub-Nyquist sampling values ​​with active user information, thereby further reducing the scale of the power spectrum that needs to be reconstructed before fine-grained detection, thereby greatly reducing the computational complexity.

[0187] [Example 3]

[0188] This embodiment provides a broadband signal detection device, including:

[0189] At least one processor; a memory coupled to the at least one processor, the memory storing executable instructions, wherein the executable instructions, when executed by the at least one processor, implement the method steps of embodiment 1 of the present invention.

[0190] In a broadband signal detection device provided by an embodiment of the present invention, a processor and a memory may be provided separately or integrated together.

[0191] For example, the memory may include a random access memory, a flash memory, a read-only memory, a programmable read-only memory, a non-volatile memory or a register, etc. The processor may be a central processing unit (CPU), etc. or a graphic processing unit (GPU). The memory may store executable instructions. The processor may execute the executable instructions stored in the memory to implement the various processes described herein.

[0192] It can be understood that the memory in this embodiment can be a volatile memory or a non-volatile memory, or can include both volatile and non-volatile memories. Among them, the non-volatile memory can be a ROM (Read-Only Memory), a PROM (Programmable ROM), an EPROM (Erasable PROM), an EEPROM (Electrically EPROM), or a flash memory. The volatile memory can be a RAM (Random Access Memory), which is used as an external cache. By way of example but not limitation, many forms of RAM are available, such as SRAM (Static RAM), DRAM (Dynamic RAM), SDRAM (Synchronous DRAM), DDR SDRAM (Double Data Rate SDRAM), ESDRAM (Enhanced SDRAM), SLDRAM (Synchlink DRAM), and DRRAM (Direct Rambus RAM). The memories described herein are intended to include, but are not limited to, these and any other suitable types of memories.

[0193] In some embodiments, the memory stores the following elements, upgrade packages, executable units or data structures, or a subset thereof, or an extended set thereof: an operating system and an application program.

[0194] The operating system includes various system programs, such as a framework layer, a core library layer, a driver layer, etc., which are used to implement various basic services and process hardware-based tasks. The application program includes various application programs, which are used to implement various application services. The program for implementing the method of the embodiment of the present invention can be included in the application program.

[0195] In the embodiment of the present invention, the processor calls a program or instruction stored in the memory, specifically, a program or instruction stored in an application program, and the processor is used to execute the method steps of the embodiment 1 of the present invention.

[0196] [Example 4]

[0197] This embodiment provides a chip for executing the method of the above-mentioned embodiment 1 of the present invention. Specifically, the chip includes: a processor for calling and running a computer program from a memory, so that a device equipped with the chip is used to execute the method of the embodiment 1 of the present invention.

[0198] [Example 5]

[0199] This embodiment provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method of Embodiment 1 of the present invention are implemented.

[0200] For example, machine-readable storage media may include, but are not limited to, various known and unknown types of non-volatile memory.

[0201] In summary, embodiments 1-5 of the present invention provide a broadband signal detection method and system, which performs multi-way Nyquist uniform sampling on the broadband signal, and then performs coarse-grained detection and fine-grained detection on the sampled superposition spectrum, and only calculates part of the power spectrum, thereby reducing the computational complexity and alleviating computing resources.

[0202] It will be appreciated by those skilled in the art that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art may use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0203] In the embodiments of the present application, the disclosed systems, devices and methods can be implemented in other ways. For example, the division of units is only a logical function division, and there can be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system. In addition, the coupling between the various units can be direct coupling or indirect coupling. In addition, the various functional units in the embodiments of the present application can be integrated into a processing unit, or can be separate physical existences, etc.

[0204] It should be understood that in the various embodiments of the present application, the size of the serial number of each process does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0205] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a machine-readable storage medium. Therefore, the technical solution of the present application can be embodied in the form of a software product, which can be stored in a machine-readable storage medium, which can include several instructions to enable an electronic device to perform all or part of the technical solution described in the embodiments of the present application. The above-mentioned storage medium may include various media that can store program codes, such as ROM, RAM, removable disk, hard disk, magnetic disk or optical disk.

[0206] The above contents are only specific implementation methods of the present application, and the protection scope of the present application is not limited thereto. Those skilled in the art may make changes or substitutions within the technical scope disclosed in the present application, and these changes or substitutions shall be within the protection scope of the present application.

Claims

1. A broadband signal detection method, characterized in that: include: Perform multi-channel Nyquist uniform sampling on the broadband signal to obtain the superposition spectrum; Detecting the superimposed sub-frequency bands in the superimposed spectrum that contain active sub-channels; A power value is estimated for the active sub-channel in the superimposed sub-frequency band.

2. The broadband signal detection method according to claim 1, characterized in that: The performing multi-channel Nyquist uniform sampling on the signal to obtain a superposition spectrum comprises: The broadband signal is uniformly sampled at a sub-Nyquist rate by multiple sets of parallel analog-to-digital converters.

3. The broadband signal detection method according to claim 2, characterized in that: The continuous time sampling signal of the broadband signal is expressed by the following equation 1: y t (t) = x(t) ∑ k∈Z δ(t - (kN + c i ))T) Equation 1; Among them, δ(t) is the impulse function, T is the sampling interval, and i is the sampling end sequence number; y i (t) is Fourier transformed and expressed by the following equation 2: Where i is the sampling end serial number.

4. The broadband signal detection method according to claim 3, characterized in that: The discrete time signal obtained at the sampling end is expressed by the following equation 3: Among them, each Nyquist sampling value in the discrete time signal sequence of each sampling end contains N-1 0 values; suppose is a sub-Nyquist sampling sequence with zero value removed, The y value of point F can be obtained by fast Fourier transform i The K-point fast Fourier transform of (n), K=FN, is expressed by the following equation 4: Among them, y i (n) is a discrete time signal, and i is the sampling end sequence number.

5. The broadband signal detection method according to claim 1, characterized in that: The detecting whether the superimposed sub-frequency band in the superimposed spectrum contains an active sub-channel comprises: The superimposed spectrum is subjected to coarse-grained detection, and the test is set for each superimposed sub-band by the following formula 5: Assuming that the signals transmitted on the broadband spectrum are all zero-mean and independent of each other, and the noise is Gaussian white noise, it can be obtained by the following formula 6: Wherein, a is the number of the occupied subchannel, and 0≤a≤N-1.

6. The broadband signal detection method according to claim 1, characterized in that: The estimating the power value of the active sub-channel in the superposition sub-frequency band includes: Fine-grained detection is performed on the active sub-channels in the superimposed sub-frequency band, and each sub-channel to be detected is represented by the following formula 7: in, is the power value of the occupied subchannel, for The variance of .

7. A broadband signal detection system, characterized in that: include: A sampling module is used to perform multi-channel Nyquist uniform sampling on broadband signals to obtain a superposition spectrum; A detection module, used to detect the superimposed sub-frequency band containing active sub-channels in the superimposed spectrum; The power value estimation module is used to estimate the power value of the active sub-channel in the superposition sub-frequency band.

8. A broadband signal detection device, comprising at least one processor; a memory coupled to the at least one processor, the memory storing executable instructions, characterized in that: The executable instructions, when executed by the at least one processor, enable the steps of the method according to any one of claims 1 to 6 to be implemented.

9. A chip, characterized in that: It comprises a processor, which is used to call and run a computer program from a memory, so that a device equipped with the chip executes the steps of the method as claimed in any one of claims 1 to 6.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images