Method and device for estimating effective burst length of TDMA (Time Division Multiple Access) signal based on amplitude spectrum accumulation

Abstract

The invention discloses a TDMA signal effective burst length estimation method and device based on amplitude spectrum accumulation. Comprising the following steps: carrying out resampling on IQ complex signal data obtained by a TDMA signal to obtain resampling data; performing sliding correlation and peak detection on the resampling data to obtain a peak initial position vector, and obtaining a burst period of the TDMA signal according to the peak initial position vector; according to the peak initial position vector and the burst period, extracting symbol data of each burst in the TDMA signal from the resampling data; and performing accumulation smoothing processing on each symbol data to obtain amplitude spectrum accumulation data, and comparing the amplitude spectrum accumulation data with an amplitude threshold to obtain an effective burst length of the TDMA signal. According to the method, high-precision estimation of the burst period and the effective burst length of the TDMA signal under the condition of low signal-to-noise ratio can be simultaneously realized on the basis of assistance of the synchronous symbol sequence and amplitude spectrum accumulation.

Description

Technical Field

[0001] This invention belongs to the field of communication technology, and in particular relates to a method and apparatus for estimating the effective burst length of a TDMA signal based on amplitude spectrum accumulation. Background Technology

[0002] Time Division Multiple Access (TDMA) divides time into non-overlapping time slots, allowing multiple users to share communication resources. It boasts advantages such as large communication capacity and high frequency utilization, and is therefore widely used in various communication systems. In the analysis and processing of TDMA signals, in addition to parameters such as frequency, bandwidth, symbol rate, and modulation pattern, burst period and effective burst length are also crucial signal parameters, providing a foundation for subsequent protocol analysis.

[0003] Methods for estimating the burst period and effective burst length of TDMA signals generally employ signal energy-based estimation methods. These methods determine the start time of each burst by detecting the signal energy within a detection window in the time domain (or frequency domain). Then, statistical analysis is performed based on the distribution of these burst start times to estimate the burst period and effective burst length. While these methods can estimate the burst period and effective burst length, their accuracy is low, and they are sensitive to noise, making them unsuitable for low signal-to-noise ratio (SNR) conditions. Alternatively, synchronization code-based estimation methods can be used. These methods utilize the autocorrelation properties of the synchronization code to detect and estimate the start time, and then statistical analysis of the start time distribution estimates the burst period of the TDMA signal. While these methods can estimate the burst period at lower SNR conditions, they cannot estimate the effective burst length. Summary of the Invention

[0004] In view of this, embodiments of the present invention provide a method and apparatus for estimating the effective burst length of a TDMA signal based on amplitude spectrum accumulation, with the aim of achieving high-precision estimation of the effective burst length of a TDMA signal.

[0005] To achieve the above objectives, according to a first aspect of the present invention, a method for estimating the effective burst length of a TDMA signal based on amplitude spectrum accumulation is provided, comprising:

[0006] The IQ complex signal data obtained from the received TDMA signal is resampled to obtain resampled data;

[0007] The resampled data is subjected to sliding correlation and peak detection to obtain the peak start position vector, and the burst period of the TDMA signal is obtained based on the peak start position vector;

[0008] Based on the peak start position vector and the burst period, the symbol data of each burst in the TDMA signal is extracted from the resampled data;

[0009] The amplitude spectrum accumulation data is obtained by performing accumulation and smoothing processing on each symbol data, and the amplitude spectrum accumulation data is compared with the amplitude threshold to obtain the effective burst length of the TDMA signal.

[0010] Furthermore, the IQ complex signal data obtained by channel filtering of the TDMA signal according to the signal bandwidth is represented as x(n1), n1 = 1, 2, ..., N, where N is the length of the IQ complex signal data;

[0011] The resampled data y(n2) is obtained according to the following formula:

[0012]

[0013] Where n1=1+(n2-1)·ξ, is the resampling factor. F s and R s These are the sampling rate and symbol rate of the TDMA signal, respectively, and α is the oversampling rate.

[0014] Furthermore, the resampled data is subjected to sliding correlation and peak detection to obtain the peak start position vector, including: performing sliding correlation on the resampled data using a preset synchronization symbol sequence;

[0015] The specific steps of performing sliding correlation on the resampled data using a preset synchronization symbol sequence are as follows:

[0016] Data is extracted from the resampled data according to the oversampling rate to obtain multiple first extracted data sequences;

[0017] The synchronization symbol sequence is used to perform sliding correlation processing with each of the first extracted data sequences to obtain the first correlated data sequence corresponding to each of the first extracted data sequences;

[0018] Compare the data at each position in the first related data sequence, and select the maximum value at each position to form the second related data sequence.

[0019] Furthermore, performing sliding correlation and peak detection on the resampled data to obtain the peak start position vector includes: performing peak detection on the second correlated data sequence to obtain the peak start position vector;

[0020] Peak detection is performed on the second related data sequence to obtain the peak start position vector, specifically:

[0021] Data exceeding a preset detection threshold are detected from the second related data sequence, and the positions corresponding to the data exceeding the preset detection threshold are used to form a first position vector. The adjacent positions in the first position vector are differentially divided to obtain a first position difference vector.

[0022] Differential components that are greater than a preset minimum peak interval threshold are detected from the first position differential vector, and the positions corresponding to the differential components are selected from the first position vector to construct the peak start position vector.

[0023] Furthermore, the burst period of the TDMA signal is obtained based on the peak start position vector, including:

[0024] The second position difference vector is obtained by differentiating the adjacent positions in the peak starting position vector;

[0025] The maximum and minimum values ​​of the positions are selected from the second position difference vector. The search range of the burst cycle is determined based on the maximum and minimum values. The distribution of the second position difference vector within the search range is statistically analyzed. The burst cycle is determined based on the distribution.

[0026] Furthermore, the maximum value and the minimum value are R. max and R min The search range is [R] min ,R min +1,....,R max -1,R max ];

[0027] The distribution of the second position difference vector within the search range is statistically analyzed, and the burst cycle is determined based on the distribution, including:

[0028] Statistical analysis of each component R in the search range min ,R min +1,....,R max -1,R max The number of components in the second position difference vector is used to select the component with the largest number from the search range as the burst period.

[0029] Furthermore, based on the peak start position vector and the burst period, symbol data for each burst in the TDMA signal is extracted from the resampled data, including:

[0030] Based on the peak start position vector and the burst period, determine the start and end positions of the signal data of each burst in the IQ complex signal data, and extract the signal data of each burst from the IQ complex signal data according to the start and end positions;

[0031] Power detection is performed on the signal data of each burst to obtain the symbol data of the burst.

[0032] Furthermore, power detection is performed on the signal data of each burst to obtain the symbol data of the burst, including;

[0033] The signal data of one of the bursts is extracted according to the oversampling rate to obtain multiple second extracted data sequences;

[0034] Calculate the average power of each of the second extracted data sequences, and select the second extracted data sequence with the largest average power as the symbol data of the burst.

[0035] Furthermore, the burst signal contains H symbol data, namely y1, y2, ..., y H The burst period is nLen;

[0036] Accumulated smoothing is performed on each of the symbol data to obtain amplitude spectrum accumulation data. The amplitude spectrum accumulation data is compared with an amplitude threshold to obtain the effective burst length of the TDMA signal, including:

[0037] Extract each of the symbol data y1, y2, ..., y H From the first nLen data, obtain the corresponding symbol truncation data y1′, y′2, ..., y′ H ;

[0038] Extract data y1′, y′2, ..., y′ for all the symbols at each position. H The amplitude spectrum accumulation data at each of the aforementioned positions is obtained by performing accumulation and moving average processing.

[0039] The values ​​in the accumulated amplitude spectrum that are greater than the amplitude threshold are detected sequentially, and the position corresponding to the last detected value is taken as the effective burst length.

[0040] According to a second aspect of the present invention, an apparatus for estimating the effective burst length of a TDMA signal based on amplitude spectrum accumulation is provided, comprising:

[0041] The resampling module is used to resample the IQ complex signal data obtained from the received TDMA signal to obtain resampled data;

[0042] A burst period estimation module is used to perform sliding correlation and peak detection on the resampled data to obtain the peak start position vector, and to obtain the burst period of the TDMA signal based on the peak start position vector;

[0043] The symbol data extraction module is used to extract symbol data of each burst in the TDMA signal from the resampled data based on the peak start position vector and the burst period;

[0044] The effective burst length estimation module is used to perform cumulative smoothing processing on each symbol data to obtain amplitude spectrum accumulation data, and compare the amplitude spectrum accumulation data with the amplitude threshold to obtain the effective burst length of the TDMA signal.

[0045] The embodiments of the present invention have at least one of the following advantages or beneficial effects:

[0046] The effective burst length estimation method for TDMA signals based on amplitude spectrum accumulation in this invention first achieves effective detection of the start time of each burst through the synchronization symbol sequence and estimates the signal burst period using a global search method. Then, it obtains amplitude spectrum accumulation data by accumulating and smoothing the symbol data of each detected burst, thereby achieving high-precision estimation of the effective burst length of TDMA signals.

[0047] The embodiments of the present invention, with the assistance of a synchronization symbol sequence, can simultaneously achieve high-precision estimation of the burst period and effective burst length of TDMA signals under low signal-to-noise ratio conditions. In particular, the effective burst length is estimated based on the amplitude spectrum data obtained by accumulated smoothing, which can achieve superior estimation performance under extremely low signal-to-noise ratio conditions (signal-to-noise ratio < 3dB).

[0048] In this invention, the above-described technical solutions can be combined with each other to achieve more preferred combinations. Other features and advantages of this invention will be set forth in the following description, and some advantages may become apparent from the description or be learned by practicing the invention. The objects and other advantages of this invention can be realized and obtained from what is particularly pointed out in the description and drawings. Attached Figure Description

[0049] The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Throughout the drawings, the same reference numerals denote the same parts.

[0050] Figure 1 This is a schematic diagram of the main process of the TDMA signal effective burst length estimation method based on amplitude spectrum accumulation according to an embodiment of the present invention.

[0051] Figure 2This is a schematic diagram of the main process of the TDMA signal effective burst length estimation method based on amplitude spectrum accumulation according to another embodiment of the present invention.

[0052] Figure 3 This is a schematic diagram of the amplitude spectrum of resampled data in an embodiment of the present invention.

[0053] Figure 4 This is a schematic diagram of the second related data in an embodiment of the present invention.

[0054] Figure 5 This is a schematic diagram of amplitude spectrum accumulation data in an embodiment of the present invention.

[0055] Figure 6 This is a schematic diagram of the main modules of the TDMA signal effective burst length estimation device based on amplitude spectrum accumulation according to an embodiment of the present invention. Detailed Implementation

[0056] Preferred embodiments of the present invention will now be described in detail with reference to the accompanying drawings, which form part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not intended to limit the scope of the present invention.

[0057] Example 1

[0058] Figure 1 This is a schematic diagram of the main flow of the TDMA signal effective burst length estimation method based on amplitude spectrum accumulation according to an embodiment of the present invention, as shown below. Figure 1 As shown, the effective burst length estimation method for TDMA signals based on amplitude spectrum accumulation in this embodiment of the present invention includes the following steps S101 to S104.

[0059] Step S101: Resample the IQ complex signal data obtained from the received TDMA signal to obtain resampled data;

[0060] Step S102: Perform sliding correlation and peak detection on the resampled data to obtain the peak start position vector, and obtain the burst period of the TDMA signal based on the peak start position vector;

[0061] Step S103: Extract symbol data of each burst in the TDMA signal from the resampled data based on the peak start position vector and the burst period;

[0062] Step S104: Accumulate and smooth each symbol data to obtain amplitude spectrum accumulation data, compare the amplitude spectrum accumulation data with the amplitude threshold to obtain the effective burst length of the TDMA signal.

[0063] Specifically, in this embodiment and some embodiments of the present invention, the IQ complex signal data obtained by channel filtering of a TDMA signal according to the signal bandwidth is represented as x(n1), n1 = 1, 2, ..., N, where N is the length of the IQ complex signal data.

[0064] The resampled data y(n2) is obtained according to the following formula:

[0065]

[0066] Where n1=1+(n2-1)·ξ, is the resampling factor. F s and R s These are the sampling rate and symbol rate of the TDMA signal, respectively, and α is the oversampling rate.

[0067] Specifically, in this embodiment and some embodiments of the present invention, step S102, which involves performing sliding correlation and peak detection on the resampled data to obtain the peak start position vector, includes: performing sliding correlation on the resampled data using a preset synchronization symbol sequence. Specifically, this includes steps S102a1 to S103a3.

[0068] Step S102a1: Extract data from the resampled data according to the oversampling rate α to obtain α first extracted data sequences; the starting points of the multiple first extracted data sequences are different, and the starting points are time positions 1, 2, ..., α respectively.

[0069] Step S102a2: The synchronization symbol sequence is used to perform sliding correlation processing with each of the first extracted data sequences to obtain the first correlation data sequence corresponding to each of the first extracted data sequences; the synchronization symbol sequence is s(n), n = 1, 2, ..., K. It should be noted that the synchronization symbol sequence is the IQ symbol obtained by bit mapping and signal modulation using the known synchronization 01 bit sequence.

[0070] Step S102a3: Compare the data at each position of the first related data sequence, and select the maximum value at each position to form the second related data sequence.

[0071] Specifically, in this embodiment and some embodiments of the present invention, step S102, which involves performing sliding correlation and peak detection on the resampled data to obtain the peak start position vector, includes: performing peak detection on the second correlated data sequence to obtain the peak start position vector. This specifically includes steps S102b1 to S103b2.

[0072] S102b1, detect data greater than a preset detection threshold from the second related data sequence, and form a first position vector by combining the positions corresponding to the data greater than the preset detection threshold, and differentiate the adjacent positions in the first position vector to obtain a first position difference vector.

[0073] S102b2, detect differential components that are greater than a preset minimum peak interval threshold from the first position differential vector, filter out the positions corresponding to the differential components from the first position vector, and construct the peak start position vector.

[0074] Specifically, in this embodiment and some embodiments of the present invention, the step S102, which obtains the burst period of the TDMA signal based on the peak start position vector, specifically includes steps S102c1 to S103c2.

[0075] Step S102c1: Divide the adjacent positions in the peak starting position vector to obtain the second position difference vector.

[0076] Step S102c2: Filter out the maximum and minimum values ​​of the position from the second position difference vector, determine the search range of the burst cycle based on the maximum and minimum values, and statistically analyze the distribution of the second position difference vector within the search range, and determine the burst cycle based on the distribution.

[0077] Specifically, in this embodiment and some embodiments of the present invention, the maximum value and the minimum value are R. max and R min The search range is [R] min ,R min +1,....,R max -1,R max Step S102c2 involves statistically analyzing the distribution of the second position difference vector within the search range and determining the burst cycle based on the distribution, including: statistically analyzing each component R within the search range. min ,R min +1,....,R max -1,R max The number of components in the second position difference vector is used to select the component with the largest number from the search range as the burst period.

[0078] Specifically, in this embodiment and some embodiments of the present invention, step S103 extracts symbol data of each burst in the TDMA signal from the resampled data based on the peak start position vector and the burst period, including steps S103a1 and S103a2.

[0079] Step S103a1: Based on the peak start position vector and the burst period, determine the start and end positions of the signal data of each burst in the IQ complex signal data, and extract the signal data of each burst from the IQ complex signal data according to the start and end positions;

[0080] Step S103a2: Perform power detection on the signal data of each burst to obtain the symbol data of the burst.

[0081] Specifically, in this embodiment and some embodiments of the present invention, step S103a2 specifically includes: extracting the signal data of a burst according to an oversampling rate to obtain multiple second extracted data sequences; calculating the average power of each second extracted data sequence, and selecting the second extracted data sequence with the largest average power as the symbol data of the burst.

[0082] Specifically, in this embodiment and some embodiments of the present invention, the symbol data of the burst signal consists of H elements, namely y1, y2, ..., y H The burst period is nLen; step S104 specifically includes steps S104a1 to S104a3.

[0083] Step S104a1: Extract each of the symbol data y1, y2, ..., y H From the first nLen data, obtain the corresponding symbol truncation data y1′, y′2, ..., y′ H ;

[0084] Step S104a2: Extract data y1′, y′2, ..., y′ from all the symbols at each position. H The amplitude spectrum accumulation data at each of the aforementioned positions is obtained by performing accumulation and moving average processing.

[0085] Step S104a3: Sequentially detect the values ​​in the accumulated amplitude spectrum that are greater than the amplitude threshold, and take the position corresponding to the last detected value as the effective burst length.

[0086] As described in the detailed explanation above, the TDMA signal effective burst length estimation method based on amplitude spectrum accumulation in this embodiment of the invention first achieves effective detection of the start time of each burst through a synchronization symbol sequence and estimates the signal burst period using a global search method. Then, it obtains amplitude spectrum accumulation data by accumulating and smoothing the symbol data of each detected burst, thus achieving high-precision estimation of the effective burst length of the TDMA signal. This embodiment of the invention, through the assistance of a synchronization symbol sequence and amplitude spectrum accumulation, can simultaneously achieve high-precision estimation of the burst period and effective burst length of TDMA signals under low signal-to-noise ratio conditions.

[0087] Example 2

[0088] Figure 2 This is a schematic diagram of the main flow of a TDMA signal effective burst length estimation method based on amplitude spectrum accumulation according to another embodiment of the present invention, as shown below. Figure 2 As shown, the effective burst length estimation method for TDMA signals based on amplitude spectrum accumulation in this embodiment of the present invention includes the following steps S201 to S20x.

[0089] Step S201: Receive the TDMA signal and perform channel filtering processing according to the signal bandwidth to obtain IQ complex signal data.

[0090] Specifically, in this embodiment and some embodiments of the present invention, the IQ complex signal data obtained after processing each TDMA signal data is represented as x(n1), n1 = 1, 2, ..., N, where N is the length of the IQ complex signal data.

[0091] S202, the IQ complex signal data is resampled according to the resampling factor obtained by the oversampling rate α to obtain resampled data y(n2), n2=1,2,…,M, as follows. Figure 3 As shown.

[0092] Specifically, in this embodiment and some embodiments of the present invention, resampling processing based on linear interpolation is used to save computational resources. The resampling data y(n2) is obtained according to the following formula:

[0093]

[0094] Where n1=1+(n2-1)·ξ, is the resampling factor. F s and R s These are the sampling rate and symbol rate of the TDMA signal, respectively, and α is the oversampling rate. α is an integer, for example, a value of 4. Indicates rounding up. This indicates rounding down to the nearest integer.

[0095] S203, extract data from the resampled data according to the oversampling rate to obtain multiple first extracted data sequences. The multiple first extracted data sequences have different starting points, and the starting points are time positions 1, 2, ..., α.

[0096] Specifically, in this embodiment and some embodiments of the present invention, the resampled data y(n2) is extracted according to the oversampling rate α to obtain α first extracted data sequences z1(n3), z2(n3), ..., z α (n3), where n3 = 1, 2, ..., M′, and The specific first extracted data sequences are as follows:

[0097]

[0098] Where y(i:α:M) represents the sequence obtained by extracting one data point from the resampled data y(n2) every α positions from position i to position M, i = 1, 2, ..., α.

[0099] S204, the synchronization symbol sequence is used to perform sliding correlation processing with each of the first extracted data sequences to obtain the first correlated data sequence corresponding to each of the first extracted data sequences.

[0100] Specifically, in this embodiment and some embodiments of the present invention, the synchronization symbol sequence is represented as s(n), n = 1, 2, ..., K1, where K1 is the length of the synchronization symbol sequence. This synchronization symbol sequence is an IQ symbol obtained by bit mapping and signal modulation using a known synchronization 01 bit sequence.

[0101] Specifically, in this embodiment and some embodiments of the present invention, the synchronization symbol sequence and the decimation signals z1(n3), z2(n3), ..., z are used. α (n3) Perform sliding correlation processing on each of the n3 to obtain α first correlation data sequences Y1(n4), Y2(n4), ..., Y α (n4), where n4 = 1, 2, ..., M′-K1+1.

[0102] Specifically, the processing methods for each first relevant data sequence are as follows:

[0103]

[0104] Among them, s * (i) represents the conjugate signal of s(i). It is understood that the above processing includes normalization; therefore, in this embodiment and some embodiments of the present invention, the first related data sequences Y1(n4), Y2(n4), ..., Y... obtained by the above processing are... α The value of (n4) is in the range [0,1].

[0105] S205, compare the data at each position of the first related data sequence, and filter out the maximum value at each position to form the second related data sequence.

[0106] Specifically, in this embodiment and some embodiments of the present invention, the maximum value of the first relevant data sequence at each position is selected to obtain the second relevant data sequence Y(n4), see [link to relevant documentation]. Figure 4 The specific handling method is as follows:

[0107] Y(n4)=max([Y1(n4) Y2(n4) … Y α (n4)]),

[0108] n4=1,2,…,M′-K1+1.

[0109] S206, based on the preset detection threshold β and minimum peak interval threshold γ, peak detection is performed on the second related data sequence Y(n4) to obtain the starting position of each peak and construct the peak starting position vector locn. The specific steps include steps S206a to S206b as follows:

[0110] Step S206a: Detect data greater than a preset detection threshold β from the second related data sequence Y(n4), and form a first position vector by combining the positions corresponding to the data greater than the preset detection threshold β. Differentiate adjacent positions in the first position vector to obtain a first position difference vector. For example, if the first point Y(1) of Y(n4) is greater than β, its position is 1, and so on, obtaining the first position vector as L1 = [loc1, loc2, loc3, ...]. i This represents the position of the i-th data point that is greater than the preset detection threshold β. Correspondingly, the first position difference vector L2 = [l1, l2, ... l...]. i ,…],l i =loc i+1 -loc i .

[0111] Step S206b: Differential components exceeding a preset minimum peak interval threshold γ are detected from the first position differential vector L2; positions corresponding to these differential components are selected from the first position vector L1 to construct the peak start position vector. Specifically, for example, if the differential components exceeding the preset minimum peak interval threshold γ detected from the first position differential vector L2 are sequentially... The peak starting position vector is

[0112] Specifically, in this embodiment and some embodiments of the present invention, the detection threshold β and the minimum peak interval threshold γ can be selected according to the actual situation, and the typical values ​​are generally set as β = 0.7 and γ = 32.

[0113] S207, the burst period nlen of the TDMA signal is obtained based on the peak start position vector locn. This specifically includes steps S207a to S207c.

[0114] S207a, the adjacent positions in the peak starting position vector locn are subtracted to obtain the second position difference vector.

[0115]

[0116] S207b, Filter out the maximum value R of the position from the second position difference vector. max and minimum value R min The search range for the burst cycle is determined to be [R]. min ,R min +1,....,R max -1,R max ].

[0117] S207c, Statistically calculate the distribution vector Q of the second position difference vector within the search range, and determine the burst period based on the distribution vector Q. Specifically, in this embodiment and some embodiments of the present invention, the various components R within the search range are statistically analyzed. min ,R min +1,....,R max -1,R max The number of components in the second position difference vector R is used to select the component with the largest number from the search range as the burst period. For example, R in R min There are 10, R min +1 represents 5, ...,R max -1 represents 7, R max If there are 6, then the distribution vector Q = [10, 5, ..., 7, 6]. If the search range is [R... min ,R min +1,....,R max -1,R max The most numerous is R. min Then the burst period nlen = R min .

[0118] S208, based on the peak start position vector locn and the burst period nlen, extract the symbol data of each burst in the TDMA signal from the resampled data y(n2). Specifically, this includes steps S208a to S208c.

[0119] Step S208a: Calculate the starting position p of each burst in the resampled data y(n2). s and the final position p e , where p s =1 + α·(locn-1), p e =p s +α·nLen-1, where α is the oversampling rate. For ease of description, let the starting positions of all bursts of a TDMA signal be p in sequence. s =[p s1 ,p s2 ,...,p sH ], p e=[p e1 ,p e2 ,...,p eH H represents the number of outbreaks.

[0120] Step S208b: Obtain signal data x1, x2, ..., x for each burst. H , where x1=y(p s1 :p e1 ), x2=y(p s2 :p e2 ), ..., x H =y(p sH :p eH Each signal data has a length of D = α·nLen.

[0121] Step S208c: Utilize the signal data x1, x2, ..., x of each burst. H Obtain the symbolic data y1, y2, ..., y for each burst. H To illustrate this step, let's take the i-th signal data x as an example. i To symbolic data y i The following is an example of the processing procedure:

[0122] Step S208c1, for signal data x i Decimation is performed at an oversampling rate of α to obtain α decimated signals x. i1 (n), x i2 (n), ..., x iα (n), where the signal data x i The length of the second extracted data sequence is D, and the length of the second extracted data sequence is...

[0123]

[0124] Step S208c2: Calculate the average power p of each second extracted data sequence. i1 p i2 ... p iα The details are as follows:

[0125]

[0126] Step S208c3: Select the second extracted data sequence with the highest average power as the symbol data y. i For example, if the average work p i1 p i2 ... p iα The maximum value in is p i1 Then the symbolic data y i =x i1 (n).

[0127] For each signal data x i By repeating the above process, other symbolic data y1, y2, ..., yH can be obtained.

[0128] S209, accumulating and smoothing the symbol data of each symbol to obtain amplitude spectrum accumulation data. Specifically, in this embodiment and some embodiments of the present invention, steps S209a to S209c are included.

[0129] Step S209a, retrieve symbolic data y1, y2, ..., y H The first nlen data are padded with zeros if the data length is less than nlen, resulting in symbolic data y1′, y′2, ..., y′ of length nlen. H ;

[0130] Step S209b: Transfer symbolic data y1′, y′2, ..., y′ H By summing, we get

[0131] Step S209c: Perform a moving average processing on g′(n) to obtain the amplitude spectrum accumulation data g(n), as follows: Figure 5 As shown.

[0132]

[0133] Where k is the window length of the moving average, which can generally be set to 4 to 8.

[0134] S210, the amplitude spectrum accumulation data g(n) is compared with the amplitude threshold to obtain the effective burst length codeLen of the TDMA signal. Specifically, in this embodiment and some embodiments of the present invention, steps S210a to S210c are included.

[0135] Step S210a: Set the threshold η. Specifically, there are two methods for setting the threshold in this embodiment of the invention. The first method is manual threshold setting: observe the amplitude spectrum accumulation data g(n), and set η to approximately the midpoint between the signal magnitude and the noise magnitude. The second method is adaptive threshold setting: utilize g(n) to determine the final threshold value U. n Noise magnitude is estimated using a set of values, and the intermediate U of g(n) is used. m The signal magnitude is estimated using a set of values, and the adaptive calculation formula for the threshold η is as follows:

[0136]

[0137] Specifically, in this embodiment and some embodiments of the present invention, a typical value U is generally set. n =8, U m =32.

[0138] Step S210b: Detect the positions of points in the accumulated amplitude spectrum data g(n) that are greater than the threshold η. For example, if the first point in g(n) is greater than η, its position is 1, and so on, to obtain the position vector pos = [np1, np2, ..., np]. Ln ];

[0139] In step S210c, the effective burst length codeLen is the last value of the position vector pos, i.e., codeLen = np. Ln .

[0140] It is understood that in this embodiment and some embodiments of the present invention, the estimated burst period nLen and effective burst length codeLen are measured in symbols. If it is necessary to convert the unit of measurement to time, it can be based on the symbol rate R. s Perform the conversion, i.e., nLen / R s and codeLen / R s .

[0141] A deeper understanding can be gained from the detailed description of the specific embodiments of the present invention above. The TDMA signal effective burst length estimation method and apparatus based on amplitude spectrum accumulation first achieves effective detection of the start time of each burst through a synchronization symbol sequence and estimates the signal burst period using a global search method. Then, amplitude spectrum accumulation data is obtained by accumulating and smoothing the symbol data of each detected burst, thus achieving high-precision estimation of the TDMA signal effective burst length. Furthermore, the advantages of the embodiments of the present invention can be more clearly understood as follows:

[0142] The embodiments of the present invention achieve precise matching between signal sampling rate and symbol rate through resampling, oversampling and threshold detection, thereby improving processing efficiency and providing a high-precision data foundation for subsequent estimation of burst period and effective burst length signal feature extraction.

[0143] The embodiments of the present invention utilize synchronous symbol sequence and sliding correlation technology to accurately locate the starting position of burst signals. Even in low signal-to-noise ratio environments with significant noise interference, the starting point of burst signals can be accurately captured, significantly enhancing the noise resistance performance of burst period detection.

[0144] This invention optimizes the symbol data extraction process by decimating the signal and detecting power according to the oversampling rate, providing high-quality symbol data for subsequent effective burst length estimation, thus ensuring the accuracy and effectiveness of effective burst length estimation.

[0145] This invention achieves high-precision estimation of the effective burst length through amplitude spectrum accumulation and moving average processing techniques. By accumulating amplitude data from multiple bursts and performing smoothing processing, this invention successfully reduces noise interference and significantly enhances the effective characteristics of the signal, thereby accurately determining the effective burst length. Even under low signal-to-noise ratio conditions, it ensures high accuracy and high reliability of the estimation results.

[0146] The embodiments of the present invention, with the assistance of a synchronization symbol sequence, can simultaneously achieve high-precision estimation of the burst period and effective burst length of TDMA signals under low signal-to-noise ratio conditions. In particular, the effective burst length is estimated based on the amplitude spectrum data obtained by accumulated smoothing, which can achieve superior estimation performance under extremely low signal-to-noise ratio conditions (signal-to-noise ratio < 3dB).

[0147] The burst period and effective burst length information obtained by the embodiments of the present invention can directly provide a basis for subsequent protocol analysis and provide reliable support for the operation of the entire communication system.

[0148] Example 3

[0149] Figure 6 This is a schematic diagram of the main modules of the TDMA signal effective burst length estimation device based on amplitude spectrum accumulation according to an embodiment of the present invention. Figure 6 As shown, the TDMA signal effective burst length estimation device based on amplitude spectrum accumulation in this embodiment of the present invention includes:

[0150] The resampling module is used to resample the IQ complex signal data obtained from the received TDMA signal to obtain resampled data;

[0151] A burst period estimation module is used to perform sliding correlation and peak detection on the resampled data to obtain the peak start position vector, and to obtain the burst period of the TDMA signal based on the peak start position vector;

[0152] The symbol data extraction module is used to extract symbol data of each burst in the TDMA signal from the resampled data based on the peak start position vector and the burst period;

[0153] The effective burst length estimation module is used to perform cumulative smoothing processing on each symbol data to obtain amplitude spectrum accumulation data, and compare the amplitude spectrum accumulation data with the amplitude threshold to obtain the effective burst length of the TDMA signal.

[0154] Furthermore, those skilled in the art will understand that all or part of the processes of the methods described in the above embodiments can also be implemented by a computer program instructing related hardware, and the program can be stored in a computer-readable storage medium. The computer-readable storage medium may be a disk, optical disk, read-only memory, or random access memory, etc.

[0155] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for estimating the effective burst length of a TDMA signal based on amplitude spectrum accumulation, characterized in that, include: The IQ complex signal data obtained from the received TDMA signal is resampled to obtain resampled data; The resampled data is subjected to sliding correlation and peak detection to obtain the peak start position vector, and the burst period of the TDMA signal is obtained based on the peak start position vector; Based on the peak start position vector and the burst period, the symbol data of each burst in the TDMA signal is extracted from the resampled data; The amplitude spectrum accumulation data is obtained by performing accumulation and smoothing processing on each symbol data, and the amplitude spectrum accumulation data is compared with the amplitude threshold to obtain the effective burst length of the TDMA signal.

2. The method according to claim 1, characterized in that, The IQ complex signal data obtained by channel filtering of the TDMA signal according to the signal bandwidth is represented as x(n1), n1=1,2,…,N,N is the length of the IQ complex signal data; The resampled data y(n2) is obtained according to the following formula: Where n1=1+(n2-1)·ξ, is the resampling factor. F s and R s These are the sampling rate and symbol rate of the TDMA signal, respectively, and α is the oversampling rate.

3. The method according to claim 1, characterized in that, To obtain the peak start position vector by performing sliding correlation and peak detection on the resampled data, the method includes: performing sliding correlation on the resampled data using a preset synchronization symbol sequence; The specific steps of performing sliding correlation on the resampled data using a preset synchronization symbol sequence are as follows: Data is extracted from the resampled data according to the oversampling rate to obtain multiple first extracted data sequences; The synchronization symbol sequence is used to perform sliding correlation processing with each of the first extracted data sequences to obtain the first correlated data sequence corresponding to each of the first extracted data sequences; Compare the data at each position in the first related data sequence, and select the maximum value at each position to form the second related data sequence.

4. The method according to claim 3, characterized in that, To obtain a peak start position vector by performing sliding correlation and peak detection on the resampled data, the method includes: performing peak detection on the second correlated data sequence to obtain a peak start position vector; Peak detection is performed on the second related data sequence to obtain the peak start position vector, specifically: Data exceeding a preset detection threshold are detected from the second related data sequence, and the positions corresponding to the data exceeding the preset detection threshold are used to form a first position vector. The adjacent positions in the first position vector are differentially divided to obtain a first position difference vector. Differential components that are greater than a preset minimum peak interval threshold are detected from the first position differential vector, and the positions corresponding to the differential components are selected from the first position vector to construct the peak start position vector.

5. The method according to claim 4, characterized in that, The burst period of the TDMA signal is obtained based on the peak start position vector, including: The second position difference vector is obtained by differentiating the adjacent positions in the peak starting position vector; The maximum and minimum values ​​of the positions are selected from the second position difference vector. The search range of the burst cycle is determined based on the maximum and minimum values. The distribution of the second position difference vector within the search range is statistically analyzed. The burst cycle is determined based on the distribution.

6. The method according to claim 5, characterized in that, The maximum value and the minimum value are R. max and R min The search range is [R] min ,R min +1,....,R max -1,R max ]; The distribution of the second position difference vector within the search range is statistically analyzed, and the burst cycle is determined based on the distribution, including: Statistical analysis of each component R in the search range min ,R min +1,....,R max -1,R max The number of components in the second position difference vector is used to select the component with the largest number from the search range as the burst period.

7. The method according to claim 5 or 6, characterized in that, Based on the peak start position vector and the burst period, symbol data for each burst in the TDMA signal is extracted from the resampled data, including: Based on the peak start position vector and the burst period, determine the start and end positions of the signal data of each burst in the IQ complex signal data, and extract the signal data of each burst from the IQ complex signal data according to the start and end positions; Power detection is performed on the signal data of each burst to obtain the symbol data of the burst.

8. The method according to claim 7, characterized in that, Power detection is performed on the signal data of each burst to obtain the symbol data of the burst, including; The signal data of one of the bursts is extracted according to the oversampling rate to obtain multiple second extracted data sequences; Calculate the average power of each of the second extracted data sequences, and select the second extracted data sequence with the largest average power as the symbol data of the burst.

9. The method according to claim 1, characterized in that, The burst signal consists of H symbol data, namely y1, y2, ..., y H The burst period is nlen; Accumulated smoothing is performed on each of the symbol data to obtain amplitude spectrum accumulation data. The amplitude spectrum accumulation data is compared with an amplitude threshold to obtain the effective burst length of the TDMA signal, including: Extract each of the symbol data y1, y2, ..., y H From the first nlen data, obtain the corresponding symbol truncation data y1′, y′2, ..., y′. H ; Extract data y1′, y′2, ..., y′ for all the symbols at each position. H The amplitude spectrum accumulation data at each of the aforementioned positions is obtained by performing accumulation and moving average processing. The values ​​in the accumulated amplitude spectrum that are greater than the amplitude threshold are detected sequentially, and the position corresponding to the last detected value is taken as the effective burst length.

10. A device for estimating the effective burst length of a TDMA signal based on amplitude spectrum accumulation, characterized in that, include: The resampling module is used to resample the IQ complex signal data obtained from the received TDMA signal to obtain resampled data; A burst period estimation module is used to perform sliding correlation and peak detection on the resampled data to obtain the peak start position vector, and to obtain the burst period of the TDMA signal based on the peak start position vector; The symbol data extraction module is used to extract symbol data of each burst in the TDMA signal from the resampled data based on the peak start position vector and the burst period; The effective burst length estimation module is used to perform cumulative smoothing processing on each symbol data to obtain amplitude spectrum accumulation data, and compare the amplitude spectrum accumulation data with the amplitude threshold to obtain the effective burst length of the TDMA signal.

Images