A non-ground network system coarse frequency offset and primary synchronization signal joint detection method

Abstract

The present invention proposes a method for jointly detecting coarse frequency offset and primary synchronization signal in a non-terrestrial network system, belonging to the field of 5G non-terrestrial network communication technology. The method comprises: receiving a signal to obtain a sequence of sampling points to be detected; performing uniform time division extraction to obtain n-channel parallel data; performing frequency domain equalization, using 50% overlapping block processing to transform the data into the frequency domain; performing frequency domain equalization to obtain a time domain mutual correlation sequence for each block; discarding the first half of the sequence and retaining the second half of the sequence, concatenating the second half of the sequence of n blocks of one channel of data into a correlation value sequence group, and obtaining an n-channel parallel mutual correlation sequence group; converting it into a single-channel mutual correlation sequence through parallel-serial conversion; detecting the peak of the single-channel mutual correlation sequence, completing primary synchronization signal detection, and obtaining the synchronization point position and coarse frequency offset value. The present invention can realize PSS detection and coarse frequency offset estimation with extremely low complexity in a 5G non-terrestrial network system.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of 5G non-terrestrial network communication, in particular to a non-terrestrial network (NTN) system coarse frequency offset and primary synchronization signal (PSS) joint detection method, which is suitable for high-speed train, high-speed unmanned aerial vehicle, satellite communication and other scenarios. BACKGROUND

[0002] Synchronization plays a crucial role in wireless communication, and only accurate system synchronization can make the whole communication system work normally. PSS detection, as an important process of synchronization in 5G communication, its detection performance and complexity will affect the performance of the whole system. However, in the 5G non-terrestrial network, there is an influence of multipath fading channel, especially under the influence of ultra-high noise environment and severe Doppler frequency offset, accurate detection of PSS has extremely high challenge.

[0003] Currently, the commonly used PSS detection algorithm is the sliding cross-correlation algorithm. This algorithm uses the good correlation characteristics of the PSS sequence to use the local PSS sequence for cross-correlation, obtains the cross-correlation sequence, and detects the PSS by detecting the correlation peak to complete the synchronization of the whole system. However, under the influence of large Doppler frequency offset and low signal-to-noise ratio in the 5G non-terrestrial network system, multiple parallel detections are required to ensure the detection accuracy of the traditional algorithm, which leads to rapid increase of the complexity of the whole system, high consumption of logic resources, and inability to be deployed on various types of hardware platforms. SUMMARY

[0004] To solve the above problems, the present application provides a non-terrestrial network system coarse frequency offset and primary synchronization signal joint detection method. The present application can effectively improve the estimation accuracy of the channel signal-to-noise ratio in an environment of ultra-low signal-to-noise ratio, multipath deep fading and large Doppler frequency offset, thereby greatly reducing the complexity under the premise of ensuring the detection performance, and completing the estimation of the coarse frequency offset at the same time of completing the PSS detection.

[0005] The technical scheme adopted by the present application is as follows:

[0006] A non-terrestrial network system coarse frequency offset and primary synchronization signal joint detection method, comprising the following steps:

[0007] S1: The terminal receives the signal and performs analog-to-digital conversion, and obtains a to-be-detected sampling point sequence after anti-aliasing filtering and down-sampling;

[0008] S2: The to-be-detected sampling point sequence is uniformly time-division decimated to complete serial-parallel conversion, and n parallel data are obtained, each of which is delayed differently;

[0009] S3: Frequency domain equalization is performed on each data respectively, and 50% overlap block processing is adopted, so that each data is changed into n blocks, and after block, data is transformed into frequency domain through fast Fourier transform;

[0010] S4: Based on the equivalence of time domain cross-correlation and frequency domain point multiplication operation, frequency domain equalization operation is performed on each block data respectively, and time domain cross-correlation sequence of each block is obtained through inverse Fourier transform, and there is a time delay relationship between n block time domain cross-correlation sequences of each data;

[0011] S5: For the time domain cross-correlation sequence of each block, the first half sequence is discarded and the second half sequence is reserved, then the n block second half sequences of 1 data are concatenated to form a correlation value sequence group, and n parallel cross-correlation sequence groups are obtained;

[0012] S6: The n parallel cross-correlation sequence groups are converted into a single cross-correlation sequence through parallel-serial conversion according to different delays;

[0013] S7: The peak value of the single cross-correlation sequence is detected to complete the primary synchronization signal detection, and the synchronization point position and coarse frequency offset value are obtained.

[0014] Further, in step S2, the to-be-detected sampling point sequence is uniformly time-division decimated in the following manner:

[0015]

[0016] Wherein, y(m) is the to-be-detected sampling point sequence, m is the point position in the sequence, y n (k) is the nth data after serial-parallel conversion.

[0017] Further, in step S3, the specific way of 50% overlap block processing is as follows:

[0018] First, a data block is taken according to the length of the fast Fourier transform window for each data, and the length of each subsequent data block is the length of the fast Fourier transform window, but the starting point of the data block is the center point of the previous data block, and the 50% overlap block is completed in turn.

[0019] Further, the specific way of step S4 is as follows:

[0020] Based on the equivalence of cross-correlation operation and frequency domain point multiplication operation, frequency domain equalization is performed, the cross-correlation sequence is calculated through inverse fast Fourier transform after frequency domain point multiplication, and a series of parallel segmented correlation value sequence groups are obtained through parallel detection of a plurality of groups of frequency domain primary synchronization signal sequences with preset frequency offset, and the calculation formula is as follows:

[0021] Y n,b (k) = FFT[y n,b (k)]

[0022]

[0023] Among them, y n,b (k) is the b-th block data of the n-th path, Y n,b (k) is y n,b (k) is the frequency domain sequence, is the p-th type primary synchronization signal sequence with a preset frequency offset of f, p = 0, 1, 2, for The frequency domain sequence, c n,b,f,p (k) is a segmented correlation value sequence of the b-th block data of the n-th channel to the p-th type primary synchronization signal sequence with a preset frequency offset of f. FFT is a fast Fourier transform, and IFFT is an inverse fast Fourier transform.

[0024] Furthermore, the correlation value sequence group in step S5 is:

[0025]

[0026] in, c b,n,f,p The second half of the sequence (k), C n,f,p (k) is the parallel correlation value sequence group after splicing.

[0027] Furthermore, in step S6, the single-path cross-correlation sequence after parallel-to-serial conversion is:

[0028] C f,p (k)=[C 1,f,p (0),C 2,θ,ω (0),…,C n,θ,ω (0),C 1,f,p (1),C 2,f,p (1),…,C n,f,p (1),…,C n,f,p (k)]

[0029] Among them, C f,p (k) is the single-path cross-correlation sequence after parallel-to-serial conversion.

[0030] Furthermore, the specific method of step S7 is:

[0031] By using the single-path cross-correlation sequence C f,p (k) Perform parallel peak-threshold detection to obtain the synchronization point position and the magnitude of the coarse frequency deviation.

[0032] The beneficial effects of the present invention are as follows:

[0033] 1. The application can effectively improve the estimation accuracy of channel signal-to-noise ratio in an environment with ultra-low signal-to-noise ratio, deep multipath fading and large Doppler frequency offset, thereby significantly reducing the complexity under the premise of ensuring detection performance.

[0034] 2. The application can complete the estimation of coarse frequency offset while completing PSS detection.

[0035] 3. The application is designed based on the 5G non-ground network multipath fading channel and large Doppler frequency offset scenario, solving the problem of high complexity of traditional algorithms in the 5G non-ground network scenario. BRIEF DESCRIPTION OF DRAWINGS

[0036] The specific embodiments of the application will be further described in detail below with reference to the accompanying drawings.

[0037] Figure 1 The schematic diagram of the principle of the application. DETAILED DESCRIPTION

[0038] In order to more clearly illustrate the application, the application will be further described below in conjunction with the embodiments and drawings. Those skilled in the art should understand that the specific description below is illustrative rather than limiting, and should not limit the protection scope of the application.

[0039] A non-ground network system coarse frequency offset and primary synchronization signal joint detection method, as shown in Figure 1 includes the following steps:

[0040] S1: The terminal samples and digitizes the baseband received signal through ADC. In order to reduce the complexity and chip area of the subsequent digital signal processing algorithm, the sampled signal is subjected to an anti-aliasing filter to suppress high-frequency signal interference, and a low sampling rate detection sequence is obtained after downsampling and buffering.

[0041] The method first performs aliasing filtering on the original sampling points sampled by the ADC. The information of the entire SSB block is obtained through anti-aliasing filtering, and a ±400KHz guard interval is added compared to the bandwidth of the SSB when designing the anti-aliasing filter bandwidth to prevent the loss of SSB information due to large Doppler frequency offset. After anti-aliasing filtering and downsampling, the subcarriers will not be aliased, thereby affecting the system performance. The downsampling multiple determines the complexity of the subsequent frequency domain equalization algorithm. Downsampling without affecting performance can significantly reduce the complexity of the system.

[0042] S2: In order to further reduce the complexity of the cross-correlation operation and maintain a high synchronization time resolution, the to-be-detected sequence is uniformly time-division decimated by a serial-parallel conversion module to convert the input sequence into n parallel data sequences, and the data delays between the n data sequences are different. The time synchronization resolution can be improved by detecting the maximum value of the n cross-correlation operations.

[0043] The down-sampled data is then sent to a serial-parallel conversion module to divide the to-be-detected sampling point sequence into n data in the following manner, and the serial-parallel conversion module uniformly time-division decimates the to-be-detected sampling points in the following manner,

[0044]

[0045] wherein y(m) is the to-be-detected sampling point sequence, y n (k) is the nth data after serial-parallel conversion.

[0046] S3: Each data is subjected to frequency domain equalization, and a 50% overlap block processing is adopted. After the block processing, the data is converted to the frequency domain by fast Fourier transform. The specific manner is as follows:

[0047] Each data is first taken as a data block according to the length of the fast Fourier transform window, and the length of each subsequent data block is the length of the fast Fourier transform window, but the starting point of the data block is the center point of the previous data block. The 50% overlap block processing is completed in sequence, and the data after the block processing is subjected to fast Fourier transform to convert to the frequency domain.

[0048] S4: Based on the equivalence of time domain cross-correlation and frequency domain point multiplication operations, each block data is subjected to frequency domain equalization operation, and the time domain cross-correlation sequence value is obtained by inverse Fourier transform. The n cross-correlation sequences have a time delay relationship. The specific manner is as follows:

[0049] Based on the equivalence of the cross-correlation operation and the frequency domain point multiplication operation, the frequency domain equalization is performed, the correlation value sequence is calculated by inverse fast Fourier transform after the frequency domain point multiplication, a series of parallel segmented correlation value sequence groups are obtained by parallel detection of a plurality of groups of frequency domain PSS sequences with preset frequency offset, and the calculation formula is as follows,

[0050] Y n,b (k) = FFT[y n,b (k)]

[0051]

[0052] wherein y n,b (k) is the nth data, Y n,b (k) is the frequency domain sequence of y n,b (k). PSS sequence of the pth(p=0,1,2) type with a preset frequency offset f, is a frequency domain sequence, c n,b,f,p (k) is a segment correlation value sequence of the pth type PSS sequence with a preset frequency offset f of the nth bth block data, FFT is a fast Fourier transform, and IFFT is an inverse fast Fourier transform.

[0053] S5: The first half of the correlation value sequence obtained through frequency domain block processing is affected by the cyclic convolution aliasing error. The first half of the cross-correlation sequence is discarded, and only the cross-correlation values output in the second half are retained. The cross-correlation sequence data blocks output in each of the n paths are concatenated to obtain a continuous cross-correlation sequence output.

[0054] For c n,b,f,p (k), due to the use of 50% overlap block, the second half is valid data, and the first half sequence after each IFFT needs to be discarded, and the second half is retained and spliced in turn to obtain the correlation value sequence. The formula is as follows,

[0055]

[0056] wherein, is the first half sequence, is the second half sequence, and c n,f,p (k) is a parallel correlation value sequence group after splicing.

[0057] S6: The n parallel cross-correlation sequences are converted into a single cross-correlation sequence through a time division multiplexing and serial conversion module according to different delays. The time synchronization resolution of the single cross-correlation sequence is improved compared with each sequence in the n paths.

[0058] The parallel correlation value sequence group after splicing is converted into a single correlation value sequence group through a parallel-serial conversion, and the formula is as follows,

[0059] C f,p (k) = [C 1,f,p (0), C 2,θ,ω (0), …, C n,θ,ω (0), C 1,f,p (1), C 2,f,p (1), …, C n,f,p (1), …, C n,f,p (k)]

[0060] wherein, C f,p (k) is a sequence group after merging by the parallel-serial conversion module. The position of the synchronization point and the size of the coarse frequency offset are obtained through parallel peak value threshold detection on the correlation value sequence group C f,p (k).

[0061] S7: detecting the peak value of the single-path cross-correlation sequence, completing the main synchronization signal detection, and obtaining the synchronization point position and the coarse frequency offset value.

[0062] Finally, by analyzing the C f,p (k) performing parallel peak threshold detection on the correlation value sequence group, and obtaining the synchronization point position and the coarse frequency offset value.

[0063] The present application is based on the correlation between the cross-correlation operation and the frequency domain point multiplication operation, adopts the overlapping frequency domain equalization method to perform core PSS detection, and performs multi-path parallel detection for the extremely low signal-to-noise ratio and large Doppler frequency offset scene in the non-ground network system. The present application is based on the design of the 5G non-ground network system, is suitable for all 5G non-ground network systems, can realize the detection and coarse frequency offset estimation of PSS in the 5G non-ground network system with extremely low complexity, and solves the problem of high complexity of the traditional sliding cross-correlation algorithm in the extremely low signal-to-noise ratio and large Doppler frequency offset scene in the 5G non-ground network system.

[0064] The above is only the preferred specific implementation of the present application, but the protection scope of the present application is not limited to this. Any person skilled in the art can easily think of changes or replacements within the technical range disclosed by the present application, which should be covered in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method for jointly detecting coarse frequency offset and primary synchronization signal in a non-terrestrial network system, characterized in that: The following steps are involved: S1: The terminal receives the signal and performs analog-to-digital conversion, and obtains the sampling point sequence to be detected through anti-aliasing filtering and downsampling; S2: Evenly time-division-sample the sampling point sequence to be detected, complete the serial-to-parallel conversion, and obtain n parallel data channels, each with a different delay; S3: Perform frequency domain equalization on each data channel and perform 50% overlapping block processing to convert each data channel into n blocks. After the blocks are divided, the data is transformed into the frequency domain through fast Fourier transform; S4: Based on the equivalence of time domain cross-correlation and frequency domain dot product operations, frequency domain equalization is performed on each block of data, and the time domain cross-correlation sequence of each block is obtained through inverse Fourier transform. The time domain cross-correlation sequences of the n blocks of each data channel have a time delay relationship; S5: For each block's time-domain cross-correlation sequence, discard the first half of the sequence and retain the second half, then concatenate the second half sequences of the n blocks of one channel of data into a correlation value sequence group, thereby obtaining an n-channel parallel cross-correlation sequence group; S6: converting the n-way parallel cross-correlation sequence group into a single-way cross-correlation sequence by parallel-to-serial conversion according to different delays; S7: Detect the peak value of the single-channel cross-correlation sequence, complete the main synchronization signal detection, and obtain the synchronization point position and coarse frequency offset value.

2. The method for joint detection of coarse frequency deviation and primary synchronization signal in a non-terrestrial network system according to claim 1, characterized in that: In step S2, the sequence of sampling points to be detected is uniformly time-divided and extracted as follows: Among them, y(m) is the sampling point sequence to be detected, m is the point in the sequence, and y n (k) is the nth data after serial-to-parallel conversion.

3. The method for joint detection of coarse frequency deviation and primary synchronization signal in a non-terrestrial network system according to claim 1, characterized in that: In step S3, the specific method of using 50% overlapping block processing is as follows: For each data channel, a data block is first taken according to the length of the fast Fourier transform window. The length of each subsequent data block is the length of the fast Fourier transform window, but the starting point of the data block is the center point of the previous data block, and 50% overlapping blocks are completed in sequence.

4. The method for joint detection of coarse frequency deviation and primary synchronization signal in a non-terrestrial network system according to claim 1, wherein: The specific method of step S4 is: Frequency domain equalization is performed based on the equivalence of cross-correlation operation and frequency domain dot multiplication operation. The cross-correlation sequence is calculated by frequency domain dot multiplication followed by inverse fast Fourier transform. At the same time, multiple sets of frequency domain main synchronization signal sequences with preset frequency offsets are detected in parallel to obtain a series of parallel segmented correlation value sequence groups. The calculation formula is as follows: Y n,b (k)=FFT[y n,b (k)] Among them, y n,b (k) is the b-th block data of the n-th path, Y n,b (k) is y n,b (k) is the frequency domain sequence, is the p-th type primary synchronization signal sequence with a preset frequency offset of f, p = 0, 1, 2, for The frequency domain sequence, c n,b,f,p (k) is a segmented correlation value sequence of the b-th block data of the n-th channel to the p-th type primary synchronization signal sequence with a preset frequency offset of f. FFT is a fast Fourier transform, and IFFT is an inverse fast Fourier transform.

5. The method for joint detection of coarse frequency deviation and primary synchronization signal in a non-terrestrial network system according to claim 4, characterized in that: The correlation value sequence group in step S5 is: in, c b,n,f,p The second half of the sequence (k), C n,f,p (k) is the parallel correlation value sequence group after splicing.

6. The method for joint detection of coarse frequency offset and primary synchronization signal in a non-terrestrial network system according to claim 5, characterized in that: In step S6, the single-path cross-correlation sequence after parallel-to-serial conversion is: C f，p (k)=[C 1，f，p (0),C 2,θ，ω (0),…,C n，θ，ω (0),C 1，f， p(1), C 2，f，p (1),…,C n，f，p (1),…,C n，f，p (k)] Among them, C f，p (k) is the single-path cross-correlation sequence after parallel-to-serial conversion.

7. The method for joint detection of coarse frequency deviation and primary synchronization signal in a non-terrestrial network system according to claim 6, characterized in that: The specific method of step S7 is: By using the single-path cross-correlation sequence C f，p (k) Perform parallel peak-threshold detection to obtain the synchronization point position and the magnitude of the coarse frequency deviation.

Images