A multi-element spread spectrum calibration method and device based on preamble reconstruction
By reconstructing the preamble and performing partitioning processing using the PMF-FFT algorithm, the problem of frequency offset correction in multi-element spread spectrum scattering communication was solved, enabling frequency offset detection and correction under low signal-to-noise ratio conditions, thereby improving communication stability and transmission efficiency.
Patent Information
- Application Number
- CN202211660754.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-12-23
- Publication Date
- 2024-11-19
- Estimated Expiration
- 2042-12-23
AI Technical Summary
In multi-element spread spectrum scattering communication, the demodulation performance degrades and the spectrum analysis resource consumption increases due to frequency offset, especially under low signal-to-noise ratio conditions where frequency offset is difficult to correct, leading to communication interruption.
By reconstructing the preamble, combining the PMF-FFT algorithm and partitioning, spectrum analysis and frequency offset correction are achieved. A continuous window peak search module is used for frequency offset detection, and the preamble in the existing frame structure is used for frequency offset correction.
It improves the accuracy and range of frequency offset detection under low signal-to-noise ratio conditions, reduces resource consumption, and ensures communication stability and transmission efficiency.
Smart Images

Figure CN115987738B_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the field of scattering communication, and in particular relates to a multi-element spread spectrum calibration method and device based on preamble code reconstruction. Background Art
[0002] The demodulation threshold of multivariate spread spectrum is low, and it can cope with high fading rates. It also has a certain ability to resist internal and external interference, and is usually used for low-speed communication. In the scattering communication system, due to the limitation of clock source stability and the existence of Doppler frequency shift, the receiver carrier will produce a frequency offset. The correlation peak obtained by time domain correlation at the receiving end of spread spectrum communication will drop sharply with the increase of frequency deviation, resulting in poor demodulation performance. Large frequency deviation may even cause the correlation peak to be completely offset, resulting in communication interruption. Since multivariate spread spectrum corresponds to multiple PN codes according to different service code words, it is a non-periodic code for a period of time, and can only perform spectrum analysis on multiple PN codes in parallel in units of symbols. Scattering communication usually resists channel fading with multiple diversity, and the resources required for spectrum analysis increase exponentially, which is huge. When the signal-to-noise ratio of the received signal is low, the spectrum lines obtained by the commonly used spectrum analysis will be submerged in the noise, resulting in the failure of frequency deviation detection. How to correct the frequency deviation of multivariate spread spectrum scattering communication to improve demodulation performance is a difficult point, and it is even more difficult under low signal-to-noise ratio. Summary of the invention
[0003] The purpose of the present invention is to avoid the deficiencies in the above-mentioned background technology and provide a multi-element spread spectrum calibration method and device based on preamble code reconstruction, so as to realize correction and tracking of a large frequency deviation range and ensure the demodulation performance of multi-element spread spectrum scattering communication.
[0004] The technical solution adopted by the present invention is:
[0005] A multi-element spread spectrum calibration method based on preamble reconstruction comprises the following steps:
[0006] Step 1: Spread the preamble and oversample it by A times, and partition it into segments of length X to form a segment of length L = A × N × B / 2 M The reconstructed preamble data is stored in the local storage module; where N is the code length of the preamble, M is the number of spread spectrum elements, B is the code length of the PN code corresponding to the spread spectrum, and the value of X is determined by the calibration range and accuracy, A≥4;
[0007] Step 2: partition the down-converted I and Q signals of the oversampled diversity branch into segments of length X and store them in a local cache;
[0008] Step 3: Each time a sampling point is written, the data in the storage module and the cache are read, and a serial correlation operation is performed on the reconstructed preamble code and the down-converted I and Q signals, and L / X partial correlation values are obtained for the I and Q signals of each diversity branch;
[0009] Step 4: pad the I and Q signals of the L / X partial correlation values of each diversity branch with zeros, perform P-point FFT operation, and then take the modulus value; wherein P is an integer power of 2;
[0010] Step 5: Combining the spectrum modulus values of multiple diversity branches at the maximum ratio;
[0011] Step 6: Perform peak search on the combined spectrum analysis results within the period of the preamble code, and obtain the frequency point corresponding to the FFT maximum value as the carrier frequency deviation value obtained this time;
[0012] Step 7: Compare the carrier frequency deviation value obtained in this frame with the frequency deviation values obtained in previous frames. If it is less than the threshold, the frequency deviation value of this frame is valid, and the process goes to step 8. Otherwise, the process goes back to step 2.
[0013] Step 8: Adjust the output frequency of the local carrier according to the obtained effective frequency deviation value, and obtain a zero intermediate frequency signal after down-conversion to complete frequency correction.
[0014] Further, it includes a preamble reconstruction module 1, a first to Kth digital orthogonal down-conversion modules 2-1 to 2-K, a first to Kth PMF-FFT modules 3-1 to 3-K, a first to Kth modulus value modules 4-1 to 4-K, a maximum ratio merging module 5, a continuous window peak search module 6 and a frequency deviation calculation and confirmation module 7;
[0015] The preamble reconstruction module 1 is used to perform M-ary spreading, oversampling and partition reconstruction on the preamble, and send the reconstructed preamble to the first to Kth PMF-FFT modules 3-1 to 3-K;
[0016] The first to Kth digital orthogonal down-conversion modules 2-1 to 2-K are respectively used to perform orthogonal down-conversion on each digital signal to form an I signal and a Q signal, and send them to the first to Kth PMF-FFT modules 3-1 to 3-K;
[0017] The first to Kth PMF-FFT modules 3-1 to 3-K are respectively used to perform partial correlation, zero padding and FFT operations on the I and Q signals of the orthogonal down-conversion to form complex results of spectrum analysis, and send them to the first to Kth modulus value modules 4-1 to 4-K;
[0018] The first to Kth modulus value taking modules 4-1 to 4-K are respectively used to take modulus values of complex number results of spectrum analysis and send them to the maximum ratio combining module 5;
[0019] The maximum ratio merging module 5 is used to perform maximum ratio merging of the spectrum modulus values of the K branches, and send the merging result to the continuous window peak search module 6;
[0020] The continuous window peak search module 6 is used to search for the peak value in the continuous window and mark the position, and send the peak position to the frequency deviation calculation and confirmation module 7;
[0021] The frequency deviation calculation and confirmation module 7 is used to obtain the frequency deviation estimation result according to the peak position calculation, and output the valid result of the frequency deviation to the first to Kth orthogonal down-conversion modules 2-1 to 2-K after comparison and confirmation to correct the local carrier frequency.
[0022] Further, the preamble reconstruction module 1 includes an M-ary spreading module 9, an oversampling module 10 and a partitioning module 11;
[0023] The M-ary spreading module 9 is used to spread the preamble code by symbol and send it to the oversampling module 10;
[0024] The oversampling module 10 is used to perform A-times oversampling on the chips, forming a length of L=A×N×B / 2 M The reconstructed preamble data is sent to the partition module 11; wherein N is the code length of the preamble, M is the number of spread spectrum elements, B is the code length of the PN code corresponding to the spread spectrum, and the value of X is determined by the calibration range and accuracy, A≥4;
[0025] The partition module 11 is used to partition according to the code length and the required frequency calibration accuracy, and send the obtained reconstructed preamble code to the first to Kth PMF-FFT modules 3-1 to 3-K respectively.
[0026] Further, the first to the KPMF-FFT modules 3-1 to 3-K respectively include a storage module 12, a first and a second cache modules 13-1 and 13-2, a first to a second L / X sub-correlation modules 14-1 to 14-2L / X, a zero padding module 15, a P-point FFT module 16, and a read-write control module 16;
[0027] The storage module 12 is used to store the reconstructed preamble sequence in a local memory, and output it to the first to second L / X sub-correlation modules 14-1 to 14-2L / X under the control of the read-write control module 16;
[0028] The first and second buffer modules 13-1 and 13-2 are respectively used to buffer the down-converted I and Q signals and send them to the first to second L / X sub-correlation modules 14-1 to 14-2L / X under the control of the read-write control module 16;
[0029] The first to second L / X sub-correlation modules 14-1 to 14-2L / X are respectively used to read the signals of the storage module 12 and the first and second cache modules 13-1 and 13-2 under the control of the read-write control module 16, perform sub-correlation, and send the L / X path sub-correlation peak complex signals obtained by correlation to the zero-filling module 15;
[0030] The zero-filling module 15 is used to fill zero after each result of the sub-correlation and send it to the P-point FFT module 16;
[0031] The P-point FFT module 16 is used to perform P-point FFT operation on the zero-filled correlation results using the IP core to obtain complex I and Q results, and send them to the first to Kth modulus value modules 4-1 to 4-K respectively.
[0032] Compared with the background technology, the present invention has the following advantages:
[0033] 1. The present invention reconstructs the preamble code to obtain long-period codewords for spectrum analysis, thereby improving the detection signal-to-noise ratio. Depending on the length of the preamble code, the present invention has a gain of more than 10 dB relative to the communication waveform and can accurately realize frequency calibration under low signal-to-noise ratio.
[0034] 2. The present invention combines the PMF-FFT algorithm partition to perform matched filtering on the preamble code and then performs spectrum analysis, thereby increasing the detection range of frequency deviation and realizing detection of a large frequency deviation range. At the same time, the present invention converts the parallel correlation operation part in each diversity branch into a serial one, greatly reducing the on-chip resource consumption.
[0035] 3. The present invention performs maximum ratio combining on each diversity branch after completing spectrum analysis, thereby obtaining corresponding diversity gain and ensuring the stability of frequency calibration.
[0036] 4. The present invention uses the preamble code part in the existing frame structure for reconstruction to detect frequency deviation, and no additional pilot signal needs to be inserted, so the frequency calibration function can be realized without reducing the transmission efficiency.
[0037] 5. The present invention adopts a continuous window peak search module, which does not require timing information and has low complexity; it does not require a fixed decision threshold and can adapt to scattering channels with large fluctuations in receiving levels under different communication distances and different environmental conditions. BRIEF DESCRIPTION OF THE DRAWINGS
[0038] Figure 1 The present invention is a flow chart of a multi-element spread spectrum calibration method based on preamble code reconstruction.
[0039] Figure 2 The invention discloses a block diagram of the electrical principle of a multi-element spread spectrum frequency calibration device based on preamble reconstruction.
[0040] Figure 3 It is the electrical schematic diagram of the preamble reconstruction module of the present invention.
[0041] Figure 4 It is the electrical schematic diagram of the i-th PMF-FFT module of the present invention. DETAILED DESCRIPTION
[0042] The present invention is further described in detail below with reference to the accompanying drawings.
[0043] like Figure 1 As shown, a multi-element spread spectrum calibration method based on preamble reconstruction includes the following steps:
[0044] Step 1: Perform M-ary spread on the preamble with a code length of N. The M-ary spread corresponds to 2 M The PN code with a barcode length of B is oversampled by A times and partitioned into segments with a length of X, forming a PN code with a length of L = A × N × B / 2. M The preamble reconstruction data is stored in the local storage module, A≥4;
[0045] Step 2: partition the down-converted I and Q signals of the A-times oversampled multiple diversity branches into segments of length X, and store them in a local cache;
[0046] Step 3: Each time a sampling point is written, the data in the storage module and the cache are read, and the reconstructed preamble code is serially correlated with the down-converted I and Q signals. L / X partial correlation values can be obtained for the I and Q of each diversity branch.
[0047] Step 4: pad the I and Q signals of the L / X partial correlation values of each diversity branch with zeros, perform P-point FFT operation, and then take the modulus value; wherein P is an integer power of 2;
[0048] Step 5: Combining the spectrum modulus values of multiple diversity branches at the maximum ratio;
[0049] Step 6: Perform peak search on the combined spectrum analysis results within the period of the preamble code, and obtain the frequency point corresponding to the FFT maximum value as the carrier frequency deviation value obtained this time;
[0050] Step 7: Compare and confirm the frequency deviation value obtained in this frame with the three frequency deviation values obtained in the previous three frames. If it is less than the threshold, the frequency deviation value of this frame is considered valid and is output;
[0051] Step 8: Adjust the output frequency of the local carrier according to the obtained effective frequency deviation value, and obtain a zero intermediate frequency signal after down-conversion to complete frequency correction.
[0052] like Figure 2 As shown, a multivariate spread spectrum calibration device based on preamble reconstruction includes a preamble reconstruction module 1, first to K-th digital orthogonal down-conversion modules 2-1 to 2-K, first to K-th PMF-FFT modules 3-1 to 3-K, first to K-th modulus value modules 4-1 to 4-K, a maximum ratio merging module 5, a continuous window peak search module 6 and a frequency deviation calculation and confirmation module 7;
[0053] Among them, the preamble reconstruction module 1 is used to perform M-ary spread spectrum, oversampling and partition reconstruction on the preamble, and send the reconstructed preamble to the first to K-th PMF-FFT modules 3-1 to 3-K; the first to K-th digital orthogonal down-conversion modules 2-1 to 2-K are respectively used to perform orthogonal down-conversion on each digital signal to form an I-way signal and a Q-way signal, and send them to the first to K-th PMF-FFT modules 3-1 to 3-K; the first to K-th PMF-FFT modules 3-1 to 3-K are respectively used to perform partial correlation, zero padding and FFT operation on the orthogonal down-converted I and Q-way signals to form a complex number result of spectrum analysis, and send them to the first to K-th modulus value taking modules 4-1 to 4 -K; the first to Kth modulus value taking modules 4-1 to 4-K are respectively used to take the modulus values of the complex results of the spectrum analysis and send them to the maximum ratio merging module 5; the maximum ratio merging module 5 is used to perform maximum ratio merging on the spectrum modulus values of the K branches, and send the merged result to the continuous window peak search module 6; the continuous window peak search module 6 is used to search for the peak value in the continuous window and mark the position, and send the peak position to the frequency deviation calculation and confirmation module 7; the frequency deviation calculation and confirmation module 7 is used to obtain the frequency deviation estimation result according to the peak position calculation, and after comparison and confirmation, output the valid result of the frequency deviation to the first to Kth orthogonal down-conversion modules 2-1 to 2-K to correct the local carrier frequency.
[0054] The embodiment of the preamble reconstruction module 1, the first to Kth digital orthogonal down-conversion modules 2-1 to 2-K, the first to Kth PMF-FFT modules 3-1 to 3-K, the first to Kth modulus value modules 4-1 to 4-K, the maximum ratio merging module 5, the continuous window peak search module 6 and the frequency deviation calculation and confirmation module 7 are all made of K7 series FPGA chips produced by Xilinx Corporation of the United States.
[0055] like Figure 3 As shown, the preamble reconstruction module 1 includes an M-ary spreading module 9, an oversampling module 10 and a partitioning module 11;
[0056] The M-ary spreading module 9 is used to spread the preamble code according to the symbol and send it to the oversampling module 10; the oversampling module 10 is used to oversample the code chip by A (A≥4) times to form a length of L=A×N×B / 2 M The reconstructed preamble data is sent to the partition module 11; wherein N is the code length of the preamble, M is the number of spread spectrum elements, and B is the code length of the PN code corresponding to the spread spectrum; the partition module 11 is used to perform partitioning according to the code length and the required frequency correction accuracy, and send the obtained reconstructed preamble code to the first to KPMF-FFT modules 3-1 to 3-K respectively.
[0057] The preamble generation module 8, the M-ary spectrum spreading module 9, the oversampling module 10, and the partitioning module 11 of the embodiment are all made of K7 series FPGA chips produced by Xilinx Corporation of the United States.
[0058] like Figure 4 As shown, the first to the KPMF-FFT modules 3-1 to 3-K respectively include a storage module 12, a first and a second cache modules 13-1 and 13-2, a first to a second L / X sub-correlation modules 14-1 to 14-2L / X, a zero padding module 15, a P-point FFT module 16 and a read-write control module 16;
[0059] The storage module 12 is used to store the reconstructed preamble sequence in the local memory, and output it to the first to second L / X sub-correlation modules 14-1 to 14-2L / X under the control of the read-write control module 16; the first and second cache modules 13-1 and 13-2 are respectively used to cache the down-converted I and Q signals and send them to the first to second L / X sub-correlation modules 14-1 to 14-2L / X under the control of the read-write control module 16; the first to second L / X sub-correlation modules 14-1 to 14-2L / X are respectively used to cache the down-converted I and Q signals and send them to the first to second L / X sub-correlation modules 14-1 to 14-2L / X under the control of the read-write control module 16. Under the control of the write control module 16, the signals of the storage module 12 and the first and second cache modules 13-1 and 13-2 are read, sub-correlation is performed, and the L / X path sub-correlation peak complex signal obtained by correlation is sent to the zero-filling module 15; the zero-filling module 15 is used to fill zero after each result of the sub-correlation and send it to the P-point FFT module 16; the P-point FFT module 16 is used to use the IP core to perform P-point FFT operation on the zero-filled correlation result to obtain complex I and Q results, and send them to the first to Kth modulus value modules 4-1 to 4-K respectively.
[0060] The storage module 12, the first to second cache modules 13-1 to 13-2, the first to second L / X sub-correlation modules 14-1 to 14-2L / X, the zero padding module 15, the P-point FFT module 16 and the read-write control module 16 are all made of K7 series FPGA chips produced by Xilinx Corporation of the United States.
[0061] The power supply 17 of the present invention provides a DC working voltage for the entire modulation and demodulation device. The embodiment is made of a commercially available general-purpose integrated voltage-stabilized DC power supply module, and its output +V voltage is +5V.
[0062] The brief working principle of a multi-element spread spectrum calibration device based on preamble reconstruction of the present invention is as follows:
[0063] The receiving end performs orthogonal down-conversion on the analog-to-digital converted signal to obtain K baseband signals, where K is determined by the diversity multiplicity, and each channel is a diversity branch. The preamble with a code length of N is spread again by M-ary, and M-ary spreading corresponds to 2 M The PN code with a barcode length of B is oversampled by A times and partitioned into segments with a length of X, forming a PN code with a length of L = 4 × N × B / 2. MThe preamble reconstruction data is stored in the local storage module. Sliding correlation is performed with the down-converted signal on each diversity branch. The matching results are partitioned and accumulated to reconstruct a new sequence according to the required frequency deviation detection range and accuracy. This sequence is sent to the FFT module for N-point FFT, and the modulus value is taken to obtain the spectrum analysis result. This result is sent to the continuous window peak search module to search for the peak position, and then the frequency deviation is calculated and compared with the results of the previous frequency deviation values. Invalid results are eliminated, and the valid values are sent to the frequency deviation correction down-conversion module for frequency correction to obtain a down-converted signal without frequency deviation, and then demodulated.
[0064] The installation structure of the present invention is as follows: Figure 1 , Figure 2 All circuits are installed on a printed board with a length×width of 70×90 mm, and then the printed board is installed in a chassis of a scattering communication device to assemble the present invention.
Claims
1. A multi-element spread spectrum calibration method based on preamble reconstruction, characterized in that: The following steps are involved: Step 1: Spread the preamble and oversample it by A times, and partition it into segments of length X to form a segment of length L = A × N × B / 2 M The reconstructed preamble data is stored in the local storage module; where N is the code length of the preamble, M is the number of spread spectrum elements, B is the code length of the PN code corresponding to the spread spectrum, and the value of X is determined by the calibration range and accuracy, A≥4; Step 2: partition the down-converted I and Q signals of the oversampled diversity branch into segments of length X and store them in a local cache; Step 3: Each time a sampling point is written, the data in the storage module and the cache are read, and a serial correlation operation is performed on the reconstructed preamble code and the down-converted I and Q signals, and L / X partial correlation values are obtained for the I and Q signals of each diversity branch; Step 4: pad the I and Q signals of the L / X partial correlation values of each diversity branch with zeros, perform P-point FFT operation, and then take the modulus value; wherein P is an integer power of 2; Step 5: Combining the spectrum modulus values of multiple diversity branches at the maximum ratio; Step 6: Perform peak search on the combined spectrum analysis results within the period of the preamble code, and obtain the frequency point corresponding to the FFT maximum value as the carrier frequency deviation value obtained this time; Step 7: Compare the carrier frequency deviation value obtained in this frame with the frequency deviation values obtained in previous frames. If it is less than the threshold, the frequency deviation value of this frame is valid, and the process goes to step 8. Otherwise, the process goes back to step 2. Step 8: Adjust the output frequency of the local carrier according to the obtained effective frequency deviation value, and obtain a zero intermediate frequency signal after down-conversion to complete frequency correction.
2. A multi-element spread spectrum calibration device based on preamble reconstruction, characterized in that: It comprises a preamble reconstruction module (1), first to K-th digital orthogonal down-conversion modules (2-1) to (2-K), first to K-th PMF-FFT modules (3-1) to (3-K), first to K-th modulus value modules (4-1) to (4-K), a maximum ratio merging module (5), a continuous window peak search module (6), and a frequency deviation calculation and confirmation module (7); The preamble reconstruction module (1) is used to perform M-ary spreading, oversampling and partition reconstruction on the preamble, and send the reconstructed preamble to the first to Kth PMF-FFT modules (3-1) to (3-K); The first to K-th digital orthogonal down-conversion modules (2-1) to (2-K) are respectively used to perform orthogonal down-conversion on each digital signal to form an I-channel signal and a Q-channel signal, and send them to the first to K-th PMF-FFT modules (3-1) to (3-K); The first to Kth PMF-FFT modules (3-1) to (3-K) are respectively used to perform partial correlation, zero padding and FFT operations on the I and Q path signals of the orthogonal down-conversion to form complex results of spectrum analysis, and send them to the first to Kth modulus value modules (4-1) to (4-K); The first to Kth modulus value taking modules (4-1) to (4-K) are used to take modulus values of complex number results of spectrum analysis and send them to the maximum ratio combining module (5); The maximum ratio merging module (5) is used to perform maximum ratio merging on the spectrum modulus values of the K branches, and send the merging result to the continuous window peak search module (6); The continuous window peak search module (6) is used to search for a peak value in the continuous window and mark the position, and send the peak position to the frequency deviation calculation and confirmation module (7); The frequency deviation calculation and confirmation module (7) is used to obtain a frequency deviation estimation result according to the peak position calculation, and after comparison and confirmation, output the valid frequency deviation result to the first to Kth digital orthogonal down-conversion modules (2-1) to (2-K) to correct the local carrier frequency; The preamble reconstruction module (1) includes an M-ary spreading module (9), an oversampling module (10) and a partitioning module (11); The M-ary spreading module (9) is used to spread the preamble code according to the symbol and send it to the oversampling module (10); The oversampling module (10) is used to perform A-fold oversampling on the code chip to form a length of L=A×N×B / 2 M The reconstructed preamble data is sent to the partition module (11); wherein N is the code length of the preamble, M is the number of spread spectrum elements, B is the code length of the PN code corresponding to the spread spectrum, and A≥4; The partitioning module (11) is used to partition according to the code length and the required frequency calibration accuracy, and send the obtained reconstructed preamble code to the first to Kth PMF-FFT modules (3-1) to (3-K) respectively.
3. The multi-element spread spectrum calibration device based on preamble reconstruction according to claim 2, characterized in that: The first to Kth PMF-FFT modules (3-1) to (3-K) respectively include a storage module (12), a first and a second buffer module (13-1) and (13-2), a first to a second L / X sub-correlation module (14-1) to (14-2L / X), a zero padding module (15), a P-point FFT module (16) and a read / write control module (17); The storage module (12) is used to store the reconstructed preamble sequence in a local memory, and output it to the first to second L / X sub-correlation modules (14-1) to (14-2L / X) under the control of the read / write control module (17); The first and second buffer modules (13-1) and (13-2) are respectively used to buffer the down-converted I and Q signals and then send them to the first to second L / X sub-correlation modules (14-1) to (14-2L / X) under the control of the read-write control module (17); The first to second L / X sub-correlation modules (14-1) to (14-2L / X) are respectively used to read signals from the storage module (12) and the first and second cache modules (13-1) and (13-2) under the control of the read-write control module (17), perform sub-correlation, and send the L / X path sub-correlation peak complex signals obtained by correlation to the zero-filling module (15); The zero-filling module (15) is used to fill zero after each result of the sub-correlation and send it to the P-point FFT module (16); The P-point FFT module (16) is used to perform a P-point FFT operation on the zero-filled correlation result using the IP core to obtain complex I and Q results, and send them to the first to Kth modulus value modules (4-1) to (4-K) respectively.
