A low earth orbit satellite communication timing advance estimation method based on affine frequency division multiplexing waveform
Through the low-orbit satellite communication timing advance estimation method based on the simulated radio frequency division multiplexing waveform, the problem of high frequency offset complexity in low-orbit satellite communication is solved, and the timing advance estimation accuracy and random access efficiency are improved under low complexity.
Patent Information
- Application Number
- CN202411194273.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-28
- Publication Date
- 2025-10-10
- Estimated Expiration
- 2044-08-28
AI Technical Summary
In existing low-orbit satellite communications, the timing advance estimation method has the problem of high frequency offset estimation complexity, which increases the algorithm complexity and makes it difficult to meet the requirements of low-energy operation.
A timing advance estimation method for low-orbit satellite communications based on simulated radio frequency division multiplexing waveform is adopted. By generating a preamble code and performing simulated radio frequency division multiplexing modulation and demodulation, combined with power delay spectrum and peak detection, a low-complexity detection algorithm is designed to eliminate frequency offset and improve the accuracy of timing advance estimation.
It achieves robustness to carrier frequency offset under low complexity, improves the accuracy of timing advance estimation and random access efficiency, and shows superior performance especially under low signal-to-noise ratio and multi-user access conditions.
Smart Images
Figure CN119071122B_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the field of communication technology, more particularly, to a low earth orbit satellite communication timing advance estimation method based on affine frequency division multiplexing waveform. BACKGROUND
[0002] In recent years, with the growing demand for wireless communication, the society has put forward higher requirements for low latency, high reliability and large coverage communication. Although low power wide area network has been widely deployed, it cannot fully meet these needs. In addition, it is inefficient and costly to build ground base stations in remote areas. Therefore, the research on satellite communication is of great significance to realize seamless coverage communication in the global range. According to the different orbital heights, there are various types of satellite communication: low earth orbit, medium earth orbit and geostationary orbit. The low earth orbit satellite communication has lower cost and shorter delay time, which makes it more valuable. Therefore, the industry and academia are interested in low earth orbit satellite communication. However, the high-speed movement of low earth orbit satellites leads to a large frequency offset in the preamble transmission process, which needs to be studied for the timing advance estimation method.
[0003] There have been many studies on timing advance estimation methods, mainly from three aspects: preamble sequence composition, preamble structure and timing advance detection algorithm. In the preamble sequence composition structure, Cui and He proposed a method based on conjugate symmetric preamble sequence, which eliminates the interference of integer frequency offset by averaging two groups of peak positions, but cannot solve the problem caused by fractional frequency offset. In the preamble structure, the existing design uses a preamble containing two root indexes, and cooperates with a correlation peak search method to eliminate frequency offset. This method improves the utilization rate of preamble index, leading to the reduction of available preambles, and the search algorithm uses the exhaustive method with high complexity. In the detection algorithm, it mainly relies on increasing the sequence length of correlation operation in timing advance detection to improve the noise performance. However, the correlation of the preamble sequence is weakened, so it is necessary to make a trade-off between complexity and detection rate.
[0004] The existing timing advance estimation methods still have some deficiencies, the most important of which is that they all need to estimate or compensate the frequency offset, which inevitably increases the complexity of the algorithm, bringing challenges to the low earth orbit satellite communication which needs to run with low energy consumption. Therefore, it is necessary to design a preamble with low complexity and carrier frequency offset robustness for low earth orbit satellite communication and its detection algorithm to improve the accuracy of timing advance estimation and random access efficiency. SUMMARY
[0005] The present invention addresses the issues of carrier frequency offset and complexity in low-orbit satellite communications, focusing on the discrete Fourier transform-based pseudo-RFD multiplexing waveform for random access preamble transmission in low-orbit satellite communications. This waveform is robust to dual-selection channels. First, pseudo-RFD multiplexing modulation can eliminate the frequency offset of the preamble sequence. Then, in response to the needs of low-orbit satellite communications, since only a single-segment subsequence is required for timing advance estimation, the length of the correlation operation is short, and the detection complexity of the preamble code is also reduced. In addition, in order to improve the accuracy of timing advance estimation, a detection algorithm is proposed to distinguish the main path peak.
[0006] To achieve the above objectives, the technical solutions of the present invention are as follows:
[0007] A method for estimating timing advance of low-orbit satellite communications based on a simulated radio frequency division multiplexing waveform comprises the following steps:
[0008] S1: At the transmitting end, generate the preamble x for timing advance detection u (n),n=0,1,……,N-1;
[0009] S2: Modulate the preamble using a simulated RF division multiplexing waveform and add a cyclic prefix to obtain the transmitted signal t(n). Then, the signal is sent to the wireless channel via RF to obtain the received signal r(n).
[0010] S3: At the receiving end, the cyclic prefix of the received signal is removed and demodulated by a pseudo-RFD demodulator to obtain a demodulated signal z(n);
[0011] S4: Determine the preamble sequence x u (n), and obtain the local preamble sequence s(n) corresponding to the index number;
[0012] S5: Calculate the power delay spectrum, perform peak detection, and obtain the peak index set
[0013] S6: In the process, each peak is judged and eliminated for multipath peaks and pseudo peaks to obtain the processed peak index set
[0014] S7: Calculate the timing advance d in the processed peak index set and output the result.
[0015] Preferably, the design of the preamble in step S1 is specifically as follows:
[0016] The ZC sequence is used as the leading sequence constituting the preamble code, and its expression is:
[0017]
[0018] Where V is the sequence length, u is the root sequence index, and N is the preamble length.
[0019] Preferably, in step S2, the preamble is modulated by RFDM to obtain the transmission signal t(n), specifically:
[0020] S2.1: Multiply the preamble code by each modulation matrix, which is expressed as:
[0021] t(n)=FMFJS·x u (n)
[0022] Where, c1 and c2 are the RFDM modulation parameters, sign(c1) is the sign function of c1; is the IDFT matrix, specifically:
[0023]
[0024] The expression of the preamble structure is:
[0025] x u (n)=[c(n),c(n),……,c(n)]
[0026] Where, As a subsequence of the preamble, is a cyclic shift set, any N that satisfies the condition cs is a prime number and [N cs,1 ,N cs,2 ]>V;
[0027] S2.2: Send the modulated signal to the wireless channel for transmission via radio frequency;
[0028] Preferably, in step S2.2, the received signal is sent to the wireless channel and is expressed as:
[0029]
[0030] Where L is the number of multipaths, h l is the channel gain, d is the timing advance, d l is the multipath delay; ε=f / Δf RA is the normalized carrier frequency offset, f is the carrier frequency offset in the channel, Δf RA is the subcarrier spacing of the transmitted signal; w(n) is the additive white Gaussian noise.
[0031] Preferably, in step S3, the received signal is demodulated using RF division multiplexing, specifically:
[0032] S3.1: The calculation formula for the demodulated signal is:
[0033] z(n)=S H F H RM H G H F H ·r(n)
[0034] Where the superscript H represents the Hermitian matrix of the matrix, is a diagonal matrix with elements on the diagonal
[0035] S3.2: Demodulate the received signal according to the calculation formula to derive the preamble code expression at the receiving end;
[0036] Preferably, the demodulated signal obtained in step S3.2 is expressed as:
[0037]
[0038] Where, X 1,0 is x u (n) is the value of IDFT at index 0 when u=1, and w′(n) is the noise after RFDM demodulation. The demodulated signal can completely eliminate the influence of frequency offset ε on the preamble, and the cyclic shift of the preamble only reflects d and d. l , which can be obtained through correlation operations.
[0039] Preferably, the index number of the preamble is determined in step S4, specifically:
[0040] S4.1: All local sequences x(k) are correlated with the demodulated signal z(n) once to obtain the power delay spectrum, which is expressed as
[0041] S4.2: Sort all the obtained power delay profiles according to their variances, and use the preamble code index number s(n) corresponding to the power delay profile with the largest variance as the preamble code index number of the signal received by the current receiving end.
[0042] Preferably, step S5 performs peak detection on the preamble of the local corresponding index number, specifically:
[0043] The peak value of the power delay spectrum is recorded as T. If T < γ, it is determined that the index number of the local sequence is wrong, and the next local sequence correlation operation is performed until there is a correlation peak T i >γ, store all peaks above the threshold and their corresponding indexes into the collection
[0044] The setting of the threshold γ is as follows:
[0045]
[0046] In the formula, a is a threshold constant, which varies in the range of [1, 1.1, 1.2, …, 19.9, 20], and a is traversed from small to large in the detection of the preamble index number until the minimum threshold constant satisfying the set detection false alarm rate is obtained.
[0047] Preferably, the peaks are subjected to multipath peak and pseudo-peak decision and elimination in step S6, and the decision expression is:
[0048]
[0049] First, the first peak d first above the threshold value is found in step S5. Then, the peaks in the maximum multipath delay range of the first peak are searched, the peak with the maximum peak value is found as the first path, and the remaining peaks in the range are eliminated to obtain the peak index set
[0050] Preferably, the timing advance d is calculated in step S7, and the calculation is specifically as follows:
[0051] S7.1: Calculate the interval of all peaks in the set
[0052] S7.2: Traverse all values N cs in the cyclic shift set, and compare d p [i] with N cs If d p [i] ∈ {N cs -1, N cs , N cs +1}, it is determined that the correlation peak corresponding to the preamble is found.
[0053] S7.3: Take out the peaks with the same peak interval, and determine the first peak according to the maximum peak interval, and the peak index of the first peak is the timing advance d. BRIEF DESCRIPTION OF DRAWINGS
[0054] Figure 1 FIG. 1 is a flowchart of the method of the present application.
[0055] Figure 2 FIG. 2 is a comparison diagram of timing advance estimation performance of different algorithms in a Gaussian white noise channel provided by the embodiment.
[0056] Figure 3 FIG. 3 is a comparison diagram of timing advance estimation performance of different algorithms in a doublet channel provided by the embodiment. DETAILED DESCRIPTION
[0057] The accompanying drawings are for illustrative purposes only and are not to be construed as limiting this patent;
[0058] In order to better illustrate this embodiment, some parts in the drawings may be omitted, enlarged, or reduced, and do not represent the actual product size;
[0059] It is understandable to those skilled in the art that some well-known structures and descriptions thereof may be omitted in the drawings.
[0060] The technical solution of the present invention will be further described below with reference to the accompanying drawings and embodiments.
[0061] Example 1
[0062] This example provides a low-orbit satellite communication timing advance estimation method based on simulated radio frequency division multiplexing waveform, such as Figure 1 As shown, the following steps are included:
[0063] S1: At the transmitting end, generate the preamble x for timing advance detection u (n),n=0,1,……,N-1;
[0064] S2: Modulate the preamble using a simulated RF division multiplexing waveform and add a cyclic prefix to obtain the transmitted signal t(n). Then, the signal is sent to the wireless channel via RF to obtain the received signal r(n).
[0065] S3: At the receiving end, the cyclic prefix of the received signal is removed and demodulated by a pseudo-RFD demodulator to obtain a demodulated signal z(n);
[0066] S4: Determine the preamble sequence x u (n), and obtain the local preamble sequence s(n) corresponding to the index number;
[0067] S5: Calculate the power delay spectrum, perform peak detection, and obtain the peak index set
[0068] S6: In the process, each peak is judged and eliminated for multipath peaks and pseudo peaks to obtain the processed peak index set
[0069] S7: Calculate the timing advance d in the processed peak index set and output the result.
[0070] Example 2
[0071] This embodiment, based on the first embodiment, further discloses the following contents:
[0072] The design of the preamble in step S1 is specifically as follows:
[0073] The ZC sequence is used as the leading sequence constituting the preamble code, and its expression is:
[0074]
[0075] Where V is the sequence length, u is the root sequence index, and N is the preamble length.
[0076] In step S2, the preamble is modulated by radio frequency division multiplexing to obtain a transmission signal t(n), specifically:
[0077] S2.1: Multiply the preamble code by each modulation matrix, which is expressed as:
[0078] t(n)=FMFJS·x u (n)
[0079] Where, c1 and c2 are the RFDM modulation parameters, sign(c1) is the sign function of c1; is the IDFT matrix, specifically:
[0080]
[0081] The expression of the preamble structure is:
[0082] x u (n)=[c(n),c(n),……,c(n)]
[0083] Where, As a subsequence of the preamble, is a cyclic shift set, any N that satisfies the condition cs is a prime number and [N cs,1 ,N cs,2 ]>V;
[0084] S2.2: Send the modulated signal to the wireless channel for transmission via radio frequency;
[0085] In step S2.2, the received signal is sent to the wireless channel and expressed as:
[0086]
[0087] Where L is the number of multipaths, h l is the channel gain, d is the timing advance, d l is the multipath delay; ε=f / Δf RA is the normalized carrier frequency offset, f is the carrier frequency offset in the channel, Δf RAis the subcarrier spacing of the transmitted signal; w(n) is the additive white Gaussian noise.
[0088] In step S3, the received signal is demodulated using RF division multiplexing, specifically:
[0089] S3.1: The calculation formula for the demodulated signal is:
[0090] z(n)=S H F H RM H G H F H ·r(n)
[0091] Where the superscript H represents the Hermitian matrix of the matrix, is a diagonal matrix with elements on the diagonal
[0092] S3.2: Demodulate the received signal according to the calculation formula to derive the preamble code expression at the receiving end;
[0093] The demodulated signal obtained in step S3.2 is expressed as:
[0094]
[0095] Where, X 1,0 is x u (n) is the value of IDFT at index 0 when u=1, and w′(n) is the noise after RFDM demodulation. The demodulated signal can completely eliminate the influence of frequency offset ε on the preamble, and the cyclic shift of the preamble only reflects d and d. l , which can be obtained through correlation operations.
[0096] The index number of the preamble is determined in step S4, specifically:
[0097] S4.1: All local sequences x(k) are correlated with the demodulated signal z(n) once to obtain the power delay spectrum, which is expressed as
[0098] S4.2: Sort all the obtained power delay profiles according to their variances, and use the preamble code index number s(n) corresponding to the power delay profile with the largest variance as the preamble code index number of the signal received by the current receiving end.
[0099] Step S5 performs peak detection on the preamble code of the local corresponding index number, specifically:
[0100] The peak value of the power delay spectrum is recorded as T. If T < γ, it is determined that the index number of the local sequence is wrong, and the next local sequence correlation operation is performed until there is a correlation peak T i>γ, store all peaks above the threshold and their corresponding indexes into the collection
[0101] The setting of the threshold γ is as follows:
[0102]
[0103] Where α is a threshold constant, ranging from [1, 1.1, 1.2, …, 19.9, 20]. α is traversed from small to large in the detection of the preamble index number until the minimum threshold constant that satisfies the set detection missed alarm rate is obtained.
[0104] In step S6, each peak is judged and eliminated as a multipath peak and a pseudo peak. The judgment expression is:
[0105]
[0106] First in Find the first peak d above the threshold first Then, search for the peak within the maximum multipath delay range of the first peak, find the peak with the largest peak value as the first path, eliminate the remaining peaks within the range, and obtain the peak index set
[0107] In step S7, the timing advance d is calculated as follows:
[0108] S7.1: Computational Sets All peak intervals in
[0109] S7.2: Cyclic shift all values N in the set cs Traverse and d p [i] and N cs For comparison, if d p [i]∈{N cs -1,N cs ,N cs +1}, it is determined that the correlation peak corresponding to the preamble code is found;
[0110] S7.3: Take all peaks with the same peak interval, determine the first peak based on the maximum peak interval, and use the peak index as the timing advance d.
[0111] Example 3
[0112] This embodiment provides the following specific embodiments based on Embodiment 1 and Embodiment 2:
[0113] The simulation parameters are set as follows: the channels used are additive white Gaussian noise (AWGN) and dual-selection channels, the dual-selection channel gains are -0.394dB and -23.373dB, the normalized multipath delay is 14.8124, the round-trip delay difference range is [0,4.8]ms, the sampling frequency is 15.46MHz, the signal-to-noise ratio (SNR) setting range is -20dB to 0dB, the bandwidth is 1.08MHz, the subcarrier spacing is 1.25kHz, the carrier frequency is 1.621GHz, the satellite orbit altitude is 780km, the elevation angle is 10°, the carrier frequency offset is 625Hz, and the Monte Carlo number is 10,000 times.
[0114] The detection performance of the new method proposed in this invention (hereinafter referred to as SSP) is compared with the existing timing advance estimation algorithms LDCC, TPC and LNRP in the AWGN channel. When the result differs from the actual value by one sampling point, it is considered a successful detection. The details are as follows: Figure 2 As shown, the detection probability of the four timing advance estimation schemes increases with increasing signal-to-noise ratio (SNR), and all ultimately achieve complete detection, demonstrating the effectiveness of the timing advance estimation schemes. Under the same conditions, the proposed scheme's detection performance far outperforms existing detection algorithms, with a performance improvement of approximately 3-5 dB. This is due to the advantages of the new preamble structure, where the statistical averaging of multiple subsequences improves the preamble's noise resistance. Furthermore, the SSP of this scheme can only achieve timing advance estimation at low SNRs, with a detection probability improvement of approximately 0.3 compared to other schemes. When the SNR exceeds -8 dB, the timing advance estimation performance of each scheme remains similar. This is due to the reduced noise interference on the preamble, allowing almost complete timing advance estimation.
[0115] The above examples show that the scheme proposed in the present invention can realize timing advance estimation under AWGN channels. The multi-segment subsequence repeated cascade preamble code structure used has advantages under low signal-to-noise ratio, and the proposed scheme has better detection performance.
[0116] On this basis, while keeping the simulation parameters unchanged, the preamble code in the method proposed by the present invention is simulated in a dual-selection channel, and there are two users initiating random access at the same time, as follows: Figure 3 As shown, the detection probability of the four timing advance estimation schemes increases with increasing SNR, demonstrating the effectiveness of the estimation schemes in dual-selection channels. Furthermore, TPC and LDCC reach detection probability saturation as the SNR increases, failing to achieve full detection. LNRP achieves full detection only at a SNR of 0 dB. In contrast, the SSP proposed in this invention achieves a 10 dB gain, achieving full detection at a SNR of -10 dB, resulting in better timing advance estimation performance.
[0117] The above examples show that when the channel is a dual-selection channel and multiple users are accessing, the method of the present invention can achieve significantly better timing advance estimation effect, and the superiority is particularly obvious under low signal-to-noise ratio.
[0118] The same or similar reference numerals correspond to the same or similar components;
[0119] The positional relationships described in the drawings are for illustrative purposes only and should not be construed as limiting this patent.
[0120] Obviously, the above examples of the present invention are merely illustrative examples for the purpose of clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention. Those skilled in the art will appreciate that other variations or modifications can be made based on the above description. It is not necessary and impossible to enumerate all embodiments here. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention are intended to be included within the scope of protection of the claims of the present invention.
Claims
1. A method for estimating timing advance of low-orbit satellite communications based on simulated radio frequency division multiplexing waveform, characterized in that: The following steps are involved: S1: At the transmitting end, generate the preamble x for timing advance detection u (n), n=0,1,……,N-1, N is the preamble length; the ZC sequence is used as the preamble sequence constituting the preamble, and its expression is: Where V is the sequence length and u is the root sequence index; The expression of the preamble structure is: z u (n)=[c(n),c(n),……,c(n)] Where, As a subsequence of the preamble, is a cyclic shift set, any N that satisfies the condition cs is a prime number and [N cs,1 ,N cs,2 ]>V; S2: Modulate the preamble using a simulated RF division multiplexing waveform and add a cyclic prefix to obtain the transmitted signal t(n). Then, the signal is sent to the wireless channel via RF to obtain the received signal r(n). S3: At the receiving end, the cyclic prefix of the received signal is removed and demodulated by a pseudo-RFD demodulator to obtain a demodulated signal z(n); S4: Determine the preamble sequence x u (n), and obtain the local preamble sequence s(n) corresponding to the index number; S5: Calculate the power delay spectrum, perform peak detection, and obtain the peak index set S6: In the process, each peak is judged and eliminated for multipath peaks and pseudo peaks to obtain the processed peak index set S7: Calculate the timing advance d in the processed peak index set and output the result.
2. The method for estimating timing advance of low-orbit satellite communication based on simulated radio frequency division multiplexing waveform according to claim 1, characterized in that: In step S2, the preamble is modulated by radio frequency division multiplexing to obtain a transmission signal t(n), specifically: S2.1: Multiply the preamble code by each modulation matrix, which is expressed as: t(n)=FMFJS·x u (n) Where, c1 and c2 are the RFDM modulation parameters, sign(c1) is the sign function of c1; is the IDFT matrix, specifically: S2.2: Send the modulated signal to the wireless channel for transmission via radio frequency.
3. The method for estimating timing advance of low-orbit satellite communication based on simulated radio frequency division multiplexing waveform according to claim 2, characterized in that: In step S2.2, the received signal is sent to the wireless channel and expressed as: Where L is the number of multipaths, h l is the channel gain, d is the timing advance, d l is the multipath delay; ε=f / Δf RA is the normalized carrier frequency offset, f is the carrier frequency offset in the channel, Δf RA is the subcarrier spacing of the transmitted signal; w(n) is the additive white Gaussian noise.
4. The method for estimating timing advance of low-orbit satellite communication based on simulated radio frequency division multiplexing waveform according to claim 3, characterized in that: In step S3, the received signal is demodulated using RF division multiplexing, specifically: S3.1: The calculation formula for the demodulated signal is: z(n)=S H F H RM H G H F H ·r(n) Where the superscript H represents the Hermitian matrix of the matrix, is a diagonal matrix with elements on the diagonal S3.2: Demodulate the received signal according to the calculation formula to derive the preamble code expression at the receiving end.
5. The method for estimating timing advance of low-orbit satellite communication based on simulated radio frequency division multiplexing waveform according to claim 4, characterized in that: The demodulated signal obtained in step S3.2 is expressed as: Where, X 1,0 is x u (n) is the value of IDFT at index 0 when u=1, and w′(n) is the noise after RFDM demodulation. The demodulated signal can completely eliminate the influence of frequency offset ε on the preamble, and the cyclic shift of the preamble only reflects d and d. l , which can be obtained through correlation operations.
6. The method for estimating timing advance of low-orbit satellite communication based on simulated radio frequency division multiplexing waveform according to claim 5, characterized in that: The index number of the preamble is determined in step S4, specifically: S4.1: All local sequences x(k) are correlated with the demodulated signal z(n) once to obtain the power delay spectrum, which is expressed as S4.2: Sort all the obtained power delay profiles according to their variances, and use the preamble code index number s(n) corresponding to the power delay profile with the largest variance as the preamble code index number of the signal received by the current receiving end.
7. The method for estimating timing advance of low-orbit satellite communication based on simulated radio frequency division multiplexing waveform according to claim 6, characterized in that: Step S5 performs peak detection on the preamble code of the local corresponding index number, specifically: The peak value of the power delay spectrum is recorded as T. If T < γ, it is determined that the index number of the local sequence is wrong, and the next local sequence correlation operation is performed until there is a correlation peak T i >γ, store all peaks above the threshold and their corresponding indexes into the collection The setting of the threshold γ is as follows: Where α is a threshold constant, ranging from [1, 1.1, 1.2, …, 19.9, 20]. α is traversed from small to large in the detection of the preamble index number until the minimum threshold constant that satisfies the set detection missed alarm rate is obtained.
8. The method for estimating timing advance of low-orbit satellite communication based on simulated radio frequency division multiplexing waveform according to claim 7, characterized in that: In step S6, each peak is judged and eliminated as a multipath peak and a pseudo peak. The judgment expression is: First in Find the first peak d above the threshold first Then, search for the peak within the maximum multipath delay range of the first peak, find the peak with the largest peak value as the first path, eliminate the remaining peaks within the range, and obtain the peak index set 9. The method for estimating timing advance of low-orbit satellite communication based on simulated radio frequency division multiplexing waveform according to claim 8, characterized in that: In step S7, the timing advance d is calculated as follows: S7.1: Computational Sets All peak intervals in S7.2: Cyclic shift all values N in the set cs Traverse and d p [i] and N cs For comparison, if d p [i]∈{N cs -1,N cs ,N cs +1}, it is determined that the correlation peak corresponding to the preamble code is found; S7.3: Take all peaks with the same peak interval, determine the first peak based on the maximum peak interval, and use the peak index as the timing advance d.