A random access preamble detection method and device suitable for 5G low-orbit satellite
By using time domain-related computing and multi-segment superposition methods in 5G low-orbit satellite communication, the problem of leading sequence detection failure caused by the maximum round-trip delay difference is solved, and the function of correctly detecting the leading ID and estimating the timing advance amount is realized, which improves the robustness and compatibility of the system.
Patent Information
- Application Number
- CN202211638966.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-12-20
- Publication Date
- 2025-04-25
- Estimated Expiration
- 2042-12-20
AI Technical Summary
In 5G low-orbit satellite communication, due to the large maximum round-trip delay difference, traditional detection algorithms cannot correctly detect the leading sequence sent by the user equipment, resulting in access failure.
A random access preamble detection method suitable for 5G low-orbit satellites is proposed. By shifting the length of the minimum round-trip delay backward at the access time by the on-site 5G base station, opening the detection window to receive the preamble sequence, and using time domain correlation operations and multi-stage superposition to obtain the power delay spectrum, randomly select the power delay spectrum generated by the root sequence, calculate the noise power estimate value and detection threshold value, and judge the preamble ID and estimate the timing advance amount.
Without relying on the navigation and positioning system, the preamble in the 5G format of the ground can be correctly detected, which improves the system's robustness and compatibility with 5G on the ground, and meets the maximum detection success rate under the 0.1% false alarm probability.
Smart Images

Figure CN116033457B_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the field of mobile communication technology, and specifically relates to a random access preamble detection method and device suitable for 5G low-orbit satellites in satellite mobile communications. Background Art
[0002] Limited by the cost of laying and the geographical environment, it is difficult for terrestrial networks to achieve global coverage of mobile communications. 3GPP has begun to propose the use of non-terrestrial networks (NTN) as a supplement to terrestrial networks and has proposed 5G-related standards for NTN. The use of low-orbit satellite communications integrated with 5GNR can achieve seamless switching between terrestrial and non-terrestrial networks.
[0003] In 5G NR, the user equipment needs to achieve uplink synchronization before interacting with the network. At this time, the terminal needs to initiate a random access process to inform the base station and obtain the timing advance. The first step in initiating the random access process is that the UE needs to send a preamble code. After the base station correctly detects the preamble ID and estimates the timing advance (TA), the subsequent random access process of Msg2 to Msg4 can be carried out.
[0004] NR uses Zadoff-Chu sequence to generate preamble. ZC is used to generate random access (RA) preamble due to its excellent autocorrelation and low peak-to-average ratio characteristics. The preamble sequence consists of cyclic prefix (CP), sequence part (SEQ) and guard interval (GT), and the sequence length is the sum of the lengths of the three. The role of CP is to ensure the integrity of the data intercepted in the preamble detection window, so that the sequence can be demodulated correctly, so that the impact of round-trip delay difference and delay spread is equivalent to the cyclic shift of ZC sequence. SEQ is the ZC sequence part. The length of ZC sequence is 139 and 839, and the length of 571 and 1151 is increased in the 3GPP R17 standard. SEQ can be composed of multiple segments of ZC repeated sequences to improve coverage performance. The role of GT is to prevent the sequence from interfering with the next time slot. There are 64 sequences that can be used for RA in each cell in the NR system. These sequences can be generated by cyclic shift based on the same root sequence, or by cyclic shift using different root sequences.
[0005] Under low-orbit satellite conditions, the radius of the service cell can reach hundreds of kilometers, and the communication delay between the terminal and the satellite at different locations in the cell may vary greatly, which causes the preamble code or other uplink signals sent by the terminal at the same time to arrive at the satellite base station at different times. The maximum round-trip delay difference (2×(maximum one-way delay-minimum one-way delay)) is much larger than the value in the ground system, and even exceeds the length of the sequence itself, that is, the distribution of the arrival time of the preamble code sequence sent by the user equipment is large, which leads to the ground 5G preamble detection scheme. The detection window may not be able to intercept the complete preamble sequence, that is, the PRACH signal sent by the user equipment cannot be correctly detected, causing the user equipment to fail to access.
[0006] From the above content, it can be seen that in the scenario of large maximum round-trip delay difference of 5G low-orbit satellite and without changing the transmission structure of the preamble code, the traditional detection algorithm of ground 5G cannot correctly detect the preamble sequence sent by the user equipment. It is necessary to propose an appropriate preamble detection method under the above scenarios and conditions to meet the requirements of correctly detecting the preamble and accurately estimating the timing advance. Summary of the invention
[0007] In order to solve the above problems, the present invention proposes a random access preamble detection method and device suitable for 5G low-orbit satellites, which can solve the problem of large maximum round-trip delay difference in cells.
[0008] The technical solution adopted by the present invention is:
[0009] A random access preamble detection method applicable to 5G low-orbit satellites comprises the following steps:
[0010] Step 1: The satellite-borne 5G base station gNB configures the access timing and cyclic shift step size of the random access preamble sequence, and broadcasts it through the system message SIB1;
[0011] Step 2: The user terminal UE sends the 5GNR preamble code at the corresponding access opportunity according to the SIB1 information;
[0012] Step 3: The satellite-borne 5G base station gNB shifts back the length of the minimum round-trip delay at the access timing, opens the detection window to receive the leading sequence, performs time domain correlation operations on the local sequences corresponding to all root sequences and the received sequence, and performs multiple superpositions according to the number of repetitions of sequences in different formats to obtain the final power delay spectrum sequence;
[0013] Step 4: Randomly select a power delay spectrum sequence generated by a root sequence to obtain the noise power estimate, and then calculate the noise power estimate based on the preset threshold coefficient T coe And the signal-to-noise ratio information obtains the corresponding detection threshold value Threshold;
[0014] Step 5: Compare the maximum value of the superimposed power delay spectrum sequence corresponding to all root sequences with the detection threshold Threshold. If it is greater than the detection threshold, determine the preamble ID and estimate the timing advance TA; if it is less than the detection threshold, determine that no preamble is sent.
[0015] Furthermore, the access timing of the preamble sequence in step 1 satisfies T PRACH <T w , where T PRACH is the PRACH access opportunity period, T w is the detection window length; the cyclic shift step size N CS Set to 0.
[0016] Furthermore, the specific method of step three is:
[0017] (301) Calculate the minimum round trip delay RTD of the current satellite cell min , the detection window length is set to the sum of the CP length, the sequence length SEQ and the maximum round-trip delay difference of the cell;
[0018] (302) performing a correlation operation on the local sequence and the received sequence directly in the time domain to obtain a correlated sequence;
[0019] (303) taking a modulus value of the correlated sequence, and then superimposing the sequence according to the number of sequence repetitions in different NR preamble code formats to obtain a superimposed power delay spectrum sequence;
[0020] (304) Determine whether all root sequences have generated corresponding power delay spectrum sequences. If not, select the next root sequence and return to step (302). Otherwise, exit the current step.
[0021] Furthermore, the specific method of step four is:
[0022] (401) setting the detection window length to be the sum of the CP length, the sequence length SEQ and the maximum round-trip delay difference of the cell;
[0023] (402) performing a correlation operation on the local sequence and a noise sequence of the detection window length in the time domain to obtain a correlated sequence;
[0024] (403) taking a modulus value of the correlated sequence, and then superimposing the sequence according to the number of sequence repetitions in different NR preamble code formats to obtain a superimposed power delay spectrum sequence;
[0025] (404) Counting the maximum values in the power delay spectrum and arranging these values in descending order to obtain the corresponding value of 0.1% false alarm probability;
[0026] (405) Change the signal-to-noise ratio, return to step (402), obtain the detection threshold under different signal-to-noise ratios, and use the ratio of the detection threshold value to the noise estimation value of the power delay spectrum sequence in step (403) as the threshold coefficient T coe ;
[0027] (6) The threshold coefficients under different signal-to-noise ratios are fitted to obtain a fitting curve of the threshold coefficient changing with the signal-to-noise ratio.
[0028] A random access preamble detection device applicable to a 5G low-orbit satellite, characterized in that it is used to implement the random access preamble detection method applicable to a 5G low-orbit satellite as described in any one of the above items, including a random access preamble sending device and a random access preamble receiving device, wherein:
[0029] A random access preamble sending device, used to send a corresponding 5G random access preamble code on a PRACH channel according to configuration information;
[0030] The preamble receiving device is used to receive the preamble code sequence, detect the preamble code ID and estimate the timing advance.
[0031] Furthermore, the random access preamble receiving device includes a time domain correlation operation unit, a multi-segment superposition unit and a detection threshold unit, wherein:
[0032] A time domain correlation operation unit, used for performing time domain correlation operation between a local sequence and a received sequence;
[0033] A multi-segment superposition unit is used for superposition of multiple segments of the sequence after time domain correlation to obtain the final power delay spectrum sequence;
[0034] The detection threshold unit is used to obtain the threshold coefficient required for correlation peak judgment according to different signal-to-noise ratios.
[0035] Compared with the prior art, the present invention has the following advantages:
[0036] 1. The ground 5G format preamble code can be directly detected without relying on the navigation and positioning system: The present invention takes the traditional ground 5G preamble code as the detection object, expands the preamble detection window and uses time domain correlation operation instead of frequency domain correlation. The ground 5G preamble can be detected without relying on ephemeris and positioning information, thereby improving the system's robustness and compatibility with ground 5G.
[0037] 2. A method for calculating the detection threshold under different signal-to-noise ratios is given to maximize the detection success rate under the premise of meeting the false alarm probability: The present invention uses the noise sequence as the receiving sequence, calculates and simulates the detection threshold of different preamble formats under various signal-to-noise ratios, and can ensure the maximum detection success rate under the condition of meeting the false alarm probability of 0.1%. BRIEF DESCRIPTION OF THE DRAWINGS
[0038] Figure 1 It is the overall flow chart of the present invention;
[0039] Figure 2 is a schematic diagram of a low-orbit satellite cell of the present invention;
[0040] Figure 3 It is a sequence processing schematic diagram of the present invention;
[0041] Figure 4 It is a system model diagram of the leading detection device of the present invention. DETAILED DESCRIPTION
[0042] The present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments.
[0043] A random access preamble detection method applicable to 5G low-orbit satellites comprises the following steps:
[0044] Step 1: The gNB configures the RACH Occasion (RO) and cyclic shift step size of the random access preamble sequence and broadcasts it through the system message SIB1.
[0045] Step 2: The UE sends the 5G NR preamble at the corresponding access opportunity according to the SIB1 configuration information;
[0046] Step 3: The gNB onboard shifts the minimum round-trip delay backward at the RO, opens the detection window to receive the leading sequence, performs time domain correlation operations on the local sequences corresponding to all root sequences and the received sequence, and performs multiple superpositions according to the number of repetitions of sequences in different formats to obtain the final power delay spectrum sequence;
[0047] Step 4: Randomly select a power delay spectrum generated by a root sequence to obtain a noise power estimate, and then obtain a corresponding detection threshold value Threshold according to the preset threshold coefficient and signal-to-noise ratio information;
[0048] Step 5: Compare the maximum value of the superimposed power delay spectrum sequence corresponding to all root sequences with the detection threshold T. If it is greater than the detection threshold Threshold, determine the preamble ID and estimate the timing advance TA; if it is less than the detection threshold value, determine that no preamble is sent.
[0049] A random access preamble detection device suitable for a 5G low-orbit satellite comprises a random access preamble sending device and a random access preamble receiving device, wherein:
[0050] A random access preamble sending device, used to send a corresponding 5G random access preamble code on a PRACH channel according to configuration information;
[0051] The preamble receiving device is used to receive the preamble code sequence, detect the preamble code ID and estimate the timing advance.
[0052] Here are more specific examples:
[0053] Example 1
[0054] A random access preamble detection method suitable for 5G low-orbit satellites, see Figure 1 , including the following steps:
[0055] Step 1: The base station broadcasts PRACH configuration information;
[0056] The ground terminal sends the preamble format, subcarrier spacing, root sequence, access timing, i.e., time-frequency position, and cyclic shift N. CS The base station needs to inform the base station of the configuration information before sending the preamble. The base station broadcasts the configuration information on the downlink channel through the parameter RACH-ConfigCommon in the system message SIB1.
[0057] The format and time domain position of the preamble are configured using the parameter prach-ConfigurationIndex.
[0058] The frequency domain RO and starting position are passed into msg1-FDM and msg1-FrequencyStart respectively;
[0059] Subcarrier spacing is passed into msg1-SubcarrierSpacing;
[0060] The logical root sequence index is passed into the parameter prach-RootSequenceIndex, which is used to look up the table to obtain the corresponding root sequence u;
[0061] The zeroCorrelationZoneConfig parameter indicates the zero correlation zone. The cyclic shift step length N can be obtained by looking up the table based on this parameter. CS .
[0062] Step 2: The UE receives and generates and sends a preamble according to the SIB1 configuration information;
[0063] After receiving and demodulating the PRACH configuration information in SIB1, the ground terminal directly obtains the sequence parameters according to the instructions of the above-mentioned high-level parameters or indirectly obtains the sequence parameters by looking up the table, generates the preamble sequence and sends it. The preamble sequence with the root sequence u can be expressed as
[0064]
[0065] Among them, L RA -point discrete Fourier transform (DFT) is expressed as To generate the PRACH signal, an N-point IDFT is required, where N ≥ L RA , so before doing IDFT, you need to change L RA subcarriers are mapped to N subcarriers.
[0066] In the NR system, each cell has a total of 64 sequences that can be used for RA. These sequences can be generated by cyclic shifting the same root sequence or by cyclic shifting multiple root sequences.
[0067] The cyclic shift sequence generation method is
[0068] x u,v (n) = x u ((n+c v )modL RA )
[0069] in, represents cyclic shift, N CS is the cyclic shift step size. In NR, the cyclic shift step size is related to the maximum round-trip delay difference of the cell, delay spread, safety margin and initial downlink timing error, which can be expressed as
[0070]
[0071] in, represents the maximum round-trip delay difference, τ max is the delay spread, s and σ represent the safety margin and initial downlink timing error respectively. CS The reason for this setting is to ensure that UEs at different distances in the cell can all receive the same N when sending preambles at the same RO. CS The preamble ID is detected within the detection window, thus correctly detecting the preamble ID. Except for N, the other three items can be set to 0. CS Similarly, the design of GT needs to meet the requirements of greater than or equal to
[0072] Step 3: The base station receives and processes the preamble within the detection window;
[0073] In low-orbit satellite communication systems, satellites are used as base stations, and the coverage of a single beam is considered a cell. Unlike ground-based cellular mobile communication systems, in addition to the greatly increased transmission delay between UE and satellite, the round-trip delay difference between users at the cell edge and users close to the satellite is also greatly increased. Figure 2In the remote cell covered by LEO, H is the orbit height of the low-orbit satellite, which can be set to 1200km, R represents the radius of the earth, which is 6371km, θ is the minimum communication angle of the cell, which is set to 20°, L is the maximum distance from the edge user in the cell to the satellite, and d represents the minimum distance from the user in the cell to the satellite.
[0074] By geometric relationship:
[0075]
[0076] The maximum distance from the satellite to the edge cell can be obtained, and the minimum distance is:
[0077]
[0078] The maximum round-trip delay difference is obtained as Substituting it into the calculation, it is 4.71ms. The long formats of the preamble sequence are format0~3, among which the longest sequence is format2 of 3.5ms. It can be seen that the maximum round-trip delay difference of the edge cell of the low-orbit satellite is much larger than the ground system and the length of the preamble sequence itself.
[0079] like Figure 3 As shown, the CP and SEQ of the leading sequence transmitter are not changed, the CP and SEQ parts of the leading sequence remain unchanged, and the number of sequence repetitions is 2. The minimum round trip delay RTD is shifted backward at the corresponding RO min Set to the starting position of the detection window, and the length of the detection window is set to The length of GT is set to the maximum round-trip delay difference of the satellite cell The GT length is adjusted to cooperate with the base station in adjusting the RO length, and the RO is scheduled to avoid interference with other UEs transmitting data services in the uplink.
[0080] The maximum round-trip delay difference is much larger than the sequence length, so 64 root sequences are used to generate 64 preamble codes, that is, the root sequence is used to distinguish users. The preamble sequence signal received by the base station is a time domain signal after the ZC sequence is transformed by FFT and then modulated by OFDM. Therefore, the local time domain sequence can be directly used for sliding correlation within the detection window. The calculation formula is:
[0081]
[0082] Among them, N R is the number of points in the received sequence, N R =T w ·R sample We get , which is the product of the detection window duration and the sampling rate.
[0083] The modulus value of the correlated sequence is taken, and then multiple sections are superimposed at each sample point. The number of superpositions is the same as the number of repetitions, and then the superimposed power delay spectrum is obtained.
[0084] Step 4: Determine the detection threshold Threshold;
[0085] The detection threshold of the preamble is related to the false alarm probability of the system. The false alarm probability is defined as the probability that the decision value φ(m) is greater than the detection threshold when there is no preamble transmission, where the decision value is the maximum value of all power delay spectra. If the detection threshold is too high, the probability of missed detection will increase, and if it is too low, the probability of false alarm will increase. It is necessary to set an appropriate threshold value for preamble detection.
[0086] Set the detection window length to be the same as that in step 3, then perform correlation operation on the local sequence and the noise sequence of the detection window length in the time domain to obtain the correlated sequence, take the modulus value of the correlated sequence and perform multiple superpositions. The number of superpositions is also the same as the number of repetitions of the corresponding format sequence to obtain the power delay spectrum.
[0087] The maximum value in the power delay spectrum is used as the decision value and counted in large quantities, and these values are arranged in descending order. 3GPP stipulates that the false alarm probability of the random access preamble code shall not be greater than 0.1%. The false alarm probability is the probability that the base station still detects the transmission of the preamble code when the terminal does not send the preamble code. The corresponding value of the 0.1% false alarm probability. Change the signal-to-noise ratio and repeat the above process to obtain the detection threshold under different signal-to-noise ratios, and use the ratio of the detection threshold value to the noise estimation value of the power delay spectrum sequence as the threshold coefficient T coe Finally, the threshold coefficients under different signal-to-noise ratios are fitted to obtain the fitting curve of the threshold coefficient changing with the signal-to-noise ratio.
[0088] Step 5: Determine the preamble ID based on the correlation peak and estimate the timing advance TA.
[0089] In the peak detection of the preamble code, the maximum value in the power delay spectrum is first obtained As the decision value, the decision value is compared with the preset threshold. Only when it is greater than the threshold can it be used as a valid peak to determine the preamble ID and TA. Since the cyclic shift set in this method is 0, the position of the correlation peak is only determined by the path delay.
[0090] The power delay spectrum of the received sequence obtained in step 4 is used to estimate the noise power to obtain the noise power estimate value, which is then compared with the corresponding signal-to-noise ratio lower threshold coefficient T coeMultiply them to get the detection threshold Threshold0 under the current conditions. If the decision value is greater than the detection threshold Threshold0, the preamble ID is determined by judging the root sequence index, and the timing advance can be estimated by calculating the offset between the correlation peak position and the start position of the detection window; if the decision value is less than the detection threshold Threshold0, it is determined that no preamble is sent.
[0091] Example 2
[0092] A random access preamble detection device suitable for 5G low-orbit satellites, used for detecting 5G random access preamble sequences in low-orbit satellite scenarios, see Figure 4 , comprising a random access preamble sending device and a preamble receiving device, wherein:
[0093] A random access preamble sending device, used to send a corresponding 5G random access preamble code on a PRACH channel according to configuration information;
[0094] The preamble receiving device is used to receive the preamble code sequence, and detect the preamble code ID and estimate the timing advance through steps such as correlation operation between the local sequence and the received sequence.
[0095] The preamble receiving device includes a time domain correlation operation unit, a multi-segment superposition unit and a detection threshold unit, wherein:
[0096] A time domain correlation operation unit, used for performing time domain correlation operation between a local sequence and a received sequence;
[0097] A multi-segment superposition unit is used for superposition of multiple segments of the sequence after time domain correlation to obtain the final power delay spectrum;
[0098] The detection threshold unit is used to obtain the threshold coefficient required for correlation peak judgment according to different signal-to-noise ratios.
[0099] In summary, the present invention sends the ground 5G corresponding format preamble code according to the configuration information of PRACH in the broadcast SIB1 according to the ground terminal; wherein, the satellite gNB sets the detection window length according to the maximum round-trip delay difference of the low-orbit satellite cell, and receives the preamble sequence; the satellite gNB uses the local sequence and the received sequence to perform correlation operations in the time domain, and performs multiple superpositions according to the number of repetitions of sequences in different formats to obtain a power delay spectrum containing a correlation peak; through the correlation and superposition operations of the local sequence and noise, the detection threshold value under the corresponding signal-to-noise ratio is obtained; the satellite gNB compares the correlation peak with the detection threshold, determines the preamble ID of the currently received preamble code sequence, and obtains the estimated value of the timing advance. The present invention solves the problem that the traditional detection algorithm cannot correctly detect the preamble ID and cannot estimate the timing advance when the maximum round-trip delay difference of the 5G low-orbit satellite cell is large.
Claims
1. A random access preamble detection method applicable to 5G low-orbit satellites, characterized in that: The steps include: Step 1: The satellite-borne 5G base station gNB configures the access timing and cyclic shift step size of the random access preamble sequence, and broadcasts it through the system message SIB1; Step 2: The user terminal UE sends the 5G NR preamble code at the corresponding access opportunity according to the SIB1 information; Step 3: The satellite-borne 5G base station gNB shifts back the length of the minimum round-trip delay at the access timing, opens the detection window to receive the leading sequence, performs time domain correlation operations on the local sequences corresponding to all root sequences and the received sequence, and performs multiple superpositions according to the number of repetitions of sequences in different formats to obtain the final power delay spectrum sequence; the specific method is as follows: (301) Calculate the minimum round trip delay RTD of the current satellite cell min , the detection window length is set to the sum of the CP length, the sequence length SEQ and the maximum round-trip delay difference of the cell; (302) directly performing a correlation operation on the local sequence and the received sequence in the time domain to obtain a correlated sequence; (303) taking a modulus value of the correlated sequence, and then superimposing the sequence according to the number of sequence repetitions in different NR preamble code formats to obtain a superimposed power delay spectrum sequence; (304) Determine whether all root sequences have generated corresponding power delay spectrum sequences. If not, select the next root sequence and return to step (302). Otherwise, exit the current step. Step 4: Randomly select a power delay spectrum sequence generated by a root sequence to obtain the noise power estimate, and then calculate the noise power estimate based on the preset threshold coefficient T coe The corresponding detection threshold Threshold is obtained by using the signal-to-noise ratio information; the specific method is as follows: (401) setting the detection window length to be the sum of the CP length, the sequence length SEQ and the maximum round-trip delay difference of the cell; (402) performing a correlation operation on the local sequence and a noise sequence of the detection window length in the time domain to obtain a correlated sequence; (403) taking a modulus value of the correlated sequence, and then superimposing the sequence according to the number of sequence repetitions in different NR preamble code formats to obtain a superimposed power delay spectrum sequence; (404) Counting the maximum values in the power delay spectrum, and arranging these values in descending order, to obtain the corresponding value of 0.1% false alarm probability; (405) Change the signal-to-noise ratio and return to step (402) to obtain the detection threshold under different signal-to-noise ratios, and use the ratio of the detection threshold value to the noise estimation value of the power delay spectrum sequence in step (403) as the threshold coefficient T coe ; (406) Fitting the threshold coefficients under different signal-to-noise ratios to obtain a fitting curve of the threshold coefficient changing with the signal-to-noise ratio; Step 5: Compare the maximum value of the superimposed power delay spectrum sequence corresponding to all root sequences with the detection threshold Threshold. If it is greater than the detection threshold, determine the preamble ID and estimate the timing advance TA; if it is less than the detection threshold, determine that no preamble is sent.
2. A random access preamble detection method applicable to 5G low-orbit satellites according to claim 1, characterized in that: The access timing of the preamble sequence in step 1 satisfies in the time domain ,in is the PRACH access opportunity period, is the detection window length; the cyclic shift step size N CS Set to 0.