Satellite internet of things joint detection and estimation method based on pre-compensation

Abstract

This invention pertains to wireless communication technology, specifically a pre-compensated joint detection and estimation method for satellite IoT, comprising three parts: frame structure design, terminal preprocessing, and preamble detection and frequency offset estimation. In scenarios with multiple random satellite terminal access, this invention employs a preamble sequence composed of two root sequences, which, compared to the traditional preamble sequence frame structure with a single root sequence, increases the number of local sequence sets. The proposed joint detection and estimation algorithm combines the characteristics of the preamble sequence and analyzes how, under different positioning error conditions, it improves the detection success rate of simultaneously accessing terminals compared to traditional detection methods.

Description

Technical Field

[0001] This invention pertains to wireless communication technology, specifically a pre-compensated satellite IoT joint detection and estimation method. Background Technology

[0002] Satellite communications have long been a focal point in the fields of science and technology and communications, providing crucial support for connecting remote regions, providing global communication services, and responding to challenges such as disasters. Satellite communication technologies encompass various methods, among which direct-connect satellite and non-direct-connect satellite applications have attracted significant attention. Direct-connect satellite refers to a direct communication connection between a satellite and ground terminal equipment. The greatest advantage of direct-connect satellite communication is its global coverage capability, eliminating reliance on ground infrastructure. Therefore, it can provide vital communication support for remote areas, aerospace, shipping, and emergency rescue.

[0003] However, the advantages of satellite direct communication come with a series of challenges. First, the high dynamic range of satellites leads to large communication delays and frequency offsets, which can cause the detection of information sent by the terminal to fail. Second, the area covered by satellite cells is significantly larger than that of terrestrial cells, and the number of terminals in each satellite cell is much greater than that in each terrestrial cell. The more terminals there are in a cell, the more strained communication resources become, thus increasing the probability of collisions in the terminal preamble sequence.

[0004] Based on this, this paper studies a pre-compensated joint detection and estimation method for satellite IoT, which can meet the needs of multiple terminals communicating with the satellite concurrently, and can correctly detect terminals and complete time-frequency synchronization even with time delay and frequency offset. Summary of the Invention

[0005] To address the problem of concurrent satellite communication among multiple terminals, this invention proposes a frame structure design to reduce the probability of preamble sequence collisions caused by concurrent transmission of multiple terminals. Secondly, to address the problem of terminal information detection failure due to time delay and frequency offset, this invention proposes a joint detection and estimation method based on preamble sequence, which can detect terminals and estimate time delay and frequency offset under the premise of time delay and frequency offset.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0007] A pre-compensated joint detection and estimation method for satellite IoT comprises three parts: frame structure design, terminal preprocessing, and joint detection and estimation. The specific steps are as follows:

[0008] Step 1, Frame Structure Design: Frame structure design includes the design of the preamble sequence and the data frame. The preamble sequence is formed by adding and combining the subsequences generated by the two root sequences. The data frame includes the frame header, data information, and cyclic redundancy check code.

[0009] Step 2, Terminal Preprocessing: The low-orbit satellite broadcasts information to the ground terminal according to the antenna angle. After receiving the satellite broadcast, the ground terminal performs time and frequency pre-compensation based on ephemeris information and its own positioning information. After pre-compensation, it selects a preamble sequence according to the broadcast information and completes framing with the data frame at the corresponding frequency and time.

[0010] Step 3, Joint Detection and Estimation: For the mixed signals received by the satellite, the signals are separated in the spatial and frequency domains. Cyclic correlation detection is performed on the received signals under the same time and frequency resources to find the peak position and complete the timing synchronization, identity detection and frequency offset estimation of the terminal.

[0011] In a further improvement of the present invention, in step one, the leader sequence is composed of the addition of two ZC (ZaddoffChu) sequences generated from two different roots, one of which is a ZaddoffChu sequence and the other is a conjugate of the ZaddoffChu sequence.

[0012] The Zaddoff Chu sequence is transmitted on an odd number of subcarriers in a 2·N subcarrier pattern. After inverse Fourier transform, two identical sequences A are generated. The second sequence is conjugated to obtain the preamble sequence [AA]. * ], respectively x(n) and x * (n+N), · * The conjugate sequence of · is represented by: where the A sequence is:

[0013]

[0014]

[0015]

[0016] Where N is the sequence length, x u (n), x v (n) denotes a Zaddoff Chu sequence with roots u and v, where u and v are constants coprime to N, and u ≠ v, x u (np) represents x u (n) The subsequence generated by cyclic shift p, x v (nq) represents x v (n) The subsequence generated by cyclic shift q, where p and q represent cyclic shifts of the Zaddoff Chu sequence, p∈[0,N-1], q∈[0,N-1], n∈[0,N-1];

[0017] The data frame mainly includes a frame header, data information, and a Cyclic Redundancy Check (CRC) check code. The frame header indicates the start position of the data frame, and the CRC check code is used at the decoding end to determine whether the decoding is correct.

[0018] In a further improvement of the present invention, in step two, the satellite IoT terminal completes preprocessing based on satellite broadcast information. The specific terminal preprocessing process is as follows:

[0019] Procedure 2.1: The satellite divides its area into different spatial frequency groups according to the antenna angle and broadcasts information, including the frequency range, time slot plan, satellite ephemeris, and preamble sequence range. After receiving the broadcast information, the ground terminal performs time and frequency pre-compensation based on the satellite ephemeris and its own positioning information with errors.

[0020]

[0021]

[0022] Where t and fd are the time delay and Doppler frequency offset of the IoT terminal arriving at the satellite based on the actual positioning information, respectively; t' and fd' are the time delay and Doppler frequency offset calculated by the IoT terminal based on positioning information with errors, respectively; and T... s Let τ be the sampling interval and Δf be the OFDM subcarrier interval. Then τ and ε are the normalized time delay and frequency offset residual after pre-compensation, respectively, and 0≤τ<N;

[0023] Procedure 2.2: The ground terminal selects the root sequence and performs cyclic shifting based on the range of the preamble sequence to determine the preamble sequence and completes the framing with the data frame.

[0024] In a further improvement of the present invention, in step three, after the satellite receives the mixed signal from the ground terminal, it performs joint detection and estimation on each terminal individually. The specific process is as follows:

[0025] Procedure 3.1: The satellite intercepts the signal one symbol time before the arrival of the terminal signal to estimate the noise power P in the current time slot. n ;

[0026] Procedure 3.2: The satellite end filters the received mixed signal on the i-th frequency group. The filtered signal is:

[0027]

[0028] Where i is the frequency group index, i = 1, 2, ..., I, and I is the number of frequency groups. The total number of users under the current frequency is M, and the number of active users at the same time is M. c=μ*M, where 0 < μ < 1, representing the active user factor, x m (n) represents the transmitted signal of the m-th user, ρ m Let τ be the channel coefficient for the m-th user. m Let ε be the latency residual for the m-th user. m Z(n) represents the normalized frequency offset residual of the m-th user, and Z(n) represents noise, n∈[0,N-1].

[0029] Step 3.3: Multiply the filtered signal by the beamforming vector to obtain the following signal:

[0030]

[0031] Where b is the beam subscript index, b = 1, 2, ..., B, B is the number of beams, and w b M is the beamforming vector. b This represents the number of users under the same time-frequency resource for the current beam.

[0032] Step 3.4: The satellite uses the signal detection preamble sequence from Step 3.3 to complete timing synchronization;

[0033] Procedure 3.5: The satellite uses the signal from Procedure 3.3 to estimate the residual frequency offset;

[0034] Procedure 3.6: After the satellite end eliminates serial interference from the detected terminal, calculate the remaining signal power P. s Is it greater than the noise threshold P? n If the noise threshold is greater than 3, proceed to step 3.4. If the noise threshold is less than 3, continue to determine whether the beam traversal has ended. If it has not ended, switch beams and proceed to step 3.3. Otherwise, determine whether the frequency traversal has ended. If it has not ended, proceed to step 3.2.

[0035] A further improvement to this invention is that, in process 3.4, the specific steps for timing synchronization and terminal preamble detection are as follows:

[0036] (3.4.1) Extracting signal y b The part corresponding to the leading sequence A in the middle is used to obtain y. m (n):

[0037]

[0038]

[0039]

[0040] (3.4.2) y m (n) and Cyclic correlation, the correlation output is represented as:

[0041]

[0042] in, For autocorrelation output, The output is the cross-correlation interference of the two root sequences. Therefore, due to noise interference output, The peak value is So, The peak value p corresponds to: (2up-(2uτ) m +ε m )) 2N =0, where ((·)) N Let x be modulo N, and u be a Zadoff Chu sequence. u The root index of (n), x u (np) represents x u (n) The subsequence generated by cyclic shift p, where p is x u Circular shift of (n), p∈[0,N-1], · * The conjugate sequence of ·;

[0043] (3.4.3) y k (n) and x v (n) Cyclic dependency, the dependency output is represented as:

[0044]

[0045] in, For autocorrelation output, Corr Z (q) represents the noise interference output. The peak value q corresponds to: (-2vq+(2vτ) m -Δε m )) 2N =0; v is a Zaddoff Chu sequence x v The root index of (n), x v (nq) represents x v (n) The subsequence generated by a cyclic shift q, where q is x v Circular shift of (n), q∈[0,N-1].

[0046] (3.4.4) Determine the residual delay τ of the current terminal based on the two peak points. m From the peak indices of p and q, we can obtain:

[0047] 2up=2uτ m +ε m +2aN

[0048] 2vq=2vτ m -εm +2bN

[0049] Where a and b are determined by the frequency offset ε m The determined integers, when summed, yield:

[0050]

[0051] In 0≤τ m Under the constraint <N, find the unique τ m .

[0052] In a further improvement to the present invention, the specific steps of frequency offset estimation in process 3.5 are as follows:

[0053] (3.5.1) Extracting signal y b The leading sequences A and A * The corresponding part yields:

[0054]

[0055]

[0056] (3.5.2) Connect y1 with Multiplying them together yields Y(n).

[0057]

[0058] (3.5.3) Connect Y(n) with Related, the related output is represented as:

[0059]

[0060] (3.5.4) Taking the phase of Corr, we can obtain the frequency offset estimate ε. m .

[0061] The beneficial effects of this invention are as follows: In the scenario of random access to multiple satellite terminals, this invention adopts a preamble sequence composed of two root sequences, which can increase the number of local sequence sets compared to the traditional preamble sequence frame structure of a single root sequence; this invention proposes a joint detection and estimation algorithm that combines the characteristics of the preamble sequence and analyzes the terminal under different positioning error conditions compared to traditional detection methods, thereby improving the detection success rate of simultaneously accessing terminals. Attached Figure Description

[0062] Figure 1 This is a flowchart of the joint detection and estimation method in the present invention.

[0063] Figure 2 This is a schematic diagram comparing the detection success rates of 20 active terminals provided in this invention example.

[0064] Figure 3 This is a schematic diagram comparing the detection success rates of 60 active terminals provided in this invention example. Detailed Implementation

[0065] To enhance understanding of the present invention, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. These embodiments are only used to explain the invention and do not limit the scope of protection of the invention.

[0066] Example 1: As Figure 1 As shown, a pre-compensated joint detection and estimation method for satellite IoT includes three parts: frame structure design, terminal preprocessing, and joint detection and estimation. The specific steps are as follows:

[0067] Step 1: Frame Structure Design

[0068] The frame structure design includes the design of the preamble sequence and the data frame. The preamble sequence is formed by adding and combining the subsequences generated from two root sequences. The data frame includes a frame header, data information, and a cyclic redundancy check (CRC) code. The preamble sequence is formed by adding two Zaddoff-Chu (ZC) sequences generated from two different roots, one of which is a Zaddoff-Chu sequence, and the other is a conjugate of the Zaddoff-Chu sequence. The preamble sequence is generated as follows:

[0069] The Zaddoff Chu sequence is transmitted on an odd number of subcarriers in a 2·N subcarrier pattern. After inverse Fourier transform, two identical sequences A are generated. The second sequence is conjugated to obtain the preamble sequence [AA]. * ], respectively x(n) and x * (n+N), · * The conjugate sequence of · is represented by: where the A sequence is:

[0070]

[0071]

[0072]

[0073] Where N is the sequence length, x u (n), x v (n) denotes a Zaddoff Chu sequence with roots u and v, where u and v are constants coprime to N, and u ≠ v, x u (np) represents x u (n) The subsequence generated by cyclic shift p, x v (nq) represents x v(n) The subsequence generated by cyclic shift q, where p and q represent cyclic shifts of the Zaddoff Chu sequence, p∈[0,N-1], q∈[0,N-1], n∈[0,N-1];

[0074] The data frame mainly includes a frame header, data information, and a Cyclic Redundancy Check (CRC) check code. The frame header indicates the start position of the data frame, and the CRC check code is used at the decoding end to determine whether the decoding is correct.

[0075] Table 1: Collision Probability Analysis of Leader Sequences

[0076]

[0077] Table 1 shows the collision probability when two terminals choose the same preamble sequence, where N ZC This represents the number of subsequences generated by cyclically shifting a root sequence. Since subsequences can be combined with each other, the number of leading sequences generated from two root sequences is... ·! represents the factorial operation of ·.

[0078] Step 2: Terminal Preprocessing

[0079] Low-Earth orbit (LEO) satellites broadcast information to ground terminals according to their antenna angles. Upon receiving the satellite broadcast, the ground terminal performs time-frequency pre-compensation based on ephemeris information and its own positioning information. After pre-compensation, it selects a preamble sequence according to the broadcast information and frames the data frames at the corresponding frequency and time. The satellite IoT terminal completes its preprocessing based on the satellite broadcast information. The specific terminal preprocessing process is as follows:

[0080] Procedure 2.1: The satellite divides its area into different spatial frequency groups according to the antenna angle and broadcasts information, including the frequency range, time slot plan, satellite ephemeris, and preamble sequence range. After receiving the broadcast information, the ground terminal performs time and frequency pre-compensation based on the satellite ephemeris and its own positioning information with errors.

[0081]

[0082]

[0083] Where t and fd are the time delay and Doppler frequency offset of the IoT terminal arriving at the satellite based on the actual positioning information, respectively; t' and fd' are the time delay and Doppler frequency offset calculated by the IoT terminal based on positioning information with errors, respectively; and T... s Let τ be the sampling interval and Δf be the OFDM subcarrier interval. Then τ and ε are the normalized time delay and frequency offset residual after pre-compensation, respectively, and 0≤τ<N;

[0084] Procedure 2.2: The ground terminal selects the root sequence and performs cyclic shifting based on the range of the preamble sequence to determine the preamble sequence and completes the framing with the data frame.

[0085] Step 3: Joint Detection and Estimation

[0086] For the mixed signals received from the satellite, the signals are separated in the spatial and frequency domains. Cyclic correlation detection is performed on the received signals under the same time and frequency resources to find the peak position, and to complete the timing synchronization, identity detection, and frequency offset estimation of the terminal. After receiving the mixed signals from the ground terminal, the satellite performs joint detection and estimation on each terminal. The specific process is as follows:

[0087] Procedure 3.1: The satellite intercepts the signal one symbol time before the arrival of the terminal signal to estimate the noise power P in the current time slot. n ;

[0088] Procedure 3.2: The satellite end filters the received mixed signal on the i-th frequency group. The filtered signal is:

[0089]

[0090] Where i is the frequency group index, i = 1, 2, ..., I, and I is the number of frequency groups. The total number of users under the current frequency is M, and the number of active users at the same time is M. c =μ*M, where 0 < μ < 1, representing the active user factor, x m (n) represents the transmitted signal of the m-th user, ρ m Let τ be the channel coefficient for the m-th user. m Let ε be the latency residual for the m-th user. m Z(n) represents the normalized frequency offset residual of the m-th user, and Z(n) represents noise, n∈[0,N-1].

[0091] Step 3.3: Multiply the filtered signal by the beamforming vector to obtain the following signal:

[0092]

[0093] Where b is the beam subscript index, b = 1, 2, ..., B, B is the number of beams, and w b M is the beamforming vector. b This represents the number of users under the same time-frequency resource for the current beam.

[0094] Step 3.4: The satellite uses the signal detection preamble sequence from Step 3.3 to complete timing synchronization;

[0095] Procedure 3.5: The satellite uses the signal from Procedure 3.3 to estimate the residual frequency offset;

[0096] Procedure 3.6: After the satellite end eliminates serial interference from the detected terminal, calculate the remaining signal power P. s Is it greater than the noise threshold P? n If the noise threshold is greater than 3, proceed to step 3.4. If the noise threshold is less than 3, continue to determine whether the beam traversal has ended. If it has not ended, switch beams and proceed to step 3.3. Otherwise, determine whether the frequency traversal has ended. If it has not ended, proceed to step 3.2.

[0097] The specific steps for timed synchronization and terminal preamble detection in process 3.4 above are as follows:

[0098] (3.4.1) Extracting signal y b The part corresponding to the leading sequence A in the middle is used to obtain y. m (n):

[0099]

[0100]

[0101]

[0102] (3.4.2) y m (n) and Cyclic correlation, the correlation output is represented as:

[0103]

[0104] in, For autocorrelation output, The output is the cross-correlation interference of the two root sequences. Therefore, due to noise interference output, The peak value is So, The peak value p corresponds to: (2up-(2uτ) m +ε m )) 2N =0, where ((·)) N Let x be modulo N, and u be a Zadoff Chu sequence. u The root index of (n), x u (np) represents x u (n) The subsequence generated by cyclic shift p, where p is x u Circular shift of (n), p∈[0,N-1], · * The conjugate sequence of ·;

[0105] (3.4.3) y k(n) and x v (n) Cyclic dependency, the dependency output is represented as:

[0106]

[0107] in, For autocorrelation output, Corr Z (q) represents the noise interference output. The peak value q corresponds to: (-2vq+(2vτ) m -Δε m )) 2N =0; v is a Zaddoff Chu sequence x v The root index of (n), x v (nq) represents x v (n) The subsequence generated by a cyclic shift q, where q is x v Circular shift of (n), q∈[0,N-1].

[0108] (3.4.4) Determine the residual delay τ of the current terminal based on the two peak points. m From the peak indices of p and q, we can obtain:

[0109] 2up=2uτ m +ε m +2aN

[0110] 2vq=2vτ m -ε m +2bN

[0111] Where a and b are determined by the frequency offset ε m The determined integers, when summed, yield:

[0112]

[0113] In 0≤τ m Under the constraint <N, find the unique τ m .

[0114] In process 3.5, the specific steps for frequency offset estimation are as follows:

[0115] (3.5.1) Extracting signal y b The leading sequences A and A * The corresponding part yields:

[0116]

[0117]

[0118] (3.5.2) Connect y1 with Multiplying them together yields Y(n).

[0119]

[0120] (3.5.3) Connect Y(n) with Related, the related output is represented as:

[0121]

[0122] (3.5.4) Taking the phase of Corr, we can obtain the frequency offset estimate ε. m .

[0123] Figure 2 , 3 The figures show the trends in detection success rate based on the preamble sequence when 20 and 60 terminals access the satellite, respectively, under terminal positioning errors with radii of R = 0.1 km and R = 1 km. It can be seen that the preamble sequence and joint detection and estimation method proposed in this paper have a higher user detection success rate than the traditional random access detection method. Furthermore, the detection success rate decreases as the number of accessing terminals increases.

[0124] In summary, the pre-compensated joint detection and estimation method for satellite IoT proposed in this invention designs a frame structure that reduces the collision probability of multiple terminals in preamble sequence selection; terminal preprocessing introduces time-frequency pre-compensation on the terminal side, which significantly reduces latency and frequency offset in satellite IoT scenarios; the joint detection and estimation method proposes a detection and estimation method based on a preamble sequence formed by the addition and combination of two root sequences, which improves the detection success rate of terminals compared with traditional detection methods.

[0125] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A joint detection and estimation method for satellite IoT based on pre-compensation, characterized in that, It includes three parts: frame structure design, time-frequency pre-compensation, and joint detection and estimation. The specific steps are as follows: Step 1, Frame Structure Design: Frame structure design includes the design of the preamble sequence and the data frame. The preamble sequence is formed by adding and combining the subsequences generated by the two root sequences. The data frame includes the frame header, data information, and cyclic redundancy check code. Step 2, Terminal Preprocessing: The low-orbit satellite broadcasts information to the ground terminal according to the antenna angle. After receiving the satellite broadcast, the ground terminal performs time and frequency pre-compensation based on ephemeris information and its own positioning information. After pre-compensation, a preamble sequence is selected according to the broadcast information and framed with the data frame at the corresponding frequency and time. Step 3, Joint Detection and Estimation: For the mixed signals received by the satellite, the signals are separated in the spatial and frequency domains. Cyclic correlation detection is performed on the received signals under the same time and frequency resources to find the peak position and complete the timing synchronization, identity detection and frequency offset estimation of the terminal. In step one, the leader sequence is formed by adding two Zaddoff-Chu sequences generated from two different roots, one of which is a Zaddoff-Chu sequence and the other is the conjugate of the Zaddoff-Chu sequence; In step one, the leader sequence is generated as follows: Zaddoff Chu sequence in The odd number of subcarriers out of the total number of subcarriers are transmitted, and after inverse Fourier transform, two identical sequences are generated. The second sequence is conjugated to obtain the leading sequence. ], respectively and , represent The conjugate sequences, where sequence A is: ， ， ， in For sequence length, , The root sequence is , The Zaddoff Chu sequence, and To and Coprime constants, and , express Circular shift The resulting subsequence, express Circular shift The resulting subsequence, , This represents a cyclic shift of the Zaddoff Chu sequence. , , ; The frame header in a data frame indicates the start of the data frame, and the cyclic redundancy check code is used at the decoding end to determine whether the decoding is correct. In step two, the satellite IoT terminal completes preprocessing based on satellite broadcast information. The specific terminal preprocessing process is as follows: Procedure 2.1: The satellite divides its area into different spatial frequency groups according to the antenna angle and broadcasts information, including the frequency range, time slot plan, satellite ephemeris, and preamble sequence range. After receiving the broadcast information, the ground terminal performs time and frequency pre-compensation based on the satellite ephemeris and its own positioning information with errors. ， in, , These are the time delay and Doppler frequency offset of the IoT terminal reaching the satellite based on its actual location information. , These are the latency and Doppler frequency offset calculated by the IoT terminal based on the positioning information, which contains errors. The sampling interval is... For OFDM subcarrier spacing, then , These are the normalized time delay and frequency offset residual after pre-compensation, respectively. ; Procedure 2.2: The ground terminal selects the root sequence and performs cyclic shift based on the range of the preamble sequence to determine the preamble sequence, and completes the framing with the data frame; In step three, after receiving the mixed signal from the ground terminal, the satellite performs joint detection and estimation on each terminal. The specific process is as follows: Procedure 3.1: The satellite intercepts the signal one symbol before it arrives at the terminal to estimate the noise power in the current time slot. ; Procedure 3.2: The satellite will receive the mixed signal at the... Filtering is performed on each frequency group, and the filtered signal is: ， in, For frequency grouping subscript index, , The number of frequency groups; the total number of users under the current frequency is... The number of active users at the same time is ,in Represents active user factors. For the first Signals sent by individual users For the first Channel coefficients for each user For the first Latency residual for each user For the first Normalized frequency offset residual for individual users For noise, ; Step 3.3: Multiply the filtered signal by the beamforming vector to obtain the following signal: ， in, For beam subscript index, , For the number of beams, For beamforming vectors, This represents the number of users under the same time-frequency resource for the current beam. Step 3.4: The satellite uses the signal detection preamble sequence from Step 3.3 to complete timing synchronization; Procedure 3.5: The satellite uses the signal from Procedure 3.3 to estimate the residual frequency offset; Procedure 3.6: After the satellite eliminates serial interference from the detected terminal, calculate the remaining signal power. Is it greater than the noise threshold? If the noise threshold is greater than 3, proceed to step 3.

4. If the noise threshold is less than 3, continue to determine whether the beam traversal has ended. If it has not ended, switch beams and proceed to step 3.

3. Otherwise, determine whether the frequency traversal has ended. If it has not ended, proceed to step 3.

2.

2. The satellite IoT joint detection and estimation method based on pre-compensation according to claim 1, characterized in that: In process 3.4, the specific steps for timed synchronization and terminal preamble detection are as follows: (3.4.1) Signal interception Middle leader sequence The corresponding part is obtained : ， ， ， (3.4.2) will and Cyclic correlation, the correlation output is represented as: ， in, For autocorrelation output, The output is the cross-correlation interference of the two root sequences. Therefore, due to noise interference output, The peak value is ,So, peak The corresponding is: ,in For model , Zaddoff Chu sequence The root index, express Circular shift The resulting subsequence, for Circular shift, , represent The conjugate sequence; (3.4.3) will and Cyclic correlation, the correlation output is represented as: ， in, For autocorrelation output, Output to prevent noise interference. peak The corresponding is: ; Zaddoff Chu sequence The root index, express Circular shift The resulting subsequence, for Circular shift, ; (3.4.4) Determine the residual delay of the current terminal based on the two peak points. ,Depend on , The peak index can be obtained as follows: ， in , Due to frequency offset The determined integers, when summed, yield: ， exist Under the constraints, find the unique .

3. The satellite IoT joint detection and estimation method based on pre-compensation according to claim 2, characterized in that: In process 3.5, the specific steps for frequency offset estimation are as follows: (3.5.1) Signal interception Preamble and The corresponding part yields: ， ， (3.5.2) will and Multiply to get , ， (3.5.3) will and Related, the related output is represented as: ， (3.5.4) to The frequency offset estimate can be obtained by taking the phase. .

Images