Time and frequency synchronization method and system for one-bit quantized cell-free massive MIMO system

Abstract

The application discloses a kind of time and frequency synchronization method and system of one bit quantization cell-free massive MIMO system based on positioning speed measurement, method as follows: step 1, the symbol timing offset STO and carrier frequency offset CFO are estimated using Schmidl-Cox synchronization algorithm;Step 2, according to the estimated value of STO and CFO, the distance between user UE and access point AP and the movement speed of user UE are estimated;Step 3, the distance and movement speed of step 2 are used to estimate STO and CFO again.The technical scheme of the present application has good realizability, and can obtain lower system synchronization error through test.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of wireless communication, and specifically aims at the time and frequency synchronization technology in a cell-free massive multiple-input multiple-output (MIMO) system, and combines a one-bit quantization technology and radio frequency (RF) structure optimization to propose a time and frequency synchronization method and system based on positioning and speed measurement. BACKGROUND

[0002] With the rapid development of wireless communication technology, especially in 5G and future 6G networks, cell-free massive MIMO technology, as one of the key technologies to improve system capacity and communication quality, has attracted widespread attention. Traditional synchronization methods mostly rely on relatively complex hardware devices and high-precision clock systems, resulting in a significant increase in system cost and complexity. In a cell-free massive MIMO system, due to factors such as user motion, signal multipath propagation, and frequency offset, time synchronization and frequency synchronization become particularly important. In a dynamic scenario, the relative motion between users (UEs) and access points (APs) will cause a large Doppler shift, further increasing the difficulty of synchronization. Accurate time and frequency synchronization is the basis for ensuring stable operation and communication performance of the system, and how to reduce hardware complexity and power consumption while ensuring system synchronization accuracy has become a key issue in current wireless communication technology. SUMMARY

[0003] To solve the above technical problems, the application, in a cell-free massive MIMO system with one-bit quantization and radio frequency (RF) structure, effectively solves the contradiction between synchronization accuracy and hardware complexity in the system by using a time and frequency synchronization technology based on positioning and speed measurement, and reduces synchronization error. The technical scheme of the application provides an efficient and stable synchronization solution.

[0004] To achieve the above application purposes, the application adopts the following technical scheme:

[0005] Consider a cell-free massive MIMO uplink system, which is composed of U users (UEs) and B access points (APs). Specifically, in the distributed MIMO architecture involved in the present application, the UEs transmit radio frequency (RF) signals to the APs, and the received RF signals at the single antenna port of each AP are first passed through a band-pass filter (BPF) and then added with a dithering noise in the signal processing. The signal is quantized by a zero threshold comparator to generate a binary waveform, which is then transmitted to a central processing unit (CPU) through an optical fiber. There exists an unknown symbol timing offset (STO) and carrier frequency offset (CFO) between each UE and each AP, which affects the quality of the received signal and correct demodulation, and reduces the reliability and performance of the system. At the CPU, the STO and CFO are estimated to reduce the synchronization error, and the signal is then digitally down-converted, spatially processed, and subjected to subsequent demodulation and decoding operations after compensation.

[0006] In the distributed system involved in the present application, the STO and CFO between a single UE and B APs are considered, and a time and frequency synchronization method based on positioning and speed measurement is used to accurately estimate and reduce the synchronization error caused by the STO and CFO.

[0007] In the cell-free massive MIMO system, a three-dimensional spatial coordinate system (x, y, z) is established, and the distances from the UE to the B APs are given as (d1, d2,..., dB), respectively. B The position coordinates of the bth AP are p b = (x b , y b , h), and the position coordinates of the UE are p (u) = (x (u) , y (u) , 0). The motion speed of the UE is v = (v x , v y , 0).

[0008] It is assumed that the signal propagates in free space, and the propagation time t b can be obtained by the distance d b and the speed of light c as follows:

[0009]

[0010] The STO between the UE and the bth AP is δ b , which is the normalized STO of the sampled period, and δ b can be given by:

[0011]

[0012] where f s is the sampling frequency.

[0013] Since the UE is located in the xoy plane, its projection velocity is also limited to this plane. By ignoring the z-axis component, the projection d of the direction vector from the UE to the b-th AP onto the xoy plane is calculated. b，proj , is represented as:

[0014] d b，proj =(x b -x (u) y b -y (u) ,0)

[0015] Its unit projection vector for:

[0016]

[0017] In three-dimensional space, the projected velocity v of the UE's motion velocity along the unit projection vector direction. b，proj for:

[0018]

[0019] According to the Doppler frequency shift formula, the frequency shift f b It can be represented as:

[0020]

[0021] Among them, f c For carrier frequency.

[0022] ε b ε is the CFO between the UE and the b-th AP after subcarrier spacing normalization. b With frequency shift f b The relationship between them is:

[0023]

[0024] Where N is the length of an Orthogonal Frequency Division Multiplexing (OFDM) symbol, f s The sampling frequency.

[0025] For a fixed symbol length N, consider the time interval T = N / f s The received signal is processed by a one-bit analog-to-digital converter (ADC) at the CPU, i.e., a one-bit quantization operation is performed on the signal. The sample of the i-th OFDM symbol of the output signal of the b-th one-bit ADC at the CPU is represented as:

[0026]

[0027] in, denotes the n th sample of the discrete-time RF signal of the i th OFDM symbol received at the b th AP after BP filtering and sampling, but before one-bit quantization. Perturbation signal Assuming it is independent of and across n, the function sgn(·) is applied element-wise to the input vector.

[0028] Received signal is expressed as:

[0029]

[0030] Baseband signal is expressed in time domain as:

[0031]

[0032] where h b is the channel gain from the UE to the b th AP, is the signal transmitted by the UE, where E s denotes the average energy of each sample, is the noise, where N 0 denotes the power spectral density of the noise.

[0033] The signal s[n] transmitted by the b th AP is composed of consecutive time-domain OFDM symbols and a cyclic prefix (CP), and can be expressed as:

[0034]

[0035] where G ≥ L-1 is the length of the CP, and the relationship between i and n is -G ≤ n-i(n+G) ≤ N-1.

[0036] s (i) is the time-domain representation of the i th OFDM symbol, and can be expressed by the inverse discrete Fourier transform (IDFT) as:

[0037]

[0038] The one-bit quantization time and frequency synchronization method for the cell-free massive MIMO system of the application has the following specific steps:

[0039] Step 1: Estimate the symbol timing offset STO and the carrier frequency offset CFO using the Schmidl-Cox synchronization algorithm.

[0040] Step 2: Estimate the distance between the user UE and the access point AP and the motion speed of the user UE according to the estimated values of STO and CFO.

[0041] Step 3: Estimate STO and CFO again using the distance and motion speed of step 2.

[0042] Preferably, step 1, Schmidl-Cox synchronization algorithm estimation;

[0043] CPU uses time-frequency synchronization, and uses Schmidl-Cox synchronization algorithm to synchronize time and carrier frequency of the quantized signal , and estimates STO and CFO. The estimated values of STO and CFO between the UE and the bth AP are respectively denoted as

[0044] Step 1.1, estimation of STO:

[0045] Calculate the autocorrelation value of the signal at different time offsets:

[0046]

[0047] Where N is the fixed OFDM symbol length, n is the index variable, δ is the time offset, and * represents the complex conjugate;

[0048] Calculate the energy distribution of the signal at different offsets:

[0049]

[0050] Calculate the ratio of the correlation of the signal to the energy of the signal:

[0051]

[0052] Where G is the length of the cyclic prefix;

[0053] Estimate the STO at the bth AP:

[0054]

[0055] Step 1.2, CFO estimation:

[0056] Calculate the relevant measure Use the function arg{·} to obtain the phase and estimate the frequency offset of the signal:

[0057]

[0058] Preferably, step 2, positioning and speed measurement;

[0059] The estimated values of STO and CFO obtained by Schmidl-Cox algorithm are used to further calculate the estimated value of the distance between UE and AP and the estimated value of the projection speed of UE in xoy plane. The position information coordinates of UE and the motion speed of UE are estimated by using least square (LS) algorithm, and the time and frequency synchronization of the system is more accurately estimated.

[0060] Step 2.1, estimating the distance between UE and AP and the projection speed of UE:

[0061] Using the estimated value of STO The distance estimation is performed to calculate the estimated value of propagation time:

[0062]

[0063] Wherein, f b is the frequency shift;

[0064] The estimated value of the distance between UE and AP is calculated:

[0065]

[0066] Wherein, c is the speed of light;

[0067] Using the estimated value of CFO The projection speed estimation is performed to calculate the estimated value of frequency shift:

[0068]

[0069] According to the Doppler shift formula, the projection estimation value of the motion speed of UE is calculated:

[0070]

[0071] Step 2.2, estimating the position coordinates of UE and the motion speed of UE:

[0072] Using the estimated value of the distance between UE and each AP obtained in step 2.1 The position of UE is estimated by LS algorithm.

[0073] The matrix is constructed:

[0074]

[0075]

[0076] Wherein, (x b ,y b ) is the position coordinates of the bth AP in the horizontal plane;

[0077] Solved by LS algorithm:

[0078]

[0079] The matrix of the coordinate estimation value of the UE, the estimation value of the position coordinate of the UE can be expressed as Wherein, and are the estimation value of the x coordinate and the y coordinate of the UE in the horizontal plane respectively.

[0080] According to the estimation value of the position coordinate of the UE and the projection estimation value of the motion speed Solve the estimation value of the motion speed of the UE Wherein, and Respectively represent the estimation value of the speed of the UE in the x axis and y axis direction in the horizontal plane.

[0081] For the UE to the b-th AP in the unit projection vector direction, the speed projection estimation value is calculated:

[0082]

[0083] The matrix is constructed:

[0084]

[0085]

[0086] Wherein, B contains the unit projection vector of each AP direction, and c contains the speed projection estimation value to each AP.

[0087] Solve by LS algorithm:

[0088]

[0089] The estimation value of the motion speed of the UE, the estimation value of the motion speed of the UE can be expressed as

[0090] Preferably, step 3, the distance and the speed are used to estimate the system synchronization;

[0091] According to the estimation value of the UE coordinate And the position coordinate of each AP, the distance between the UE and the b-th AP is calculated:

[0092]

[0093] Wherein, h represents the vertical height of the AP;

[0094] The distance estimation value is used Calculate the propagation time estimation value:

[0095]

[0096] Calculate the time synchronization offset estimation value between the UE and the bth AP:

[0097]

[0098] According to the motion speed estimation value of the UE Calculate the projection speed estimation value in the unit projection vector Direction:

[0099]

[0100] Calculate the frequency shift estimation value:

[0101]

[0102] Calculate the frequency synchronization offset estimation value between the UE and the bth AP:

[0103]

[0104] Where N is the length of the OFDM symbol, f s The sampling frequency.

[0105] The application also discloses a one-bit quantization time and frequency synchronization system of a cell-free massive MIMO system, which is used for executing the above method and comprises the following modules:

[0106] The SC algorithm estimates the STO / CFO module: the symbol timing offset STO and the carrier frequency offset CFO are estimated by using the Schmidl-Cox synchronization algorithm;

[0107] The positioning and speed measurement module: according to the estimation values of the STO and the CFO, the distance between the user UE and the access point AP and the motion speed of the user UE are estimated;

[0108] The STO / CFO estimation module based on positioning and speed measurement: the STO and the CFO are estimated again by using the obtained distance and motion speed.

[0109] Compared with the prior art, the time and frequency synchronization method based on positioning and speed measurement of the application estimates the STO and the CFO, and more accurate time and frequency synchronization can be obtained, which helps to reduce the estimation error and improve the receiving performance of the system. BRIEF DESCRIPTION OF DRAWINGS

[0110] Figure 1 It is the structure diagram of the cell-free massive MIMO system related to the preferred embodiment of the application.

[0111] Figure 2 This is a simulation experiment diagram of the user transmit power and the root mean square error of STO estimation in a preferred embodiment of the present invention.

[0112] Figure 3 This is a simulation experiment diagram of the user transmit power and the root mean square error of CFO estimation in a preferred embodiment of the present invention.

[0113] Figure 4 This is a block diagram of a time and frequency synchronization system for a one-bit quantized non-cellular massive MIMO system based on positioning and velocity measurement, according to a preferred embodiment of the present invention. Detailed Implementation

[0114] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to specific embodiments.

[0115] like Figure 1 As shown, in the cellular-free massive MIMO uplink system of the preferred embodiment of the present invention, the system consists of U UEs and B APs. Specifically, in the distributed MIMO architecture of the preferred embodiment of the present invention, the UE sends an RF signal to the AP. The RF signal received by each AP's single antenna port first passes through the BPF, and then jitter noise is added during signal processing. The signal is quantized by a zero-threshold comparator to generate a binary waveform, which is then transmitted to the CPU via optical fiber. There are unknown STOs and CFOs between each UE and each AP, which affect the quality of the received signal and correct demodulation, and reduce the reliability and performance of the system. At the CPU, the STOs and CFOs are estimated to reduce synchronization errors, and after compensation, the signal is digitally down-converted, spatially processed, and then demodulated and decoded.

[0116] In the distributed system of the preferred embodiment of the present invention, the STO and CFO between a single UE and B APs are considered, and the synchronization error caused by STO and CFO is accurately estimated and reduced by a time and frequency synchronization method based on positioning and speed measurement.

[0117] In this cellular-free massive MIMO system, a three-dimensional spatial coordinate system (x, y, z) is established, where the distances from the UE to B APs are (d1, d2, ..., d). B The position coordinates of the b-th AP are p. b =(x b ,y b The position coordinates of the UE are p, h). (u) =(x (u) ,y (u) ,0), the velocity of the UE is v=(v x ,vy ,0).

[0118] Assuming the signal propagates in free space, the propagation time t b By distance d b and light speed c, we can get:

[0119]

[0120] δ b is the sampled period normalized STO between UE and the bth AP, δ b can be given by:

[0121]

[0122] Where f s is the sampling frequency.

[0123] Since the UE is located in the xoy plane, its projection velocity is also limited in the plane. By ignoring the z-axis component, the projection d b,proj of the direction vector from UE to the bth AP in the xoy plane is represented as:

[0124] d b,proj = (x b -x (u) , y b -y (u) , 0)

[0125] Its unit projection vector is:

[0126]

[0127] In three-dimensional space, the projection velocity v b,proj of the UE's movement velocity in the direction of the unit projection vector is:

[0128]

[0129] According to the Doppler shift formula, the frequency shift f b can be represented as:

[0130]

[0131] Where f c is the carrier frequency.

[0132] ε b is the subcarrier spacing normalized CFO between UE and the bth AP, ε b and the relationship between the frequency shift f b is:

[0133]

[0134] where N is the length of the OFDM symbol, f s is the sampling frequency.

[0135] For a fixed symbol length N, consider the received signal in the time interval T = N / f s The sample of the i-th OFDM symbol of the output signal of the b-th one-bit ADC at the CPU is denoted as:

[0136]

[0137] where denotes the discrete-time RF signal of the i-th OFDM symbol received at the b-th AP, after BP filtering and sampling, the n-th sample before one-bit quantization. The perturbation signal is assumed to be independent of and independent across n. The function sgn(·) is applied element-wise to the input vector.

[0138] The received signal is denoted as:

[0139]

[0140] The time-domain expression of the baseband signal is:

[0141]

[0142] where h b is the channel gain from the UE to the b-th AP, is the signal transmitted by the UE, where E s denotes the average energy per sample, is the noise, where N0denotes the power spectral density of the noise.

[0143] The signal s[n] transmitted by the b-th AP is composed of consecutive time-domain OFDM symbols and CPs, and can be denoted as:

[0144]

[0145] where G > L - 1 is the length of the CP, and the relationship between i and n is -G < n - i < N - 1.

[0146] s (i) [n] is the time-domain representation of the i-th OFDM symbol, and can be denoted by IDFT as:

[0147]

[0148] In the embodiment, the number of UEs U = 1, the number of APs B = 4, the number of sampling points N = 4050, the number of occupied subcarriers S = 1201, the carrier frequency f c = 2.4 GHz, the signal bandwidth is 24 MHz, the number of channel taps L = 1, and the application scenario is a cell-free massive MIMO uplink system.

[0149] Specifically, the one-bit quantization time and frequency synchronization method for a cell-free massive MIMO system in the embodiment includes the following specific steps:

[0150] Step 1, estimate the symbol timing offset STO and the carrier frequency offset CFO using the Schmidl-Cox synchronization algorithm.

[0151] The CPU uses time-frequency synchronization, and the Schmidl-Cox synchronization algorithm is used to synchronize the time and carrier frequency of the quantized signal , and estimate the STO and CFO. The estimated values of the STO and CFO between the UE and the bth AP are denoted as Specifically as follows:

[0152] Step 1.1, estimation of STO:

[0153] Calculate the autocorrelation value of the signal at different time offsets:

[0154]

[0155] Calculate the energy distribution of the signal at different offsets:

[0156]

[0157] Calculate the ratio of the correlation of the signal to the energy of the signal:

[0158]

[0159] Estimate the STO at the bth AP:

[0160]

[0161] Step 1.2, estimation of CFO:

[0162] Calculate the relevant measure Use the function arg{·} to obtain the phase and estimate the frequency offset of the signal:

[0163]

[0164] Step 2, positioning and speed measurement;

[0165] The estimated values of STO and CFO obtained by the Schmidl-Cox algorithm are used to further calculate the estimated value of the distance between the UE and the AP and the estimated value of the projection speed of the UE in the xoy plane. The least square (LS) algorithm is used to estimate the position coordinate of the UE and the motion speed of the UE, and more accurate estimation of the time and frequency synchronization of the system is performed. Specifically, the following steps are taken:

[0166] Step 2.1, estimating the distance between the UE and the AP and the projection speed of the UE:

[0167] Using the estimated value of STO Distance estimation is performed to calculate the estimated value of the propagation time:

[0168]

[0169] The estimated value of the distance between the UE and the AP is calculated:

[0170]

[0171] Using the estimated value of CFO Projection speed estimation is performed to calculate the estimated value of the frequency shift:

[0172]

[0173] According to the Doppler shift formula, the projection estimate of the motion speed of the UE is calculated:

[0174]

[0175] Step 2.2, estimating the position coordinate of the UE and the motion speed of the UE:

[0176] Using the estimated value of the distance between the UE and each AP obtained in step 2.1 The position of the UE is estimated by the LS algorithm.

[0177] Constructing a matrix:

[0178]

[0179] Solving by the LS algorithm:

[0180]

[0181] The matrix for the estimated value of the coordinate of the UE is

[0182] According to the estimated value of the position coordinate of the UE and the projection estimate of the motion speed Information, solve the UE's motion speed estimate

[0183] For the UE to the b-th AP between the unit projection vector direction, calculate the speed projection estimate value:

[0184]

[0185] Construct the matrix:

[0186]

[0187]

[0188] Where B contains the unit projection vector of each AP direction, and c contains the speed projection estimate value to each AP.

[0189] Solve with LS algorithm:

[0190]

[0191] Contains the estimate of the UE's motion speed, then the UE's motion speed estimate can be expressed as

[0192] Step 3, use the distance and speed to estimate STO and CFO again;

[0193] According to the estimate of the UE coordinate And the location coordinates of each AP, calculate the distance between the UE and the b-th AP:

[0194]

[0195] Use the distance estimate Calculate the propagation time estimate:

[0196]

[0197] Calculate the time synchronization offset estimate between the UE and the b-th AP:

[0198]

[0199] According to the estimate of the UE's motion speed Calculate the projection speed estimate value in the unit projection vector Direction:

[0200]

[0201] Calculate the frequency shift estimate:

[0202]

[0203] Calculate the frequency synchronization offset estimation value between the UE and the bth AP:

[0204]

[0205] Figure 2 is a simulation experiment diagram of the root mean square error of the STO estimation of the system user transmission power and the STO estimation in the application, the horizontal coordinate represents the user transmission power (dBm), and the vertical coordinate represents the root mean square error of the STO estimation. In the diagram, there are two curves, wherein the “positioning speed estimation based synchronization estimation” is the root mean square error simulation curve of the STO estimation of the synchronization estimation method of the application, and the “SC algorithm estimation” is the root mean square error simulation curve of the STO estimation of the synchronization estimation method of the single Schmidl-Cox algorithm. From Figure 2 It can be seen that under different user transmission powers, the root mean square error of the CFO estimation of the synchronization estimation method based on positioning speed is obviously lower than that of the synchronization estimation method of the single SC algorithm. The application method can obtain a lower CFO estimation error result through software testing, and its performance is obviously better than that of the single Schmidl-Cox algorithm estimation.

[0206] Figure 3 is a simulation experiment diagram of the root mean square error of the STO estimation of the system user transmission power and the STO estimation in the application, the horizontal coordinate represents the user transmission power (dBm), and the vertical coordinate represents the root mean square error of the STO estimation. In the diagram, there are two curves, wherein the “positioning speed estimation based synchronization estimation” is the root mean square error simulation curve of the STO estimation of the synchronization estimation method of the application, and the “SC algorithm estimation” is the root mean square error simulation curve of the STO estimation of the synchronization estimation method of the single Schmidl-Cox algorithm. From Figure 3 It can be seen that under different user transmission powers, the root mean square error of the CFO estimation of the synchronization estimation method based on positioning speed is obviously lower than that of the synchronization estimation method of the single SC algorithm. The application method can obtain a lower CFO estimation error result through software testing, and its performance is obviously better than that of the single Schmidl-Cox algorithm estimation.

[0207] As Figure 4 shown, the embodiment discloses a one-bit quantization time and frequency synchronization system of a positioning speed-based large-scale MIMO system without cells, which comprises the following modules:

[0208] The SC algorithm estimation STO / CFO module estimates the symbol timing offset STO and the carrier frequency offset CFO by using the Schmidl-Cox synchronization algorithm;

[0209] Positioning and speed measurement module: according to the estimated values of STO and CFO, the distance between the user UE and the access point AP and the motion speed of the user UE are estimated;

[0210] STO / CFO estimation based on positioning and speed measurement module: the STO and CFO are accurately estimated again by using the distance and speed to reduce the synchronization error.

[0211] In summary, the present application proposes a time and frequency synchronization method and system based on positioning and speed measurement, which combines one-bit quantization and RF structure optimization technology, aiming to achieve high-precision synchronization estimation in a large-scale MIMO system without cells. By using the one-bit quantization method, the present application can significantly reduce the hardware complexity and power consumption of the system, and simplify the signal processing process. By combining the radio frequency signal structure, the present application can further improve the flexibility and reliability of the system, and maintain efficient signal transmission and synchronization performance in a dynamic environment. The present application can significantly improve the synchronization accuracy while reducing the system cost, and effectively improve the overall communication performance of the system, which has important technical advantages and broad application prospects.

[0212] For those skilled in the art, various changes, modifications, replacements and variations of the above embodiments can be made without departing from the principles and spirits of the present application. The scope of the present application is defined by the appended claims and their equivalents, and still belongs to the scope of the method described in the present application, and is still considered to be within the protection scope of the present application.

Claims

1. A time and frequency synchronization method for a one-bit quantized, cellular-free massive MIMO system, characterized in that, The specific steps are as follows: Step 1: Use the Schmidl-Cox synchronization algorithm to estimate the symbol timing offset (STO) and carrier frequency offset (CFO); Step 2: Based on the estimated values ​​of STO and CFO, estimate the distance between the user UE and the B access points (APs) and the movement speed of the user UE; Step 3: Use the distance and speed of movement from Step 2 to estimate STO and CFO again.

2. The time and frequency synchronization method for a one-bit quantized non-cellular massive MIMO system as described in claim 1, characterized in that, In step 1, the CPU uses time-frequency synchronization and employs the Schmidl-Cox synchronization algorithm to synchronize the i-th OFDM symbol of the b-th bit ADC output signal at the CPU. Synchronize time and carrier frequency, and estimate symbol timing offset (STO) and carrier frequency offset (CFO); where n represents the index variable.

3. The time and frequency synchronization method for a one-bit quantized non-cellular massive MIMO system as described in claim 2, characterized in that, User UE and the The estimated values ​​of STO and CFO among the access points (APs) are denoted as follows: , Step 1 is as follows: Step 1.1, Estimation of STO: Calculate the autocorrelation value of the signal at different time offsets: in, For fixed OFDM symbol length, It is the time offset. Indicates complex conjugation; Calculate the energy distribution of the signal at different offsets: Calculate the ratio of signal correlation to signal energy: in, The length of the cyclic prefix; Estimate in the first STO at each AP: Step 1.2, CFO estimation: calculate Related metrics Using functions Obtain the phase and estimate the frequency shift of the signal: 。 4. The time and frequency synchronization method for a one-bit quantized non-cellular massive MIMO system as described in claim 3, characterized in that, In step 2, based on the obtained estimates of STO and CFO, the distance from UE to AP and the speed of UE movement are estimated using the least squares algorithm.

5. The time and frequency synchronization method for a one-bit quantized non-cellular massive MIMO system as described in claim 4, characterized in that, Step 2 is as follows: Step 2.1: Estimate the distance from the UE to the AP and the projection velocity of the UE: Using the STO estimate Perform distance estimation and calculate the estimated propagation time: Among them, f s The sampling frequency; Calculate the estimated distance from UE to AP: in, It's the speed of light; Using CFO estimates Perform projection velocity estimation and calculate the estimated frequency shift: Based on the Doppler frequency shift formula, calculate the projected estimate of the UE's motion velocity: Step 2.2: Estimate the UE's position coordinates and UE's movement speed: Using the estimated distances between the UE and each AP obtained in step 2.1 The LS algorithm is used to estimate the UE's position; Construct the matrix: in, It is the first The position coordinates of each AP in the horizontal plane; Solve using the LS algorithm: Let be the matrix of coordinate estimates for the UE, then the estimated position coordinates of the UE are expressed as: ,in, and These are the UE on the horizontal plane. coordinates and Estimated coordinates; Based on the estimated location coordinates of the UE Projected estimates of motion velocity Solve for the estimated motion velocity of the UE, where, and These represent the UE in the horizontal plane. shaft and Estimated velocity along the axial direction; For UE to the Calculate the velocity projection estimate along the unit projection vector direction between APs: Construct the matrix: in, Includes the unit projection vector for each AP direction. Includes velocity projection estimates for each AP; Solve using the LS algorithm: Including the estimated motion speed of the UE, the estimated motion speed of the UE is expressed as: .

6. The time and frequency synchronization method for a one-bit quantized non-cellular massive MIMO system as described in claim 5, characterized in that, Step 3 is as follows: Based on the estimated values ​​of the UE coordinates And the location coordinates of each AP, calculate the UE to the number Distance between APs: in, Indicates the vertical height of AP; Using distance estimates Calculate the estimated propagation time: Calculate UE to the Estimated time synchronization offset between APs: Based on the estimated motion speed of the UE Calculate the unit projection vector Projected velocity estimate in the direction: Calculate the frequency shift estimate: Calculate UE to the Estimated frequency synchronization offset between APs: 。 7. A time and frequency synchronization system for a one-bit quantized, cellular-free massive MIMO system, used to perform the method as described in any one of claims 1-6, characterized in that, Includes the following modules: The SC algorithm estimates the STO / CFO module: It uses the Schmidl-Cox synchronization algorithm to estimate the symbol timing offset (STO) and carrier frequency offset (CFO). Positioning and speed measurement module: Based on the estimated values ​​of STO and CFO, estimate the distance between the user UE and B access points (APs) and the speed of the user UE. The STO / CFO estimation module based on positioning and velocity measurement: uses the obtained distance and velocity to estimate STO and CFO again.

Images