Method and system for time and carrier frequency synchronization based on joint correlation of preamble and prefix

Abstract

The application discloses a time and carrier frequency synchronization method and system based on joint correlation of a preamble and a prefix. Aiming at a MIMO-OFDM downlink system with one-bit quantization, the application provides a time and carrier frequency synchronization method based on joint correlation of a preamble and a prefix, which is used for realizing accurate synchronization of time and carrier frequency of the system and reducing estimation errors of time offset (STO) and carrier frequency offset (CFO). The method is as follows: step 1, calculating correlation values and ratios of a preamble symbol and prefixes of OFDM symbols except for a first symbol, the preamble symbol and a last symbol, to obtain an estimated value of the STO; and step 2, estimating a phase shift of the preamble symbol and the prefixes of the OFDM symbols except for the first symbol, the preamble symbol and the last symbol to obtain an estimated value of the CFO. The application can effectively reduce the estimation errors of the STO and the CFO, reduce a bit error rate of the system, improve system performance and has good realizability.

Description

Technical Field

[0001] The present invention belongs to the field of wireless communication technology, and is directed to a system that combines Multiple-Input Multiple-Output (MIMO) and Orthogonal Frequency Division Multiplexing (OFDM) technologies under one-bit quantization, and specifically relates to a method and system for synchronizing time and carrier frequency of a MIMO-OFDM system. Background Art

[0002] In recent years, with the rapid development of wireless communication technology, MIMO and OFDM, two key technologies in communications, have made significant contributions to improving system performance and spectrum efficiency. MIMO technology can increase communication system capacity and spectrum efficiency, enhance signal reliability, overcome channel fading, and reduce bit error rates. OFDM, on the other hand, can transform frequency-selective fading into flat fading, thereby reducing the impact of multipath fading. Therefore, to reduce power consumption, the industry has proposed the use of MIMO-OFDM systems with one-bit quantization, which has made significant contributions to modern wireless communication systems. However, while MIMO-OFDM systems offer significant advantages, they also face the complex challenges of time and carrier frequency synchronization.

[0003] In a MIMO-OFDM system, time offset is caused by the start moment of an unknown frame and propagation delay. Time synchronization ensures that the receiving end can correctly detect the start position of the signal frame and enable OFDM symbols to be accurately demodulated. Carrier frequency offset is caused by the instability of the oscillator and Doppler frequency shift, which greatly affects the demodulation performance of the system. Carrier frequency synchronization ensures that the receiving end can correctly estimate and compensate for these offsets, which is very important for maintaining the orthogonality between OFDM symbols. In response to this situation, the present invention proposes a method and system for time synchronization and carrier frequency synchronization of a MIMO-OFDM system under one-bit quantization based on joint correlation of the preamble and prefix. The present invention aims to achieve precise synchronization of the system time and carrier frequency, reduce the estimation error of time offset and carrier frequency offset, and reduce the system bit error rate. Summary of the Invention

[0004] In the existing MIMO-OFDM system under one-bit quantization, the present invention introduces an innovative method based on joint correlation of preamble and prefix, aiming to achieve precise synchronization of the system time and carrier frequency, reduce the estimation error of time offset and carrier frequency offset, and lower the system bit error rate.

[0005] Consider a MIMO-OFDM downlink system using one-bit quantization. The transmitter is a base station (BS) with a total of B antennas, and the receiver has U single-antenna users (UEs). The transmitter transmits an OFDM signal to the receiver. The BS processes the signal using precoding, inverse discrete Fourier transform, adding a cyclic prefix (CP), and a one-bit digital-to-analog converter (DAC). Each BS performs digital-to-analog conversion using a pair of one-bit DACs. There are unknown time and frequency offsets between each UE and the BS. These offsets are called symbol-timing offset (STO) and carrier-frequency offset (CFO), and will degrade system performance.

[0006] In the presence of time and frequency offset, for the u-th UE, the n-th time slot The received signal is:

[0007]

[0008] Among them, ε u ∈R is the CFO at the u-th UE normalized by subcarrier spacing, τ u ∈Z is the STO at the u-th UE normalized by the sampling period, N is the number of subcarriers in each OFDM symbol, L is the length of the multipath channel tap, is the channel vector of the lth channel tap between the uth UE and the BS. Q(·) is a quantization function defined as:

[0009]

[0010] Corresponding to the operation of a one-bit DAC; is the signal transmitted by the u-th terminal in the n-th time slot; sgn is a step function, sgn(x)={-1,x<0;1,x≥0}; ω u [n]~CN(0,N0) is the receiving noise of the u-th UE, where CN(0,N0) is represented by a complex Gaussian distribution with a mean of 0 and a variance of N0.

[0011] The signal transmitted from the u-th UE consists of continuously precoded time-domain OFDM symbols and CP, which can be expressed as:

[0012]

[0013] Where G≥L-1 is the length of CP, and the relationship between m and n is -G≤nm(n+G)≤N-1;

[0014] is the m-th precoded time-domain OFDM signal with CP, which can be expressed as:

[0015]

[0016] where is the m-th frequency-domain precoded OFDM signal, and the subcarrier range When some subcarriers are not occupied, i.e., S < N, where S is the cardinality of. BS adopts linear precoding in the frequency domain, which is expressed as:

[0017]

[0018] where is the precoding vector of the u-th UE in the k-th subcarrier, is the frequency-domain OFDM signal in the k-th subcarrier of the m-th symbol of the u-th UE, and satisfies where E represents expectation.

[0019] UE adopts time-frequency synchronization to estimate and compensate STO and CFO. At the u-th UE, the estimates of STO and CFO are denoted as After compensation, the received signal is processed as:

[0020]

[0021] Then, through time shift and discrete Fourier transform, r u [n] is further transformed into:

[0022]

[0023] where the shift of is to change the range of STO to achieve interference-free communication.

[0024] To achieve the technical purpose of the present invention, that is, to achieve the precise synchronization of the system time and carrier frequency, reduce the estimation error of time offset and carrier frequency offset, and reduce the system bit error rate. The present invention adopts the following technical solutions:

[0025] The present invention introduces a method based on the joint correlation of preamble and prefix to estimate STO and CFO. Define the number of OFDM symbols as O. The data frame structure of the present invention consists of a virtual symbol, a preamble symbol, a pilot symbol, (O - 3) data symbols and a virtual symbol. is the value of the nth sampling point of the mth OFDM symbol of the uth UE. The preamble symbol consists of two identical sequences of length N / 2, such that in, It follows that y u [n] = z u [n]+w u [n], in, In the absence of Gaussian white noise, except for the phase shift caused by CFO -πε u In addition, the two corresponding received preamble sequences are identical. In addition, the presence of a one-bit DAC does not affect the symmetric structure of the preamble symbols, and this property is exploited to calculate the estimate of the STO at the u-th UE.

[0026] The following are the specific steps of the present invention:

[0027] A time and carrier frequency synchronization method based on joint correlation of preamble and prefix is ​​applied to a MIMO-OFDM downlink system using one-bit quantization. The linear array is a uniform linear array. The specific steps of the method are as follows:

[0028] Step 1: Calculate the correlation value and ratio of the leading symbol and the OFDM symbol prefix excluding the first symbol, the leading symbol, and the last symbol to obtain an estimated value of STO;

[0029] Step 2: Estimate the phase shift using the preamble symbol and the prefix of the OFDM symbol except the first symbol, the preamble symbol, and the last symbol to obtain an estimated value of CFO.

[0030] Preferably, step 1 is as follows:

[0031] 1.1. First calculate the correlation value and ratio of the leading symbol. Divide the leading symbol into five parts and calculate the cross-correlation and autocorrelation respectively. The calculation formula of the cross-correlation is:

[0032]

[0033]

[0034]

[0035] The sum of the cross-correlation values ​​is P u (τ) = a u (τ)+b u (τ)+c u (τ).

[0036] Where n represents the time slot; u represents the UE; N represents the number of subcarriers in each OFDM symbol; G represents the length of the cyclic prefix; τ represents the STO normalized by the sampling period; y u [n] represents the received signal of the u-th UE in the n-th time slot; represents the complex conjugate of the received signal of the n+τth time slot of the uth UE; a u (τ) represents the calculation result of cross-correlation when n∈(0,G-1); b u (τ) represents The calculation results of time-correlation; c u (τ) represents The calculation results of the time correlation.

[0037] Calculate autocorrelation, the formula is:

[0038]

[0039] Calculate the ratio of the two, the formula is:

[0040]

[0041] 1.2. Calculate the cross-correlation and autocorrelation of the prefix of the OFDM symbol except the first symbol, the leading symbol and the last symbol. The calculation formula of the cross-correlation is:

[0042]

[0043] Where O represents the number of all OFDM symbols; i represents the number of OFDM symbols involved in the calculation;

[0044] The formula for calculating autocorrelation is:

[0045]

[0046] The ratio of the two is:

[0047]

[0048] 1.3. Multiply the above results by their respective weights, then J u The calculation formula for (τ) is:

[0049]

[0050] 1.4. The final STO estimate obtained through the above steps is

[0051] Preferably, the second step is to estimate the value of CFO, and the specific steps are as follows.

[0052] 2.1, first use the leading symbol to estimate the phase shift. u |<1,P u The phase value range is (-π, π). The calculation formula is:

[0053]

[0054] Among them, the angle function is used to calculate the phase angle of the complex number, and obtain P u The phase angle range is (-π, π).

[0055] 2.2. Use the prefix of OFDM symbols except the first symbol, leading symbol and last symbol to estimate the phase shift. u |<1,D u The phase value range is (-2π, 2π). The calculation formula is:

[0056]

[0057] Get D u The phase angle range is (-π, π).

[0058] 2.3. Combining the above two situations, before obtaining the final CFO estimate, it is necessary to convert the value obtained in the second step to the range of (-2π, 2π) to obtain another phase estimate. The calculation formula is:

[0059]

[0060] 2.4. Take the average of the two and get the final estimated CFO value:

[0061]

[0062] The present invention also discloses a time and carrier frequency synchronization system based on joint correlation of preamble and prefix, comprising the following modules:

[0063] STO estimation value acquisition module: calculates the correlation value and ratio of the leading symbol and the OFDM symbol prefix except the first symbol, the leading symbol and the last symbol to obtain the STO estimation value;

[0064] CFO estimation value calculation module: uses the preamble symbol and the prefix of the OFDM symbol except the first symbol, the preamble symbol and the last symbol to estimate the phase shift and obtain the CFO estimation value.

[0065] The present invention proposes a method and system for time synchronization and carrier frequency synchronization of a MIMO-OFDM system under one-bit quantization based on joint correlation of a preamble and a prefix, aiming to achieve precise synchronization of the system time and carrier frequency, reduce the estimation error of the time offset and carrier frequency offset, and lower the system bit error rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] Figure 1 This is a downlink diagram of the MIMO-OFDM system under one-bit quantization.

[0067] Figure 2 It is a simulation diagram showing the change of the root mean square error (RMSE) of the estimated STO with the signal-to-noise ratio (SNR) in the preferred embodiment of the present invention.

[0068] Figure 3 3 is a simulation diagram showing how the RMSE of the estimated CFO varies with the SNR in a preferred embodiment of the present invention.

[0069] Figure 4 It is a simulation diagram of the bit error rate (BER) without coding using QPSK, including three cases: known real STO and CFO, Schmidl-Cox algorithm, and the method of the present invention.

[0070] Figure 5 This is a block diagram of a time and carrier frequency synchronization system based on joint correlation of a preamble and a prefix in a preferred embodiment of the present invention. DETAILED DESCRIPTION

[0071] In order to make the objectives, technical solutions and advantages of the present invention more clear, the present invention will be further described in detail below with reference to preferred embodiments.

[0072] This embodiment provides a time and carrier frequency synchronization method based on joint correlation of a preamble and a prefix. The specific steps of the method are as follows:

[0073] Step 1: Calculate the correlation value and ratio between the leading symbol and the OFDM symbol prefix excluding the first symbol, the leading symbol, and the last symbol, and finally obtain the estimated value of STO.

[0074] 1.1. First calculate the correlation value and ratio of the leading symbol. Divide the leading symbol into five parts and calculate the cross-correlation and autocorrelation respectively. The calculation formula of cross-correlation is:

[0075]

[0076]

[0077]

[0078] The sum of the cross-correlation values ​​is P u (τ) = a u (τ)+b u (τ)+c u (τ).

[0079] Calculate autocorrelation, the formula is:

[0080]

[0081] Calculate the ratio of the two:

[0082]

[0083] 1.2. Calculate the cross-correlation and autocorrelation of the prefix of the OFDM symbol except the first symbol, the leading symbol and the last symbol. The calculation formula of the cross-correlation is:

[0084]

[0085] The formula for calculating autocorrelation is:

[0086]

[0087] The ratio of the two is:

[0088]

[0089] 1.3. Multiply the above results by their respective weights, then J u The calculation formula for (τ) is:

[0090]

[0091] 1.4. The final STO estimate obtained through the above steps is

[0092] Step 2: Estimate the phase shift using the preamble symbol and the prefix of the OFDM symbol except the first symbol, the preamble symbol, and the last symbol, and finally obtain an estimated value of the CFO.

[0093] 2.1, using the leading symbol to estimate the phase shift. Since |ε u |<1,P u The phase value range is (-π, π). The calculation formula is:

[0094]

[0095] Among them, the angle function is used to calculate the phase angle of the complex number, and obtain Pu The phase angle range is (-π, π).

[0096] 2.2. Use the prefix of OFDM symbols except the first symbol, leading symbol and last symbol to estimate the phase shift. u |<1,D u The phase value range is (-2π, 2π). The calculation formula is:

[0097]

[0098] Get D u The phase angle range is (-π, π).

[0099] 2.3. Combining the above two situations, before obtaining the final CFO estimate, it is necessary to convert the value obtained in the second step to the range of (-2π, 2π) to obtain another phase estimate. The calculation formula is:

[0100]

[0101] 2.4. Take the average of the two to get the final estimated CFO value. The calculation formula is:

[0102]

[0103] Figure 2 This is a simulation diagram of the estimated RMSE of STO as the SNR changes in the preferred embodiment of the present invention. The horizontal axis is SNR and the vertical axis is the RMSE value of STO. There are four curves in the figure, two of which are simulation curves of the method of the present invention. "ZFI of the present invention" represents the performance of the method of the present invention under no quantization conditions, "ZFQ of the present invention" represents the performance of the method of the present invention under one-bit quantization, and the other two are simulation curves using the Schmidl-Cox algorithm. "SC algorithm ZFI" represents the performance of the Schmidl-Cox algorithm under no quantization conditions, and "SC algorithm ZFQ" represents the performance of the Schmidl-Cox algorithm under one-bit quantization. Among them, B = 128, U = 8, N = 2048, S = 1200, G = 144, L = 10. By Figure 2 It can be seen that the estimation error of the method of the present invention is smaller than the estimation error of the Schmidl-Cox algorithm.

[0104] Figure 3This is a simulation graph of the change of RMSE of CFO with SNR in the preferred embodiment of the present invention, with the horizontal axis being SNR and the vertical axis being the RMSE value of CFO. There are four curves in the figure, two of which are simulation curves of the method of the present invention. "ZFI of the present invention" represents the performance of the method of the present invention under no quantization conditions, and "ZFQ of the present invention" represents the performance of the method of the present invention under one-bit quantization. The other two are simulation curves using the Schmidl-Cox algorithm. "SC algorithm ZFI" represents the performance of the Schmidl-Cox algorithm under no quantization conditions, and "SC algorithm ZFQ" represents the performance of the Schmidl-Cox algorithm under one-bit quantization. Wherein, B = 128, U = 8, N = 2048, S = 1200, G = 144, L = 10. By Figure 3 It can be seen that the estimation error of the method of the present invention is smaller than the estimation error of the Schmidl-Cox algorithm.

[0105] Figure 4 It is a simulation diagram of the uncoded BER using QPSK, which is divided into three cases: known real STO and CFO, Schmidl-Cox algorithm, and the method of the present invention. There are six curves in the figure, two of which are simulation curves of known real STO and CFO, "(known STO and CFO) ZFI" represents the performance of known real STO and CFO under no quantization conditions, "(known STO and CFO) ZFQ" represents the performance of known real STO and CFO under one-bit quantization; two are simulation curves using the Schmidl-Cox algorithm, "SC algorithm ZFI" represents the performance of the Schmidl-Cox algorithm under no quantization conditions, "SC algorithm ZFQ" represents the performance of the Schmidl-Cox algorithm under one-bit quantization; two are simulation curves of the method of the present invention, "ZFI of the present invention" represents the performance of the method of the present invention under no quantization conditions, "ZFQ of the present invention" represents the performance of the method of the present invention under one-bit quantization. Among them, B = 128, U = 8, N = 2048, S = 1200, G = 144, L = 10, P = 1. By Figure 4 It can be seen that the bit error rate of the method of the present invention is lower than that of the Schmidl-Cox algorithm and is close to the bit error rate under known true STO and CFO conditions. Therefore, the present invention proposes a method for MIMO-OFDM system time synchronization and carrier frequency synchronization using one-bit quantization based on joint correlation of preamble and prefix, which can effectively reduce the estimation error of STO and CFO, reduce the system bit error rate, and effectively improve system performance.

[0106] like Figure 5 As shown, this embodiment discloses a time and carrier frequency synchronization system based on joint correlation of preamble and prefix, including the following modules:

[0107] STO estimation value acquisition module: calculates the correlation value and ratio of the leading symbol and the OFDM symbol prefix except the first symbol, the leading symbol and the last symbol to obtain the STO estimation value;

[0108] CFO estimation value calculation module: uses the preamble symbol and the prefix of the OFDM symbol except the first symbol, the preamble symbol and the last symbol to estimate the phase shift and obtain the CFO estimation value.

[0109] For other contents of this embodiment, please refer to the above method embodiment.

[0110] The present invention discloses a time and carrier frequency synchronization method and system based on the joint correlation of preamble and prefix. For a MIMO-OFDM downlink system using one-bit quantization, the present invention proposes a time and carrier frequency synchronization method based on the joint correlation of preamble and prefix, aiming to achieve precise synchronization of the system's time and carrier frequency, while reducing the estimation errors of time offset and carrier frequency offset. This method can effectively reduce the estimation errors of STO and CFO, lowering the system's bit error rate and improving system performance, with good feasibility.

[0111] The above describes the specific embodiments of the present invention to facilitate understanding of the present invention by those skilled in the art. However, the present invention is not limited to the scope of the specific embodiments. For those skilled in the art, as long as various changes are within the spirit and scope of the present invention as defined and determined by the appended claims, these changes are obvious, and all inventions and creations conceived by the present invention are protected.

Claims

1. A time and carrier frequency synchronization method based on joint correlation of preamble and prefix, characterized in that The specific steps are as follows: Step 1: Calculate the correlation value and ratio of the leading symbol and the OFDM symbol prefix excluding the first symbol, the leading symbol, and the last symbol to obtain an estimated value of STO; Step 2: Estimate the phase shift using the preamble symbol and the prefix of the OFDM symbols except the first symbol, the preamble symbol, and the last symbol to obtain an estimated value of CFO; Step 1 is as follows: Step 1.

1. Calculate the correlation value and ratio of the leading symbol; divide the leading symbol into five parts, and calculate the cross-correlation and autocorrelation respectively; the calculation formula of the cross-correlation is as follows: The sum of the cross-correlation values ​​is: P u (τ) = a u (τ)+b u (τ)+c u (τ); Where n represents the time slot; u represents the UE; N represents the number of subcarriers in each OFDM symbol; G represents the length of the cyclic prefix; τ represents the STO normalized by the sampling period; y u [n] represents the received signal of the u-th UE in the n-th time slot; represents the complex conjugate of the received signal of the n+τth time slot of the uth UE; a u (τ) represents the calculation result of cross-correlation when n∈(0,G-1); b u (τ) represents The calculation results of time-correlation; c u (τ) represents The calculation results of time cross-correlation; Calculate the autocorrelation value using the following formula: The ratio of the cross-correlation value to the autocorrelation value is calculated as: Step 1.2: Calculate the cross-correlation and autocorrelation of the prefix of the OFDM symbol except the first symbol, the leading symbol, and the last symbol. The calculation formula of the cross-correlation is as follows: Where O represents the number of all OFDM symbols; i represents the number of OFDM symbols involved in the calculation; The formula for calculating autocorrelation is: The ratio of the cross-correlation value to the autocorrelation value is: Step 1.3: Multiply the above two ratios by their respective weights, then J u The calculation formula for (τ) is: Step 1.4: The final STO estimate is: Step 2 is as follows: Step 2.1, use the leading symbol to estimate the phase shift; |ε u |<1,P u The phase value range is (-π, π); the calculation formula is: Among them, the angle function is used to calculate the phase angle of the complex number, and obtain P u The phase angle range is (-π, π); Step 2.2, use the prefix of the OFDM symbol except the first symbol, the leading symbol and the last symbol to estimate the phase shift; |ε u |<1,D u The phase value range is (-2π, 2π); the calculation formula is: Get D u The phase angle range is (-π, π); Step 2.3: Convert the value obtained in step 2.2 so that its value range is (-2π, 2π) to obtain another phase estimate. The calculation formula is: Step 2.4: Take the average of V1 and V3 to get the final estimated CFO value. The calculation formula is:

2. A time and carrier frequency synchronization system based on joint correlation of preamble and prefix, characterized in that: Includes the following modules: STO estimation value acquisition module: calculates the correlation value and ratio of the leading symbol and the OFDM symbol prefix except the first symbol, the leading symbol and the last symbol to obtain the STO estimation value; CFO estimation value calculation module: uses the preamble symbol and the prefix of the OFDM symbol except the first symbol, the preamble symbol and the last symbol to estimate the phase shift and obtain the estimated value of CFO; The STO estimated value acquisition module is as follows: Calculate the correlation value and ratio of the leading symbol; divide the leading symbol into five parts, and calculate the cross-correlation and autocorrelation respectively; the calculation formula of the cross-correlation is as follows: The sum of the cross-correlation values ​​is: P u (τ) = a u (τ)+b u (τ)+c u (τ); Where n represents the time slot; u represents the UE; N represents the number of subcarriers in each OFDM symbol; G represents the length of the cyclic prefix; τ represents the STO normalized by the sampling period; y u [n] represents the received signal of the u-th UE in the n-th time slot; represents the complex conjugate of the received signal of the n+τth time slot of the uth UE; a u (τ) represents the calculation result of cross-correlation when n∈(0,G-1); b u (τ) represents The calculation results of time-correlation; c u (τ) represents The calculation results of time cross-correlation; Calculate the autocorrelation value using the following formula: The ratio of the cross-correlation value to the autocorrelation value is calculated as: The cross-correlation and autocorrelation are calculated for the prefix of the OFDM symbol except the first symbol, the leading symbol, and the last symbol. The cross-correlation calculation formula is as follows: Where O represents the number of all OFDM symbols; i represents the number of OFDM symbols involved in the calculation; The formula for calculating autocorrelation is: The ratio of the cross-correlation value to the autocorrelation value is: Multiply the above two ratios by their respective weights, then J u The calculation formula for (τ) is: The final STO estimate is: The CFO estimated value calculation module is as follows: Use the leading symbol to estimate the phase shift; |ε u |<1,P u The phase value range is (-π, π); the calculation formula is: Among them, the angle function is used to calculate the phase angle of the complex number, and obtain P u The phase angle range is (-π, π); The phase shift is estimated using the prefix of the OFDM symbol except the first symbol, the leading symbol, and the last symbol; |ε u |<1,D u The phase value range is (-2π, 2π); the calculation formula is: Get D u The phase angle range is (-π, π); The obtained value is converted so that its value range is (-2π, 2π), and another phase estimate is obtained. The calculation formula is: Take the average of V1 and V3 to get the final estimated value of CFO, the calculation formula is:

Images