MIMO system channel estimation method and device based on ODDM modulation

Abstract

The present invention discloses a MIMO system channel estimation method and device based on ODDM modulation, comprising: generating a pilot data signal, performing ODDM modulation on the pilot signal, and then sending the modulated signal to a receiving module through a transmitting module. The receiving module performs matched filtering on the received pilot signal to obtain a received signal; an equivalent input-output relationship is derived based on the generated pilot signal and the channel received by the receiving end, and a low-complexity approximate Bayesian optimal algorithm is proposed to obtain information such as the channel delay, Doppler, gain and angle. The present invention can ensure that the accuracy of channel estimation is improved in high-speed mobile scenarios. At the same time, the present invention also avoids the process of matrix inversion in channel estimation, thereby significantly reducing the computational complexity.

Description

Technical Field

[0001] The present invention relates to the field of mobile communication technology, and in particular to a MIMO system channel estimation method and device based on ODDM modulation. Background Art

[0002] With the rapid development of wireless communication technology, future 6G mobile communication systems need to provide reliable communications in highly dynamic scenarios, including high-speed railways, drones, and low-Earth orbit satellites. Existing 4G and 5G mobile communication systems widely adopt the efficient and robust Orthogonal Frequency Division Multiplexing (OFDM) modulation method. However, in high-speed mobile scenarios, the Doppler spread associated with the time-varying channel can cause severe inter-carrier interference (ICI). This ICI phenomenon significantly degrades the performance of OFDM systems and poses a major challenge to maintaining a reliable communication link.

[0003] To address these issues, the recently proposed Orthogonal Time Frequency Space (OTFS) multi-carrier modulation has attracted widespread attention due to its significant advantages in time-varying channels. However, the OTFS waveform is still an orthogonal pulse in the TF domain, essentially acting as a precoded OFDM. In OTFS, the ideal pulse is assumed to have dual orthogonality. Unfortunately, such an ideal pulse does not exist in practice. This non-ideality faces challenges in practical implementation, such as high out-of-band emission (OOBE) and severe interference. In order to obtain achievable orthogonal pulses in the delay-Doppler (DD) plane, orthogonal delay-Doppler division multiplexing (ODDM) modulation is proposed. The characteristic of this method is to design a set of pulse sequences that can maintain orthogonality relative to the DD plane. ODDM modulation can achieve excellent coupling between the modulated signal and the DD channel, and its performance is better than OTFS.

[0004] As a key technology of communication system, channel estimation has always been a hot topic in the field of communication and has attracted the attention of many scholars. In large-scale multiple-input multiple-output (MIMO) communication systems based on ODDM modulation, the channel information changes rapidly due to the high-speed movement of devices. How to quickly and accurately estimate the channel and design a low-complexity channel estimation algorithm are urgent problems to be solved.

[0005] Based on the above problems, in order to quickly estimate channel information in high-speed mobile scenarios and improve the accuracy of channel estimation, a low-complexity approximate Bayesian optimal algorithm is proposed to estimate the time-varying multipath channel to improve the comprehensive performance of the synaesthesia system. Summary of the invention

[0006] The objective of the present invention is achieved through the following technical scheme: a MIMO system channel estimation method based on ODDM modulation, the method comprising:

[0007] S1. Generate a pilot signal and perform ODDM modulation on the pilot signal;

[0008] S2, sending the modulated signal to the sending device, which sends it to the receiving device through a multipath channel;

[0009] S3, the receiving device performs matched filtering on the received data to obtain a received signal, and calculates an equivalent input-output relationship of the pilot signal;

[0010] S4. According to the equivalent input-output relationship of the pilot signal, the memory approximate message passing algorithm is used to estimate the delay, Doppler and channel gain of the multipath channel. The angle rotation matrix of the sending steering vector is obtained through a one-dimensional search solution. The angle of the channel is obtained using discrete Fourier transform to obtain the estimated value of the multipath channel.

[0011] Furthermore, the generating of the pilot signal specifically includes: mapping the generated random signal into a delay-Doppler domain matrix, and then performing inverse Fourier transform, upsampling, and pulse train beamforming to obtain a modulated pilot signal.

[0012] Further, the ODDM modulation of the pilot signal is specifically expressed as:

[0013]

[0014] Where X[m,n] is the transmitted pilot signal, m∈{0,1,,M-1}, n∈{0,1,,N-1} represent the delay index and Doppler index respectively, M is the number of time slots, each multi-carrier symbol contains N sub-carriers, the duration is T, the spacing between each carrier is 1 / NT, and the multi-carrier symbols are evenly distributed at intervals of T / M. tx (t) is the filter at the transmitting end, which is a pulse train and is expressed as a(t) is a real-valued square root Nyquist pulse with time symmetry and duration 2QT / M, and satisfies the condition When the integer 2Q <<M g tx (t) is orthogonal in terms of delay and Doppler resolution, and the ambiguity function satisfy

[0015]

[0016] Furthermore, the matched filtering specifically includes: calculating the cross ambiguity function between the received waveform and the received signal by the matched filter

[0017]

[0018] in, hp is the channel gain, g tx (t) is the filter at the transmitting end, n c (t)∈C is noise, following Gaussian distribution and l p =v p NT are the indexes of delay and Doppler, and are the delay and Doppler of the pth path, respectively, where k p ,l p ∈ Z;

[0019] Get the discrete form of the received signal

[0020]

[0021] Where n′=[nl p ] N .

[0022] Further, the calculation of the equivalent input-output relationship of the pilot signal includes:

[0023] y=Φh+n c

[0024] in is the equivalent coefficient matrix, is the corresponding equivalent channel vector; n c For noise,

[0025] in Expressed as

[0026]

[0027] in and It is nth t The i-th row of symbols transmitted by the root antenna;

[0028] G is a matrix of dimension (2K+1)×L obtained by arranging the P paths of the delay-Doppler spatial channel, where the non-zero elements g(k+K+1,l) represent the channel gain of the corresponding path. The delay and Doppler of the path are and

[0029] Furthermore, the memory approximate message passing algorithm includes a long memory linear estimator and a nonlinear estimator,

[0030] The long memory linear estimator is used to obtain a channel estimation value based on the minimum mean square error criterion and eliminate the influence of multipath interference and Doppler shift, while the nonlinear estimator reduces Gaussian noise on the estimation value to improve the accuracy of the estimation;

[0031] The long memory linear estimator is specifically:

[0032] where κ t To speed up the convergence of the memory approximate message passing detector, θ t As a relaxation parameter, ensure The spectral radius is kept below 1; by minimizing the minimum mean square error of the long memory linear estimator Get the optimal κ t ;

[0033] The nonlinear estimator comprises:

[0034]

[0035] Among them, (t) represents the estimated mean value of the channel gain at time t,

[0036] in, Represented in κ t Selecting the optimal parameters The minimum mean square error of the long memory linear estimator when the channel vector h i = bg, where b follows a Bernoulli distribution with probability q and g follows a mean distribution of u g Variance is v g Gaussian distribution, and bg are independent of each other;

[0037] The maximum a posteriori mean and variance are

[0038]

[0039] in

[0040] is the damping vector, satisfying

[0041] The optimal damping vector is

[0042]

[0043] for The covariance of A collection of indexes for effective memory.

[0044] Furthermore, the channel angle obtained by discrete Fourier transform is specifically:

[0045] To avoid angle aliasing and satisfy the condition Nd / λ+1≤N-Nd / λ+1, the received signal is subjected to discrete Fourier transform to obtain

[0046]

[0047] in is the normalized discrete Fourier transform matrix, where the elements are is the corresponding direction vector, where d BS ,λ c , are the antenna spacing, carrier frequency wavelength and multipath channel angle respectively.

[0048] On the other hand, the present specification also provides a system for implementing the method, the system comprising:

[0049] Encoder: used to preset the pilot signal, perform ODDM modulation on the pilot signal, and then send the modulated pilot signal to the transmission module

[0050] Transmitting module: used to send the pilot signal modulated by the encoder to the receiving module

[0051] Receiving module: used to obtain the received signal after matching filtering the received pilot signal

[0052] Channel estimation module: According to the equivalent input-output relationship of the pilot signal, the memory approximate message passing algorithm is used to estimate the delay, Doppler and channel gain of the multipath channel, and the discrete Fourier transform with added phase rotation is used to obtain the channel angle and the estimated value of the multipath channel.

[0053] On the other hand, the present specification also provides a MIMO system channel estimation device based on ODDM modulation, including a memory and one or more processors, wherein the memory stores executable code, and when the processor executes the executable code, the MIMO system channel estimation method based on ODDM modulation is implemented.

[0054] On the other hand, the present specification also provides a computer-readable storage medium having a program stored thereon, and when the program is executed by a processor, the MIMO system channel estimation method based on ODDM modulation as described above is implemented.

[0055] Beneficial effects of the present invention:

[0056] (1) The ODDM modulation of the present invention can achieve the orthogonality of the delay Doppler plane, meeting the requirements of high-reliability communication and high-precision perception;

[0057] (2) The present invention derives the input-output relationship based on the generated pilot signal and the received signal, and then proposes a low-complexity algorithm to estimate the multipath channel. The algorithm can approximately achieve the Bayesian optimal channel estimation performance in a large-scale antenna scenario, thereby improving the accuracy of channel estimation.

[0058] (3) The present invention avoids the matrix inversion operation in the iteration of channel estimation, and only has the operation of matrix and vector multiplication, thereby reducing the complexity of channel estimation operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 is a flow chart of an implementation method of the present invention;

[0060] Figure 2 It is a schematic diagram of a device implemented by the present invention;

[0061] Figure 3 It is a schematic diagram of the normalized mean square error of different channel estimation methods under different signal-to-noise ratios;

[0062] Figure 4 It is a schematic diagram of the normalized mean square error of different channel estimation methods under different user speeds;

[0063] Figure 5 is a schematic diagram showing the performance comparison between the present invention and the traditional memory approximate message passing algorithm;

[0064] Figure 6 is a schematic diagram of the bit error rate of the present invention under conditions of perfect channel information and imperfect channel information;

[0065] Figure 7 It is a schematic diagram of a device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0066] The specific implementation modes of the present invention are further described in detail below with reference to the accompanying drawings.

[0067] like Figure 1 As shown, the MIMO system channel estimation method and device based on ODDM modulation provided by the present invention are specifically referred to as Figure 1 As shown, the specific implementation steps of the present invention are: S1, generating a pilot signal, and performing ODDM modulation on the pilot signal;

[0068] The pilot signal is specifically: the generated random signal is mapped into a delay-Doppler domain matrix, and then the modulated pilot signal is obtained after inverse Fourier transformation, upsampling, and pulse train beamforming.

[0069] S2, sending the modulated signal to the sending device, which sends it to the receiving device through a multipath channel;

[0070] The ODDM modulation process includes: first, mapping the pilot signal to an M×N dimensional DD domain data matrix, where M and N are the number of time slots and subcarriers, respectively, and defining X[m,n] as the transmitted pilot signal, where m∈{0,1,,M-1}, n∈{0,1,,N-1} represent the index of the delay and the index of the Doppler, respectively, so there are MN data information to be modulated. Each multi-carrier symbol contains N subcarriers and lasts for T. In the DD domain, the spacing between each carrier is set to 1 / NT, and the multi-carrier symbols are evenly distributed at intervals of T / M. The ODDM modulation process is shown below

[0071]

[0072] where g tx (t) is the filter at the transmitting end, which is a pulse train and can be expressed as In this expression, a(t) is a real-valued square root Nyquist pulse with time symmetry, its duration is 2QT / M, and it satisfies and satisfies the condition Therefore, we have When the integer 2Q <<M, it can be proved that g tx (t) is orthogonal in terms of delay and Doppler resolution, i.e., the ambiguity function satisfy

[0073]

[0074] in Yes tx The conjugate function of (t).

[0075] Multipath fading channel modeling: In the time-varying channel scenario, define W s and T s are the sampling rate and duration of the received signal, respectively. Therefore, the delay and Doppler resolution are 1 / W respectively. s and 1 / T s , the equivalent delay-Doppler channel can be written as

[0076]

[0077] Where P is the number of paths in the channel, h p , and are the channel gain, delay and Doppler of the pth path respectively, where k p ,l p ∈Z.

[0078] S3, the receiving device performs matched filtering on the received data to obtain a received signal, and calculates an equivalent input-output relationship of the pilot signal;

[0079] Specifically, after receiving the data, the ODDM demodulation process includes: when the modulated signal x(t) passes through the time-varying multipath channel, the signal received by the receiving end is

[0080]

[0081] where n c (t)∈C is noise, following Gaussian distribution and l p =v p NT are the indices of delay and Doppler respectively. Assuming the channel gain h p Also obeys Gaussian distribution. In ODDM demodulation, the received waveform g rx The cross ambiguity function between y(t) and the received signal y(t) It is calculated by matching filter, as follows

[0082]

[0083] Then the discrete form of the received signal can be obtained

[0084]

[0085] in

[0086]

[0087] Where n′=[nl p ] N .

[0088] Furthermore, considering that the maximum delay index and Doppler index are L and K respectively, then k p ∈[0,L-1] and l p ∈[-K,K]. For a massive MIMO system, assume that the transmitter has N t uniform linear array antennas, the corresponding received signal is

[0089]

[0090] in By the nth t The symbol sent by each antenna is where d BS ,λ c , are the antenna spacing, carrier frequency wavelength and multipath channel angle respectively, is the corresponding Delay Doppler Space (DDS) channel, and

[0091]

[0092] Where n" = [nl p ] N .

[0093] Specifically, the derivation of the equivalent input-output relationship is as follows: As shown above, the maximum delay and Doppler of the channel are and We organize the P paths into a matrix G of dimension (2K+1)×L, where the non-zero elements g(k+K+1,l) represent the channel gain of the corresponding path, i.e. h p =g(k+K+1,l), the delay and Doppler of this path are and Since the mth received ODDM symbol in ODDM modulation is delayed by The path is interfered by the (ml)th symbol. In this case, the resulting Doppler shift is Will interfere with the (ml)th symbol in Periodic shifts begin to occur, and the Doppler effect causes Phase rotation. Therefore, the ISI from the (ml)th ODDM symbol can be expressed as

[0094]

[0095] Where C is the circulant transposed matrix. In addition, when ml<0, the phase rotation term Used for x [m-l]M .

[0096] Based on the above analysis, we have the following vector form

[0097]

[0098] where y = vec(Y)∈C MN×1 , Expressed as

[0099]

[0100] in and It is nth t The i-th row of symbols transmitted by the root antenna.

[0101] make The above formula can be transformed into the following equivalent channel model

[0102] y=Φh+n c

[0103] where y∈C MN×1 , is the equivalent coefficient matrix, is the corresponding equivalent channel vector. The channel information h contains the delay, Doppler, gain and angle information of the multipath channel. Since there are only P non-zero elements in the channel vector, h is a sparse vector. Therefore, in the massive MIMO system based on ODDM modulation, the channel estimation problem can be regarded as a sparse recovery problem. Next, we will analyze the characteristics of the equivalent coefficient matrix Φ and the equivalent channel vector h.

[0104] In the equivalent coefficient matrix Φ, each (2K+1) column is a repeated element, so there is a certain correlation between the elements in the equivalent coefficient matrix, rather than being completely random. However, when the number of antennas tends to infinity and satisfies N t >>K, these elements approach completely random. In the channel vector h, its non-zero elements correspond to the product of the channel gain and the elements of the direction vector, where the channel gain follows a Bernoulli-Gaussian distribution. However, since the elements of the channel vector vary between different antennas, the elements of the channel vector h do not follow a Bernoulli-Gaussian distribution. Fortunately, the amplitude of the direction vector is 1, making the amplitude of the elements in the channel vector equal to the channel gain, which is distributed according to a Bernoulli-Gaussian distribution.

[0105] S4. According to the equivalent input-output relationship of the pilot signal, the memory approximate message passing algorithm is used to estimate the delay, Doppler and channel gain of the multipath channel, and the discrete Fourier transform with added phase rotation is used to obtain the angle of the channel to obtain the estimated value of the multipath channel.

[0106] Based on the above characteristics, considering that the traditional sparse signal recovery algorithm will face relatively high computational complexity. Therefore, it is not feasible in high-speed mobile scenarios. Based on this, the present invention provides a two-stage low-complexity channel estimation algorithm, which uses a memory approximate message passing algorithm to estimate the delay, Doppler and channel gain of the multipath channel, and uses the discrete Fourier transform method to estimate the angle of the multipath channel. The algorithm steps of the present invention are described in detail below:

[0107] Estimation of channel delay, Doppler and gain: The present invention adopts a memory approximate message passing algorithm to estimate channel delay, Doppler and gain, wherein the memory approximate message passing algorithm includes two estimators, a long memory linear estimator and a nonlinear estimator. The long memory linear estimator can obtain a channel estimation value based on the minimum mean square error criterion and can eliminate the influence of multipath interference and Doppler shift, while the nonlinear estimator can reduce Gaussian noise and thus improve the accuracy of the estimation. Based on the minimum mean square error estimation criterion, the a posteriori estimate of h is expressed as

[0108]

[0109] in is an orthogonal factor. In order to obtain channel information, it is necessary to perform a matrix inversion operation. Therefore, in order to significantly reduce the computational complexity, the above formula needs to be simplified to

[0110]

[0111] where κ t is selected to speed up the convergence of the memory-approximate message passing detector, θ t As a relaxation parameter, ensure The spectral radius of θ is kept below 1. t This can be determined by

[0112]

[0113] in λ min and λ max They are ΦΦ H The minimum and maximum eigenvalues ​​of min and λ max When it is difficult to obtain, the corresponding lower and upper bounds can be used instead.

[0114] Define B as So we have θ t B=I-θ t (ρ t I+ΦΦ H ), so we can get θ t The B spectrum radius is less than 1, so the convergence condition is satisfied. Furthermore, the minimum mean square error can be approximated as

[0115]

[0116] in R t,k =-η t,k Φ H B t-k Φ H ,and

[0117]

[0118] As can be seen above, channel estimation requires information from all previous moments, so the current output estimation error must be orthogonal to all previous input estimates to ensure strict orthogonality conditions. t After that, we need to optimize to minimize the minimum mean square error of the long memory linear estimator Before optimizing, define Can get

[0119]

[0120] in is the covariance of the nonlinear estimator,

[0121] Then, by minimizing Get κ t Optimal parameters When t>2, we have

[0122]

[0123] About Nonlinear Estimators: Nonlinear Estimator Gamma t It is necessary to reduce Gaussian noise and improve the quality of h estimation. The output of the nonlinear estimator includes the maximum a posteriori mean and variance. Since the elements in h follow the Bernoulli-Gaussian distribution, that is, h i = bg, where b follows a Bernoulli distribution with probability q and g follows a mean distribution of u g Variance is v g Gaussian distribution, and bg are independent of each other, the maximum a posteriori mean and variance can be derived as

[0124]

[0125] in

[0126] In obtaining h i After the maximum a posteriori mean and variance, the present invention needs to ensure the orthogonality between the input error and the output error, and then update the external mean and variance to

[0127]

[0128] Finally, in order to ensure the convergence of the memory approximate message passing estimator and improve the efficiency, the damping vector

[0129] satisfy So the nonlinear estimator can be rewritten as

[0130]

[0131] where h (t) It represents the mean value of the channel gain estimation at time t. Definition is a set of effectively memorized indexes, so is the corresponding complement. In order to avoid the increase of invalid memory during the damping process, set Next, define h i , The optimal damping vector can be derived as

[0132]

[0133] Next, update according to the following formula

[0134]

[0135] Angle estimation: Since there are many antennas at the transmitting end, the present invention applies discrete Fourier transform to the received signal to estimate the angle, thereby studying the sparsity of the angle domain. In order to avoid angle aliasing, the condition N needs to be met. t d / λ+1≤N t -N t d / λ+1, using discrete Fourier transform for the received signal, we can get

[0136]

[0137] in is the normalized discrete Fourier transform matrix, where the elements are is the corresponding direction vector.

[0138] After discrete Fourier transform, the received signal becomes a progressively sparse vector with P non-zero elements, and P angles can be estimated through the non-zero elements. t is finite, so the estimated angle is not necessarily an integer, which will cause power leakage and lead to inaccurate angle estimation. In order to improve the accuracy of angle estimation, an angle rotation scheme is adopted, where the angle rotation matrix is

[0139]

[0140] in represents the phase rotation parameters. These parameters are used to adjust the angle in the transmit steering vector to ensure minimum power leakage during angle estimation. The optimal phase rotation parameters can be obtained by solving the following problem in one-dimensional search

[0141]

[0142] in Get the optimal phase rotation parameters, and the angle can be obtained as follows

[0143]

[0144] in

[0145] The functions and effects of the present invention are further illustrated by the following simulation experiments:

[0146] Simulation conditions: The carrier frequency is set to 5 GHz, the subcarrier spacing is 15 kHz, the time length M is set to 512, the number of subcarriers N is set to 32, the modulation method used is 4-QAM, and the user's mobile speed is 250 km / h. In ODDM modulation, a(t) is set to square root raised cosine pulse, and Q is set to 20. The channel is characterized by the extended vehicle model in the 3GPP standard. In the simulation comparison, the normalized mean square error (NMSE) is used to show the estimation performance, specifically in is the estimated channel information. In simulation, the present invention is compared with the existing sparse signal recovery algorithms Orthogonal Matching Pursuit (OMP) and 3D Orthogonal Matching Pursuit (3D-OMP) to highlight the performance of the algorithm proposed by the present invention.

[0147] Simulation results: Figure 3 In this paper, we show the NMSE performance comparison of various algorithms under different signal-to-noise ratios (SNRs). It can be clearly seen that the NMSE performance of all algorithms shows an upward trend under higher SNRs, and the reason is also clear that high SNR can improve the performance of signal detection. In addition, the estimation algorithm of the present invention outperforms the traditional OMP and 3D-OMP schemes. This superior performance is attributed to the ability of the proposed algorithm to achieve Bayesian optimal results, indicating its potential in practical applications.

[0148] Figure 4The NMSE performance at different user speeds is demonstrated. It can be clearly seen that the NMSE performance of the three algorithms remains basically stable at different user speeds, reflecting the robustness of the proposed ODDM modulation method. In addition, the channel estimation algorithm of the present invention is significantly better than the traditional OMP and 3D-OMP algorithms in NMSE performance, highlighting its effectiveness in different user speed scenarios. This highlights the potential of the algorithm of the present invention to achieve reliable and robust channel estimation in practical applications.

[0149] Figure 5 The NMSE performance of the algorithm proposed in the present invention and the traditional memory approximate message passing algorithm under different numbers of antennas are compared, among which the traditional memory approximate message passing algorithm can achieve the Bayesian optimal result (represented by random in the figure). In the traditional memory approximate message passing algorithm scheme, the elements in the effective channel matrix are completely randomly generated. From the results, as the number of antennas increases, the NMSE performance of the algorithm proposed in the present invention gradually converges to the performance of the traditional memory approximate message passing algorithm scheme. This convergence trend means that as the number of antennas tends to infinity, the effective channel matrix can be approximated to be completely random. Therefore, the proposed algorithm shows robustness and efficiency in scenarios with different numbers of antennas, and ultimately closely matches the performance characteristics of the traditional memory approximate message passing algorithm scheme.

[0150] Finally, we Figure 6 The comparison of bit error rate (BER) at different SNR and speed is shown in . In this scenario, we consider the case of channel state information (CSI) obtained by the proposed algorithm and perfect CSI, and use the orthogonal approximate message passing (OAMP) method for symbol detection, which can also achieve Bayesian optimality. In addition, we also compare the BER performance of ODDM and OTFS modulation. Figure 6 In this paper, we consider the BER performance at different speeds. It can be seen that due to the rapid change of channel state information (CSI), the BER of ODDM modulation and OTFS modulation increases with the increase of speed. Compared with OTFS modulation, ODDM can achieve better BER performance and robustness at different speeds. In addition, perfect CSI can achieve better BER performance, and the gap between the estimated CSI and the perfect CSI is very small, which verifies the superior performance of the algorithm proposed in this invention.

[0151] Corresponding to the above-mentioned embodiment of the MIMO system channel estimation method based on ODDM modulation, the present invention also provides an embodiment of the MIMO system channel estimation device based on ODDM modulation.

[0152] See also Figure 7The MIMO system channel estimation device based on ODDM modulation provided in an embodiment of the present invention includes a memory and one or more processors, wherein the memory stores executable code, and when the processor executes the executable code, it is used to implement the MIMO system channel estimation method based on ODDM modulation in the above embodiment.

[0153] The embodiment of the MIMO system channel estimation device based on ODDM modulation provided by the present invention can be applied to any device with data processing capabilities, and the device with data processing capabilities can be a device or apparatus such as a computer. The device embodiment can be implemented through software, or through hardware or a combination of software and hardware. Taking software implementation as an example, as a device in a logical sense, it is formed by the processor of any device with data processing capabilities in which it is located reading the corresponding computer program instructions in the non-volatile memory into the internal memory for execution. From the hardware level, if Figure 7 As shown in FIG. 1 , a hardware structure diagram of any device with data processing capability in which the MIMO system channel estimation device based on ODDM modulation provided by the present invention is located, except Figure 7 In addition to the processor, memory, network interface, and non-volatile memory shown, any device with data processing capabilities in which the apparatus in the embodiments is located may also include other hardware, generally based on the actual functions of the device with data processing capabilities, which will not be described in detail.

[0154] The implementation process of the functions and effects of each unit in the above-mentioned device is specifically described in the implementation process of the corresponding steps in the above-mentioned method, and will not be repeated here.

[0155] For the device embodiment, since it basically corresponds to the method embodiment, the relevant parts can refer to the partial description of the method embodiment. The device embodiment described above is only schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of the present invention. Ordinary technicians in this field can understand and implement it without paying creative work.

[0156] An embodiment of the present invention further provides a computer-readable storage medium on which a program is stored. When the program is executed by a processor, the MIMO system channel estimation method based on ODDM modulation in the above embodiment is implemented.

[0157] The computer-readable storage medium may be an internal storage unit of any device with data processing capability described in any of the aforementioned embodiments, such as a hard disk or a memory. The computer-readable storage medium may also be an external storage device of any device with data processing capability, such as a plug-in hard disk, a smart media card (SMC), an SD card, a flash card, etc. equipped on the device. Furthermore, the computer-readable storage medium may also include both an internal storage unit and an external storage device of any device with data processing capability. The computer-readable storage medium is used to store the computer program and other programs and data required by any device with data processing capability, and may also be used to temporarily store data that has been output or is to be output.

[0158] The present invention also provides a computer program product, comprising a computer program, and when the computer program is executed by a processor, the MIMO system channel estimation method based on ODDM modulation is implemented.

[0159] Those skilled in the art will readily appreciate other embodiments of the present application after considering the description and practicing the contents disclosed herein. The present application is intended to cover any modification, use or adaptation of the present application, which follows the general principles of the present application and includes common knowledge or customary techniques in the art that are not disclosed in the present application. The description and examples are intended to be exemplary only, and the true scope and spirit of the present application are indicated by the claims.

[0160] It should be understood that the above general description and the detailed description below are only exemplary and explanatory and cannot limit the present application. The present application is not limited to the precise structure described above and shown in the drawings, and various modifications and changes can be made without departing from the scope thereof. The scope of the present application is limited only by the attached claims.

Claims

1. A MIMO system channel estimation method based on ODDM modulation, characterized in that: The method includes: S1. Generate a pilot signal and perform ODDM modulation on the pilot signal; S2, sending the modulated signal to the sending device, which sends it to the receiving device through a multipath channel; S3, the receiving device performs matched filtering on the received data to obtain a received signal, and calculates an equivalent input-output relationship of the pilot signal; S4, using a memory approximate message passing algorithm to estimate the delay, Doppler and channel gain of the multipath channel according to the equivalent input-output relationship of the pilot signal, obtaining the angle rotation matrix of the sending steering vector through a one-dimensional search solution, using discrete Fourier transform to obtain the angle of the channel, and obtaining the estimated value of the multipath channel; The memory approximate message passing algorithm includes a long memory linear estimator and a nonlinear estimator, The long memory linear estimator is used to obtain a channel estimation value based on the minimum mean square error criterion and eliminate the influence of multipath interference and Doppler shift, while the nonlinear estimator reduces Gaussian noise on the estimation value to improve the accuracy of the estimation; The long memory linear estimator is specifically: where κ t To speed up the convergence of the memory approximate message passing detector, θ t As a relaxation parameter, ensure The spectral radius is kept below 1; by minimizing the minimum mean square error of the long memory linear estimator Get the optimal κ t ; The nonlinear estimator comprises: Among them, (t) represents the estimated mean value of the channel gain at time t, in, Represented in κ t Selecting the optimal parameters When , the minimum mean square error of the long memory linear estimator; the channel vector Where b follows a Bernoulli distribution with probability q, and g follows a mean distribution with u g Variance is v g Gaussian distribution, and bg are independent of each other; The maximum a posteriori mean and variance are in is the damping vector, satisfying The optimal damping vector is for The covariance of A collection of indexes for effective memory.

2. The MIMO system channel estimation method based on ODDM modulation according to claim 1, characterized in that: The generating of the pilot signal specifically includes: mapping the generated random signal into a delay-Doppler domain matrix, and then performing inverse Fourier transform, up-sampling, and pulse train beamforming to obtain a modulated pilot signal.

3. The MIMO system channel estimation method based on ODDM modulation according to claim 1, characterized in that: The ODDM modulation of the pilot signal is specifically expressed as follows: Where X[m,n] is the transmitted pilot signal, m∈{0,1,…,M-1}, n∈{0,1,…,N-1} represents the delay index and Doppler index respectively, M is the number of time slots, each multi-carrier symbol contains N sub-carriers, the duration is T, the spacing between each carrier is 1 / NT, and the multi-carrier symbols are evenly distributed at intervals of T / M. tx (t) is the filter at the transmitting end, which is a pulse train and is expressed as a(t) is a real-valued square root Nyquist pulse with time symmetry and duration 2QT / M, and satisfies the condition When the integer 2Q <<M g tx (t) is orthogonal in terms of delay and Doppler resolution, and the ambiguity function satisfy Yes tx The conjugate function of (t).

4. The MIMO system channel estimation method based on ODDM modulation according to claim 1, characterized in that: The matched filtering specifically includes: calculating the cross ambiguity function between the received waveform and the received signal through the matched filter in, h p is the channel gain, g tx (t) is the filter at the transmitting end, Yes tx The conjugate function of (t), n c (t)∈C is noise, following Gaussian distribution and l p =v p NT are the indexes of delay and Doppler, and are the delay and Doppler of the pth path, respectively, where k p ,l p ∈ Z; Get the discrete form of the received signal Where n′=[nl p ] N , X[m,n] is the transmitted pilot signal.

5. The MIMO system channel estimation method based on ODDM modulation according to claim 1, characterized in that: The calculation of the equivalent input-output relationship of the pilot signal includes: in is the equivalent coefficient matrix, is the corresponding equivalent channel vector; n c For noise, in Expressed as in z and It is nth t The i-th row of symbols transmitted by the root antenna; G is a matrix of dimension (2K+1)×L obtained by arranging the P paths of the delay-Doppler spatial channel. The largest delay index and Doppler index are L and K respectively. The non-zero element g(k+K+1,l) represents the channel gain of the corresponding path. The delay and Doppler of the path are and C is the circulant transposed matrix and D is the phase rotation term.

6. The MIMO system channel estimation method based on ODDM modulation according to claim 1, characterized in that: The channel angle obtained by discrete Fourier transform is: To avoid angle aliasing and satisfy the condition Nd / λ+1≤N-Nd / λ+1, the received signal is subjected to discrete Fourier transform to obtain Among them U Nt is the normalized discrete Fourier transform matrix, where the elements are is the corresponding direction vector, where d BS ,λ c , are the antenna spacing, carrier frequency wavelength and multipath channel angle, respectively, t is the number of antennas.

7. A channel estimation system for implementing the method according to any one of claims 1 to 6, characterized in that: The system includes: Encoder: used to preset the pilot signal, perform ODDM modulation on the pilot signal, and then send the modulated pilot signal to the sending module; Sending module: used to send the pilot signal modulated by the encoder to the receiving module; Receiving module: used to obtain a received signal after matching filtering the received pilot signal; Channel estimation module: According to the equivalent input-output relationship of the pilot signal, the memory approximate message passing algorithm is used to estimate the delay, Doppler and channel gain of the multipath channel, and the discrete Fourier transform with added phase rotation is used to obtain the channel angle and the estimated value of the multipath channel.

8. A MIMO system channel estimation device based on ODDM modulation, comprising a memory and one or more processors, wherein the memory stores executable code, characterized in that: When the processor executes the executable code, the MIMO system channel estimation method based on ODDM modulation as described in any one of claims 1 to 6 is implemented.

9. A computer-readable storage medium having a program stored thereon, characterized in that: When the program is executed by a processor, the MIMO system channel estimation method based on ODDM modulation as described in any one of claims 1 to 6 is implemented.

Images