A signal detection method and system for a multiple-input multiple-output antenna system

Abstract

The application discloses a kind of signal detection method and system of multiple input-output antenna system, it is related to signal detection technical field, including the establishment of target antenna system's received signal model, consider some nonlinear factors, superimposed to linear combination model, can better explain the real situation of target antenna system's received signal and description.Analysis channel uncertainty, interference and noise three aspects content between each other coupling, consider the channel uncertainty, interference and noise three aspects between each other coupling of target antenna system involved, establish multi-angle coupling model, improve the adaptability of signal processing.Through the first suppression algorithm, channel uncertainty is suppressed, and the second suppression algorithm is used to suppress interference and noise, and signal processing is carried out in a targeted manner.According to the signal quality, the first suppression algorithm and the second suppression algorithm are optimized, the calculation complexity and the energy consumption are balanced, and the real-time demand of signal dynamic change is met.

Description

Technical Field

[0001] This invention relates to the field of signal detection technology, and in particular to a signal detection method and system for a multi-input output antenna system. Background Technology

[0002] In signal detection schemes for multiple-input multiple-output (MIMO) antenna systems, traditional methods often treat channel uncertainty, interference, and noise as independent factors. While this approach may perform well in simple channel environments, its performance deteriorates significantly in complex electromagnetic scenarios because it ignores the coupling relationships between these three factors. Specifically, channel uncertainties (such as multipath effects and time-varying characteristics) affect the statistical properties of interference and noise, while interference (such as co-channel interference and adjacent-channel interference) and noise (such as thermal noise and phase noise) can, in turn, exacerbate channel estimation errors. Because traditional signal detection schemes do not establish such a multi-angle coupling model, they struggle to jointly suppress channel uncertainty, interference, and noise, leading to a deterioration in key performance indicators such as bit error rate (BER) and signal-to-noise ratio (SNR) in complex channel environments.

[0003] In existing technologies, the coupling between channel uncertainty, interference, and noise is not taken into account, resulting in poor accuracy and adaptability of signal detection in antenna systems, which cannot meet complex signal transmission requirements.

[0004] Therefore, improving the accuracy and adaptability of signal detection in antenna systems is a technical problem that needs to be solved. Summary of the Invention

[0005] The purpose of this invention is to address the problem of poor accuracy and adaptability of signal detection in existing antenna systems due to the lack of consideration for the coupling between channel uncertainty, interference, and noise. Therefore, this invention proposes a signal detection method for a multi-input / output antenna system, comprising:

[0006] Collect basic information about the multi-input output target antenna system and establish a received signal model of the target antenna system;

[0007] Based on the received signal model of the target antenna system, the channel uncertainty, interference and noise are analyzed independently, and the coupling relationship between the three aspects is also analyzed.

[0008] A multi-angle coupling model of the target antenna system is established based on the coupling relationship between channel uncertainty, interference, and noise.

[0009] The system receives real-time signals from the target antenna system. Based on the multi-angle coupling model of the target antenna system, it uses a first suppression algorithm to suppress channel uncertainty and a second suppression algorithm to suppress interference and noise.

[0010] The signal is detected, the signal quality is evaluated, and the first and second suppression algorithms are optimized based on the signal quality.

[0011] In some embodiments of this application, a received signal model of the target antenna system is established, including...

[0012] An initial received signal model under linear combination is established based on the channel matrix, interference matrix, and noise matrix. The channel matrix reflects the influence of the wireless channel on the transmitted signal, the interference matrix reflects the interference signals from other wireless sources or systems, and the noise matrix reflects the thermal noise and other random noise during the signal reception process.

[0013] The hardware nonlinearity, channel nonlinearity, interference nonlinearity, and noise nonlinearity of the target antenna system are identified. The performance and calculation indicators under the hardware nonlinearity, channel nonlinearity, interference nonlinearity, and noise nonlinearity are obtained through simulation software. Based on the performance and calculation indicators, the key and non-key items among the hardware nonlinearity, channel nonlinearity, interference nonlinearity, and noise nonlinearity are selected.

[0014] Key terms are described using a nonlinear model, while non-key terms are described using a linear approximation model.

[0015] The model descriptions of key and non-key terms are superimposed on the initial received signal model under linear combination to form the received signal model of the target antenna system.

[0016] In some embodiments of this application, channel uncertainty, interference, and noise are analyzed independently, including:

[0017] Channel uncertainty includes channel estimation error, channel time-varying properties, spatial correlation, and nonlinear terms, and is quantified by the channel estimation error covariance matrix.

[0018] The interference information includes the source of interference, type of interference, power of interference, and spatial characteristics of interference, and the interference situation is quantified by the interference covariance matrix.

[0019] The noise content includes noise type, noise power, and noise correlation, and the noise situation is quantified by noise power spectral density and autocorrelation function.

[0020] In some embodiments of this application, the coupling relationship between channel uncertainty, interference, and noise is analyzed, including...

[0021] The channel estimation error covariance matrix, interference covariance matrix, noise power spectral density, and autocorrelation function are combined to quantify the coupling relationship between channel uncertainty, interference, and noise.

[0022] In some embodiments of this application, a multi-angle coupling model of the target antenna system is established based on the coupling relationship between channel uncertainty, interference, and noise, including:

[0023] A mathematical model is established based on the coupling relationship between the three aspects of quantized channel uncertainty, interference, and noise, and the input layer, coupling layer, and output layer of the multi-angle coupling model are designed.

[0024] The objective function of the multi-angle coupling model is defined, and the parameters in the objective function are optimized using channel data, interference data, and noise data, thereby establishing the multi-angle coupling model of the target antenna system.

[0025] In some embodiments of this application, after receiving the real-time signal from the target antenna system, the method further includes,

[0026] The channel uncertainty, interference, and noise of the real-time signal are extracted. These conditions are then input into the multi-angle coupling model of the target antenna system, and the target values ​​for suppressing the channel uncertainty, interference, and noise are output.

[0027] In some embodiments of this application, channel uncertainty suppression is performed using a first suppression algorithm, and interference and noise suppression is performed using a second suppression algorithm, including:

[0028] The first suppression algorithm includes robust channel estimation and deep learning predistortion; the second suppression algorithm includes MVDR beamforming, wavelet transform denoising, MMSE detection and nonlinear denoising.

[0029] Channel uncertainty suppression is performed using robust channel estimation and deep learning predistortion, based on the target value for channel uncertainty suppression.

[0030] Interference suppression is achieved using MVDR beamforming and wavelet transform denoising based on the target interference suppression value.

[0031] Noise suppression is achieved using MMSE detection and nonlinear noise reduction based on the noise suppression target value.

[0032] In some embodiments of this application, signal quality is evaluated, and a first suppression algorithm and a second suppression algorithm are optimized based on the signal quality, including...

[0033] Multiple evaluation parameters are generated to evaluate signal quality, and these evaluation parameters are divided into two categories: single evaluation parameters and multiple evaluation parameters.

[0034] The suppression strength of the first and second suppression algorithms is adjusted based on two categories: single evaluation parameters and multiple evaluation parameters, thereby optimizing the first and second suppression algorithms.

[0035] Correspondingly, this application also provides a signal detection system for a multiple input / output antenna system, including,

[0036] The first module is used to collect basic information about the multi-input output target antenna system and establish a received signal model of the target antenna system.

[0037] The second module is used to independently analyze the channel uncertainty, interference, and noise based on the received signal model of the target antenna system, and to analyze the coupling relationship between the three aspects of channel uncertainty, interference, and noise.

[0038] The third module is used to establish a multi-angle coupling model of the target antenna system based on the coupling relationship between the three aspects of channel uncertainty, interference and noise.

[0039] The fourth module is used to receive real-time signals from the target antenna system. Based on the multi-angle coupling model of the target antenna system, it uses the first suppression algorithm to suppress channel uncertainty and the second suppression algorithm to suppress interference and noise.

[0040] The fifth module is used to detect the signal, evaluate the signal quality, and optimize the first and second suppression algorithms based on the signal quality.

[0041] The present invention has the following beneficial effects:

[0042] 1. A received signal model of the target antenna system is established, taking into account some nonlinear factors, which are superimposed on a linear combination model, thus better illustrating and describing the actual received signal of the target antenna system. The coupling relationships between channel uncertainty, interference, and noise are analyzed. Considering the coupling of these three aspects involved in the target antenna system, a multi-angle coupling model is established, improving adaptability to signal processing and demonstrating strong robustness in complex environments.

[0043] 2. Channel uncertainty is suppressed using a first suppression algorithm, and interference and noise are suppressed using a second suppression algorithm. Targeted signal processing ensures signal robustness. The first and second suppression algorithms are optimized based on signal quality, balancing computational complexity and energy consumption, and meeting the real-time requirements of dynamically changing signals. Attached Figure Description

[0044] Figure 1This is a schematic flowchart of a signal detection method for a multi-input output antenna system proposed in this invention.

[0045] Figure 2 This is a schematic diagram of the structure of a signal detection system for a multi-input output antenna system proposed in this invention. Detailed Implementation

[0046] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0047] Reference Figure 1 A signal detection method for a multi-input output antenna system includes the following steps:

[0048] Step S101: Collect basic information about the multi-input output target antenna system and establish a received signal model of the target antenna system.

[0049] In this embodiment, the Multiple-Input Multiple-Output (MIMO) antenna system is a core architecture that utilizes multi-antenna technology to improve the performance of wireless communication systems. By deploying multiple antennas at both the transmitter and receiver, it leverages spatial resources to achieve efficient signal transmission and reception, significantly improving the capacity, reliability, and spectral efficiency of the communication system. Spatial multiplexing: At the transmitter, the MIMO system divides the high-speed data stream into multiple low-speed substreams, which are transmitted simultaneously through different antennas. At the receiver, the differences in signals received by multiple antennas (such as phase and amplitude) are used to separate and reconstruct the original data. Advantages: Under ideal channel conditions, the channel capacity of a MIMO system is proportional to the number of antennas (e.g., the theoretical capacity of 4×4 MIMO is four times that of SISO), significantly improving spectral efficiency. Diversity gain: By receiving different copies of the same signal through multiple antennas (spatial diversity), the MIMO system can combat channel fading. For example, in Rayleigh fading channels, the receiver can significantly reduce the bit error rate by using maximum ratio combining (MRC). Advantages: Diversity gain enables the system to maintain reliable communication even in low signal-to-noise ratio (SNR) environments, extending communication distance. Beamforming, principle: MIMO systems adjust the phase and amplitude of each antenna to form a narrow beam directed towards a specific user, concentrating signal energy. For example, in the millimeter-wave band, beamforming can compensate for path loss, achieving directional transmission. Advantages: Beamforming improves the quality of received signals while reducing interference to other users.

[0050] In this embodiment, the basic information includes antenna configuration, channel characteristics, hardware characteristics, signal type, and modulation method. Specifically:

[0051] Antenna configuration

[0052] Antenna quantity and arrangement: MIMO systems utilize spatial dimensional resources to improve communication performance by deploying multiple antennas (such as 4×4, 8×8, 64×8, etc.) at the transmitter and receiver. The arrangement of the antennas (such as uniform linear array ULA, uniform circular array UCA, or three-dimensional array) also affects the spatial characteristics of the signal.

[0053] Antenna spacing: Antenna spacing affects channel correlation. Larger antenna spacing (typically greater than half a wavelength) can reduce channel correlation, thereby increasing diversity gain and spatial multiplexing gain.

[0054] Channel characteristics

[0055] Multipath effect: Wireless signals travel through multiple paths to reach the receiver during propagation, resulting in multipath effect. This leads to signal delay spread and frequency-selective fading.

[0056] Time-varying nature: The time-varying nature of the channel is caused by the relative motion between the transmitter and receiver, resulting in Doppler frequency shift. This is particularly noticeable in high-speed mobile scenarios (such as vehicle-mounted communications).

[0057] Spatial correlation: Due to differences in antenna spacing and scattering environment, the channels between different antenna pairs may exhibit correlation. Spatial correlation can affect the performance and capacity of MIMO systems.

[0058] Hardware features

[0059] Power amplifier (PA) nonlinearity: Power amplifiers produce nonlinear distortion when amplifying signals, which affects the amplitude and phase of the signal, thereby reducing the quality of the received signal.

[0060] RF front-end characteristics include the frequency response of filters and the noise figure of low-noise amplifiers (LNAs). These characteristics affect the signal-to-noise ratio (SNR) and dynamic range of the received signal.

[0061] Signal type and modulation method

[0062] Signal types include Orthogonal Frequency Division Multiplexing (OFDM) and Single-Carrier Frequency Division Multiple Access (SC-FDMA). Different signal types have different spectral characteristics and interference immunity.

[0063] Modulation methods: such as QPSK, 16QAM, 64QAM, etc. The choice of modulation method will affect the signal transmission rate and bit error rate performance.

[0064] In some embodiments of this application, a received signal model of the target antenna system is established, including...

[0065] An initial received signal model under linear combination is established based on the channel matrix, interference matrix, and noise matrix. The channel matrix reflects the influence of the wireless channel on the transmitted signal, the interference matrix reflects the interference signals from other wireless sources or systems, and the noise matrix reflects the thermal noise and other random noise during the signal reception process.

[0066] The hardware nonlinearity, channel nonlinearity, interference nonlinearity, and noise nonlinearity of the target antenna system are identified. The performance and calculation indicators under the hardware nonlinearity, channel nonlinearity, interference nonlinearity, and noise nonlinearity are obtained through simulation software. Based on the performance and calculation indicators, the key and non-key items among the hardware nonlinearity, channel nonlinearity, interference nonlinearity, and noise nonlinearity are selected.

[0067] Key terms are described using a nonlinear model, while non-key terms are described using a linear approximation model.

[0068] The model descriptions of key and non-key terms are superimposed on the initial received signal model under linear combination to form the received signal model of the target antenna system.

[0069] In this embodiment, the initial received signal model under linear combination is:

[0070] ;

[0071] in, In order to receive signals, For the channel matrix, In order to send a signal, For the interference matrix, This is the noise matrix.

[0072] Simulation tools: Use Keysight SystemVue or MATLAB Communications Toolbox.

[0073] Hardware nonlinearity, channel nonlinearity, interference nonlinearity, and noise nonlinearity include, for example, the following:

[0074] Hardware nonlinearity: AM-AM / AM-PM distortion of the power amplifier (PA) (using the Rapp model).

[0075] Channel nonlinearity: Oxygen absorption loss in millimeter-wave channels (attenuation of 15 dB / km at frequencies > 60 GHz).

[0076] Interference nonlinearity: Phase noise of the interference signal (Rice distribution, k=10 dB).

[0077] Noise nonlinearity: 1 / f noise component of thermal noise (power spectral density S(f)∝1 / f).

[0078] Performance metrics include bit error rate, capacity, and latency. The impact of a module (non-linear term) on core system performance metrics (such as bit error rate, capacity, and latency) is considered. For example, performance analysis involves quantifying the impact of module parameter changes on performance through simulation or experimentation (e.g., the proportion of capacity reduction caused by channel estimation error). Threshold setting: If a 10% decrease in module performance leads to a system capacity reduction exceeding 5%, it is classified as a critical module.

[0079] The computational resources required for implementing the module (non-linear term) are determined (e.g., number of floating-point operations, memory usage, etc.). Two metrics are used to filter out critical and non-critical components. Complexity quantification: Module complexity is evaluated using FLOPs (floating-point operations) or MACs (multiply-accumulate operations). Threshold setting: If the module complexity exceeds 30% of the total system budget, it needs to be optimized to a linear approximation. A mixed-precision model design is adopted: high-precision non-linear models are used in critical modules (such as channel estimation), while linear approximations are used in less critical modules.

[0080] Superposition of nonlinear functions:

[0081] For hardware nonlinear terms, such as the nonlinear distortion of a power amplifier, nonlinear functions (such as the Rapp model, memory polynomial model, MPM, etc.) can be used to describe their impact on the transmitted signal.

[0082] These nonlinear functions can be embedded into the initial received signal model, for example, by performing nonlinear predistortion processing on the signal at the transmitting end, or by performing nonlinear compensation on the received signal at the receiving end.

[0083] Nonlinear channel model superposition:

[0084] For channel nonlinearities, such as oxygen loss in millimeter-wave channels, nonlinear channel models (such as the ITU-R P.676 model) can be used to describe their impact on signal propagation.

[0085] These nonlinear channel models can be combined with the channel matrix H in the initial received signal model to more accurately describe the channel characteristics.

[0086] Nonlinear description of interference and noise:

[0087] Interference and noise may also have nonlinear characteristics, such as phase noise and non-Gaussian noise.

[0088] These nonlinear characteristics can be described by introducing nonlinear interference and noise terms into the initial received signal model.

[0089] Step S102: Based on the received signal model of the target antenna system, analyze the three aspects of channel uncertainty, interference and noise independently, and analyze the coupling relationship between the three aspects of channel uncertainty, interference and noise.

[0090] In this embodiment, channel uncertainty refers to the uncertainty caused by various factors (such as multipath effect, time-varying nature, spatial correlation, etc.) during the propagation of a wireless channel. A wireless channel is the path a signal takes from the transmitter to the receiver, and its characteristics are affected by various factors such as environment, movement speed, and frequency. Interference refers to signals from other wireless sources or systems that interfere with the received signal of the target antenna system. Interference can be co-channel interference (same frequency as the target signal), adjacent-channel interference (similar frequency to the target signal), or co-channel interference (from other users on the same channel). Noise refers to random interference introduced during signal reception, mainly including thermal noise and phase noise. Noise is an unavoidable factor in wireless communication systems, originating from the thermal motion of electronic devices, phase instability, etc. These three factors can interact and couple, requiring consideration to aid subsequent signal processing.

[0091] In some embodiments of this application, channel uncertainty, interference, and noise are analyzed independently, including:

[0092] Channel uncertainty includes channel estimation error, channel time-varying properties, spatial correlation, and nonlinear terms, and is quantified by the channel estimation error covariance matrix.

[0093] The interference information includes the source of interference, type of interference, power of interference, and spatial characteristics of interference, and the interference situation is quantified by the interference covariance matrix.

[0094] The noise content includes noise type, noise power, and noise correlation, and the noise situation is quantified by noise power spectral density and autocorrelation function.

[0095] In this embodiment, channel uncertainty is addressed by estimating the channel estimation error covariance matrix using pilot signals. Time-varying characteristics are addressed by calculating the Doppler frequency shift. Spatial correlation is addressed by calculating the correlation coefficient of the antenna array. Interference is addressed by obtaining the interference covariance matrix through spectrum sensing. Noise is addressed by measuring the thermal noise floor (N0 = -174 dBm / Hz) and phase noise power (-100 dBc / Hz @ 1 MHz). The autocorrelation function is calculated by determining the time-domain correlation of the noise (R(τ) = 0.9e^{-|τ| / 10}, where τ is the time delay).

[0096] In some embodiments of this application, the coupling relationship between channel uncertainty, interference, and noise is analyzed, including...

[0097] The channel estimation error covariance matrix, interference covariance matrix, noise power spectral density, and autocorrelation function are combined to quantify the coupling relationship between channel uncertainty, interference, and noise.

[0098] In this embodiment, channel-interference coupling is calculated: the correlation coefficient between channel estimation error and interference power (r=0.6) indicates that channel uncertainty amplifies the impact of interference.

[0099] Channel-noise coupling: The mutual information of the correlation between channel time-varying characteristics and noise (I(H;N)=0.8 bits) is calculated, indicating that time-varying channels introduce memory effect noise.

[0100] Interference-noise coupling: The joint probability density function (PDF) of interference power and noise power was calculated, and it was found that the noise power increased by 20% in high interference scenarios.

[0101] Coupling analysis reveals that channel uncertainty exacerbates the effects of interference and noise, requiring joint suppression rather than independent handling.

[0102] Step S103: Establish a multi-angle coupling model of the target antenna system based on the coupling relationship between the three aspects of channel uncertainty, interference and noise.

[0103] In some embodiments of this application, a multi-angle coupling model of the target antenna system is established based on the coupling relationship between channel uncertainty, interference, and noise, including:

[0104] A mathematical model is established based on the coupling relationship between the three aspects of quantized channel uncertainty, interference, and noise, and the input layer, coupling layer, and output layer of the multi-angle coupling model are designed.

[0105] The objective function of the multi-angle coupling model is defined, and the parameters in the objective function are optimized using channel data, interference data, and noise data, thereby establishing the multi-angle coupling model of the target antenna system.

[0106] In this embodiment, the input layer consists of: a channel estimation error covariance matrix (32×32), an interference covariance matrix (8×8), and a noise power spectral density (1×1024). The coupling layer uses a graph neural network (GNN), where nodes represent the channel, interference, and noise, and edges represent coupling relationships. The output layer contains the channel uncertainty suppression target value (CEER), interference suppression target value (ISRPR), and noise suppression target value (NSNRG). The objective function is a multi-objective optimization balance (channel, interference, and noise). The dataset used is the DeepMIMO dataset, containing 10,000 channel implementations, interference scenarios, and noise samples. The loss function is a weighted MSE loss, dynamically adjusted. The optimizer is the Adam optimizer with a learning rate of 0.001 and 100 training epochs. GNNs can capture the topological relationships between the channel, interference, and noise, making them suitable for modeling complex couplings; the dynamic weighted loss function can adapt to different service requirements.

[0107] Step S104: Receive the real-time signal from the target antenna system. Based on the multi-angle coupling model of the target antenna system, suppress channel uncertainty using the first suppression algorithm and suppress interference and noise using the second suppression algorithm.

[0108] In this embodiment, the suppression target values ​​for each of the three directions are output according to the multi-angle coupling model, and the poor suppression of the signal is achieved through the suppression algorithm.

[0109] In some embodiments of this application, after receiving the real-time signal from the target antenna system, the method further includes,

[0110] The channel uncertainty, interference, and noise of the real-time signal are extracted. These conditions are then input into the multi-angle coupling model of the target antenna system, and the target values ​​for suppressing the channel uncertainty, interference, and noise are output.

[0111] In some embodiments of this application, channel uncertainty suppression is performed using a first suppression algorithm, and interference and noise suppression is performed using a second suppression algorithm, including:

[0112] The first suppression algorithm includes robust channel estimation and deep learning predistortion; the second suppression algorithm includes MVDR beamforming, wavelet transform denoising, MMSE detection and nonlinear denoising.

[0113] Channel uncertainty suppression is performed using robust channel estimation and deep learning predistortion, based on the target value for channel uncertainty suppression.

[0114] Interference suppression is achieved using MVDR beamforming and wavelet transform denoising based on the target interference suppression value.

[0115] Noise suppression is achieved using MMSE detection and nonlinear noise reduction based on the noise suppression target value.

[0116] In this embodiment, channel uncertainty refers to the uncertainty caused by factors such as multipath effects, time-varying nature, and spatial correlation during the propagation of a wireless channel. To suppress channel uncertainty, the system employs a first suppression algorithm, which includes robust channel estimation and deep learning predistortion.

[0117] Robust Channel Estimation Principle: Robust channel estimation uses pilot signals to estimate channel state information (CSI) while considering channel uncertainties such as multipath effects and time-varying characteristics. It employs robust estimation methods, such as least squares (LS) or minimum mean square error (MMSE) estimation, combined with a channel uncertainty model to improve the accuracy of channel estimation. Implementation: At the receiver, the system performs correlation operations between the known pilot signal and the received signal to obtain the channel estimate. Simultaneously, the system corrects the channel estimate based on the channel uncertainty model to more accurately reflect the actual channel state. Deep Learning Predistortion Principle: Deep learning predistortion technology uses deep learning models (such as neural networks) to compensate for the nonlinear distortion of the transmitted signal. Due to the nonlinear characteristics of hardware such as power amplifiers, the transmitted signal will be distorted during transmission. Deep learning predistortion learns this distortion characteristic by training the model and preprocesses the signal at the transmitter to compensate for subsequent distortion. Implementation: At the transmitter, the system inputs the signal to be transmitted into the trained deep learning model, and the model outputs the predistorted signal. After passing through hardware such as a power amplifier, the signal's distortion characteristics are compensated, thereby improving the quality of the received signal.

[0118] Interference and noise are two other important factors affecting signal reception quality. To suppress interference and noise, the system employs a second suppression algorithm, which includes MVDR beamforming, wavelet transform denoising, MMSE detection, and nonlinear denoising.

[0119] MVDR beamforming

[0120] Principle: MVDR (Minimum Variance Distortionless Response) beamforming is an adaptive beamforming technique that adjusts the weights of the antenna array to align the main lobe with the direction of the desired signal and the null with the direction of interference, thereby suppressing interference and enhancing the desired signal.

[0121] Implementation: At the receiving end, the system calculates the optimal weights of the antenna array based on the received signal and the known direction of the desired signal. Then, the system uses these weights to perform a weighted summation of the received signal to obtain the enhanced desired signal.

[0122] Wavelet transform noise reduction

[0123] Principle: Wavelet transform is a time-frequency analysis method that can decompose a signal into different time-frequency scales. Wavelet transform denoising utilizes the multi-scale analysis characteristics of wavelet transform to extract signal features and suppress noise.

[0124] Implementation: At the receiving end, the system performs wavelet transform on the received signal to obtain wavelet coefficients at different scales. Then, based on the characteristics of the noise, the system performs thresholding or soft thresholding on the wavelet coefficients to suppress noise components. Finally, the system performs inverse wavelet transform on the processed wavelet coefficients to obtain the denoised signal.

[0125] MMSE testing

[0126] Principle: MMSE (Minimum Mean Square Error) detection is a signal detection method based on minimizing the mean square error. It optimizes signal detection performance by minimizing the mean square error between the received signal and the desired signal.

[0127] Implementation: At the receiver, the system calculates the weights of the MMSE detector based on the channel estimate and the received signal. Then, the system uses these weights to linearly combine the received signal to obtain the detected signal. MMSE detection can effectively suppress noise and interference, improving the accuracy of signal detection.

[0128] Nonlinear noise reduction

[0129] Principle: Nonlinear noise reduction technology is used to process nonlinear noise components in signals. Unlike traditional linear noise reduction methods, nonlinear noise reduction methods can better handle nonlinear noise and improve signal quality.

[0130] Implementation: At the receiving end, the system can employ various nonlinear noise reduction methods, such as neural network-based noise reduction methods and nonlinear filter-based noise reduction methods. These methods can effectively suppress nonlinear noise based on the characteristics of the signal.

[0131] In MIMO systems, these suppression algorithms work together to improve signal reception quality. Robust channel estimation provides accurate channel state information for deep learning predistortion, making predistortion processing more precise. Simultaneously, MVDR beamforming, wavelet transform denoising, MMSE detection, and nonlinear denoising work together to suppress interference and noise, improving the signal-to-noise ratio and accuracy of the received signal.

[0132] It should be noted that the above suppression algorithms can be used individually or in combination, and can be modified according to the actual situation.

[0133] Step S105: Detect the signal, evaluate the signal quality, and optimize the first suppression algorithm and the second suppression algorithm based on the signal quality.

[0134] In some embodiments of this application, signal quality is evaluated, and a first suppression algorithm and a second suppression algorithm are optimized based on the signal quality, including...

[0135] Multiple evaluation parameters are generated to evaluate signal quality, and these evaluation parameters are divided into two categories: single evaluation parameters and multiple evaluation parameters.

[0136] The suppression strength of the first and second suppression algorithms is adjusted based on two categories: single evaluation parameters and multiple evaluation parameters, thereby optimizing the first and second suppression algorithms.

[0137] In this embodiment, single evaluation parameters are parameters that evaluate only one direction (CEE, IP, NP), while multi-evaluation parameters (SINR, BER, EVM) can reflect multiple directions. These parameters are normalized. The adjustment formula for the suppression intensity is:

[0138] ;

[0139] in, For the adjusted number The suppressive force in one of the three directions. , 2 represents the number of single and multiple evaluation parameters involved in this direction, respectively. For the first The combined weight of individual evaluation parameters, For the first The first direction The size of a single evaluation parameter For the first The combined weights of multiple evaluation parameters For the first The first direction The size of multiple evaluation parameters For the first The constants corresponding to each direction To adjust the coefficient, This represents the adjustment coefficient mapped from both single and multiple evaluation parameters. This indicates that multiple evaluation parameters correct for a single evaluation parameter, collectively reflecting the quality in that direction. This represents the suppression level before adjustment. The suppression level can be represented by different parameters in different algorithms, such as the following parameters:

[0140] Robust channel estimation

[0141] Regularization coefficient (ε)

[0142] Purpose: To balance the bias and variance of channel estimation. Increasing ε can reduce estimation bias, but may increase variance.

[0143] Deep learning predistortion

[0144] Training epochs

[0145] Purpose: To control the model's ability to learn nonlinear distortions. Increasing the number of epochs can improve compensation accuracy, but may lead to overfitting.

[0146] Learning rate (η)

[0147] Effect: Affects the model's convergence speed. An excessively large η may cause oscillations, while an excessively small η will result in slow convergence.

[0148] MVDR beamforming

[0149] Diagonal loading (δ)

[0150] Function: To prevent numerical instability during the inversion of the covariance matrix and to balance interference suppression with noise enhancement. Increasing δ can enhance interference suppression, but may introduce noise.

[0151] Wavelet transform noise reduction

[0152] Threshold (λ)

[0153] Function: Determines whether wavelet coefficients are retained or set to zero. Increasing λ can reduce noise, but may result in loss of signal details.

[0154] MMSE testing

[0155] Noise variance estimate (σ²)

[0156] Function: Updates noise statistics, affecting detection weights. Increasing σ² can improve noise suppression, but may lead to overcompensation.

[0157] Nonlinear noise reduction

[0158] Nonlinear order (L)

[0159] Function: To control the suppression intensity of nonlinear noise. Increasing L can improve the nonlinear noise reduction capability, but it also increases the computational complexity.

[0160] Correspondingly, this application also provides a signal detection system for a multiple input / output antenna system, such as... Figure 2 As shown, including,

[0161] The first module is used to collect basic information about the multi-input output target antenna system and establish a received signal model of the target antenna system.

[0162] The second module is used to independently analyze the channel uncertainty, interference, and noise based on the received signal model of the target antenna system, and to analyze the coupling relationship between the three aspects of channel uncertainty, interference, and noise.

[0163] The third module is used to establish a multi-angle coupling model of the target antenna system based on the coupling relationship between the three aspects of channel uncertainty, interference and noise.

[0164] The fourth module is used to receive real-time signals from the target antenna system. Based on the multi-angle coupling model of the target antenna system, it uses the first suppression algorithm to suppress channel uncertainty and the second suppression algorithm to suppress interference and noise.

[0165] The fifth module is used to detect the signal, evaluate the signal quality, and optimize the first and second suppression algorithms based on the signal quality.

[0166] The present invention has the following beneficial effects:

[0167] A received signal model of the target antenna system is established, taking into account some nonlinear factors, which are superimposed on a linear combination model, thus providing a better description of the actual received signal of the target antenna system. The coupling relationships between channel uncertainty, interference, and noise are analyzed, and a multi-angle coupling model is established, considering the coupling of these three aspects involved in the target antenna system. This improves the adaptability to signal processing and enhances robustness in complex environments.

[0168] Channel uncertainty is suppressed using a first suppression algorithm, while interference and noise are suppressed using a second suppression algorithm. Targeted signal processing ensures signal robustness. The first and second suppression algorithms are optimized based on signal quality, balancing computational complexity and energy consumption, and meeting the real-time requirements of dynamically changing signals.

[0169] Through the above description of the embodiments, those skilled in the art can clearly understand that the present invention can be implemented in hardware or by means of software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solution of the present invention can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) and includes several instructions to cause a computer device (such as a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0170] Those skilled in the art will understand that the accompanying drawings are merely schematic diagrams of a preferred embodiment, and the modules or processes shown in the drawings are not necessarily essential for implementing the present invention.

[0171] Those skilled in the art will understand that the modules in the system of the implementation scenario can be distributed throughout the system of the implementation scenario as described, or they can be modified to reside in one or more systems different from this implementation scenario. The modules of the above-mentioned implementation scenario can be merged into one module, or they can be further divided into multiple sub-modules.

[0172] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A signal detection method for a multi-input output antenna system, characterized in that, include, Collect basic information about the multi-input output target antenna system and establish a received signal model of the target antenna system; Based on the received signal model of the target antenna system, the channel uncertainty, interference and noise are analyzed independently, and the coupling relationship between the three aspects is also analyzed. A multi-angle coupling model of the target antenna system is established based on the coupling relationship between channel uncertainty, interference, and noise. The system receives real-time signals from the target antenna system. Based on the multi-angle coupling model of the target antenna system, it uses a first suppression algorithm to suppress channel uncertainty and a second suppression algorithm to suppress interference and noise. The suppressed signal is detected, the quality of the suppressed signal is evaluated, and the first and second suppression algorithms are optimized based on the signal quality. in, A multi-angle coupling model of the target antenna system is established based on the coupling relationships between channel uncertainty, interference, and noise, including: A mathematical model is established based on the coupling relationship between the three aspects of quantized channel uncertainty, interference, and noise, and the input layer, coupling layer, and output layer of the multi-angle coupling model are designed. The objective function of the multi-angle coupling model is defined, and the parameters in the objective function are optimized by channel data, interference data and noise data, thereby establishing the multi-angle coupling model of the target antenna system. After receiving the real-time signal from the target antenna system, the method further includes, The channel uncertainty, interference, and noise of the real-time signal are extracted. These conditions are then input into the multi-angle coupling model of the target antenna system, and the target values ​​for suppressing the channel uncertainty, interference, and noise are output. Channel uncertainty is suppressed using a first suppression algorithm, and interference and noise are suppressed using a second suppression algorithm, including... The first suppression algorithm includes robust channel estimation and deep learning predistortion; the second suppression algorithm includes MVDR beamforming, wavelet transform denoising, MMSE detection and nonlinear denoising. Channel uncertainty suppression is performed using robust channel estimation and deep learning predistortion, based on the target value for channel uncertainty suppression. Interference suppression is achieved using MVDR beamforming and wavelet transform denoising based on the target interference suppression value. Noise suppression is achieved using MMSE detection and nonlinear noise reduction based on the noise suppression target value.

2. The signal detection method for a multi-input output antenna system according to claim 1, characterized in that, Establish a received signal model for the target antenna system, including: An initial received signal model under linear combination is established based on the channel matrix, interference matrix, and noise matrix. The channel matrix reflects the influence of the wireless channel on the transmitted signal, the interference matrix reflects the interference signals from other wireless sources or systems, and the noise matrix reflects the thermal noise and other random noise during the signal reception process. The hardware nonlinearity, channel nonlinearity, interference nonlinearity, and noise nonlinearity of the target antenna system are identified. The performance and calculation indicators under the hardware nonlinearity, channel nonlinearity, interference nonlinearity, and noise nonlinearity are obtained through simulation software. Based on the performance and calculation indicators, the key and non-key items among the hardware nonlinearity, channel nonlinearity, interference nonlinearity, and noise nonlinearity are selected. Key terms are described using a nonlinear model, while non-key terms are described using a linear approximation model. The model descriptions of key and non-key terms are superimposed on the initial received signal model under linear combination to form the received signal model of the target antenna system.

3. The signal detection method for a multi-input output antenna system according to claim 1, characterized in that, The channel uncertainty, interference, and noise are analyzed independently, including... Channel uncertainty includes channel estimation error, channel time-varying properties, spatial correlation, and nonlinear terms, and is quantified by the channel estimation error covariance matrix. The interference information includes the source of interference, type of interference, power of interference, and spatial characteristics of interference, and the interference situation is quantified by the interference covariance matrix. The noise content includes noise type, noise power, and noise correlation, and the noise situation is quantified by noise power spectral density and autocorrelation function.

4. The signal detection method for a multi-input output antenna system according to claim 3, characterized in that, Furthermore, the coupling relationship between channel uncertainty, interference, and noise is analyzed, including... The channel estimation error covariance matrix, interference covariance matrix, noise power spectral density, and autocorrelation function are combined to quantify the coupling relationship between channel uncertainty, interference, and noise.

5. The signal detection method for a multi-input output antenna system according to claim 1, characterized in that, Evaluate the quality of the suppressed signal, and optimize the first and second suppression algorithms based on the signal quality, including: Multiple evaluation parameters are generated to evaluate signal quality, and these evaluation parameters are divided into two categories: single evaluation parameters and multiple evaluation parameters. The suppression strength of the first and second suppression algorithms is adjusted based on two categories: single evaluation parameters and multiple evaluation parameters, thereby optimizing the first and second suppression algorithms.

6. A signal detection system for a multi-input output antenna system, characterized in that, A signal detection method for implementing a multi-input output antenna system as described in any one of claims 1-5, the system comprising, The first module is used to collect basic information about the multi-input output target antenna system and establish a received signal model of the target antenna system. The second module is used to independently analyze the channel uncertainty, interference, and noise based on the received signal model of the target antenna system, and to analyze the coupling relationship between the three aspects of channel uncertainty, interference, and noise. The third module is used to establish a multi-angle coupling model of the target antenna system based on the coupling relationship between the three aspects of channel uncertainty, interference and noise. The fourth module is used to receive real-time signals from the target antenna system. Based on the multi-angle coupling model of the target antenna system, it uses the first suppression algorithm to suppress channel uncertainty and the second suppression algorithm to suppress interference and noise. The fifth module is used to detect the suppressed signal, evaluate the quality of the suppressed signal, and optimize the first and second suppression algorithms based on the signal quality.

Images