A multi-antenna based communication coverage enhancement optimization system

Abstract

The application discloses a kind of communication coverage enhancement optimization systems based on multiple antennas, belong to wireless communication processing technical field;By introducing multipath delay spectrum and space arrival angle, so that the system can accurately identify that the several antenna points that cause "time delay similar" are which several, to provide interference source space fingerprint for subsequent dynamic logic port mapping from the root cause;By introducing coherence bandwidth parameter and correlation coefficient as the basis for decision, ensure that only when signal meets strict frequency domain coherence condition and time domain alignment condition, start coherence combination, from the physical layer source, eliminate the harsh superposition of non-correlated signal;In interference competition state, by merging all physical array into a single second type logic port, abandon the space multiplexing intention, instead focus on constructing extremely deep notch, interference energy is cancelled in spatial dimension, rather than accumulated, from the root of algorithm, solve the problem that coverage enhancement becomes interference enhancement.

Description

Technical Field

[0001] This invention relates to the field of wireless communication processing technology, and more specifically to a communication coverage enhancement and optimization system based on multiple antennas. Background Technology

[0002] When existing communication coverage enhancement and optimization schemes are implemented, if a large number of distributed micro-cells are deployed indoors or in hotspot areas to enhance coverage, the terminals will receive signals from several points with similar time delays due to the different physical locations of the antenna points. This will cause frequent hard handovers or inter-cell interference, rather than MIMO gain, and turn coverage enhancement into interference enhancement. Summary of the Invention

[0003] The purpose of this invention is to provide a communication coverage enhancement and optimization system based on multiple antennas to solve the technical problems mentioned in the background art.

[0004] The objective of this invention can be achieved through the following technical solutions: A communication coverage enhancement and optimization system based on multiple antennas includes: Signal holographic extraction module: Based on several distributed micro-station antenna points, it acquires the uplink detection signal of the target terminal and extracts the multi-dimensional feature information of the detection signal. The multi-dimensional feature information includes at least the multipath delay spectrum, spatial angle of arrival, and Doppler frequency shift. Coherent spectrum clustering discriminator: Calculates the coherent bandwidth parameters between the detection signals received by each antenna point, and determines whether the current channel state is coherent merging state or interference contention state based on the coherent bandwidth parameters; Logical port dynamic mapping module: Based on the channel state determination result, the antenna arrays physically distributed in different geographical locations are dynamically reconstructed into at least two types of logical antenna ports. The two types of logical antenna ports include: a first type of port for obtaining spatial multiplexing gain and a second type of port for constructing dynamic virtual nulls to actively cancel co-channel interference. Delay slope tracking compensator: Based on the rate of change of Doppler frequency shift and spatial angle of arrival, predict the slope of signal propagation delay change of each distributed antenna point relative to the same target terminal. Based on the slope of delay change, perform independent small-step closed-loop lead / lag compensation on the baseband transmission time of each distributed antenna point, so that the path delay difference of each path to the target terminal after compensation dynamically converges within the preset delay window, and the starting position of the delay window avoids the boundary region of the cyclic prefix.

[0005] Furthermore, the received signal is subjected to analog-to-digital conversion to obtain a discrete sampling sequence with the sampling period. The discrete sampling sequence is then subjected to a fast Fourier transform to convert it to the frequency domain, thus obtaining the frequency domain received signal. By performing point-by-point complex multiplication between the frequency domain sequence of the locally stored probe reference signal and the frequency domain received signal, channel frequency response estimation is performed to obtain the frequency domain channel estimate. Perform an inverse fast Fourier transform on the frequency domain channel estimate to transform it back to the time domain, obtaining the time domain channel impulse response sequence of the i-th antenna point; calculate the squared magnitude of the sequence and coherently accumulate it along the time dimension to obtain the multipath power delay spectrum.

[0006] Furthermore, by utilizing the relative spatial phase difference between distributed antenna points, the main reference antenna point with the strongest received signal-to-noise ratio is selected as the phase reference point, and the relative phase difference between the i-th antenna point and the main reference point is calculated. By simultaneously solving the phase difference equations of all antenna points, an overdetermined set of equations is formed. Then, a two-dimensional spectral peak search is performed on the direction vector of arrival using a multi-signal classification algorithm based on eigenvalue decomposition, and finally the spatial angle of arrival is calculated.

[0007] Furthermore, using the channel estimation results of several consecutive uplink time slots j, j=1,2,…,J, where J is the total number of uplink time slots, for the strongest principal path of the i-th antenna point, when tracking the change of its complex amplitude on the time axis, the autocorrelation value of the principal path complex amplitude between the j-th time slot and the (j-1)-th time slot is calculated. The phase difference average of the autocorrelation results over J consecutive time slots is taken, and the Doppler frequency shift is calculated using the following expression. : ;in, This indicates taking the principal value of the complex phase angle, with the range limited to... between; The Nyquist sampling frequency; This is the time interval between adjacent uplink time slots.

[0008] Furthermore, for each antenna point, the root mean square delay spread of that antenna point is calculated based on its own power delay spectrum. Based on the inverse relationship between channel coherence bandwidth and root mean square delay spread, the local coherence bandwidth estimate for each antenna point is calculated using the root mean square delay spread.

[0009] Furthermore, when performing time-domain discreteness discrimination, the global minimum delay and global maximum delay detected by all distributed antenna points are extracted, and the global maximum relative delay difference is calculated; In addition, the system reads the preset cyclic prefix length parameter and pre-sets the first and second time domain thresholds; when implementing joint state machine determination, it integrates the pairwise correlation coefficients between all antenna points and calculates the global average correlation coefficient; it performs state determination according to the logical rule group, which includes determination rule A, determination rule B, and determination rule C; determination rule A corresponds to the coherent merging state triggering condition, determination rule B corresponds to the interference contention state triggering condition, and determination rule C corresponds to the transition state maintenance mechanism.

[0010] Furthermore, to avoid frequent state jumps in the discriminator due to sudden noise or momentary occlusion, the coherent spectrum clustering discriminator incorporates a first-order hysteresis filter: before each decision update, the globally average correlation coefficient calculated in this iteration is used as the filter element. Global average correlation coefficient with the previous time step For smoothing, the corresponding expression is: ;in, This is the global average correlation coefficient after first-order lag filtering; This is a preset forgetting factor, with a value range of (0,1).

[0011] Furthermore, when implementing the second type of logic port mapping under the interference contention state, when determining the status flag bit... At that time, the interference suppression mapping strategy is enforced.

[0012] Furthermore, to avoid frequent switching between coherent merging and interference contention states at the logic ports, a minimum dwell time timer is set internally within the logic port dynamic mapping module; during the timer's operation, even the status flag bit output by the coherent spectrum clustering discriminator is updated. If changes occur, the current mapping configuration remains unchanged. Only after the timer times out and the status flag remains stable in the new state for more than two consecutive measurement cycles will a new round of logic port remapping calculation be triggered, thereby effectively suppressing system oscillations.

[0013] Furthermore, to ensure that each signal meets the time-domain conditions for coherent combining or interference cancellation at the target terminal, after each transmission compensation, the absolute propagation delay estimate of the signal at each distributed antenna point to the target terminal is calculated. Design a dynamically biased delay window whose starting position actively avoids the boundary region of the cyclic prefix, so as to prevent delay jitter caused by channel time variation from crossing the cyclic prefix boundary and thus causing inter-symbol interference.

[0014] Compared to existing solutions, the beneficial effects achieved by this invention are: This invention introduces multipath delay spectrum and spatial angle of arrival, enabling the system to accurately identify which antenna points cause "similar delays," thus providing a spatial fingerprint of the interference source for subsequent dynamic logic port mapping and avoiding interference enhancement caused by blind merging. By introducing coherence bandwidth parameters and correlation coefficients as decision criteria, it ensures that coherent merging is only initiated when the signal meets strict frequency domain coherence and time domain alignment conditions, eliminating the hard superposition of unrelated signals at the physical layer source and fundamentally solving the problem of misjudgment of interference enhancement. In interference contention states, through... By merging all physical arrays into a single Type II logical port, the spatial reuse intention is abandoned, and instead the focus is placed on constructing extremely deep null traps. This strategy causes the originally interfering multi-point signals to undergo phase-destructive interference at the target terminal, canceling the interference energy in the spatial dimension rather than accumulating it. This solves the problem of coverage enhancement turning into interference enhancement from the root of the algorithm. By jointly predicting the delay slope through Doppler frequency shift and angle change rate, the compensation changes from a passive response to an active prediction, eliminating the inter-symbol interference caused by the abrupt change in delay in the traditional scheme and reducing the probability of link loss of synchronization by an order of magnitude. Attached Figure Description

[0015] The invention will now be further described with reference to the accompanying drawings.

[0016] Figure 1 This is a flowchart illustrating the operation of a communication coverage enhancement and optimization system based on multiple antennas according to the present invention.

[0017] Figure 2 This is a block diagram of a communication coverage enhancement and optimization system based on multiple antennas according to the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] like Figures 1 to 2 As shown, the present invention is a communication coverage enhancement and optimization system based on multiple antennas, including a signal holographic extraction module, a coherent spectrum clustering discriminator, a logic port dynamic mapping module, and a delay slope tracking compensator; Signal holographic extraction module: Based on several distributed micro-station antenna points, it acquires the uplink detection signal of the target terminal and extracts the multi-dimensional feature information of the detection signal. The multi-dimensional feature information includes at least the multipath delay spectrum, spatial angle of arrival, and Doppler frequency shift. The signal holographic extraction module is physically coupled to several micro-station antenna points distributed in indoor or hotspot areas. The several micro-station antenna points are marked as i, i=1,2,...,N, where N is the total number of antenna points participating in the collaborative coverage. Each antenna point is equipped with at least one set of dual-polarized antenna arrays and is synchronously connected to the central baseband processing pool through high-speed optical fiber or radio frequency remote unit. When performing multi-dimensional feature extraction, this module, based on the reciprocity of uplink and downlink channels in a time-division duplex (TDD) system, captures the sounding reference signal (SRS) transmitted by the target terminal (user equipment) within a preset uplink pilot time slot. , k is the current active user number; the received baseband complex baseband signal model expression is set as: ;in, For receiving signals, t is a continuous time variable. For signal propagation delay, The time-domain waveform of a known pilot sequence transmitted by the target terminal. The time-varying channel impulse response between the i-th antenna point and the target terminal. This is the additive white Gaussian noise term. This represents the convolution operation; When processing the received signal and extracting the multipath delay spectrum, the received signal... Perform analog-to-digital conversion, based on the sampling period. Obtain discrete sampling sequence , For discrete sampling point numbers; for discrete sampling sequences Perform a Fast Fourier Transform (FFT) to convert it to the frequency domain, and obtain the frequency domain received signal. , This is a discrete subcarrier frequency index; the analog-to-digital conversion and fast Fourier transform performed are conventional technical solutions in this field, and the specific implementation steps will not be elaborated here. Frequency domain sequence of the probe reference signal stored locally and Perform pointwise complex multiplication to estimate the channel frequency response and obtain the frequency domain channel estimate. The corresponding expression is: ; Frequency domain channel estimation Perform an inverse fast Fourier transform (IFFT) to transform it back to the time domain, obtaining the time-domain channel impulse response sequence for the i-th antenna point. , The time-delay tap index is used; the modulus squared of the sequence is calculated, and coherent accumulation is performed along the time dimension, with the number of accumulations denoted as . The multipath power delay spectrum was obtained. The corresponding expression is: ;in, The multipath power delay spectrum of the i-th antenna point; , indicating the first The absolute propagation delay value corresponding to each tap; For the first The first coherent accumulation time The complex amplitude value of each tap; For coherent cumulative index; Extracting multipath power delay spectrum The mid-amplitude exceeds the noise floor threshold. The time delays corresponding to several peak points constitute a multipath time delay spectrum set. , This represents the total number of effective paths detected at the i-th antenna point.

[0020] To accurately extract the spatial angle of arrival, the relative spatial phase difference between distributed antenna points is utilized, and based on the previous steps, the main reference antenna point with the strongest received signal-to-noise ratio is selected. As the phase reference point; for any i-th cooperative antenna point, its frequency domain phase on the line-of-sight path is... , The preset reference subcarrier frequency, This represents the phase angle when taken as a complex number; Let the three-dimensional spatial coordinate vector of the i-th antenna point be... The spatial coordinate vector of the main reference point is And calculate the relative phase difference The corresponding expression is: ;in, The phase unambiguity de-evaluation integer factor is determined by the phase unwrapping algorithm based on the Chinese Remainder Theorem embedded in the module, and is used to eliminate periodicity. The resulting phase winding; This is the frequency domain channel estimate of the reference antenna point on the reference subcarrier; according to the array signal propagation principle, the spatial angle of arrival is determined by the direction vector of the angle of arrival. The following relationship between phase difference and spatial position is satisfied: ;in, The speed of light; By simultaneously solving the phase difference equations for all N antenna points, an overdetermined system of equations is formed. Then, the Multiple Signal Classification (MUSIC) algorithm based on eigenvalue decomposition is used to classify the signals. A two-dimensional spectral peak search is performed, and the spatial angle of arrival is finally calculated, which is derived from the azimuth angle. and pitch angle The unique determination makes ; Among them, MUSIC (Multiple Signal Classification) is a subspace-based DOA (Direction of Arrival) estimation algorithm. Its core relies on the eigenvalue decomposition (EVD) of the array receiver covariance matrix, splitting the observation space into two orthogonal subspaces: a signal subspace and a noise subspace. Specifically: Signal subspace: spanned by eigenvectors corresponding to large eigenvalues, containing information about the incident signal; Noise subspace: spanned by eigenvectors corresponding to small eigenvalues, containing only Gaussian noise; The signal steering vector is strictly orthogonal to the noise subspace. This orthogonality is used to construct a spatial spectral function, with the spectral peaks corresponding to the signal incident angles, achieving high-precision direction finding. A multi-signal classification algorithm based on eigenvalue decomposition is also employed. The two-dimensional spectral peak search and calculation performed is a conventional technical solution in this field, and the specific implementation steps will not be elaborated here.

[0021] To extract the Doppler frequency shift, channel estimation results from several consecutive uplink time slots j are used, j = 1, 2, ..., J, where J is the total number of uplink time slots; for the strongest principal path of the i-th antenna point, i.e., the delay... When tracking the change of complex amplitude of the corresponding tap on the time axis, calculate the autocorrelation value of the principal diameter complex amplitude between the j-th time slot and the (j-1)-th time slot. The corresponding expression is: ;in, The time interval between adjacent uplink time slots is a preset system constant. ; The tap index corresponding to the strongest primary path; Let be the complex amplitude of the main diameter tap in the j-th time slot; The conjugate of the complex amplitude of the main path tap in the (j-1)th time slot is given; the phase difference average of the autocorrelation results for J consecutive time slots is taken, and the Doppler frequency shift is calculated using the following expression. : ;in, This indicates taking the principal value of the complex phase angle, with the range limited to... between; The Nyquist sampling frequency is used; considering the frequency aliasing that may occur due to terminal movement, when the calculated frequency is... When the frequency exceeds the Nyquist sampling frequency, a frequency tracking loop based on a Kalman predictor is automatically activated for correction, outputting the final accurate Doppler frequency shift estimate. The correction based on the frequency tracking loop of the Kalman predictor is a conventional technical solution in this field, and the specific implementation steps will not be elaborated here.

[0022] The signal holographic extraction module has a built-in feature timestamp synchronizer, which ensures the quality of the extracted multipath delay spectrum. Angle of arrival in space and Doppler shift They all carry the same system frame number marker and symbol index marker, thus ensuring strict time alignment; The local features extracted from each antenna point are aggregated into a global multidimensional feature tensor. The corresponding expression is: This tensor serves as the sole input to the subsequent coherent spectrum clustering discriminator.

[0023] It should be noted that the core function of the signal holographic extraction module is to deconstruct the one-dimensional time-domain uplink pilot signal received by the distributed micro-station antenna points into holographic feature parameters of time delay distribution (distance dimension), angle pointing (spatial dimension), and frequency offset (motion dimension) in three-dimensional space through parallel time-domain, frequency-domain, and spatial-domain joint processing. It is not only responsible for extracting the original data, but more importantly, it completes the strict synchronization and structured encapsulation of heterogeneous features on the time baseline, providing high-dimensional, low-redundancy physical basis data for the subsequent decision-maker to distinguish between "coherent merging state" and "interference competition state".

[0024] Traditional coverage enhancement schemes rely solely on scalar metrics such as reference signal received power, failing to distinguish between interference and signals. In this embodiment, by introducing multipath delay spectrum and spatial angle of arrival, the system can accurately identify which antenna points cause "similar delays," thus providing a spatial fingerprint of interference sources for subsequent dynamic logic port mapping and avoiding interference enhancement caused by blind merging. The extracted spatial angle of arrival directly serves the subsequent active parasitic array driving module, ensuring that the main lobe pointing has a clear target vector. Simultaneously, the extraction of Doppler frequency shift provides the system with time change rate information for predicting terminal movement trends, providing indispensable input variables for subsequent Kalman filter channel prediction correction factors and delay slope tracking compensators, ensuring that the system can maintain beam alignment stability even in highly dynamic scenarios. By meticulously characterizing the multipath delay spectrum, the system can clearly distinguish between the merging gain path and the interference countermeasure path, avoiding the power waste caused by transmitting signals from all antenna points in the same way. Combined with high-precision spatial angle of arrival, energy can be accurately directed towards the target terminal. At the same time, it lays the physical layer measurement foundation for identifying and avoiding grating lobes pointing to non-user areas, thus alleviating the problem of neighboring cell interference degradation from the source.

[0025] Coherent spectrum clustering discriminator: Calculates the coherent bandwidth parameters between the detection signals received by each antenna point, and determines whether the current channel state is coherent merging state or interference contention state based on the coherent bandwidth parameters; At the physical layer, the coherent spectrum clustering discriminator functions as a parallel computing unit within the baseband processing pool. Its input is connected to the aforementioned signal holography extraction module via a high-speed data bus. This discriminator reads the global multidimensional feature tensor generated by the signal holography extraction module. And extract the frequency domain channel response matrix and multipath delay spectrum set required for constructing coherent decision; Among them, the frequency domain channel response matrix Its dimensions are , This represents the total number of effective subcarriers carrying the probe reference signal within the system bandwidth; the matrix is... Line number The column elements are those calculated in the aforementioned embodiments. ; Multipath Delay Spectrum Collection , used to calculate the maximum additional time delay spread at each antenna point; To quantify the frequency correlation between the received signals at each antenna point, the discriminator calculates the frequency domain coherence bandwidth parameter and correlation coefficient for any two cooperating antenna points i and j. Calculate their normalized correlation coefficients in the frequency domain. The corresponding expression is: ;in, For the j-th antenna point at frequency point Complex channel estimates; For the j-th antenna point at frequency point The conjugate transpose of the complex channel estimate on; symbol This indicates the modulus operation for complex numbers; Let be a dimensionless real number, and its value range be 1. The closer the value is to 1, the more linearly correlated the channel frequency domain response is between the two antenna points.

[0026] Based on statistical wireless communication theory, frequency domain correlation and time domain delay spread form a Fourier transform pair. This discriminator does not directly calculate a single global coherence bandwidth, but rather, for each antenna point i, it calculates based on its own power delay spectrum. Calculate the root mean square delay spread at this antenna point. The corresponding expression is: ;in, This is the index of the number of valid multipaths detected at the i-th antenna point; For the first The absolute time delay value of the effective path; This represents the received power value along this path; The average additional time delay of the i-th antenna point is expressed as follows: ; Based on the inverse relationship between channel coherence bandwidth and root mean square delay spread, the estimated local coherence bandwidth for each antenna point is calculated, and the corresponding expression is: ;in, Let be the estimated local coherence bandwidth of the i-th antenna point, representing the width of the frequency range where the channel frequency response remains highly correlated at the i-th antenna point; Furthermore, the core of the coherent spectrum clustering discriminator lies in jointly deciding on the frequency domain correlation and the time domain dispersion. This embodiment designs a two-dimensional decision state machine, the specific logic of which is as follows: When performing time-domain discreteness discrimination, extract the global minimum time delay detected by all distributed antenna points. and global maximum latency Calculate the global maximum relative delay difference : ; In addition, read the system's preset loop prefix length parameter. This parameter is determined by the communication standard; for example, in Long Term Evolution (LTE) or New Radio (NR) standards, the typical cyclic prefix length is approximately 4.69. Or 2.34 And two time-domain decision thresholds are pre-defined: the first time-domain threshold... Second time-domain threshold ; When implementing joint state machine decision-making, the pairwise correlation coefficients between all antenna points are considered. Calculate the global average correlation coefficient The corresponding expression is: ; Furthermore, state determination is performed according to a set of logical rules, which includes decision rule A, decision rule B, and decision rule C. Decision rule A corresponds to the coherent merging state trigger condition, decision rule B corresponds to the interference and contention state trigger condition, and decision rule C corresponds to the transition state preservation mechanism. Specifically: Judgment Rule A: When the global average correlation coefficient is greater than the preset first correlation coefficient threshold and the global maximum relative delay difference is less than the first time domain threshold, the current channel state is determined to be coherent merging state; this state outputs a logical flag bit. The system indicates that the signals from each distributed antenna point are treated as highly correlated signals from an equivalent large aperture array, which are suitable for maximum ratio combining to obtain coherent combining gain; the first correlation coefficient threshold is set to 0.85 in this embodiment. Judgment Rule B: When the global average correlation coefficient is less than the preset second correlation coefficient threshold (set to 0.3 in this embodiment), and the global maximum relative delay difference is greater than the second time domain threshold, the current channel state is determined to be an interference contention state; this state outputs a logic flag. The signals at each antenna point of the indicator system have already shown frequency-selective fading in the frequency domain and are sufficient to distinguish different paths in the time domain. If forced merging is performed at this time, it will cause inter-symbol interference (ISI). Therefore, it is suitable to map the physical antennas as logic ports for active interference cancellation. The second correlation coefficient threshold is set to 0.3 in this embodiment. Judgment Rule C: When neither judgment rule A nor B is satisfied, the state is not switched immediately. Instead, the state flag of the previous time slot is maintained and a timer is triggered to accumulate the transition time. If the transition time continues to exceed the preset number of system frames, such as 5 consecutive wireless frames, decorrelation processing is forcibly executed, and the state flag is forcibly set to the interference contention state to ensure the robustness of the system in prioritizing interference suppression under fuzzy boundaries.

[0027] After completing the above state determination, the coherence spectrum clustering discriminator further utilizes pairwise correlation coefficients. Generate a Correlation distance matrix Its element definition is as follows: ;in, This represents the equivalent distance between the i-th antenna point and the j-th antenna point in terms of channel characteristics. The smaller this value, the stronger the correlation. The correlation distance matrix... Status flags They are packaged together into a decision output vector. And transmit it to the logical port dynamic mapping module in real time; To avoid frequent state jumps in the discriminator caused by sudden noise or momentary occlusion, i.e., the ping-pong effect, the discriminator incorporates a first-order hysteresis filter: before each decision update, the globally average correlation coefficient calculated in this iteration is used as the filter. Global average correlation coefficient with the previous time step For smoothing, the corresponding expression is: ;in, This is the global average correlation coefficient after first-order lag filtering; The preset forgetting factor has a value range of (0,1), and in this embodiment, a value of 0.3 is preferred; using this smoothed factor... The instantaneous value is used instead of the instantaneous value in the comparison of the above judgment rules A and B, thereby significantly improving the temporal stability of the judgment result.

[0028] It should be noted that the core function of the coherent spectrum clustering discriminator is to transform the complex frequency domain response and multipath delay raw data output by the signal holographic extraction module into a single decision state quantity with clear physical meaning. By jointly analyzing the frequency domain correlation coefficient and the time domain spread dispersion, it fundamentally determines whether the distributed antenna points are coherent partners from the same source or interfering opponents from different sources. The output of the coherent spectrum clustering discriminator does not contain complex beam weights, but rather provides a decisive basis for subsequent logical port mapping, that is, to determine which antennas should be aggregated to enhance the signal and which antennas should be split to suppress interference.

[0029] Traditional solutions rely solely on the strength of the received reference signal to determine whether an antenna should participate in cooperation. In scenarios where indoor hotspots have similar latency, antennas with strong received signal power but severe mutual interference may still be incorrectly merged, causing coverage enhancement to become interference enhancement. In this embodiment, coherent bandwidth parameters and correlation coefficients are creatively introduced as decision criteria to ensure that coherent merging is only initiated when the signal meets strict frequency domain coherence and time domain alignment conditions. This eliminates the forced superposition of unrelated signals from the physical layer source and fundamentally solves the problem of misjudgment in interference enhancement. By setting an independent second decision threshold, the interference competition state flag can be output in time when the signal shows obvious frequency selective fading. This provides a precise triggering time for the subsequent mixed precoding of the zero-forcing criterion and the minimum mean square error criterion, ensuring that the system immediately switches to the zero-trap construction mode in the interference scenario, turning the interference source into a distinguishable signal source. By introducing decision rule C and the forgetting factor, the discriminator output oscillation caused by instantaneous dynamic changes such as people moving around in the indoor environment is effectively prevented. This avoids the huge signaling overhead and beam pointing jitter caused by the frequent reconfiguration of physical antenna elements in the logic port mapping module, ensuring a smooth transition of the system when switching between dynamic and static scenarios, and significantly enhancing the robustness of engineering deployment.

[0030] Logical port dynamic mapping module: Based on the channel state determination result, the antenna arrays physically distributed in different geographical locations are dynamically reconstructed into at least two types of logical antenna ports. The two types of logical antenna ports include: a first type of port for obtaining spatial multiplexing gain and a second type of port for constructing dynamic virtual nulls to actively cancel co-channel interference. The logic port dynamic mapping module, running in the central scheduling unit of the baseband processing pool, is the "executor" that directly solves the problem. It transforms the decision result into physical actions, forcibly canceling out the originally "superimposed" interference signals in space, thereby actively eliminating frequent hard handovers and inter-cell interference. Its input is coupled to the aforementioned coherent spectrum clustering discriminator through high-speed shared memory. This module reads the decision output vector in real time. Simultaneously, it acquires the full-dimensional uplink channel state information matrix estimated by the signal holographic extraction module. Its dimensions are , The number of effective subcarriers; Based on the channel reciprocity of Time Division Duplex (TDD) systems, the uplink channel estimation matrix is... Perform transpose and conjugate operations to generate the initial reference matrix for downlink precoding: ;in, For dimension The matrix has an element in row f and column i that represents the equivalent baseband channel response from the baseband processing pool to the i-th physical antenna array in the downlink; the superscript T indicates matrix transpose. When implementing the first type of logical port mapping in coherent merging state, when determining the status flag bit At that time, read the correlation distance matrix D and set the spatial clustering radius threshold. In this embodiment, the preferred value is 0.15; the density-based noise spatial clustering algorithm (DBSCAN) is used to divide the N physical antenna elements into M clusters, where M is a dynamic variable and satisfies the following conditions: The correlation distance between any two antenna elements within each cluster All less than This indicates that the channel responses of the arrays within the cluster are strongly correlated, making them suitable for coherent combining. For the m-th cluster, Assuming the cluster contains There are physical antenna elements, and their corresponding physical element index set is denoted as . Map all physical antenna elements in the physical element index set to the m-th sub-port of the first type of logical port, denoted as . The downlink transmit weight initial vector corresponding to this logical port Based on the maximum ratio merging criterion, the corresponding expression is: ;in, The normalized weight vector has dimension . This is used to ensure that the total transmission power remains constant; Indicates from Extracting column indexes belongs to the physical array sub-index set. The submatrix consisting of all column vectors has dimension . ; The preset reference beam direction vector has a dimension of All of its elements are To achieve equal gain excitation; Represents the Euclidean norm operation for vectors; To ensure that the first type of logical port does not introduce inter-stream interference while acquiring spatial multiplexing gain, the total number of currently active terminals is calculated. For the m-th cluster, its equivalent spatial degrees of freedom Defined as And perform the following constraint determination: ;in, This represents the number of active users confirmed through the Physical Random Access Channel (PRACH) in the current time slot. If the above formula does not hold, it indicates that coherent merging has led to excessive compression of spatial degrees of freedom. In this case, some clusters are automatically split, and the split clusters are remapped to the second type of logical port until the above constraints are met, thereby ensuring that the system always has sufficient spatial dimensions to support multi-user multiplexing. When implementing the second type of logic port mapping under interference contention state, when determining the status flag bit At that time, the following interference suppression mapping strategy will be enforced: Select the highest energy from the multipath delay spectrum Interference path, this embodiment takes and its space reaches the angle The corresponding steering vector is constructed as an interference subspace matrix. Its dimensions are , This represents the total number of physical antenna elements contained in all micro-station antenna points in a distributed collaborative deployment; furthermore, the... Steering vectors corresponding to each interference path The The elements are: ;in, This is the global index number of the physical antenna array, with a value range of [value range missing]. This is used to uniquely identify each independent radiation unit participating in the collaborative coverage; The imaginary unit, ; The downlink carrier wavelength is determined by the system carrier frequency. pass Calculations show that The speed of light; Let G be the three-dimensional spatial coordinates of the g-th physical antenna array; and The first Azimuth and elevation angles of a strong interference path It should be explained that each distributed microcell antenna point can contain one or more physical antenna elements. Let the i-th antenna point contain... If there are 1 physical array, then the total number of physical arrays in the world is 1. ; Interference covariance matrix The calculation expression is: ;in, for The complex Hermitian matrix characterizes the spatial statistical properties of the interference signal; This is the conjugate transpose of the interference subspace matrix; The receiver noise floor power is pre-calibrated by the system noise figure; for The identity matrix; Using the linear constraint minimum variance criterion, all Each physical antenna array is mapped to a Class II logical port, denoted as . The downlink transmit weight vector corresponding to this port This is obtained by solving the following constrained optimization problem: ; The independent variable that minimizes the objective function; The candidate emission weight vector; This is the conjugate transpose of the candidate emission weight vector; The constraint is that the signal in the direction of the target terminal has no distortion response: ;in, The guide vector is the direction of the target terminal, with dimension [missing information]. Its expression is the same as the aforementioned The format is the same, only the azimuth angle corresponding to the target terminal is substituted. and pitch angle The closed-form solution to this constrained optimization problem is: ;in, This is the transmit weight vector for the second type of logic port; the superscript -1 indicates matrix inversion; the dimension generated by this expression is... The weight vector can ensure that the transmitted signal maintains a constant gain in the direction of the target terminal, while forming an extremely deep null in all directions of strong interference, which theoretically tends to negative infinity and can reach -40dB to -50dB in engineering implementation; To enhance the robustness of the second type of logical port in frequency-selective fading channels, in the weight vector Based on this, a time-domain cyclic shift parameter is added to the g-th physical antenna element. ;Specifically, The value of is related to the spatial coordinates of the element, and the corresponding expression is: ;in, Let g be the spatial position vector of the g-th element; The reference point position vector is preset, and in this embodiment, the geometric center point is selected; The weighting coefficient for the geographical location of the g-th element is dynamically generated based on the element in the g-th row of the correlation distance matrix, and its expression is: The range of values ​​is ; Introduction Subsequently, the equivalent baseband transmitted signal of the second type of logic port forms phase destructive interference at the target terminal, thereby widening its null bandwidth and maintaining stable interference suppression performance in a multipath environment.

[0031] After completing the above mapping, generate the logical port configuration instruction vector. The corresponding expression is: ;in, This is a port type identifier, with a value of 0 or 1, which is mapped to the first type of logical port and the second type of logical port, respectively. This is the type identifier for the second type of logical port; The set of time-domain cyclic shift parameters added to the second type of port; Let be the transmit weight vector of the m-th first-class logic port; For set-and-join operations, the instruction vector is sent to the radio frequency front-end of each distributed micro-station via a high-speed control bus, driving it to perform downlink transmission according to the new logical port affiliation.

[0032] To avoid frequent switching between coherent merging and interference contention states at the logic ports, a minimum dwell time timer is set internally in this module. In this embodiment, 5 wireless subframes are preset, corresponding to 5 milliseconds; during the timer operation, even if the status flag bit output by the coherent spectrum clustering discriminator is... Even if changes occur, the module maintains the current mapping configuration. Only after the timer times out and the status flag remains stable in the new state for more than two consecutive measurement cycles will the module trigger a new round of logic port remapping calculation, thereby effectively suppressing system oscillations.

[0033] It should be noted that the core function of the logical port dynamic mapping module is to transform the abstract channel state decision result (coherent merging state or interference contention state) into a specific executable physical element grouping scheme and transmission weight matrix. Through mathematical means, it reconstructs the numerous physical antenna elements scattered indoors into two logical ports with distinctly different electromagnetic behaviors as needed: one type is dedicated to energy focusing to obtain the maximum ratio merging gain, and the other type is dedicated to spatial nulling to destroy co-channel interference paths. In essence, it plays the role of an adaptive spatial filter architect, dynamically defining the gain distribution of the system in the spatial dimension.

[0034] Traditional solutions treat all distributed antennas as a fixed multiple-input multiple-output (MIMO) array. In scenarios with similar latency, the energy of incoherent paths is forcibly superimposed as interference. In this embodiment, under interference contention, all physical elements are merged into a single second-type logical port. This abandons the intention of spatial multiplexing and instead focuses on constructing extremely deep null traps. This strategy causes the originally interfering multi-point signals to undergo phase destructive interference at the target terminal, canceling the interference energy in the spatial dimension rather than accumulating it. This solves the problem of coverage enhancement turning into interference enhancement from the root of the algorithm.

[0035] Traditional solutions blindly pursue large apertures at the expense of degrees of freedom during coherent merging, resulting in the system being unable to serve multiple users simultaneously. This embodiment dynamically controls the aggregation granularity of the first type of logic ports, ensuring that sufficient spatial dimensions are always reserved for multi-user multiplexing while obtaining coherent gain, thus achieving the best trade-off between coverage enhancement and system capacity. The introduction of geographic location weighting coefficients and time-domain cyclic shift overcomes the limitation that traditional static nulls are only effective at a single frequency point. Even if the terminal undergoes a slight displacement indoors, causing a slight change in the interference arrival angle, the broadened null still has sufficient suppression depth, significantly improving the system's interference suppression robustness in dynamic indoor environments.

[0036] By setting a minimum dwell time timer, the frequent reconfiguration of logic ports caused by instantaneous channel fluctuations is effectively curbed. Each change in the logical affiliation of a physical array is accompanied by a state switch of the RF switch and phase shifter. This embodiment effectively reduces the number of hardware reconfigurations through time-dimensional lag constraints, which not only extends the service life of RF front-end devices, but also avoids transient beam pointing disorder caused by frequent reconfigurations, thus ensuring the continuity of the communication link.

[0037] Delay slope tracking compensator: Based on the rate of change of Doppler frequency shift and spatial angle of arrival, predict the slope of signal propagation delay change of each distributed antenna point relative to the same target terminal. Based on the slope of delay change, perform independent small-step closed-loop lead / lag compensation on the baseband transmission time of each distributed antenna point, so that the path delay difference of each path to the target terminal after compensation dynamically converges within the preset delay window, and the starting position of the delay window avoids the boundary region of the cyclic prefix.

[0038] In this embodiment, the delay slope tracking compensator, as a real-time closed-loop adjustment unit in the baseband processing pool, is a forward-looking compensation module to prevent problem recurrence. Through active time calibration, it fundamentally curbs the occurrence of frequent hard handover events caused by delay jitter. Its input is coupled to the signal holographic extraction module and the coherent spectrum clustering discriminator via a data bus. The delay slope tracking compensator operates independently on a per-distributed antenna point basis. For the i-th antenna point, it reads two key input parameters in real time: the Doppler frequency shift estimate and the spatial angle of arrival change rate. Among them, the Doppler frequency shift estimate The unit is Hertz, which is calculated by the signal holographic extraction module based on the time correlation of the uplink detection signal, and represents the radial relative motion rate between the i-th antenna point and the target terminal. Rate of change of spatial angle of arrival and The units are radians per second, and represent the time derivatives of the target terminal's azimuth and elevation angles observed at the i-th antenna point, respectively; this rate of change is obtained by linearly fitting the estimated angle of arrival values ​​of several consecutive uplink time slots by the signal holographic extraction module; Based on the above input, the delay slope tracking compensator first predicts the slope of the signal propagation delay change at the i-th antenna point relative to the target terminal. The slope is defined as the instantaneous rate of change of the delay over time, and its physical origin lies in the continuous change of the propagation path length caused by the movement of the target terminal. Specifically, the following prediction expression based on geometric kinematics is used: ;in, The downlink carrier frequency of the system is predetermined by the wireless communication standard; The instantaneous elevation angle of the target terminal at the i-th antenna point at the current moment is output in real time by the signal holographic extraction module; Let be the spatial distance between the i-th antenna point and the reference point, where the reference point is chosen as the geometric center of all distributed antenna points, and its expression is: , The geometric center coordinate vector; the first term in the above expression The second term corresponds to the time delay slope component caused by radial motion. This corresponds to the lateral motion time delay slope component caused by the angle change, and the two are combined to form a complete predicted value of the time delay change slope. Based on the predicted slope of time delay change Independent small-step closed-loop compensation is performed on the baseband transmission time of each distributed antenna point; this compensation is not a one-time large adjustment, but rather a periodic compensation for each uplink time slot, denoted as . The unit is seconds, preset by the system frame structure, and a small time advance or lag is accumulated successively; a time delay compensation state variable is maintained for the i-th antenna point. , where n is the time slot number, n=0, 1, 2, ..., represents the time offset of the baseband signal at the i-th antenna point relative to the system reference timing when transmitting in the n-th downlink time slot. A positive value indicates early transmission, and a negative value indicates delayed transmission; the recursive update expression for this state variable is: ;in, This is a preset small step size adjustment factor, with a value range of [value range missing]. In this embodiment, it is preferably set to 0.05 to control the single compensation amplitude and avoid overshoot caused by prediction error; This is the instantaneous value of the delay change slope calculated based on the input parameters of the nth time slot; This is the compensation state quantity from the previous time slot, which is initialized to 0. Compensation state variables The digital up-conversion unit of the baseband transmit link, which sends high-speed control signaling to the i-th antenna point, is specifically implemented as follows: Before baseband signal transmission, the time-domain baseband sampling sequence is... Applying a cyclic time shift is equivalent to shifting the emission time globally over a continuous time domain. For sampling rate The baseband system achieves this cyclic time shift by adjusting the starting position of the Inverse Fast Fourier Transform (IFFT) output, with its discrete sampling point offset... It is given by the following formula: ;in, This is an integer representing the offset of the number of sampling points; This indicates the rounding operation; when At that time, it is equivalent to launching in advance, when At that time, it is equivalent to delayed launch; To ensure that all signals meet the time-domain conditions for coherent combining or interference cancellation at the target terminal, after each transmission compensation, the absolute propagation delay estimate of the signal from each distributed antenna point to the target terminal is calculated; the total delay of the signal from the i-th antenna point from the baseband transmission time to the target terminal reception time is also considered. It is given by the following formula: ;in, Let be the physical space propagation delay from the i-th antenna point to the target terminal, in seconds. It is calculated based on the estimated spatial location of the target terminal at the current moment, and its specific expression is: , The estimated spatial position vector of the target terminal in the nth time slot is obtained by jointly solving the spatial angle of arrival of each antenna point through the triangulation method; Also, calculate the arrival delay difference between any two antenna points i and j. The corresponding expression is: ; Define the global maximum arrival delay difference : Through monitoring Whether it converges within a preset time delay window, the time delay window is determined by the lower limit. and upper limit Definition, unit is seconds.

[0039] The key feature of this embodiment is that it does not simply align the delays of each path to the starting boundary of the cyclic prefix (CP). Instead, it designs a dynamically offset delay window whose starting position actively avoids the boundary region of the CP, in order to prevent delay jitter caused by channel time-varying factors from crossing the CP boundary and thus causing inter-symbol interference (ISI). The preset delay window parameters specifically include: ; ; in, The duration of the system cyclic prefix, in seconds, is determined by the communication standard. For example, in the Long Term Evolution (LTE) standard, the standard cyclic prefix length is [length missing]. Second; To protect the threshold, it is defined as follows: , The preset protection ratio factor is preferably set to 0.15 in this embodiment, which is used to establish a safety isolation zone between the start position of the delay window and the start boundary of the cyclic prefix; This is a dynamic offset that is adjusted in real time according to the current channel state. Its expression is: , The preset bias adjustment factor is preferably set to 0.5 in this embodiment. This represents the global maximum arrival delay difference for the current time slot. This represents the global maximum arrival delay difference in the previous time slot; It can be positive, negative or zero. Its physical meaning is that when the time delay difference is increasing, the time delay window is shifted backward as a whole to reserve a larger time delay redundancy space in advance. Conversely, it is shifted forward to reduce the timing lead of the system. The delay window width, in seconds, is set based on the remaining length of the loop prefix, and the expression is: To ensure that the upper limit of the delay window is always lower than This ensures a safe distance from the end boundary of the cyclic prefix; Furthermore, by executing the following decision logic, it is ensured that Dynamically converges within the aforementioned time delay window: ; If the above conditions are not met, it indicates that the current compensation state deviates from the target delay window, triggering the adaptive correction mechanism: when At that time, all antenna points Unified increase in increment In this embodiment, , The sampling period is This causes an overall delay in all signals, increasing the time delay difference; when At that time, all antenna points Unified reduction This advances the overall timing of all signals, reducing the time delay difference. Through the above closed-loop adjustment, the time delay difference is ultimately guaranteed to fall stably within the preset time delay window.

[0040] It is worth emphasizing that this module provides compensation values ​​for each antenna point i. They are independent of each other, and their differences arise from the different Doppler frequency shifts at each antenna point. and rate of change of spatial angle of arrival This differentiated compensation mechanism allows the time delay curves, which were originally drifting independently due to their dispersed physical locations, to converge into a unified time delay window after closed-loop adjustment. To prevent the independent compensation accumulation at each antenna point from exceeding the system's maximum allowable timing advance. The maximum timing advance is preset by the system frame structure, and a saturation limiting operation is performed after each update. ;in, and These represent the minimum and maximum value operations, respectively; after limiting. As the final downlink transmission time offset command, it is sent to the baseband processing unit of each distributed microstation for execution via the control interface.

[0041] For example, in a typical indoor hotspot deployment scenario, with an antenna spacing of 5 meters and a terminal moving along an arc at a speed of 3 km / h, without compensation, the time delay difference between each antenna point and the target randomly fluctuates between 0 and 0.5 microseconds, frequently crossing the CP boundary and causing inter-symbol interference; after introducing this time delay slope tracking compensator, the following settings are made: , , , After the 10th time slot, it converges to the interval between 0.7 microseconds and 4 microseconds, with the lower limit of 0.7 microseconds being greater than... The upper limit is less than 4 microseconds. It completely avoids the boundary region of the cyclic prefix, namely 0 to 0.7 microseconds and 4 to 4.69 microseconds, and the inter-symbol interference probability drops from 12 before compensation to below 0.1.

[0042] It should be noted that the delay slope tracking compensator is the core of the system's fine-grained closed-loop control in the time dimension. Its core function is to address the problem of differential drift in propagation delay of each link caused by the dispersed physical locations of distributed antenna points and the continuous movement of the target terminal. By predicting the unique delay change slope of each link in real time and performing independent advance / lag compensation for the baseband transmission time of each antenna point in microsecond-level steps, the multi-path arrival delay difference that originally diverged with time and terminal movement is dynamically brought together and locked within a carefully designed delay window, thereby creating precise time synchronization conditions for coherent merging or interference cancellation.

[0043] The reason why similar time delays cause frequent hard handovers and interference lies in the dynamic change of the time delay difference between links over time, which causes the symbol boundary of the synthesized signal to drift continuously. In this embodiment, the time delay slope is predicted by jointly using Doppler frequency shift and angle change rate, so that the compensation changes from a passive response to an active prediction, eliminating the inter-symbol interference caused by sudden time delay changes in traditional schemes and reducing the probability of link loss of synchronization by an order of magnitude. Conventional schemes simply control the time delay to 0 to However, this embodiment ignores the phase discontinuity problem at the cyclic prefix boundary, especially at the beginning of the cyclic prefix. This embodiment introduces a dynamic offset to shift the delay window to the middle region of the cyclic prefix, avoiding the vulnerable 15% boundary band at the beginning and end. Through the synergistic effect of the protection scaling factor and the offset adjustment factor, the delay window can automatically migrate in advance with the time-varying trend of the channel. It can be pre-calibrated before the delay difference exceeds the boundary, which greatly enhances the time synchronization stability of the system in high-speed terminal movement scenarios and effectively avoids the sensitive area of ​​the cyclic prefix boundary. The compensation amounts of each antenna point are independent of each other, allowing each antenna point to be adjusted individually according to its own geographical location and channel dynamic characteristics; while the unified convergence constraint of the global maximum delay difference ensures that all links eventually converge within the same delay window. This divide-and-conquer strategy effectively solves the contradiction between individual differences and overall coordination in a distributed system, and significantly improves the quality of multi-antenna synthesized signals. The limiting operation ensures that the cumulative compensation amount of each antenna point will not grow indefinitely and exceed the physical limit of the system's timing advance capability, avoiding hardware failures caused by excessive compensation leading to conflicts between the transmit frame and the uplink receive frame. This robust design enables the module to maintain stable operation in various extreme mobile scenarios.

[0044] In the several embodiments provided by this invention, it should be understood that the disclosed system can be implemented in other ways. For example, the embodiments of the invention described above are merely illustrative; for example, the division of modules is only a logical functional division, and there may be other division methods in actual implementation.

[0045] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; they may be located in one place or distributed across several network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0046] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module. The integrated module can be implemented in hardware or in the form of hardware plus software functional modules.

[0047] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the essential characteristics of the present invention.

[0048] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A communication coverage enhancement and optimization system based on multiple antennas, characterized in that, include: Signal holographic extraction module: Based on several distributed micro-station antenna points, it acquires the uplink detection signal of the target terminal and extracts the multi-dimensional feature information of the detection signal. The multi-dimensional feature information includes at least the multipath delay spectrum, spatial angle of arrival, and Doppler frequency shift. Coherent spectrum clustering discriminator: Calculates the coherent bandwidth parameters between the detection signals received by each antenna point, and determines whether the current channel state is coherent merging state or interference contention state based on the coherent bandwidth parameters; Logical port dynamic mapping module: Based on the channel state determination result, the antenna arrays physically distributed in different geographical locations are dynamically reconstructed into at least two types of logical antenna ports. The two types of logical antenna ports include: a first type of port for obtaining spatial multiplexing gain and a second type of port for constructing dynamic virtual nulls to actively cancel co-channel interference. Delay slope tracking compensator: Based on the rate of change of Doppler frequency shift and spatial angle of arrival, predict the slope of signal propagation delay change of each distributed antenna point relative to the same target terminal. Based on the slope of delay change, perform independent small-step closed-loop advance / lag compensation on the baseband transmission time of each distributed antenna point, so that the path delay difference of each path to the target terminal after compensation dynamically converges within the preset delay window, and the starting position of the delay window avoids the boundary region of the cyclic prefix. In order to ensure that each signal meets the time-domain conditions for coherent combining or interference cancellation at the target terminal, the absolute propagation delay estimate of the signal at each distributed antenna point to the target terminal is calculated after each transmission compensation. Design a dynamically biased delay window whose starting position actively avoids the boundary region of the cyclic prefix, preventing delay jitter caused by channel time-varying factors from crossing the cyclic prefix boundary and thus causing inter-symbol interference; the dynamic bias amount is adjusted in real time according to the current channel state, and its expression is: ; This is a dynamic bias value; This is the preset bias adjustment factor. This represents the global maximum arrival delay difference for the current time slot. This represents the global maximum arrival delay difference in the previous time slot.

2. The communication coverage enhancement and optimization system based on multiple antennas according to claim 1, characterized in that, The received signal is converted from analog to digital to obtain a discrete sampling sequence with the sampling period. The discrete sampling sequence is then subjected to a fast Fourier transform to convert it to the frequency domain, thus obtaining the frequency domain received signal. By performing point-by-point complex multiplication between the frequency domain sequence of the locally stored probe reference signal and the frequency domain received signal, channel frequency response estimation is performed to obtain the frequency domain channel estimate. Perform an inverse fast Fourier transform on the frequency domain channel estimate to transform it back to the time domain, obtaining the time domain channel impulse response sequence of the i-th antenna point; calculate the squared magnitude of the sequence and coherently accumulate it along the time dimension to obtain the multipath power delay spectrum.

3. The communication coverage enhancement and optimization system based on multiple antennas according to claim 1, characterized in that, By utilizing the relative spatial phase difference between distributed antenna points, the main reference antenna point with the strongest received signal-to-noise ratio is selected as the phase reference point, and the relative phase difference between the i-th antenna point and the main reference point is calculated. By simultaneously solving the phase difference equations of all antenna points, an overdetermined set of equations is formed. Then, a two-dimensional spectral peak search is performed on the direction vector of the angle of arrival using a multi-signal classification algorithm based on eigenvalue decomposition, and finally the spatial angle of arrival is calculated.

4. The communication coverage enhancement and optimization system based on multiple antennas according to claim 1, characterized in that, Using the channel estimation results of several consecutive uplink time slots j, j=1,2,…,J, where J is the total number of uplink time slots, for the strongest principal path of the i-th antenna point, while tracking the change of its complex amplitude on the time axis, the autocorrelation value of the principal path complex amplitude between the j-th time slot and the (j-1)-th time slot is calculated. The phase difference average of the autocorrelation results over J consecutive time slots is taken, and the Doppler frequency shift is calculated using the following expression. : ;in, This indicates taking the principal value of the complex phase angle, with the range limited to... between; This is the phase-frequency conversion factor, representing the time... Within, the frequency is The number of radians the signal experiences; This is the time interval between adjacent uplink time slots.

5. The communication coverage enhancement and optimization system based on multiple antennas according to claim 1, characterized in that, For each antenna point, calculate the root mean square delay spread of that antenna point based on its own power delay spectrum; Based on the inverse relationship between channel coherence bandwidth and root mean square delay spread, the local coherence bandwidth estimate for each antenna point is calculated using the root mean square delay spread.

6. A communication coverage enhancement and optimization system based on multiple antennas according to claim 5, characterized in that, When performing time-domain discreteness discrimination, extract the global minimum delay and global maximum delay detected by all distributed antenna points, and calculate the global maximum relative delay difference; In addition, the system reads the preset loop prefix length parameter and presets the first time domain threshold and the second time domain threshold; When implementing joint state machine determination, the global average correlation coefficient is calculated by integrating the pairwise correlation coefficients between all antenna points; the state determination is performed according to the logical rule group, which includes determination rule A, determination rule B, and determination rule C; determination rule A corresponds to the coherent merging state triggering condition, determination rule B corresponds to the interference contention state triggering condition, and determination rule C corresponds to the transition state maintenance mechanism.

7. The communication coverage enhancement and optimization system based on multiple antennas according to claim 1, characterized in that, To avoid frequent discriminator state jumps caused by sudden noise or momentary occlusion, the coherent spectrum clustering discriminator incorporates a first-order hysteresis filter: before each decision update, the globally average correlation coefficient calculated in this iteration is used as the filter. Global average correlation coefficient with the previous time step For smoothing, the corresponding expression is: ;in, This is the global average correlation coefficient after first-order lag filtering; This is a preset forgetting factor, with a value range of (0,1).

8. The communication coverage enhancement and optimization system based on multiple antennas according to claim 1, characterized in that, When implementing the second type of logic port mapping under interference contention state, when determining the status flag bit At that time, the interference suppression mapping strategy is enforced.

9. A communication coverage enhancement and optimization system based on multiple antennas according to claim 1, characterized in that, To avoid frequent switching between coherent merging state and interference contention state of the logic port, a minimum dwell time timer is set inside the logic port dynamic mapping module; During the timer's operation, even the status flag bit output by the coherent spectrum clustering discriminator... If changes occur, the current mapping configuration remains unchanged. Only after the timer times out and the status flag remains stable in the new state for more than two consecutive measurement cycles will a new round of logic port remapping calculation be triggered, thereby effectively suppressing system oscillations.

Images