A Non-Line-of-Sight Channel Modeling Method for Intelligent Metasurface Wireless Communication in Coal Mines

Abstract

The present invention provides a non-line-of-sight channel modeling method for intelligent metasurface wireless communication in coal mines. This modeling method can be used for feature recognition and extraction of complex multipath fading and intelligent metasurface passive relay in the confined space of coal mines, as well as obtaining parameters such as path loss, scatter cluster distribution, and multipath delay. Based on actual channel measurement data and the propagation scenario space reconstruction method, the environmental structure in the original propagation scenario is used to establish a corresponding digital map of scatterers through position, texture, and reflection characteristics, and the intelligent metasurface is equivalent to a virtual scatterer, which can effectively reduce the difficulty of analysis and modeling. Then, the actual scatterers are matched with the multipath clustering results to form cluster cores, and the non-line-of-sight channel modeling process is transformed into the analysis and modeling of multiple direct paths, which can effectively meet the modeling requirements of the wireless communication system in coal mines and effectively reduce the complexity and difficulty of the modeling process.

Description

Technical Field

[0001] The present invention relates to the field of wireless communication in coal mines, and particularly to a non-line-of-sight channel modeling method for intelligent metasurface wireless communication in coal mines. Background Art

[0002] Non-line-of-sight propagation scenarios such as roadway turns, branch roads, chambers, and equipment blockages are common in coal mines, resulting in difficult edge coverage of communication cells and wireless signal coverage blind spots and dead spots. Moreover, in more important and critical production sites such as mining faces and panels, the spatial dynamic changes, equipment movement blockages, dust, water mist and other interferences are more obvious, the effective and reliable coverage of wireless signals is more difficult, and the impact of blind spots and dead spots is more serious. The traditional solutions mainly include adding relay base stations or introducing leaky cable modes, but this will bring new problems such as high construction and usage costs, difficult maintenance and usage, signal coverage overlapping interference, and frequent handovers between mobile terminals and base stations. The latter also leads to problems such as complex system structures. In recent years, the emergence of intelligent metasurfaces has made it possible to artificially reshape wireless channels, showing great potential in improving wireless transmission and coverage capabilities in non-line-of-sight propagation scenarios, attracting wide attention, and being listed as one of the key alternative technologies for next-generation mobile communication networks. In addition, as passive devices, intelligent metasurfaces have low power consumption and cost, and are easier to deploy, especially suitable for coal mine underground scenarios with difficult power supply and strict explosion-proof requirements for active devices. This provides a new approach and new idea for solving wireless coverage in non-line-of-sight propagation scenarios in coal mines.

[0003] The existing research on intelligent metasurfaces mainly focuses on classical problems such as channel estimation, modeling, beamforming, and deployment strategies. The introduction of a large-scale electromagnetic unit array in intelligent metasurfaces may make the near-field characteristics more obvious, and at the same time, it will also cause changes in the number and paths of propagation links, forming an intelligent metasurface coupling channel with a brand-new mode. However, at present, the research on the wireless propagation characteristics of intelligent metasurface coupling channels is not sufficient. Generally, the Rayleigh channel model or the standardized wireless channel models defined by 3GPP and ITU are directly used, and the phase response of the intelligent metasurface is simplified to a diagonal matrix. These methods can effectively simplify the model, but they do not fully consider the propagation link changes caused by the introduction of intelligent metasurfaces and the interaction between the new links and the original links. Especially in the confined space of coal mines with severe multipath characteristics, the difference between the above simplified model and the actual propagation characteristics is more obvious, affecting the application effect of intelligent metasurfaces. At present, there are few research results on the intelligent metasurface coupling channel in actual scenarios. The existing work generally follows classical methods such as ray tracing method and multipath clustering theory, and establishes theoretical models around large-scale fading or simple multipath channels in free space, generally only involving the far-field propagation characteristics of single-carrier narrowband signals. The existing research and practice equivalent the intelligent metasurface to an ideal reflector or a virtual scatterer cluster, and establish a free-space path loss model without considering multipath effects and shadow effects. Or use classical methods such as ray tracing method and multipath clustering theory to establish a free-space channel model with multipath effects, but greatly simplify the number of multipaths and the characteristics of scatterers, and ignore the multipath effects between the intelligent metasurface and the receiving end. In the confined space of coal mines, there are widespread roadway turns and bifurcations, chambers, large equipment and facilities blocking, and the shadow effects and multipath effects are more significant and complex. Signals on different paths will have large differences after being reflected by the intelligent metasurface due to the difference in the angle of arrival. The existing research results based on free space are difficult to accurately reflect the complex real channel state in coal mines, and the measurement and modeling of intelligent metasurface coupling channels in coal mines are still blank areas. Summary of the Invention

[0004] Object of the Invention: The object of the present invention is to provide a channel modeling method for the intelligent metasurface wireless communication system in coal mines, which is used for the feature recognition and extraction of complex multipath fading and intelligent metasurface passive relay in the confined space of coal mines, as well as the acquisition of parameters such as path loss, scatterer cluster distribution, and multipath delay, so as to eliminate or improve one or more defects existing in the prior art.

[0005] The present invention provides a non-line-of-sight channel modeling method for intelligent metasurface wireless communication in coal mines, including:

[0006] Step 1: According to the existing or future required wireless communication system in the coal mine underground environment to be measured, select the parameters of the channel measurement platform for the multiple-input multiple-output intelligent metasurface wireless communication system, and build the transmitter, intelligent metasurface, and receiver of the channel measurement platform in the ground conventional environment;

[0007] Step 2: Determine and calibrate the characteristics of the channel measurement platform itself;

[0008] Step 3: In the coal mine underground wireless communication scenario to be measured, deploy the channel measurement platform that has been determined and calibrated in the ground conventional environment, conduct on-site channel measurements, and control the working states of the transmitting antenna array, intelligent metasurface array, and receiving antenna array through the antenna high-speed switching module. Based on the time-division multiple access principle, sequentially measure the non-line-of-sight channel impulse responses between each transmitting antenna unit, each intelligent metasurface array unit, and each receiving antenna unit, and store them in the data storage unit for later data processing and analysis;

[0009] Step 4: In the coal mine underground wireless communication scenario to be measured, deploy a lidar, and use the Simultaneous Localization and Mapping (SLAM) technology to perform three-dimensional image scanning and reconstruction of the channel measurement site, and collect and calibrate the positions, texture features, and reflection characteristics of objects such as large equipment, metal protection nets, rock masses, and walls;

[0010] Step 5: After completing the on-site measurement of the coal mine underground wireless communication scenario, perform image semantic segmentation on the SLAM measurement data and results through computer vision methods on the ground, match them with each object on-site to form a scatterer digital map, and according to the previously determined characteristics of the intelligent metasurface array, equivalent the intelligent metasurface to a virtual scatterer and insert it into the scatterer digital map;

[0011] Step 6: After completing the on-site measurement of the coal mine underground wireless communication scenario, use the sliding correlation method on the ground to obtain the complete channel impulse response matrix of the multiple-input multiple-output intelligent metasurface wireless communication system and form a measurement result data set. Through the SAGE algorithm, obtain channel parameters from the measurement result data set, such as signal amplitude attenuation, multipath delay, horizontal departure angle, vertical departure angle, horizontal arrival angle, vertical arrival angle, complex polarization matrix, and Doppler frequency shift, etc. Then, perform multipath clustering through the K-nearest neighbor clustering algorithm to determine parameters such as the number of scatterer clusters, intra-cluster delay, and intra-cluster angular spread;

[0012] Step 7: Combine the multipath clustering results with the scatterer map, and through reinforcement learning, match the scatterer clusters with the scatterers to obtain a finite number of multipath scatterer clusters and an intelligent metasurface equivalent cluster core;

[0013] Step 8: Taking the cluster core as a node, the non-line-of-sight propagation link through which the wireless signal passes is decomposed into multiple logical sub-channels. By connecting the logical sub-channels to form a propagation path, an effective propagation path with the connected nodes being the transmitting antenna unit, the cluster core, and the receiving antenna unit is obtained. If the end point of the propagation path is not the receiving antenna unit, it is an invalid propagation path that can be excluded.

[0014] Step 9: Merge all the effective propagation paths to obtain the channel impulse response from the transmitter to the receiver, and obtain the non-line-of-sight channel model for intelligent metasurface wireless communication in the current coal mine underground scenario.

[0015] In Step 1, the transmitter, intelligent metasurface, and receiver of the channel measurement platform consist of a transmitting antenna array, a receiving antenna array, an intelligent metasurface, a synchronous clock, a signal generator, a signal receiver, a data storage unit, and a control terminal. Among them, the parameters of the channel measurement platform include transmitting signal parameters, transmitting antenna array parameters, intelligent metasurface parameters, and receiving antenna array parameters. The transmitting signal parameters include the frequency range to be measured, the transmitting signal power level, and the transmitting signal type. The parameters of the transmitting antenna array, intelligent metasurface, and receiving antenna array include the number, spacing distance, position, and orientation of the antenna elements or array elements.

[0016] In Step 2, the self-characteristics of the channel measurement platform include measuring and calibrating the antenna feed transmission power error, the high-frequency coaxial cable transmission loss, the adapter insertion loss, the transmitting antenna array, the intelligent metasurface array, the receiving antenna array, and the system response error of other instruments and equipment. Among them, the measurement and calibration method uses the through-reflect-line calibration method, the short-open-load-through calibration method, the through-reflect-match calibration method, or the through-open-short-match calibration method.

[0017] In Step 3, the antenna high-speed switching module is used to achieve time-division multiple access between the transmitting antenna array, the uniform rectangular RIS array, and the receiving antenna array. Among them, the transmitting antenna array has M T transmitting antenna elements, the receiving antenna array has M R receiving antenna elements, and the RIS array has MN RIS array elements. M and N are the numbers of RIS array elements on the long side and the short side of the rectangle respectively. In each time slot, only one combination of transmitting antenna elements, RIS array elements, and receiving antenna elements is measured. The non-line-of-sight channel impulse response between each transmitting antenna element, each intelligent metasurface array element, and each receiving antenna element is measured in sequence. After traversing all combinations, a channel sampling snapshot is obtained, which specifically includes:

[0018] Define the following activation time function

[0019]

[0020] The activation time function has an activation time range from 0 to time and is used to control whether to activate the transmit antenna unit, intelligent metasurface unit or receive antenna unit at time t, where 1 means activation and 0 means non-activation;

[0021] For the l-th channel sampling snapshot, according to the definition of the activation time function, the activation time function of the p-th transmit antenna unit the activation time function of the q-th receive antenna unit and the activation time function of the r-th RIS array unit are respectively:

[0022]

[0023]

[0024]

[0025] where T T is the activation time of a single transmit antenna unit, T R is the activation time of a single receive antenna unit, T RIS is the activation time of a single RIS array unit, and the period for traversing all combinations of transmit antenna units, RIS array units and receive antenna units is T cycle , M S is the number of channel sampling snapshots;

[0026] The transmitted signal u(t) is expressed as:

[0027]

[0028] where s(t) is a PN sequence signal, and the vectors and are the activation time functions of the first transmit antenna unit, the second transmit antenna unit and the M T -th transmit antenna unit respectively;

[0029] The signal after being reflected by the RIS is:

[0030]

[0031] where Φ is the RIS array response matrix, is the channel transfer matrix from the transmit antenna array to the RIS array, is the propagation delay, the superscript T is the matrix transpose operator, and the vector and are the activation time functions of the first RIS array element, the activation time function of the second RIS array element, and the activation time function of the MN-th RIS array element, respectively;

[0032] The received signal vector y(t) at the receive antenna array is:

[0033]

[0034] where, is the channel transfer matrix from the RIS array to the receive antenna array, where is the propagation delay, H NLoS (t,τ NLoS ) is the NLOS channel transfer matrix from the transmit antenna array to the receive antenna array, where τ NLoS is the transmission delay, and n(t) is a complex Gaussian white noise vector;

[0035] Finally, the received signal Y(t) obtained at each measurement is:

[0036]

[0037] where, the vectors and are the activation time functions of the first receive antenna element, the activation time function of the second receive antenna element, and the activation time function of the M R -th receive antenna element, respectively.

[0038] Step 6 includes:

[0039] The wireless signal arrives at the receive antenna array in the form of clusters. In the n-th cluster component, the channel h p,q,n (t) from the p-th transmit antenna element to the q-th receive antenna element is expressed as:

[0040]

[0041] where, M c represents the number of sub-paths in the n-th cluster component, and α n,m , Ψ n,m , v n,m and τ n,m represent the amplitude attenuation, complex polarization matrix, Doppler frequency shift, and propagation delay corresponding to the m-th sub-path in the n-th cluster component, respectively. Ω Tx,n,m ={θ Tx,n,m ,φTx,n,m} and Ω Rx,n,m = {θ Rx,n,m , φ Rx,n,m} respectively represent the angle information of the transmitting antenna array and the receiving antenna array, θ Tx,n,m and φ Tx,n,m respectively represent the elevation angle and the horizontal angle of the transmitting antenna array corresponding to the m-th sub-path in the n-th cluster component, θ Rx,n,m and φ Rx,n,m respectively represent the elevation angle and the horizontal angle of the receiving antenna array corresponding to the m-th sub-path in the n-th cluster component, Γ Tx,p (Ω Tx,n,m ) and Γ Rx,q (Ω Rx,n,m ) are the responses of the transmitting antenna array and the receiving antenna array respectively, exp is the natural exponential function, and j is the imaginary unit;

[0042] Define the parameter set of the -th sub-path received as: Define as:

[0043]

[0044] Among them, and respectively represent the corresponding amplitude attenuation, complex polarization matrix, Doppler frequency shift and propagation delay of the -th sub-path, and respectively represent the angle information of the transmitting antenna array and the receiving antenna array, and respectively represent the elevation angle and the horizontal angle of the transmitting antenna array corresponding to the -th sub-path, and respectively represent the elevation angle and the horizontal angle of the receiving antenna array corresponding to the -th sub-path, the symbol indicates that takes any value within its value range;

[0045] For the known PN sequence s(t) and the measured received signal Y(t), the likelihood function determined by the parameter set is The SAGE algorithm iteratively solves the parameter set to find a set of multi-path clustering parameters that maximize the likelihood function L. In each iteration, only a single parameter in the set is estimated, and the remaining parameters remain unchanged. After the solution is completed, the single parameter is updated and substituted into the next round of iteration for solving other parameters. For the k-th iteration, one of the iteration update orders is:

[0046]

[0047]

[0048]

[0049]

[0050]

[0051]

[0052] Among them, the function argmax is to find the variable value that maximizes the function immediately to its right, and the symbol below the function argmax is the variable to be adjusted. and represent the values at the k-th and (k - 1)-th iterations respectively and ;

[0053] The parameter sets of each sub-path are estimated through the SAGE algorithm and then form a feature vector The feature vectors of each sub-path are clustered in the vector space through the K-nearest neighbor clustering algorithm to achieve clustering, determine the number of clusters and the sub-paths within the clusters.

[0054] The channel parameter indicators obtained from the measurement result dataset through the SAGE algorithm in step 6 include:

[0055] Signal amplitude attenuation, multipath delay, horizontal departure angle, vertical departure angle, horizontal arrival angle, vertical arrival angle, complex polarization matrix, Doppler frequency shift, number of clusters and sub-paths within the clusters.

[0056] In step 7, the environment of the reinforcement learning is a wireless channel containing direct, first-order reflection and second-order reflection, the action is to randomly match the multipath scattering clusters and scatterers, and the reward is the reciprocal of the Frobenius norm of the difference between the channel impulse response matrix generated in the environment based on the ray tracing method and the actual channel impulse response matrix obtained in step 6.

[0057] In step 7, combining the multipath clustering result with the scatterer map, the scattering clusters and scatterers are matched with each other through reinforcement learning to obtain M C multipath scatterer cluster cores and 1 RIS equivalent cluster core, and define the multipath scatterer cluster core set as where C1, C2 and are the first multipath scatterer cluster core, the second multipath scatterer cluster core and the M-th c multipath scatterer cluster core respectively;

[0058] The environment of reinforcement learning is set as a wireless channel with direct, first - order reflection, and second - order reflection. Given the positions and characteristics of the transmitter, receiver, and scatterers, the channel impulse response matrix \(H\) from the transmitter to the receiver is calculated using the ray - tracing method. RT The action of reinforcement learning is to randomly match multipath scattering clusters and scatterers to form cluster cores. The position of the cluster core corresponds to the position of the scatterer in the scatterer digital map, and the characteristics of the cluster core correspond to the parameters of the multipath scattering cluster. Combining the characteristics of the transmitter and receiver obtained in Step 1 and Step 2, substitute the position and characteristics of the cluster core into the environment of reinforcement learning, and use the ray - tracing method to obtain the channel impulse response matrix \(H\) of the wireless channel. RT The reward of reinforcement learning is set as the reciprocal of the Frobenius norm of the difference between the channel impulse response matrix generated in the environment based on the ray - tracing method and the actual channel impulse response matrix \(H\) obtained in Step 6. true Calculated according to the formula \(\left\|\left|H\right|\right|\) RT -\(H\) true \(\left\|\right\|\) -1 , where the symbol \(\left\|\left|\cdot\right|\right|\) represents the Frobenius norm, and the superscript \(- 1\) represents the reciprocal.

[0059] Step 8 includes:

[0060] For the non - line - of - sight (NLOS) link without RIS reflection, the NLOS channel from the \(p\) - th transmit antenna element to the \(q\) - th receive antenna element Is successively split into the following line - of - sight (LOS) logical sub - channels according to the cluster cores:

[0061]

[0062] where \(Tx\) (p) represents the \(p\) - th transmit antenna element, \(Rx\) (q) represents the \(q\) - th receive antenna element, \(\rightarrow\) represents the signal propagation path, is the set of multipath scatterer cluster cores obtained in Step 7, and are the \(i\) - th and \(j\) - th multipath scatterer cluster cores in this set; and ;

[0063] The channel after RIS reflection is divided into a virtual line - of - sight channel that only passes through the RIS The NLOS channel with scatterers before RIS reflection The NLOS channel with scatterers after RIS reflection And the NLOS channel with scatterers before and after RIS reflection Is successively split into the following line - of - sight (LOS) logical sub - channels according to the cluster cores:

[0064] Tx (p) →RIS→Rx (q) ,

[0065]

[0066]

[0067]

[0068] Substitute the parameters obtained in Steps 6 and 7, including signal amplitude attenuation, multipath time delay, horizontal departure angle, vertical departure angle, horizontal arrival angle, vertical arrival angle, complex polarization matrix, Doppler frequency shift, number of scattering clusters, intra-cluster time delay, intra-cluster angular spread, multipath scatterer cluster core, and RIS equivalent cluster core, into the split logical sub-channels, and calculate the channel impulse response of each logical channel in sequence to obtain five channels their respective channel impulse responses and

[0069] Step 9 includes:

[0070] At the transmission time t, after experiencing the propagation delay τ, the complex channel impulse response h from the p-th transmit antenna element to the q-th receive antenna element p,q (t,τ) is expressed as:

[0071]

[0072] The corresponding multiple-input multiple-output MIMO intelligent surface RIS wireless communication system channel model is expressed as an M R ×M T complex matrix H(t,τ):

[0073]

[0074] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0075] The present invention establishes a corresponding digital map of scatterers for the environmental structure in the original propagation scenario through position, texture, and reflection characteristics, and equates the intelligent surface to a virtual scatterer, which can effectively reduce the difficulty of analysis and modeling. Then, the actual scatterers are matched with the multipath clustering results to form cluster cores, and the non-line-of-sight channel modeling process is transformed into the analysis and modeling of multiple direct paths, which can effectively meet the modeling requirements of the wireless communication system in coal mines and effectively reduce the complexity and difficulty of the modeling process. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] The following further specifically describes the present invention in conjunction with the accompanying drawings and specific embodiments, and the above and / or other advantages of the present invention will become clearer.

[0077] Figure 1 This is the flowchart of channel modeling based on cluster core formation in the present invention.

[0078] Figure 2 This is the schematic diagram of the channel measurement platform and process in the present invention.

[0079] Figure 3 This is the schematic diagram of the non-line-of-sight propagation scenario of the intelligent metasurface wireless communication system in coal mines in the present invention.

[0080] Figure 4 This is the signal timing logic diagram of the channel measurement platform in the present invention. Specific Embodiments

[0081] The following is the definition table of abbreviations and key terms used in the present invention:

[0082] Table 1

[0083]

[0084] As Figure 1 shown, the present invention provides a non-line-of-sight channel modeling method for intelligent metasurface wireless communication in coal mines, mainly including three stages: platform construction and ground tests, actual measurements in coal mines, and post-measurement data processing and channel modeling.

[0085] In a conventional ground environment, construct a channel measurement platform for the multiple-input multiple-output (MIMO) intelligent metasurface (RIS) wireless communication system in coal mines as Figure 2 shown, and conduct platform determination, calibration, and tests. The main steps include:

[0086] Step 1: According to the existing or future required wireless communication system in the coal mine environment to be measured, select the parameters of the channel measurement platform for the MIMO RIS wireless communication system, and construct the transmitter, intelligent metasurface, and receiver of the channel measurement platform as Figure 2 shown to form a Sounder system for channel measurement of the MIMO RIS wireless communication system in coal mines.

[0087] Figure 2The shown channel measurement platform mainly consists of a transmitter (Tx), a reconfigurable intelligent surface (RIS), and a receiver (Rx). The transmitter includes, but is not limited to, a transmitter control terminal, a transmitter synchronization clock, a transmitter signal generator, and a transmitting antenna array. The RIS includes, but is not limited to, a RIS control terminal, a RIS synchronization clock, a high-speed antenna switching module of the RIS, and a RIS array. The receiver includes, but is not limited to, a receiver control terminal, a receiver synchronization clock, a receiver signal receiver, a high-speed antenna switching module of the receiver, a receiving antenna array, and a data storage unit. All devices and components of the channel measurement platform need to meet the regulations on the safety of electrical equipment in underground coal mines. The control terminals of the transmitter, RIS, and receiver can use intrinsically safe portable computers that have passed safety certification, which are used to observe and control the operation of each device and component, and can also serve as data storage units at the same time. Compared with the open environment on the ground, the enclosed environment underground leads to severe refraction and reflection, thus generating a large amount of multipath information and incident waves in all directions. The transmitting antenna array can use a uniform planar antenna array, the RIS array can use a uniform rectangular planar array, and the receiving antenna array can use an omnidirectional antenna array composed of eight planes. Since it is not convenient to lay and plug in synchronization cables in the underground coal mine working environment, and when conducting channel measurement, the transmitter, RIS, and receiver need to ensure time synchronization, a rubidium atomic clock can be used as the synchronization clock for the transmitter, RIS, and receiver. Because special signal sequences and waveforms are required for channel measurement and the frequency range to be measured is large, two universal software radio peripherals (USRPs) can be used to implement the transmitter signal generator and the receiver signal receiver respectively. The frequency range and power of the signal transmitted by the transmitter are set according to the operating frequency of the existing or future wireless communication system in underground coal mines. The transmitted signal uses a pseudo-random (PN) sequence and is generated by the transmitter signal generator.

[0088] The parameters of the channel measurement platform include transmitted signal parameters, transmitting antenna array parameters, RIS parameters, and receiving antenna array parameters. The transmitted signal parameters include the frequency range to be measured, the power of the transmitted signal, and the type of the transmitted signal. The parameters of the transmitting antenna array, RIS, and receiving antenna array include the number, spacing, position, and orientation of antenna elements or array elements.

[0089] The method for parameter selection and the construction method of the channel measurement platform are mainly realized through existing technologies and will not be elaborated here.

[0090] Step 2: Determine and calibrate the characteristics of the channel measurement platform itself. Measure on the ground the antenna feed transmission power error, the transmission loss of the high-frequency coaxial cable between each interface and terminal, and the insertion loss of the adapter. In the microwave anechoic chamber environment, measure the overall frequency response of the system, the frequency response of the transmitting antenna array, the frequency response of the intelligent metasurface array, and the frequency response of the receiving antenna array. Calibrate the system and record the error characteristics and obvious interference characteristics to facilitate eliminating the error and interference effects of the measurement platform from the measurement results. The determination and calibration methods use the through-reflection-transmission line calibration method, the short-open-load-through calibration method, the through-reflection-match calibration method, or the through-open-short-match calibration method. The detailed steps of determination and calibration are mainly achieved through existing technologies and will not be elaborated here.

[0091] In a coal mine underground NLOS wireless communication scenario as Figure 3 shown, deploy the measured and calibrated channel measurement platform and SLAM device, and conduct on-site channel measurement and scenario information collection. Among them, the transmitting antenna array has a total of M T transmitting antenna elements, the uniform rectangular planar RIS array has a total of M×N elements, and the receiving antenna array has a total of M R receiving antenna elements, thus forming a MIMO RIS wireless communication system. Due to obstacles such as bends and large equipment blocking, there is no LOS path between the transmitting antenna array and the receiving antenna array. The wireless signal forms an NLOS path through the reflection and scattering of a large number of scatterer clusters C1, C2... C i . At the same time, since there is a RIS array in the scenario, the signal reflected by the RIS array will form a VLOS path. The steps of channel measurement and scenario information collection include:

[0092] Step 3: In order to accurately and independently obtain the characteristics of each sub-channel of the MIMO RIS wireless communication system, use the antenna high-speed switching module to achieve time-division multiple access between the transmitting antenna array, the RIS array, and the receiving antenna array. Measure only one combination of transmitting antenna elements, RIS array elements, and receiving antenna elements in each time slot. The measurement process follows the signal timing logic of the channel measurement platform as Figure 4 shown. Among them, T T is the activation time of a single transmitting antenna element, T R is the activation time of a single receiving antenna element, T RIS is the activation time of a single RIS array element, and the period for traversing all combinations of transmitting antenna elements, RIS array elements, and receiving antenna elements is T cycle . After execution Figure 4After the signal timing logic of the channel measurement platform shown, the channel measurement platform will traverse all combinations of transmit antenna units, RIS array units, and receive antenna units to form a channel sampling snapshot. The actual measurement process needs to repeatedly execute the above process to obtain multiple channel sampling snapshots.

[0093] During the measurement process, the signal generator at the transmitter generates a pseudo-random sequence, which is processed through modulation, AD conversion, up-conversion module, etc. Finally, the antenna high-speed switching module selects the corresponding transmit antenna unit for transmission. The wireless signal propagates in space and is reflected by the RIS array unit selected by the antenna high-speed switching module, reaches the receive antenna unit selected by the antenna high-speed switching module, and then in the signal receiver at the receiving end, after down-conversion, demodulation, and DA conversion, the non-line-of-sight channel impulse responses between each transmit antenna unit, each intelligent metasurface array unit, and each receive antenna unit are measured in sequence and stored in the data storage unit for later data processing and analysis.

[0094] Define the activation time function:

[0095]

[0096] The activation time range of this activation time function is from 0 to the moment The purpose is to control whether to activate the transmit antenna unit, intelligent metasurface unit, or receive antenna unit at time t. 1 indicates activation, and 0 indicates non-activation.

[0097] For the l-th channel sampling snapshot, according to the definition of the activation time function, the activation time functions of the p-th transmit antenna unit, the q-th receive antenna unit, and the r-th RIS array unit are respectively:

[0098]

[0099]

[0100]

[0101] where M S is the number of channel sampling snapshots.

[0102] Therefore, the transmitted signal can be expressed as:

[0103]

[0104] where s(t) is the PN sequence signal, and the vectors and are the first transmit antenna unit, the second transmit antenna unit, and the M TActivation time function of the transmitting antenna units.

[0105] The signal after being reflected by the RIS is:

[0106]

[0107] where Φ is the RIS array response matrix, is the channel transfer matrix from the transmitting antenna array to the RIS array, where is the propagation delay, and the vectors and are the activation time functions of the first RIS array unit, the second RIS array unit, and the MN-th RIS array unit, respectively;

[0108] The signal vector received by the receiving antenna array is:

[0109]

[0110] where, is the channel transfer matrix from the RIS array to the receiving antenna array, where is the propagation delay, H NLoS (t,τ NLoS ) is the NLOS channel transfer matrix from the transmitting antenna array to the receiving antenna array, where τ NLoS is the transmission delay, and n(t) is a complex Gaussian white noise vector.

[0111] Finally, the received signal obtained in each measurement is:

[0112]

[0113] where the vectors are the activation time functions of the first receiving antenna unit, the second receiving antenna unit, and the M R th receiving antenna unit, respectively.

[0114] Step 4: Use lidar and Simultaneous Localization and Mapping (SLAM) technology to perform 3D image scanning and reconstruction of the channel measurement site, and collect and calibrate the positions, texture features, and reflection characteristics of objects such as large equipment, metal protection nets, rock masses, and walls. This step is mainly implemented through existing technologies and will not be elaborated here.

[0115] After completing the measurement in the coal mine shaft, perform post-processing of the data and channel modeling on the ground. The main steps include:

[0116] Step 5: Use SLAM measurement data and technology to perform 3D image scanning and reconstruction on the channel measurement site. Then, through computer vision methods, perform image semantic segmentation on the SLAM measurement data and results, and match them with each object on site to form a scatterer digital map. According to the characteristics of the intelligent metasurface array measured previously, the intelligent metasurface is equivalent to a virtual scatterer and inserted into the scatterer digital map. The specific steps are mainly realized through existing technologies and will not be elaborated here.

[0117] Step 6: For the wireless channel data measured through the Figure 2 platform shown, use the sliding correlation method to obtain the complete channel impulse response matrix H of the MIMO RIS wireless communication system true , and form a measurement result data set. Then, regard the RIS as a special scatterer cluster, and through the SAGE algorithm, obtain channel parameters from the measurement result data set, such as signal amplitude attenuation, multipath delay, horizontal departure angle, vertical departure angle, horizontal arrival angle, vertical arrival angle, complex polarization matrix, and Doppler frequency shift, etc. Then, perform multipath clustering through the K-nearest neighbor clustering algorithm to determine parameters such as the number of scatterer clusters, intra-cluster delay, and intra-cluster angular spread.

[0118] The wireless signal arrives at the receiving antenna array in the form of clusters. In the nth cluster component, the channel from the pth transmitting antenna element to the qth receiving antenna element can be expressed as:

[0119]

[0120] where M c represents the number of sub-paths in the nth cluster component, α n,m , Ψ n,m , v n,m and τ n,m respectively represent the amplitude attenuation, complex polarization matrix, Doppler frequency shift, and propagation delay corresponding to the mth sub-path in the nth cluster component. Ω Tx,n,m ={θ Tx,n,m , φ Tx,n,m} and Ω Rx,n,m ={θ Rx,n,m , φ Rx,n,m} respectively represent the angle information of the transmitting antenna array and the receiving antenna array. θ Tx,n,m and φ Tx,n,m respectively represent the elevation angle and horizontal angle of the transmitting antenna array corresponding to the mth sub-path in the nth cluster component. θ Rx,n,m and φ Rx,n,m respectively represent the elevation angle and horizontal angle of the receiving antenna array corresponding to the mth sub-path in the nth cluster component. Γ Tx,p (Ω Tx,n,m ) and Γ Rx,q (Ω Rx,n,mare the transmitting antenna array response and the receiving antenna array response respectively, exp is the natural exponential function, and j is the imaginary unit.

[0121] Therefore, the parameter set of the th sub-path received is defined as:

[0122]

[0123] where and represent the corresponding amplitude attenuation, complex polarization matrix, Doppler frequency shift, and propagation delay of the th sub-path respectively, and represent the angle information of the transmitting antenna array and the receiving antenna array respectively, and represent the elevation angle and the horizontal angle of the th sub-path corresponding to the transmitting antenna array respectively, and represent the elevation angle and the horizontal angle of the th sub-path corresponding to the receiving antenna array respectively. The symbol means that takes any value within its value range.

[0124] For the known PN sequence s(t) and the measured received signal Y(t), the likelihood function determined by the parameter set is The SAGE algorithm solves the parameter set by multiple iterations to find a set of multipath clustering parameters that maximize the likelihood function L. At each iteration, only a single parameter in the set

[0125]

[0126]

[0127]

[0128]

[0129]

[0130]

[0131] Among them, the function argmax is to find the variable value that maximizes the function immediately to its right. The symbol below the function argmax is the variable to be adjusted. and represent the values at the k-th and (k - 1)-th iterations respectively and ;

[0132] The parameter sets of each sub-path are estimated by the SAGE algorithm, and then the feature vectors are formed The feature vectors of each sub-path are clustered in the vector space by the K-nearest neighbor clustering algorithm to achieve clustering, determine the number of clusters and the sub-paths within the clusters.

[0133] Step 7: Combine the multi-path clustering results with the scatterer map, and through reinforcement learning, match the scatterer clusters with the scatterers to obtain M C multi-path scatterer cluster cores and 1 RIS equivalent cluster core, and define the multi-path scatterer cluster core set as where C1, C2, and are the first multi-path scatterer cluster core, the second multi-path scatterer cluster core, and the M c -th multi-path scatterer cluster core respectively.

[0134] The environment of reinforcement learning is set as a wireless channel containing direct, first-order reflection, and second-order reflection. Given the positions and characteristics of the transmitter, receiver, and scatterers, the channel impulse response matrix H from the transmitter to the receiver can be calculated using the ray tracing method RT . The action of reinforcement learning is to randomly match the multi-path scatterer clusters and scatterers to form cluster cores. The position of the cluster core corresponds to the position of the scatterer in the scatterer digital map, and the characteristics of the cluster core correspond to the parameters of the multi-path scatterer cluster. Thus, combine the characteristics of the transmitter and receiver obtained in Step 1 and Step 2, and substitute the position and characteristics of the cluster core into the environment of reinforcement learning, and use the ray tracing method to obtain the channel impulse response matrix H RT . The reward of reinforcement learning is set as the reciprocal of the Frobenius norm of the difference between the channel impulse response matrix generated in the environment based on the ray tracing method and the actual channel impulse response matrix H true obtained in Step 6, that is, calculated according to the formula ||H RT - H true || -1 , where the symbol || || represents the Frobenius norm, and the superscript -1 represents the reciprocal. The specific implementation methods of reinforcement learning and the ray tracing method are mainly realized through existing technologies and will not be elaborated here.

[0135] Step 8: Taking the cluster core as a node, the non-line-of-sight propagation link through which the wireless signal passes is decomposed into multiple logical sub-channels. By connecting the logical sub-channels to form a propagation path, an effective propagation path with the connection nodes being the transmitting antenna unit, the cluster core, and the receiving antenna unit in sequence is obtained. If the end point of the propagation path is not the receiving antenna unit, it is an invalid propagation path that can be excluded.

[0136] For a scenario such as Figure 3 shown, for a non-line-of-sight (NLOS) link that has not been reflected by the RIS, the NLOS channel from the p-th transmitting antenna unit to the q-th receiving antenna unit can be sequentially split into the following line-of-sight logical sub-channels according to the cluster core:

[0137]

[0138] where, Tx (p) represents the p-th transmitting antenna unit, Rx (q) represents the q-th receiving antenna unit, → represents the signal propagation path, is the set of multipath scatterer cluster cores obtained in Step 7, and are the -th and -th multipath scatterer cluster cores in this set.

[0139] The channel that has been reflected by the RIS can be divided into a virtual line-of-sight channel that only passes through the RIS an NLOS channel with scatterers existing before the RIS reflection an NLOS channel with scatterers existing after the RIS reflection and an NLOS channel with scatterers existing both before and after the RIS reflection which are sequentially split into the following line-of-sight logical sub-channels according to the cluster core:

[0140] Tx (p) → RIS → Rx (q) ,

[0141]

[0142]

[0143]

[0144] Substitute the parameters obtained in Steps 6 and 7, namely signal amplitude attenuation, multipath time delay, horizontal departure angle, vertical departure angle, horizontal arrival angle, vertical arrival angle, complex polarization matrix, Doppler frequency shift, number of scattering clusters, intra-cluster time delay, intra-cluster angular spread, multipath scatterer cluster core, and RIS equivalent cluster core, into the split logical sub-channels, and calculate the channel impulse response of each logical channel in turn, so as to obtain the channel impulse responses of the above five channels respectively in this step. and

[0145] Step 9: Merge all the effective propagation paths to obtain the channel impulse response from the transmitter to the receiver, and obtain the non-line-of-sight channel model for the intelligent metasurface wireless communication system in the coal mine underground in the current scenario.

[0146] At the transmission time t, after experiencing the propagation delay τ, the complex channel impulse response (CIR) h p,q (t, τ) from the p-th transmit antenna element to the q-th receive antenna element can be expressed as:

[0147]

[0148] The corresponding channel response matrix of the MIMO RIS wireless communication system is the following M R ×M T complex matrix:

[0149]

[0150] In specific implementation, the present application provides a computer storage medium and a corresponding data processing unit. Among them, the computer storage medium can store a computer program, and when the computer program is executed by the data processing unit, it can run the inventive content of a non-line-of-sight channel modeling method for intelligent metasurface wireless communication in coal mine underground and some or all of the steps in each embodiment. The storage medium can be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM), etc.

[0151] Those skilled in the art can clearly understand that the technical solutions in the embodiments of the present invention can be implemented by means of a computer program and its corresponding general hardware platform. Based on such an understanding, the technical solutions in the embodiments of the present invention, in essence, or the part that contributes to the prior art can be embodied in the form of a computer program, that is, a software product. This computer program software product can be stored in a storage medium, including several instructions for causing a device (which can be a personal computer, a server, a single-chip microcomputer, a MUU or a network device, etc.) containing a data processing unit to execute the methods described in various embodiments or some parts of the embodiments of the present invention.

[0152] The present invention provides a non-line-of-sight channel modeling method for intelligent metasurface wireless communication in coal mines. There are many methods and ways to specifically implement this technical solution. The above is only the preferred embodiment of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present invention. Each component not clearly defined in this embodiment can be implemented by the prior art.

Claims

1. A non-line-of-sight channel modeling method for intelligent metasurface wireless communication in coal mines, characterized in that, It includes the following steps: Step 1: According to the existing or future required wireless communication system in the coal mine underground environment to be measured, select the parameters of the channel measurement platform for the multiple-input multiple-output intelligent metasurface wireless communication system, and build the transmitter, intelligent metasurface, and receiver of the channel measurement platform; Step 2: Measure and calibrate the characteristics of the channel measurement platform itself; Step 3: In the coal mine underground wireless communication scenario to be measured, deploy the channel measurement platform that has been measured and calibrated, conduct on-site channel measurement, and control the working states of the transmitting antenna array, intelligent metasurface, and receiving antenna array through the antenna high-speed switching module. Based on the time division multiple access principle, sequentially measure the non-line-of-sight channel impulse responses between each transmitting antenna unit, each intelligent metasurface array unit, and each receiving antenna unit, and store them in the data storage unit; Step 4: In the coal mine underground wireless communication scenario to be measured, deploy a lidar, and use simultaneous localization and mapping technology to perform three-dimensional image scanning and reconstruction of the channel measurement site, and collect and calibrate the positions, texture features, and reflection characteristics of objects; Step 5: After completing the on-site measurement of the coal mine underground wireless communication scenario, on the ground, perform image semantic segmentation on the measurement data and results of the simultaneous localization and mapping technology through computer vision methods, and match them with each object on-site to form a scatterer digital map. According to the characteristics of the intelligent metasurface array measured previously, the intelligent metasurface is equivalent to a virtual scatterer and inserted into the scatterer digital map; Step 6: After completing the on-site measurement of the coal mine underground wireless communication scenario, on the ground, use the sliding correlation method to obtain the complete channel impulse response matrix of the multiple-input multiple-output intelligent metasurface wireless communication system and form a measurement result data set. Through the SAGE algorithm, obtain the channel parameters from the measurement result data set, and then perform multipath clustering through the K-nearest neighbor clustering algorithm to determine the number of scattering clusters, intra-cluster delay, and intra-cluster angular spread parameters; Step 7: Combining the multipath clustering results with the scatterer map, through reinforcement learning, match the scattering clusters with the scatterers to obtain a finite number of multipath scatterer clusters and an intelligent metasurface equivalent cluster core; Step 8: Using the cluster core as a node, decompose the non-line-of-sight propagation link passed by the wireless signal into two or more logical sub-channels, form a propagation path by connecting the logical sub-channels to each other, and obtain an effective propagation path with the connection nodes being the transmitting antenna unit, cluster core, and receiving antenna unit in sequence. If the end point of the propagation path is not the receiving antenna unit, it is an invalid propagation path that can be excluded; Step 9: Combine all the effective propagation paths to obtain the channel impulse response from the transmitter Tx to the receiver Rx, and obtain the non-line-of-sight channel model for the intelligent metasurface wireless communication in the current scenario in the coal mine underground.

2. The method according to claim 1, wherein In step 1, the transmitter, intelligent metasurface, and receiver of the channel measurement platform consist of a transmitting antenna array, a receiving antenna array, an intelligent metasurface, a synchronous clock, a signal generator, a signal receiver, a data storage unit, and a control terminal. Among them, the parameters of the channel measurement platform include transmitting signal parameters, transmitting antenna array parameters, intelligent metasurface parameters, and receiving antenna array parameters. The transmitting signal parameters include the frequency range to be measured, the magnitude of the transmitting signal power, and the type of the transmitting signal. The parameters of the transmitting antenna array, intelligent metasurface, and receiving antenna array include the number, spacing, position, and orientation of antenna elements or array elements.

3. The method according to claim 2, characterized in that, In step 2, the self-characteristics of the channel measurement platform include measuring and calibrating the antenna feed transmission power error, the high-frequency coaxial cable transmission loss, the adapter insertion loss, the transmitting antenna array, the intelligent metasurface array, the receiving antenna array, and the system response error of other instrument devices.

4. The method according to claim 3, wherein In step 3, a time-division multiple access is implemented among the transmitting antenna array, the uniform rectangular RIS array, and the receiving antenna array by using an antenna high-speed switching module. Among them, the transmitting antenna array has a total of M T transmitting antenna elements, the receiving antenna array has a total of M R receiving antenna elements, and the RIS array has a total of MN RIS array elements. M and N are the numbers of RIS array elements on the long side and the short side of the rectangle respectively. Only one combination of transmitting antenna elements, RIS array elements, and receiving antenna elements is measured in each time slot. The non-line-of-sight channel impulse responses between each transmitting antenna element, each intelligent metasurface array element, and each receiving antenna element are measured in turn. After traversing all combinations, a channel sampling snapshot is obtained, which specifically includes: Define the activation time function as follows The activation time function has an activation time range from 0 to the moment for the purpose of controlling whether to activate the transmitting antenna unit, the intelligent metasurface unit or the receiving antenna unit at time t, where 1 indicates activation and 0 indicates non-activation; For the th channel sampling snapshot, according to the definition of the activation time function, the activation time function of the p-th transmit antenna element the activation time function of the q-th receive antenna element and the activation time function of the r-th RIS array element are respectively: Among them, T T is the activation time of a single transmit antenna element, T R is the activation time of a single receive antenna element, T RIS is the activation time of a single RIS array element, and the period for traversing all combinations of transmit antenna elements, RIS array elements, and receive antenna elements is T cycle , M S is the number of channel sampling snapshots; The transmitted signal u(t) is expressed as: where s(t) is a PN sequence signal, and the vectors and are the activation time functions of the first transmitting antenna element, the second transmitting antenna element, and the M T th transmitting antenna element, respectively; The signal after being reflected by the RIS is as follows: where, Φ is the RIS array response matrix, is the channel transfer matrix from the transmit antenna array to the RIS array, is the propagation delay, the superscript T is the matrix transpose operator, the vectors and are the activation time functions of the first RIS array element, the activation time function of the second RIS array element, and the activation time function of the MN-th RIS array element, respectively; The signal vector y(t) received by the receiving antenna array is: Among them, is the channel transfer matrix from the RIS array to the receiving antenna array, where is the propagation delay, and H NLoS (t,τ NLoS ) is the NLOS channel transfer matrix from the transmitting antenna array to the receiving antenna array, where τ NLoS is the transmission delay, and n(t) is a complex Gaussian white noise vector; Finally, the received signal Y(t) obtained each time of measurement is: Among them, the vector and are respectively the activation time function of the first receiving antenna unit, the activation time function of the second receiving antenna unit, and the activation time function of the M R th receiving antenna unit.

5. The method according to claim 4, wherein Step 6 includes: Wireless signals arrive at the receiving antenna array in the form of clusters. In the nth cluster component, the channel h p,q,n (t) from the pth transmitting antenna element to the qth receiving antenna element is expressed as: Among them, M c represents the number of sub - paths in the nth cluster component, α n,m , Ψ n,m , v n,m and τ n,m respectively represent the amplitude attenuation, complex polarization matrix, Doppler frequency shift and propagation delay corresponding to the mth sub - path in the nth cluster component. Ω Tx,n,m ={θ Tx,n,m , φ Tx,n,m} and Ω Rx,n,m ={θ Rx,n,m , φ Rx,n,m} respectively represent the angle information of the transmitting antenna array and the receiving antenna array. θ Tx,n,m and φ Tx,n,m respectively represent the elevation angle and azimuth angle of the transmitting antenna array corresponding to the mth sub - path in the nth cluster component. θ Rx,n,m and φ Rx,n,m respectively represent the elevation angle and azimuth angle of the receiving antenna array corresponding to the mth sub - path in the nth cluster component. Γ Tx,p (Ω Tx,n,m ) and Γ Rx,q (Ω Rx,n,m ) are the transmitting antenna array response and the receiving antenna array response respectively. exp is the natural exponential function and j is the imaginary unit; Define the parameter set of the received sub-path as follows: Among them, and respectively represent the corresponding amplitude attenuation, complex polarization matrix, Doppler frequency shift, and propagation delay of the th sub-path. and respectively represent the angle information of the transmitting antenna array and the receiving antenna array. and respectively represent the elevation angle and azimuth angle of the th sub-path corresponding to the transmitting antenna array. and respectively represent the elevation angle and azimuth angle of the th sub-path corresponding to the receiving antenna array. The symbol represents that takes any value within its value range; For a known PN sequence s(t) and a measured received signal Y(t), the likelihood function determined by the parameter set is The SAGE algorithm iteratively solves the parameter set to find a set of multipath clustering parameters that maximize the likelihood function L. At each iteration, only a single parameter in the set is estimated, and the remaining parameters remain unchanged. After the solution, the single parameter is updated and substituted into the next iteration to solve for other parameters. For the k-th iteration, one of the iteration update orders is as follows: Among them, the function argmax is to find the variable value that maximizes the function immediately to its right. The symbol below the function argmax is the variable to be adjusted. and represent the values at the k-th and (k - 1)-th iterations respectively and ; The parameter sets of each sub-path are estimated by the SAGE algorithm, and then the feature vectors are formed. The feature vectors of each sub-path are clustered in the vector space by the K-nearest neighbor clustering algorithm to achieve clustering, determine the number of clusters and the sub-paths within the clusters.

6. The method according to claim 5, wherein The channel parameter indicators obtained from the measurement result dataset by the SAGE algorithm in step 6 include: Signal amplitude attenuation, multipath delay, horizontal departure angle, vertical departure angle, horizontal arrival angle, vertical arrival angle, complex polarization matrix, Doppler frequency shift, number of clusters, and sub-paths within a cluster.

7. The method according to claim 6, characterized in that, In step 7, the environment of the reinforcement learning is a wireless channel containing direct, first-order reflection, and second-order reflection. The action is to randomly match multipath scattering clusters and scatterers, and the reward is the reciprocal of the Frobenius norm of the difference between the channel impulse response matrix generated in the environment based on the ray tracing method and the actual channel impulse response matrix obtained in step 6.

8. The method according to claim 7, wherein In step 7, the multipath clustering results are combined with the scatterer map, and the scatterer clusters and scatterers are matched to each other through reinforcement learning to obtain M C multipath scatterer cluster nuclei and 1 RIS equivalent cluster nucleus, and define the multipath scatterer cluster nucleus set as where C1, C2, and are the first multipath scatterer cluster nucleus, the second multipath scatterer cluster nucleus, and the M c th multipath scatterer cluster nucleus, respectively; The environment of reinforcement learning is set as a wireless channel with direct, first - order reflection, and second - order reflection. Given the positions and characteristics of the transmitter, receiver, and scatterers, the channel impulse response matrix \(H\) from the transmitter to the receiver is calculated using the ray - tracing method. RT The action of reinforcement learning is to randomly match multipath scattering clusters and scatterers to form a cluster core. The position of the cluster core corresponds to the position of the scatterer in the scatterer digital map, and the characteristics of the cluster core correspond to the parameters of the multipath scattering cluster. Combining the characteristics of the transmitter and receiver obtained in steps 1 and 2, substitute the position and characteristics of the cluster core into the environment of reinforcement learning, and use the ray - tracing method to obtain the channel impulse response matrix \(H\) of the wireless channel. RT The reward of reinforcement learning is set as the reciprocal of the Frobenius norm of the difference between the channel impulse response matrix generated in the environment based on the ray - tracing method and the actual channel impulse response matrix \(H\) obtained in step 6. true According to the formula \(\left\|\left|H\right|\right|\) RT -\(H\) true \(\left\|\left|\right|\right.\) -1 for calculation, where the symbol \(\left\|\left|\right|\right.\) represents the Frobenius norm, and the superscript \(- 1\) represents the reciprocal.

9. The method according to claim 8, wherein Step 8 includes: For the non-line-of-sight (NLOS) link without RIS reflection, the NLOS channel from the p-th transmit antenna element to the q-th receive antenna element is successively split into the following line-of-sight (LOS) logical sub-channels according to the cluster cores: where Tx (p) represents the p-th transmit antenna element, Rx (q) represents the q-th receive antenna element, and → represents the signal propagation path, is the set of multipath scatterer cluster cores obtained in step 7, and are the -th and -th multipath scatterer cluster cores in this set; The channel reflected by the RIS is divided into a virtual line-of-sight channel that only passes through the RIS The NLOS channel with scatterers before RIS reflection The NLOS channel with scatterers after RIS reflection And the NLOS channel with scatterers both before and after RIS reflection According to the cluster cores, it is successively split into the following line-of-sight logical sub-channels: Tx (p) → RIS → Rx (q) , Substitute the parameters obtained in steps 6 and 7, including signal amplitude attenuation, multipath time delay, horizontal departure angle, vertical departure angle, horizontal arrival angle, vertical arrival angle, complex polarization matrix, Doppler frequency shift, number of scattering clusters, intra-cluster time delay, intra-cluster angular spread, multipath scatterer cluster core, and RIS equivalent cluster core, into the split logical sub-channels, and calculate the channel impulse response of each logical channel in turn to obtain five channels Their respective channel impulse responses and 10. The method according to claim 9, characterized in that Step 9 includes: At the transmission time t, after experiencing the propagation delay τ, the complex channel impulse response h from the p-th transmit antenna element to the q-th receive antenna element is expressed as: p,q (t, τ) is expressed as: The corresponding multi-input multi-output (MIMO) intelligent reflecting surface (RIS) wireless communication system channel model is represented as an $M \times M$ complex matrix $\mathbf{H}(t, \tau)$: R $\times M$ T :

Images