Train and rail side communication system applied to intelligent high-speed rail network
By installing a reconfigurable refractive surface with millimeter wave XL-RRS technology on high-speed railway trains, the problems of high penetration loss and data transmission delay in high-speed railway communication systems are solved, and more efficient and reliable T2T communication is achieved, reducing the number of times the number of times the railway communication system is reduced. Equipment cost and power consumption.
Patent Information
- Application Number
- CN202411313963.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-20
- Publication Date
- 2024-12-03
AI Technical Summary
Currently, high-speed railway communication systems are facing the problems of high penetration loss and data transmission delay. Traditional MRS equipment is costly and has a large power consumption, making it difficult to meet the efficient and stable T2T communication needs.
Using millimeter wave XL-RRS technology, the reconstructible refractive surface of multi-layer composite metamaterial is installed on the shell of the high-speed rail train to realize the refractive and spatial multiplexing of signals. The intelligent high-speed rail heterogeneous network based on millimeter wave XL-RRS is designed and disassembled. The channel-separated model adopts packet multi-user IDMA and ZF precoding schemes to improve signal coverage and data transmission efficiency.
It reduces propagation penetration loss between the base station on the track and the in-vehicle users, improves signal strength and data transmission rate, reduces power consumption and delay, and achieves more reliable uplink data detection and higher downlink data throughput.
Smart Images

Figure CN119071806A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the field of communication networks, and specifically relates to a train and trackside communication system applied to an intelligent high-speed rail network. Background Art
[0002] Given the unprecedented development of contemporary railway technology, high-speed railway (HSR) is expected to become the preferred mode of transportation for people, as it can provide comfortable high-speed travel services at an attractive price. However, the current railway communication systems in operation, such as the railway integrated digital mobile communication system (GSM-R), railway long-term evolution (LTE-R), IEEE ~ 802.16 wireless broadband standard and IEEE ~ 802.20 mobile broadband network specification, are facing the pressure of overloaded long-distance wireless communication load. Therefore, in the context of the next generation of intelligent HSR communication network, it is necessary to design a more efficient T2T (Train-to-Trackside) communication network.
[0003] The current research and technical directions of advanced wireless communication technologies and network architectures mainly include the following: massive multiple-input multiple-output (MIMO) systems, millimeter wave / terahertz communications, hybrid beamforming, non-orthogonal multiple access (NOMA), and heterogeneous networks relying on cell-free architectures. These technologies can be integrated with each other, and after careful design, they are expected to provide sufficient support for the high data rate requirements of future high-speed rail communication networks, such as the fifth-generation railway communication system (5G-R) and its evolution. On the other hand, the high penetration loss of signals (especially high-frequency signals) transmitted by the trackside BS (Base Station) that provides communication services to passengers in the car is one of the main challenges that need to be solved. Deploying MRS (Mobile Relay Stations) on the top of the train is a typical solution to establish a stable and reliable communication link between the trackside BS and the users in the train. However, MRS is not only an expensive active device, but also must first process the received signal before forwarding it to the trackside BS or sending it to the UE (User Equipment) when serving as an in-car repeater to provide services to passengers in the car in the form of Ethernet or Wi-Fi. These complex signal processing and forwarding processes significantly increase data transmission delay, which may seriously affect the transmission stability and efficiency of highly dynamic T2T communication links.
[0004] The recently emerging RIS (Reconfigurable Intelligent Surfaces) can serve as a promising and low-cost alternative to MRS. This RIS is composed of a programmable metasurface with a multi-layer stacked planar structure, and the metasurface contains a large number of low-cost array elements (also known as meta-atoms) integrated with low-power electronic devices such as photodiodes. The amplitude and phase of the electromagnetic signal received by each meta-atom can be dynamically adjusted to radiate electromagnetic energy in the desired beam direction, thereby realizing an intelligent and controllable wireless communication environment. Most of the current research is dedicated to solving the many challenges faced when applying IRS (Intelligent Reflecting Surfaces) to current and future wireless communication networks. The main function of IRS is to increase the received signal power in the base station coverage blind area by reflecting the incident electromagnetic signal, thereby expanding its effective coverage range. Recently, researchers have also begun to consider the application scenarios of RIS working in refraction (or transmission) mode, which can be called RRS (Reconfigurable Refracting Surfaces). At the same time, RIS with dual functions of reflection and refraction has also received widespread attention recently and will be one of the future development directions of RIS-related research. Summary of the invention
[0005] The purpose of the present invention is to apply millimeter wave XL-RRS (Extra-Large Scale RRS) to the intelligent high-speed rail communication network by utilizing the characteristics of RRS (Reconfigurable Refracting Surfaces) that can provide many potential performance improvements for wireless communication systems, and propose a T2T (Train-to-Trackside) communication system based on millimeter wave XL-RRS. At the same time, a millimeter wave XL-RRS-assisted T2T data transmission scheme is proposed to achieve stable, reliable and efficient T2T communication in the designed intelligent high-speed rail heterogeneous network, and obtain a more reliable uplink data detection bit error rate and a higher downlink data transmission system throughput performance.
[0006] In order to achieve the above-mentioned object, the technical solution adopted by the present invention is: a train and trackside communication system applied to an intelligent high-speed rail network, including a millimeter wave XL-RRS architecture, an intelligent high-speed rail heterogeneous network based on millimeter wave XL-RRS, and a channel model between a trackside millimeter wave base station and an in-vehicle user;
[0007] The millimeter wave XL-RRS architecture includes a plurality of RRS sub-blocks, wherein each RRS sub-block is mounted at a designated position on the train shell;
[0008] The intelligent high-speed rail heterogeneous network based on millimeter wave XL-RRS includes a central unit CU, a distributed unit DU and an active antenna unit AAU, wherein the central unit CU is connected to multiple distributed units DU, and the distributed unit DU controls multiple active antenna units AAU;
[0009] The channel model between the trackside millimeter wave base station and the user in the train is split into a base station to RRS channel model and an RRS to user channel model by the millimeter wave XL-RRS installed on the train shell.
[0010] RRS (Reconfigurable Refracting Surfaces) composed of multi-layer composite metamaterials can be installed on the shell or windows of high-speed trains to use their refractive properties to connect the inside and outside spaces of the train, so as to reduce the serious propagation penetration loss between the trackside base station and the users in the train, thereby improving the strength of the indoor signal. By installing multiple RRS on the shell (car body or window) of the high-speed train and jointly controlling them, a millimeter-wave XL-RRS (Extra-Large Scale RRS) is formed, and the following intelligent high-speed rail heterogeneous network is designed based on the millimeter-wave XL-RRS.
[0011] 1. Design of intelligent high-speed rail heterogeneous network based on millimeter wave XL-RRS
[0012] The present invention proposes a millimeter wave XL-RRS-based intelligent high-speed rail heterogeneous network that is compatible with the current 5G (fifth generation) mobile communication network. The heterogeneous network includes a three-layer network structure, namely CU (centralized units), DU (distributed units) and AAU (active antenna units), wherein low-frequency massive MIMO (large-scale multiple-input multiple-output) base stations working in the frequency band below 6GHz (Gigahertz) and high-frequency massive MIMO base stations working in the millimeter wave or even terahertz frequency band respectively undertake the tasks of DU and AAU. Each low-frequency massive MIMO base station is connected to multiple high-frequency massive MIMO base stations through a fronthaul link, and multiple adjacent low-frequency massive MIMO base stations transmit their signals to CU through a midhaul link for joint processing, and then CU is connected to the core network through a backhaul link, where the midhaul link is a transmission link connecting DU and CU in the current 5G mobile communication standard networking mode. The proposed heterogeneous network adopts a cell-free network architecture, which can build a virtual cell centered on the high-speed train. Since the CU is connected to multiple DUs, and each DU can jointly control multiple AAUs, the CU can comprehensively dispatch different AAUs to track the train using prior information such as positioning. Therefore, the heterogeneous network designed by the institute can reduce the interference between cells during the movement of the train. At the same time, multiple AAUs under the control of the CU are in the virtual cell and can avoid frequent cell switching. In this heterogeneous network, the millimeter wave XL-RRS installed on the outer shell of the high-speed train (car body or window) can be used as a hub to connect the users in the car with the base station beside the track to enhance T2T communication.
[0013] 2. Millimeter wave XL-RRS architecture design
[0014] The present invention designs a specific architecture of millimeter wave XL-RRS. Specifically, a millimeter wave XL-RRS can be split into multiple RRS sub-blocks, each of which is installed at a specified position on the train shell. Accordingly, according to the specific positions of the users in the car, these users can be divided into multiple user groups, and the users in each group are served by the same RRS sub-block. A feasible example of a future intelligent high-speed rail communication system is to replace traditional window glass with a metasurface based on smart glass. The millimeter wave XL-RRS can control the phase and amplitude of its double-sided outer layer atoms in real time by integrating and processing the CSI (Channel State Information) obtained by various means (such as high-speed rail control systems, railway Internet of Things, and advanced channel estimation methods, etc.), thereby providing better signal coverage services for users in the car with considerable near-field LoS link path multiplexing gain. At the same time, the millimeter wave XL-RRS designed above can split the high dynamic fast time-varying T2T channel between the trackside base station and the user in the train into a cascade form of the channel between the BS (base station) and RRS and the channel between the RRS and the UE (user equipment), because of its ability to control the propagation channel between the trackside base station and the user in the train. The fast time-varying LoS link channel between the BS and RRS outside the vehicle can be reliably predicted based on the characteristics of the high-speed train running along a fixed track, and the quasi-static channel between the BS and RRS in the car can be obtained using some advanced channel estimation methods. In addition, the CSI obtained above can also solve the problem of pre-compensation of Doppler frequency shift.
[0015] 3. Establishment of channel model between base station and user
[0016] The present invention establishes a channel model between the millimeter wave base station beside the track and the user in the train. The millimeter wave XL-RRS installed on the train shell splits the channel model into the following base station to RRS channel and RRS to user channel. Here, the relevant parameters are first defined as follows: and are the number of sub-arrays (antennas in each sub-array) of the multi-block antenna array in the horizontal and vertical directions, respectively, and and are the number of antennas in the horizontal and vertical directions of the entire array. The total number of antennas in each subarray and the entire antenna array at the base station is and The number of RF links at the base station is the number of subarrays, that is, The spacing between adjacent sub-blocks and adjacent antennas in a multi-block antenna array is defined as Δ and d, respectively. The millimeter wave XL-RRS is also split into multiple RRS sub-blocks, so its dimensional parameters have similar definitions to the corresponding parameters at the base station, which can be expressed as as well as and
[0017] 1) Channel model between base station and RRS
[0018] The channel matrix between the millimeter wave XL-RRS and the BS-RRS between the multiple antenna arrays of the trackside base station at the jth time slot can be expressed as in is the i-th (1≤i≤I RRS ) The channel matrix between the RRS sub-block and the base station can be further expressed as follows:
[0019]
[0020] where α i ~CN(0,1) is the channel gain, and D i are the large-scale fading coefficient and signal transmission distance between the i-th RRS sub-block and the base station, λ is the wavelength, exp(2πD i / λ) corresponds to the near-field spherical wave component, f d =v d / λ and T s denote the Doppler frequency shift and the duration of a time slot, respectively, and v d is the relative radial velocity, and and are defined as the horizontal and vertical virtual angles of the base station end (i-th RRS sub-block), and and are the corresponding azimuth and elevation angles, and are the array response vectors of the base station and the i-th RRS sub-block, respectively, and (·) H is the conjugate transpose operation.
[0021] 2) Channel model from RRS to user
[0022] For K single-antenna users in a high-speed train, the channel vector between the millimeter-wave XL-RRS and the kth (1≤k≤K) user can be expressed as in is the i-th (1≤i≤I RRS ) RRS sub-block to user channel vector (including L k,i paths), that is
[0023]
[0024] where γ k,i,l~CN(0,1) is the lth (1≤i≤L k,i ) paths corresponding to the channel gain, and D k,i,l are the large-scale fading coefficient and signal transmission distance between the lth path of the kth user and the i-th RRS sub-block, and and are defined as the corresponding horizontal and vertical virtual angles (azimuth and elevation), is the corresponding array response vector.
[0025] The present invention also provides a method for transmitting data between a train and a trackside, which uses the above-mentioned train and trackside communication system applied to an intelligent high-speed rail network, and is divided into an uplink data transmission method based on group multi-user IDMA and a downlink precoding method based on ZF;
[0026] The uplink data transmission method based on packet multi-user IDMA is:
[0027] According to the channel model between the base station and RRS and between RRS and the user, the base station receives the signal vector corresponding to the jth time slot during uplink data transmission. for
[0028]
[0029] in and are the base station end analog and digital receiving merging matrices and is the i-th (1≤i≤I BS ) The simulated receiving combined vector corresponding to the subarray, the diagonal matrix and is the phase value corresponding to the double-sided outer layer in the designed millimeter-wave XL-RRS, P tx and k [j] are the user transmit power and the uplink data sent by the kth (1≤k≤K) user, and n UL [j] is zero-mean and has a covariance matrix of The complex additive Gaussian white noise is Diagonal matrix Φ UL and UL Expressed as and For the i-th (1≤i≤I RRS )Diagonal matrix and Further expressed as and Here the vector and That is, the phase value corresponding to the i-th RRS sub-block; in addition, the digital receiving merging matrix is directly considered as the unit matrix, that is,
[0030] The downlink precoding method based on ZF (Zero-Forcing) is:
[0031] In the downlink transmission process, the channel reciprocity of the time division duplex system is used to obtain The corresponding low-dimensional equivalent channel matrix Right now in is the channel matrix that combines the downlink channel vectors of K users in the car and is the RRS-UE channel vector of the kth (1≤k≤K) user, Φ DL and DL is the phase value corresponding to the double-sided outer layer of the millimeter wave XL-RRS during downlink, and the diagonal matrix Φ DL and DL The selection of UL and UL similar, is the BS-RRS channel after Doppler shift compensation in downlink, and is the analog precoding matrix at the base station; when designing downlink precoding, then, consider I BS = K, the digital precoding matrix at the base station is designed by using zero-forcing precoding Right now where f BB,k is the digital precoding vector corresponding to the kth user.
[0032] In the uplink data transmission phase, a known CSI is assumed to design the base station receiving combining matrix W RF , and the phase value Φ corresponding to the double-sided outer layer in the millimeter wave XL-RRS UL and UL At the same time, the Doppler frequency shift compensation of the BS-RRS channel is also considered; considering that the BS-RRS channel is mainly a LoS link, the W is designed through beam alignment. RF and Φ UL In order to obtain sufficient beamforming gain, the RRS-UE channel exhibits a rich scattering characteristic, so the energy focusing method is used to design Ψ UL , in order to maximize the energy on the user side; specifically, first for W RF The i-th (1≤i≤I BS ) simulates receiving the combined vector w RF,i , first initialize w RF,i for Then make Among them IBS,i It represents the antenna index set of the ith subarray at the base station in the entire array and its cardinality is M BS =|I BS,i | c ; Secondly, for Φ UL The i-th (1≤i≤I RRS ) Phase vector Can be directly ordered Finally, the users in the car are divided into M user groups according to their specific locations, and the number of users in each group is U = K / M. Then for Ψ UL The i-th (1≤i≤I RRS ) Phase vector Can be ordered To achieve energy concentration; using the above designed W RF , Φ UL And UL ,have
[0033]
[0034] in is the BS-RRS channel after Doppler shift compensation,
[0035] and Consider the channel matrix and transmission vector of K users respectively, represents the low-dimensional equivalent channel matrix, and According to the user grouping and simplified signal transmission model, the data detection process based on IDMA can be completed.
[0036] Compared with the prior art, the technical solution adopted by the present invention has the following beneficial effects:
[0037] 1. Position and deployment: In order to maintain the streamlined structure of the high-speed train, the MRS with a uniform planar array antenna array structure is usually embedded in the top of the carriage in a horizontal direction. Then, this requires the trackside base station to have a certain height to ensure that the MRS end has an appropriate grazing angle, but this may increase the construction and maintenance costs of the base station. In contrast, the location and deployment of the millimeter wave XL-RRS will be more flexible. Specifically, the millimeter wave XL-RRS can be deployed at any position of the train, including the roof, shell, and windows on both sides, by utilizing the current mature skin antenna technology, thus ensuring a favorable grazing angle, thereby realizing LoS link communication. This kind of smart glass metasurface is used to design passive millimeter wave XL-RRS equipment to ensure that it can be widely deployed on train windows, thereby realizing an intelligent high-speed rail communication network.
[0038] 2. Power consumption and transmission delay: As an active device, MRS needs to be equipped with multiple high-cost and high-energy consumption RF links to serve multiple users at the same time, and the corresponding relay processing process will inevitably introduce additional transmission delays, which is not suitable for high-dynamic fast-fading T2T communication channels. On the contrary, the millimeter-wave XL-RRS only uses low-cost and low-power meta-atoms to design a passive hardware architecture, and it only requires a small amount of power to manipulate the meta-atoms to refract electromagnetic signals to complete the signal forwarding process. Since there is no need for cumbersome baseband signal processing, the delay caused by the electromagnetic signal penetrating the millimeter-wave XL-RRS installed on the train shell can be ignored.
[0039] 3. Spatial multiplexing capability: In the high-speed rail communication scenario, since there are no scatterers between the trackside base station and the XL-RRS or MRS, T2T communication can only establish a unique LoS link. For MRS, it can only use the LoS link without spatial multiplexing gain to connect to the trackside base station, which severely limits the throughput of the entire T2T communication system, so the data rate it can provide is limited, and the number of users that can be served simultaneously is also small. In contrast, a millimeter-wave XL-RRS array is divided into multiple sub-array blocks installed at different positions in the car to form a giant dimensional array with an ultra-large effective aperture. In this way, the millimeter-wave XL-RRS not only utilizes the passive beamforming gain of the large-aperture metasurface to improve the received SNR (signal-to-noise ratio, Signal-to-Noise Ratio), but also utilizes the unique near-field LoS path intra-multiplexing characteristics in the large-aperture array to obtain spatial multiplexing gain to increase the data transmission rate. Therefore, millimeter-wave XL-RRS-assisted T2T communication is sufficient to ensure that high-speed rail passengers can enjoy the ultimate Internet surfing experience such as immersive extended reality and high-fidelity mobile holograms. BRIEF DESCRIPTION OF THE DRAWINGS
[0040] Figure 1 (a) Schematic diagram of IRS deployment scenario and basic structure.
[0041] Figure 1 (b) is a schematic diagram of the RRS deployment scenario and basic structure.
[0042] Figure 2 Schematic diagram of intelligent high-speed rail heterogeneous network based on millimeter wave XL-RRS.
[0043] Figure 3 Schematic diagram of the conceptual structure for deploying millimeter-wave XL-RRS on high-speed trains.
[0044] Figure 4 (a) Schematic diagram of the multi-antenna array structure equipped with the trackside millimeter wave base station.
[0045] Figure 4(b) Schematic diagram of the millimeter-wave XL-RRS structure installed on the train shell.
[0046] Wherein Δ and d represent the spacing between adjacent sub-blocks and adjacent antennas in a sub-block, respectively.
[0047] Figure 5 This is the flow chart of the proposed T2T data transmission scheme based on packet multi-user IDMA.
[0048] Figure 6 Schematic diagram of simulation scenarios of three T2T modes (i.e., single hop, MRS, and XL-RRS) in high-speed rail communication networks.
[0049] Figure 7 Comparison of bit error rate performance under three T2T communication modes with varying transmit power
[0050] Figure 8 The figure is a comparison chart of average and rate performance under three T2T communication modes as the transmission power changes.
[0051] The code descriptions in the figure are as follows:
[0052] BER – Bit Error Rate
[0053] IDMA - Interleaved Multiple Access
[0054] IRS – Intelligent Reflective Surface
[0055] LoS - direct diameter
[0056] MIMO – Massive Multiple Input Multiple Output
[0057] MRS——Mobile Relay Platform
[0058] RRS – Reconfigurable Refractive Surface
[0059] T2T – Train to Trackside
[0060] XL - Extra Large Aperture Large Scale
[0061] ZF——Force Zero DETAILED DESCRIPTION
[0062] The specific implementation of the present invention is described in detail below with reference to the accompanying drawings and examples.
[0063] The present invention adopts a new RIS (Reconfigurable Intelligent Surfaces) technology to improve the reliability and throughput of wireless communication. RIS is composed of a large number of controlled atoms and low-power electronic devices, which can shape electromagnetic waves without using power amplifiers or performing baseband digital signal processing operations. Figure 1 (a) shows an example of a reflective RIS, which can controllably reflect electromagnetic signals from a base station to serve users outside the base station coverage area. When the LoS (Line-of-Sight) link is blocked by obstacles in indoor and outdoor environments, the reflective RIS is very suitable for establishing reliable non-line-of-sight or virtual LoS links in high-frequency channels (such as millimeter wave and terahertz bands). In general, the typical structure of a reflective RIS consists of an external reconfigurable passive reflective layer and an internal control circuit layer, where the former layer is a reflective surface made of a large number of metal patch array elements (so-called meta-atoms) printed on an insulating substrate, and the latter layer is a control circuit board that can control the amplitude and phase of each meta-atom through an FPGA (Field-Programmable Gate Array) controller.
[0064] In contrast, Figure 1 As shown in (b), RRS (Reconfigurable Refracting Surfaces) is composed of multi-layer composite metamaterials, including a double-sided outer layer that can directly adjust and refract the incident electromagnetic wave and an inner control circuit. The double-sided outer layer is composed of dense meta-atoms, and the electronic circuit in the inner layer (such as photodiodes / varicap diodes, etc.) can control the signal response (i.e., amplitude and phase) of the double-sided outer layer meta-atoms. Obviously, this type of refractive RIS can first intercept the radiation flux of the electromagnetic signal, and then realize controllable forwarding, and its reconfigurability can be achieved by changing electronic devices such as photodiodes or varactor diodes. It is precisely because of this structure that RRS can dynamically control the refraction mode of the signal incident on the double-sided outer layer meta-atoms to achieve flexible and adaptive passive beamforming. Therefore, RRS can be installed on the walls or windows of closed spaces such as high-rise buildings and high-speed trains, and use its refractive properties to connect indoor and outdoor spaces to reduce the serious propagation penetration loss between outdoor base stations and indoor users, thereby breaking the physical barriers in the process of electromagnetic signal propagation and improving the strength of indoor signals. The present invention forms a millimeter-wave XL-RRS (Extra-LargeScale RRS) by installing multiple RRSs on the shell of a high-speed train (car body or window) and jointly controlling them, and designs the subsequent intelligent high-speed rail heterogeneous network based on the millimeter-wave XL-RRS.
[0065] 1. Design of intelligent high-speed rail heterogeneous network based on millimeter wave XL-RRS
[0066] Although T2T (Train-to-Trackside) communication networks can gain a lot of benefits by using large bandwidth millimeter wave carriers, they are also affected by adverse propagation factors such as higher Doppler shift, frequent cell switching, and more severe penetration loss of high-frequency signals. Figure 2 The intelligent high-speed rail heterogeneous network based on millimeter wave XL-RRS shown in the figure can cope with the sharp increase in communication throughput demand caused by the increase in data rate required by a large number of passengers. In order to achieve the goal of compatibility with the current 5G (Fifth Generation) mobile communication network, the designed heterogeneous network includes a three-layer network structure, namely CU (Centralized Units), DU (Distributed Units) and AAU (Active Antenna Units), among which low-frequency massive MIMO (Multiple-Input Multiple-Output) base stations working in the frequency band below 6GHz (Gigahertz) and high-frequency massive MIMO base stations working in the millimeter wave or even terahertz frequency bands undertake the tasks of DU and AAU respectively. Each low-frequency massive MIMO base station is connected to multiple high-frequency massive MIMO base stations through a fronthaul link, and multiple adjacent low-frequency massive MIMO base stations transmit their signals to the CU through a midhaul link for joint processing, and then the CU is connected to the core network through a backhaul link. Here, the midhaul link is the transmission link connecting the DU and the CU in the current 5G mobile communication standard networking mode. The proposed heterogeneous network adopts a cell-free network architecture, which can build a virtual cell centered on the high-speed train. Since the CU is connected to multiple DUs, and each DU can jointly network and control multiple AAUs, the CU can comprehensively dispatch different AAUs to track the train using prior information such as positioning. Therefore, the heterogeneous network designed by the institute can reduce the interference between cells during the train's travel, and at the same time, multiple AAUs under the control of the CU are all in the virtual cell and can avoid frequent cell switching. In addition, low-frequency massive MIMO base stations can also use their good wide-area coverage and spatial multiplexing capabilities to send wireless control signaling to achieve wireless access to massive sensor nodes in the railway Internet of Things, while high-frequency massive MIMO base stations can provide sufficiently high data rates for passengers in the car by utilizing the large bandwidth available in the millimeter wave band. In this heterogeneous network, the millimeter wave XL-RRS installed on the outer shell (car body or window) of the high-speed train can be used as a hub to connect the users in the car with the base station beside the track to enhance T2T communication.
[0067] Generally speaking, trackside base stations communicate with on-board users in a single-hop or dual-hop mode. Specifically, in the single-hop mode where the base station communicates directly with the user, the electromagnetic waves must penetrate the metal body or windows of the enclosed carriage to reach the user end. However, the penetration loss of electromagnetic signals in the high-frequency band is much more serious than that in the low-frequency band, resulting in extremely weak received signal strength. In addition, the passengers on the train move very fast with the high-speed train, which will cause the entire communication link to exhibit high-dynamic fast-varying fading characteristics. As an effective alternative to the single-hop mode, the dual-hop communication mode (such as MRS (Mobile Relay Stations) and XL-RRS) can significantly improve the service quality in T2T communication. It should be pointed out that MRS, which has been widely used in the current high-speed rail communication network, can relay the signals of the trackside base station to hundreds of on-board users, or forward the signals of on-board users to the base station. Similarly, the millimeter-wave XL-RRS installed on the outer shell of the train also has many advantages, and in the surrounding Figure 2 It plays a vital role in the heterogeneous networks designed in .
[0068] 2. Advantages of millimeter wave XL-RRS in intelligent high-speed rail communication network
[0069] 1) Location and deployment
[0070] In order to maintain the streamlined appearance of high-speed trains, the MRS with a uniform planar array antenna array structure is usually embedded in the roof of the car in a horizontal direction. This requires the trackside base station to have a certain height to ensure that the MRS end has an appropriate grazing angle, but this may increase the construction and maintenance costs of the base station. In contrast, the location and deployment of the millimeter wave XL-RRS will be more flexible. Specifically, by utilizing the current mature skin antenna technology, the millimeter wave XL-RRS can be deployed at any position of the train, including the roof, shell, and windows on both sides, thus ensuring a favorable grazing angle, thereby achieving LoS link communication. Researchers at NTT DOCOMO in Japan designed the following Figure 1 (b) and Figure 3 A passive smart glass metasurface prototype is shown in the figure, which contains a large number of subwavelength array elements arranged on a highly transparent glass substrate. Therefore, this smart glass metasurface can be used to design passive millimeter-wave XL-RRS devices to ensure that they can be widely deployed on train windows, thereby realizing smart high-speed rail communication networks.
[0071] 2) Power consumption and transmission delay
[0072] As an active device, MRS needs to be equipped with multiple high-cost and high-energy RF links to serve multiple users at the same time, and the corresponding relay processing will inevitably introduce additional transmission delays, which is not suitable for high-dynamic fast-fading T2T communication channels. On the contrary, the millimeter-wave XL-RRS only uses low-cost and low-power meta-atoms to design a passive hardware architecture, and it only needs a small amount of power to manipulate the meta-atoms to refract electromagnetic signals to complete the signal forwarding process. Since there is no need for cumbersome baseband signal processing, the delay caused by the electromagnetic signal penetrating the millimeter-wave XL-RRS installed on the train shell can be ignored.
[0073] 3) Spatial multiplexing capability
[0074] In the high-speed rail communication scenario, since there is no scatterer between the trackside base station and the XL-RRS or MRS, T2T communication can only establish a single LoS link. For MRS, it can only use the LoS link without spatial multiplexing gain to connect to the trackside base station, which severely limits the throughput of the entire T2T communication system. Therefore, the data rate it can provide is limited, and the number of users that can be served simultaneously is also small. In contrast, as shown in the attached figure, Figure 2 and attached Figure 3 As shown in the figure, a millimeter-wave XL-RRS array can be divided into multiple sub-array blocks installed at different positions in the car, forming a giant dimensional array with a very large effective aperture. In this way, the millimeter-wave XL-RRS not only utilizes the passive beamforming gain of the large-aperture metasurface to improve the received SNR (Signal-to-Noise Ratio), but also utilizes the unique near-field LoS path intra-multiplexing characteristics in the large-aperture array to obtain spatial multiplexing gain to increase the data transmission rate. Therefore, the millimeter-wave XL-RRS-assisted T2T communication is sufficient to ensure that high-speed rail passengers enjoy the ultimate Internet surfing experience such as immersive extended reality and high-fidelity mobile holograms.
[0075] 3. Millimeter wave XL-RRS architecture design
[0076] In order to make full use of the intra-path multiplexing characteristics in the above-mentioned near-field LoS link, an additional Figure 3 The specific architecture of the millimeter wave XL-RRS is shown in Figure 1. Specifically, a millimeter wave XL-RRS can be split into multiple RRS sub-blocks, each of which is installed at a designated position on the train shell. Accordingly, according to the specific locations of the users in the car, these users can be divided into multiple user groups, and the users in each group are served by the same RRS sub-block. A feasible example of a future intelligent high-speed rail communication system is to replace traditional window glass with a metasurface based on smart glass.
[0077] The millimeter wave XL-RRS can control the phase and amplitude of its double-sided outer layer atoms in real time by integrating and processing the CSI (Channel State Information) obtained by various means (such as high-speed rail control systems, railway Internet of Things, and advanced channel estimation methods, etc.), thereby providing better signal coverage services for users in the car with the help of considerable near-field LoS link path multiplexing gain. At the same time, the millimeter wave XL-RRS designed above can split the high-dynamic fast time-varying T2T channel between the trackside base station and the user in the train into a cascade form of the channel between the BS (Base Station) and RRS and the channel between RRS and UE (User Equipment), among which the fast time-varying LoS link channel between the BS-RRS outside the car can be reliably predicted according to the characteristics of the high-speed rail train running along a fixed track, and the quasi-static channel between the BS-RRS in the car can be obtained using some advanced channel estimation methods. In addition, the CSI obtained above can also solve the pre-compensation problem of Doppler frequency shift.
[0078] 4. Establishment of channel model between base station and user
[0079] The trackside millimeter wave base station uses a hybrid beamforming architecture in the form of a partial sub-connected phase-shifted network to serve K single-antenna UEs in the high-speed train. The specific form of its antenna array is considered in the attached figure. Figure 4 (a) shows a multi-block antenna array. This multi-block array form with a larger antenna effective aperture can also provide a certain near-field path multiplexing gain for the LoS communication link. According to the schematic diagram of the millimeter-wave XL-RRS architecture designed above, the millimeter-wave XL-RRS structure installed on the train shell is shown in the attached figure. Figure 4 (b) as shown.
[0080] The specific configuration of the base station antenna array and millimeter wave XL-RRS is as follows. Since the base station adopts PSN with partial sub-connection, it is defined here and are the number of sub-arrays (antennas in each sub-array) of the multi-block antenna array in the horizontal and vertical directions, respectively, and and are the number of antennas in the horizontal and vertical directions of the entire array. Then, the total number of antennas in each subarray and the entire antenna array at the base station is and Note that the number of RF links at the base station is the number of subarrays, i.e. In addition, the spacing between adjacent sub-blocks and adjacent antennas in a multi-block antenna array is defined as Δ and d, respectively. On the other hand, the millimeter wave XL-RRS is also split into multiple RRS sub-blocks, so its dimensional parameters have similar definitions to the corresponding parameters at the base station, which can be expressed as as well as and It should be noted that the spacing between adjacent RRS sub-blocks is much larger than the Δ defined by the base station, which can be set in the simulation. Considering that the trackside millimeter-wave base station needs to be aligned with the millimeter-wave XL-RRS installed on the train shell, here, without loss of generality, let and
[0081]
[0082] 1) Channel model between base station and RRS
[0083] Since the entire surface of the millimeter wave XL-RRS has a giant array aperture, the RRS sub-blocks can be considered independent of each other. Then, the channel matrix between the millimeter wave XL-RRS and the BS-RRS between the multiple antenna arrays of the trackside base station at the jth time slot can be expressed as in is the i-th (1≤i≤I RRS ) The channel matrix between the RRS sub-block and the base station can be further expressed as follows:
[0084]
[0085] where α i ~CN(0,1) is the channel gain, and D i are the large-scale fading coefficient and signal transmission distance between the i-th RRS sub-block and the base station, λ is the wavelength, exp(2πD i / λ) corresponds to the near-field spherical wave component, f d =v d / λ and T s denote the Doppler frequency shift and the duration of a time slot, respectively, and v d is the relative radial velocity, and and are defined as the horizontal and vertical virtual angles of the base station end (i-th RRS sub-block), and and are the corresponding azimuth and elevation angles respectively. In formula (1),
[0086]
[0087] is the array response vector at the base station, where and are the horizontal and vertical steering vectors respectively, and the vector and can be further written as
[0088]
[0089] And the vector and They have the same form as formula (2) and (3) respectively. In addition, in formula (1)
[0090]
[0091] is the array response vector of the ith RRS sub-block, where the horizontal and vertical steering vectors and Can be expressed as
[0092]
[0093] It should be emphasized here that the i-th (1≤i≤I RRS ) Large-scale fading coefficient between RRS sub-block and base station Its specific expression can be deduced according to the Friis formula:
[0094]
[0095] in and are the antenna array gains of the base station and the i-th RRS sub-block, respectively, and It can be further expressed as
[0096]
[0097] The function Representation and real angle The value of the relevant beam pattern is directly related to the precoding / combining vector designed by the base station and the actual array response vector of the base station. It has an expression similar to formula (7).
[0098] 2) Channel model from RRS to user
[0099] The channel matrix H between BS-RRS UL Similarly, the channel vector between the millimeter wave XL-RRS and the kth (1≤k≤K) user can be expressed as in is the i-th (1≤i≤I RRS ) RRS sub-block to user channel vector (including Lk,i paths), that is
[0100]
[0101] where γ k,i,l ~CN(0,1) is the lth (1≤i≤L k,i ) paths corresponding to the channel gain, and D k,i,l are the large-scale fading coefficient and signal transmission distance between the lth path of the kth user and the i-th RRS sub-block, respectively. Here, the antenna array gain has an expression similar to formula (7), and and are defined as the corresponding horizontal and vertical virtual angles (azimuth and elevation). In formula (8),
[0102]
[0103] is the corresponding array response vector, where the horizontal and vertical steering vectors and They have the same form as formula (4) and (5) respectively.
[0104] This embodiment also discloses a millimeter wave XL-RRS assisted T2T data transmission solution:
[0105] The proposed millimeter-wave XL-RRS-assisted T2T data transmission scheme adopts the millimeter-wave XL-RRS architecture, and further designs an uplink data transmission scheme based on packet multi-user IDMA (Interleave-Division Multiple Access) and a downlink precoding scheme based on ZF (Zero-Forcing).
[0106] 1) Uplink data transmission scheme based on packet multi-user IDMA
[0107] According to the channel model between the base station and RRS and between RRS and the user, the base station receives the signal vector corresponding to the jth time slot during uplink data transmission. for
[0108]
[0109] in and are the base station end analog and digital receiving merging matrices and is the i-th (1≤i≤I BS ) The simulated receiving combined vector corresponding to the subarray, the diagonal matrix and For attachment Figure 1 (b) and Figure 3 The phase value corresponding to the double-sided outer layer in the designed millimeter-wave XL-RRS, P tx and k [j] are the user transmit power and the uplink data sent by the kth (1≤k≤K) user, and n UL [j] is zero-mean and has a covariance matrix of Additive White Gaussian Noise (AWGN), that is In formula (9), the diagonal matrix Φ UL and UL It can be specifically expressed as and For the i-th (1≤i≤I RRS )Diagonal matrix and It can be further expressed as and Here the vector and That is, the phase value corresponding to the i-th RRS sub-block. In addition, the digital receiving merging matrix is directly considered as the unit matrix, that is,
[0110] As a non-orthogonal multiple access technology, IDMA can use a unique, user-specific pseudo-random interleaver to distinguish different users, so the number of users it supports can exceed the number of RF links equipped at the base station. At the same time, the signal detection complexity of the multi-user detector used in IDMA is proportional to the number of users. Therefore, IDMA technology is very suitable for supporting a large number of high-speed rail passengers to access real-time broadband services at the same time. According to the attached Figure 3 The millimeter wave XL-RRS designed in this paper proposes an additional Figure 5 The T2T data transmission scheme based on packet multi-user IDMA is shown to achieve reliable uplink data transmission. The packet multi-user IDMA scheme will be described in detail below.
[0111] In the uplink data transmission phase, a known CSI is assumed to design the base station receiving combining matrix W RF , and the phase value Φ corresponding to the double-sided outer layer in the millimeter wave XL-RRS UL and UL At the same time, the Doppler frequency shift compensation of the BS-RRS channel is also considered. Considering that the BS-RRS channel is mainly a LoS link, the W can be designed by beam alignment. RF and Φ ULTo obtain sufficient beamforming gain, the RRS-UE channel exhibits a rich scattering characteristic, so the energy focusing method can be used to design Ψ UL , in order to maximize the energy on the user side. Specifically, first for W RF The i-th (1≤i≤I BS ) simulates receiving the combined vector w RF,i , you can first initialize w RF,i for Then make Among them I BS,i It represents the antenna index set of the ith subarray at the base station in the entire array and its cardinality is M BS =|I BS,i | c Secondly, for Φ UL The i-th (1≤i≤I RRS ) Phase vector Can be directly ordered Finally, the users in the car are divided into M user groups according to their specific locations, and the number of users in each group is U = K / M. Then for Ψ UL The i-th (1≤i≤I RRS ) Phase vector Can be ordered To achieve energy focusing.
[0112] Using the above designed W RF , Φ UL And UL , formula (9) can be further written as
[0113]
[0114] in is the BS-RRS channel after Doppler shift compensation,
[0115] and Consider the channel matrix and transmission vector of K users respectively, represents the low-dimensional equivalent channel matrix, and According to the user grouping and simplified signal transmission model in formula (10), the following IDMA-based data detection process can be completed.
[0116] As attached Figure 5 As shown in , for the kth (1≤k≤K) user, the data bit it sends is denoted as d k , after passing through the Turbo encoder, the encoded sequence c can be obtained k , and then pass through the spread spectrum, interleaver π k And constellation modulation can be obtained Where J is the frame length of the data sequence sent by the user. For different users, their interleavers is unique. Here we assume that these interleavers are independently and randomly generated. Then, according to formula (10), the received signal matrix at the base station is It can be expressed as in is the data matrix of frame length J sent by K users, and N UL is the corresponding noise matrix. The design of the base station receiver can be divided into two processes: multi-user detector (MUD) and decoder (DEC), and these two parts iterate each other until convergence, and finally output the reconstructed bits of K users. There is a mature solution for the specific calculation process, which will not be elaborated here.
[0117] It should be pointed out that since the proposed data transmission scheme adopts a low-complexity but efficient multi-user detector, the proposed packet multi-user IDMA scheme can serve more users compared with traditional detection methods such as maximum ratio combining and zero forcing schemes. Based on the multi-user detection characteristics of IDMA, the proposed millimeter-wave XL-RRS-assisted T2T data transmission scheme can further shorten the access delay of many users in the car.
[0118] 2) Downlink precoding scheme based on ZF (Zero-Forcing)
[0119] According to the channel reciprocity of the time division duplex system, during the downlink transmission process, the same equation as in formula (10) can also be obtained. The corresponding low-dimensional equivalent channel matrix Right now in is the channel matrix that combines the downlink channel vectors of K users in the car and is the RRS-UE channel vector of the kth (1≤k≤K) user, Φ DL and DL is the phase value corresponding to the double-sided outer layer of the millimeter wave XL-RRS during downlink, and the diagonal matrix Φ DL and DL The selection of is respectively related to Φ in formula (9) UL and UL similar, is the BS-RRS channel after Doppler shift compensation in downlink, and is the analog precoding matrix at the base station. When designing downlink precoding, F RF , Φ DL and DL The beam alignment and energy focusing methods described above are also used. Then, consider I BS= K, the digital precoding matrix at the base station can be designed by using zero-forcing precoding Right now where f BB,k is the digital precoding vector corresponding to the kth user. Therefore, the system sum rate (denoted as R) can be calculated as
[0120]
[0121] Where P n is the noise power, which can be calculated based on the power spectral density of the background noise.
[0122] In order to illustrate the advantages of the present invention in uplink data detection bit error rate (BER) and downlink data transmission system throughput performance compared with the single-hop and MRS communication modes in the traditional railway communication network, the attached diagram is used here. Figure 6 to Figure 8 To illustrate the effect of the present invention. Specifically, considering the following Figure 6 The scenes of three T2T communication modes, single hop, MRS and mmWave XL-RRS, are shown. The trackside base station can serve multiple user groups using single antennas in the vehicle using multi-block antenna arrays, and for the proposed XL-RRS mode, all users can be split into M user groups and each user group corresponds to an RRS sub-block. It should be pointed out that the MRS mode uses beam alignment and single-stream transmission in the BS-MRS communication link, while multi-user zero-forcing precoding / detector is used in the MRS-UE communication link.
[0123] Attached Figure 7 The three T2T communication modes are compared with the transmission power P tx The uplink BER performance when the BER changes is used to illustrate the advantages of the proposed millimeter wave XL-RRS assisted T2T data transmission solution. In the uplink, the single-hop and XL-RRS modes use the traditional IDMA and the proposed packet multi-user IDMA schemes respectively to compare the performance difference between the two. Figure 7 It can be observed that the BER performance of the packet IDMA scheme designed in the proposed millimeter wave XL-RRS assisted solution is much better than the other two modes in the entire transmission power range, and the BER performance will be significantly improved when the number of users served is smaller. In addition, it can be noted from the figure that under the simulation parameter setting (considering QPSK, the modulation order is recorded as Q), the system throughput of the three modes of single hop (K=8), MRS (K=1) and XL-RRS (K=8 or K=16) is Kf respectively. slog2(Q) = 3.2 Gigabit per second (Gbps), 400 Megabit per second (Mbps) and 3.2 Gbps or 6.4 Gbps. When the number of users K = 16, the proposed data transmission scheme based on packet multi-user IDMA only needs to use N RF =8 RF links can support 2N simultaneously RF = Data transmission for 16 users. In other words, the proposed solution can serve twice as many users as the other two modes using the same number of RF chains.
[0124] Attached Figure 8 The three T2T communication modes are compared with the transmission power P tx The achievable downlink and rate performance when the beam width changes. In the downlink, the base station in single-hop and XL-RRS mode can perform downlink precoding design based on zero forcing according to the low-dimensional equivalent baseband channel after beam alignment. Figure 8 It can be observed that the designed millimeter-wave XL-RRS has a large effective array aperture, which can introduce considerable near-field LoS link intra-path multiplexing gain, which makes the achievable downlink sum rate of the proposed millimeter-wave XL-RRS assisted solution much higher than the other two modes in the entire transmit power range. In contrast, since the MRS mode can hardly obtain additional LoS link spatial multiplexing gain, its sum rate is lower than the transmit power P. tx In addition, the single-hop mode is affected by severe penetration loss, which will result in very weak received signal energy. Therefore, at low transmit power (i.e., P tx <6dBm), its sum rate is even worse than that of MRS mode.
[0125] In summary, the above are only preferred embodiments of the present invention and are not intended to limit the protection scope of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.
Claims
1. A train and trackside communication system for intelligent high-speed rail networks, Features: Including the mmWave XL-RRS architecture, the intelligent high-speed rail heterogeneous network based on mmWave XL-RRS, and the channel model between the trackside mmWave base station and the users in the train; The millimeter wave XL-RRS architecture includes a plurality of RRS sub-blocks, wherein each RRS sub-block is mounted at a designated position on the train shell; The intelligent high-speed rail heterogeneous network based on millimeter wave XL-RRS includes a central unit CU, a distributed unit DU and an active antenna unit AAU, wherein the central unit CU is connected to multiple distributed units DU, and the distributed unit DU controls multiple active antenna units AAU; The channel model between the trackside millimeter wave base station and the user in the train is split into a base station to RRS channel model and an RRS to user channel model by the millimeter wave XL-RRS installed on the train shell.
2. A train and trackside communication system for an intelligent high-speed rail network according to claim 1, Features: Low-frequency massive MIMO base stations operating in the frequency band below 6 GHz and high-frequency massive MIMO base stations operating in the millimeter wave or terahertz frequency bands undertake the tasks of DU and AAU respectively; Each low-frequency massive MIMO base station is connected to multiple high-frequency massive MIMO base stations through a fronthaul link, and multiple adjacent low-frequency massive MIMO base stations transmit their signals to the CU through a midhaul link for joint processing, and then the CU is connected to the core network through a backhaul link. The midhaul link is the transmission link connecting the DU and the CU in the current 5G mobile communication standard networking method.
3. A train and trackside communication system for an intelligent high-speed rail network according to claim 1, Features: The intelligent high-speed rail heterogeneous network based on millimeter wave XL-RRS adopts a cell-free network architecture, which can build a virtual cell centered on the high-speed rail train. Since the CU is connected to multiple DUs, and each DU can jointly network and control multiple AAUs, the CU can comprehensively dispatch different AAUs to use prior information to track the train.
4. A train and trackside communication system for an intelligent high-speed rail network according to claim 1, Features: In the channel model between the trackside millimeter wave base station and the user in the vehicle, the following parameters are defined: and are the number of sub-arrays (antennas in each sub-array) of the multi-block antenna array in the horizontal and vertical directions, respectively, and and are the number of antennas in the horizontal and vertical directions of the entire array respectively; the total number of antennas in each subarray and the entire antenna array at the base station is and The number of RF links at the base station is the number of subarrays, that is, The spacing between adjacent sub-blocks and adjacent antennas in a multi-block antenna array is defined as Δ and d, respectively; the millimeter wave XL-RRS is also split into multiple RRS sub-blocks, so its dimensional parameters and the corresponding parameters at the base station are expressed as as well as and Then, the channel model between the base station and the RRS is specifically: The channel matrix between the millimeter wave XL-RRS and the BS-RRS between the multiple antenna arrays of the trackside base station at the jth time slot is expressed as in is the i-th (1≤i≤I RRS ) The channel matrix between the RRS sub-block and the base station can be further expressed as follows: where α i ~CN(0,1) is the channel gain, and D i are the large-scale fading coefficient and signal transmission distance between the i-th RRS sub-block and the base station, λ is the wavelength, exp(2πD i / λ) corresponds to the near-field spherical wave component, f d =v d / λ and T s denote the Doppler frequency shift and the duration of a time slot, respectively, and v d is the relative radial velocity, and and are defined as the horizontal and vertical virtual angles of the base station end (i-th RRS sub-block), and and are the corresponding azimuth and elevation angles, and are the array response vectors of the base station and the i-th RRS sub-block, respectively, and (·) H is the conjugate transpose operation.
5. A train and trackside communication system for an intelligent high-speed rail network according to claim 4, Features: The RRS to user channel model is specifically: The channel vector between the millimeter wave XL-RRS and the kth (1≤k≤K) user is expressed as in is the i-th (1≤i≤I RRS ) RRS sub-block to user channel vector (including L k,i paths), that is where γ k,i,l ~CN(0,1) is the lth (1≤i≤L k,i ) paths corresponding to the channel gain, and D k,i,l are the large-scale fading coefficient and signal transmission distance between the lth path of the kth user and the i-th RRS sub-block, and and ( and ) are defined as the corresponding horizontal and vertical virtual angles (azimuth and elevation), is the corresponding array response vector.
6. A method for transmitting data between a train and a trackside. Features: A train and trackside communication system for an intelligent high-speed rail network according to any one of claims 1 to 5, which is divided into an uplink data transmission method based on packet multi-user IDMA and a downlink precoding method based on ZF; The uplink data transmission method based on packet multi-user IDMA is: According to the channel model between the base station and RRS and between RRS and the user, the base station receives the signal vector corresponding to the jth time slot during uplink data transmission. for in and are the base station end analog and digital receiving merging matrices and is the i-th (1≤i≤I BS ) The simulated receiving combined vector corresponding to the subarray, the diagonal matrix and is the phase value corresponding to the double-sided outer layer in the designed millimeter-wave XL-RRS, P tx and k [j] are the user transmit power and the uplink data sent by the kth (1≤k≤K) user, and n UL [j] is zero-mean and has a covariance matrix of The complex additive Gaussian white noise is Diagonal matrix Φ UL and UL Expressed as and For the i-th (1≤i≤I RRS )Diagonal matrix and Further expressed as and Here the vector and That is, the phase value corresponding to the i-th RRS sub-block; In addition, the digital receiving merging matrix is directly considered as the unit matrix, that is, W BB =I IBS ; The downlink precoding method based on ZF (Zero-Forcing) is: In the downlink transmission process, the channel reciprocity of the time division duplex system is used to obtain The corresponding low-dimensional equivalent channel matrix Right now in is the channel matrix that combines the downlink channel vectors of K users in the car and is the RRS-UE channel vector of the kth (1≤k≤K) user, Φ DL and DL is the phase value corresponding to the double-sided outer layer of the millimeter wave XL-RRS during downlink, and the diagonal matrix Φ DL and DL The selection of UL and UL similar, is the BS-RRS channel after Doppler shift compensation in downlink, and is the analog precoding matrix at the base station; when designing downlink precoding, then, consider I BS = K, the digital precoding matrix at the base station is designed by using zero-forcing precoding Right now where f BB,k is the digital precoding vector corresponding to the kth user.
7. A method for transmitting data between a train and a trackside according to claim 6, Features: In the uplink data transmission phase, a known CSI is assumed to design the base station receiving combining matrix W RF , and the phase value Φ corresponding to the double-sided outer layer in the millimeter wave XL-RRS UL and UL At the same time, the Doppler frequency shift compensation of the BS-RRS channel is also considered; considering that the BS-RRS channel is mainly a LoS link, the W is designed through beam alignment. RF and Φ UL In order to obtain sufficient beamforming gain, the RRS-UE channel exhibits a rich scattering characteristic, so the energy focusing method is used to design Ψ UL , in order to maximize the energy on the user side; specifically, first for W RF The i-th (1≤i≤I BS ) simulates receiving the combined vector w RF,i , first initialize w RF,i for Then make Among them I BS,i It represents the antenna index set of the ith subarray at the base station in the entire array and its cardinality is M BS =|I BS,i | c ; Secondly, for Φ UL The i-th (1≤i≤I RRS ) Phase vector Can be directly ordered Finally, the users in the car are divided into M user groups according to their specific locations, and the number of users in each group is U = K / M. Then for Ψ UL The i-th (1≤i≤I RRS ) Phase vector Can be ordered To achieve energy concentration; using the above designed W RF , Φ UL And UL ,have in is the BS-RRS channel after Doppler shift compensation, and Consider the channel matrix and transmission vector of K users respectively, represents the low-dimensional equivalent channel matrix, and According to the user grouping and simplified signal transmission model, the data detection process based on IDMA can be completed.
