Data demand-driven resource joint optimization method for high-speed rail train-ground communication

Abstract

The present invention relates to the field of wireless communication technology, and specifically to a resource joint optimization method driven by data demand under high-speed rail vehicle-to-ground communication, comprising: S1, constructing a high-speed rail vehicle-to-ground communication model in motion according to the data demand of the equipment and users in the carriage; S2, constructing a millimeter wave transmission channel model, and establishing an overall optimization problem with the goal of minimizing the transmission time slot; S3, performing an equivalent transformation on the overall optimization problem to reduce the dimension of the problem; S4, decomposing the equivalently transformed problem into a time planning sub-problem and a power control optimization sub-problem for alternating optimization and solving, and introducing a forcing step in the alternating optimization solution process so that the necessary conditions for the optimal solution of the problem are met after the alternating optimization is completed, and a time slot planning strategy and a power allocation strategy are obtained; S5, expanding the resource allocation strategy to consider meeting the green communication demand. The present invention can allocate time and power on demand according to user demand to achieve the purpose of minimizing transmission time and green communication.

Description

Technical Field

[0001] The present invention relates to the field of wireless communication technology, and more particularly to a data demand-driven resource joint optimization method for high-speed rail train-to-ground communication. Background Art

[0002] The rapid development of high-speed rail in recent years has highlighted its critical role in modern transportation networks. This mode of transportation is widely favored for its low energy consumption, efficient carrying capacity, comfortable travel experience, and excellent safety standards. With the iterative upgrades of high-speed rail, the demand for high-quality, low-latency broadband wireless access services is increasing rapidly to support a variety of applications, such as high-data application services, low-latency online conferencing access, reliable transmission of train control signals, and the railway Internet of Things.

[0003] Although various mature broadband wireless access technologies exist, their transmission rates are far below the levels required for future high-speed rail communications. Therefore, millimeter wave technology, with its large bandwidth advantage, is seen as a key solution for achieving high data rate transmission and supporting future communication needs. Considering that millimeter waves suffer significant penetration losses due to the metal cladding of train cars, a mobile relay (MR) has been introduced to address this issue. It is installed on the roof of each train car to provide communication between the base station and the devices within the car.

[0004] To achieve efficient resource transmission, spatial multiplexing is used, resulting in stronger interference between co-frequency signals. This means that efficient resource allocation is also key to meeting the Quality of Service (QoS) requirements of users or devices, thereby avoiding network congestion and degradation of communication quality. Better resource allocation algorithms typically require more computing resources and time to complete. In wireless communication systems, complex algorithms may require more time to generate results, which increases the response time of the communication system. The power optimization problem alone is an NP-hard problem, and the time slot allocation algorithm is an integer programming problem. Reducing the complexity of the algorithm and improving the feasibility of the solution are also key issues.

[0005] Furthermore, the objective function setting and constraints of the problem also require careful consideration. Reasonable settings can be used to guide resource allocation without wasting resources. Most existing algorithms maximize system throughput while meeting minimum user requirements. However, this optimization can lead to throughput overflow, causing some nodes with good channel conditions to have excessive communication resources, while some nodes with poor channel conditions can only meet minimum communication requirements and fail to fully utilize communication resources, resulting in waste. However, currently, in D2D scenarios such as the proposed high-speed rail scenario, few solutions exist that combine time and power optimization algorithms to strictly meet user demand constraints, and they do not consider green communication.

[0006] Therefore, in the case of resource scarcity, how to efficiently solve the joint optimization problem of resource allocation to meet the real-time data needs of future high-speed rail and the concept of green energy has become a technical problem that technical personnel in this field urgently need to solve. Summary of the Invention

[0007] In view of this, the present invention provides a resource joint optimization method based on data demand driven in high-speed rail train-to-ground communication, which can allocate time and power on demand according to user needs to achieve the purpose of minimized transmission time and green communication.

[0008] In order to achieve the above object, the present invention adopts the following technical solutions:

[0009] A data demand-driven resource joint optimization method for high-speed rail train-to-ground communication includes the following steps:

[0010] S1. Build a high-speed rail vehicle-to-ground communication model based on the data needs of the equipment and users in the carriage. In the high-speed rail vehicle-to-ground communication model, each carriage has a mobile relay on top, and the base station is deployed with multiple millimeter-wave directional antennas to exchange data with each mobile relay.

[0011] S2. Build a millimeter wave transmission channel model, analyze the impact of link mutual interference on transmission performance, and establish an overall optimization problem with the goal of minimizing the transmission time slot;

[0012] S3, performing equivalent transformation on the overall optimization problem to reduce the dimension of the problem;

[0013] S4. Decompose the problem after equivalent transformation into a time planning subproblem and a power control optimization subproblem. Alternately optimize and solve the two subproblems. In the alternating optimization process, introduce a forcing step to ensure that the necessary conditions for the optimal solution of the problem are met after the alternating optimization is completed, and obtain the time slot planning strategy and power allocation strategy.

[0014] S5. Establish a green communication problem, substitute the solution obtained in S4 as an initial value into the green communication problem, perform iterative solution, and obtain a resource allocation strategy that meets the green communication requirements.

[0015] Furthermore, in S1, in the high-speed rail train-to-ground communication model, the equipment in any car is connected to the mobile relay on the top of the car through the access points distributed inside the car. Each mobile relay communicates with the base station on behalf of the equipment in its own car and obtains data resources from the base station through the superframe transmission protocol.

[0016] Furthermore, when the superframe transmission protocol is used for resource allocation, the entire superframe includes three phases: a beacon phase, a request phase, and a transmission phase;

[0017] During the beacon phase, each mobile relay synchronizes with the base station to receive control information from the base station;

[0018] In the request phase, each mobile relay sends a request for the required data to the base station;

[0019] During the transmission phase, a superframe consists of hundreds or thousands of time slots. Within each time slot, multiple mobile relays can simultaneously access the base station for spatial multiplexing, and each mobile relay can select multiple time slots for data transmission.

[0020] During the request phase, the base station collects data requests from each mobile relay and allocates time slots and power resources based on meeting the data needs of each mobile relay. In the next beacon phase, the base station notifies the corresponding mobile relay of the control message and resource allocation results, so that the data request from the previous superframe is transmitted in the subsequent superframe.

[0021] Furthermore, in S2, the millimeter wave transmission channel model is expressed as:

[0022] The link transmitted by the base station to the jth mobile relay is link j, and the channel gain G caused by the jth link to the mth mobile relay is mj Expressed as:

[0023]

[0024] Where k represents the constant fading factor, which is related to the wavelength of the millimeter wave; G t (m,j) represents the transmit antenna gain of link j to the mth mobile relay, G r (m,j) represents the receiving antenna gain of the mth mobile relay for link j, represents the distance between the base station and the mth mobile relay, and γ represents the exponential decay factor;

[0025] According to Shannon's formula, the rate of the mth mobile relay in the sth time slot is Expressed as:

[0026]

[0027] in, in, represents the signal to interference and noise ratio of the mth mobile relay in the sth time slot; G mm represents the channel gain caused by the m-th link to the m-th mobile relay; represents the transmission power of the base station on link m to the mth mobile relay in the sth time slot; represents the transmission power of the base station's link j to the jth mobile relay in the sth time slot; It represents the multiple of the difference from the Shannon formula due to the actual situation, W represents the channel bandwidth, and N0 represents the power spectral density of Gaussian white noise.

[0028] Furthermore, the overall optimization problem P1 established in S2 is expressed as:

[0029]

[0030] Among them, L slot Indicates the time length of each time slot; S represents the number of time slots used in the superframe, which is a natural number;

[0031] represents the data demand of the mth mobile relay; the constraint C1 in problem P1 means that the data transmitted to each mobile relay is greater than To meet the data requirements of the carriage represented by the mobile relay; M represents the total number of carriages;

[0032] Constraint C2 in problem P1 represents the transmission power of link m corresponding to the mth mobile relay in the sth time slot. Less than the maximum transmit power P max ; S represents the time slot set of the entire superframe; when When it is equal to 0, that is, no transmit power, it means that the link is not planned in this time slot;

[0033] Constraint C3 in problem P1 guarantees that the number of time slots does not exceed the maximum number of time slots S. max , Expressed as a natural number.

[0034] Furthermore, S3 includes:

[0035] S3.1. Convert the overall optimization problem P1 into problem P2. Problem P2 is expressed as:

[0036]

[0037] Where N means that the time is divided into N time periods, and the value of N is equal to M; t n Indicates the nth time period; represents the data transmission rate of the mth mobile relay corresponding to time period n; Represents the nth time period t n When , the power of the mth mobile relay;

[0038] Constraint C1 in problem P2 guarantees that the sum of data obtained by the mth mobile relay in all time periods is greater than its target data demand; where the time vector is represented by t n Composition, expressed as t=[t1,t n ,…,t N ] T , when transformed into the original problem At each time period t n The number of time slots that need to be converted into corresponding duration S n To meet the transmission requirements, the calculation formula is symbol It is rounded up, and the mth mobile relay has S n The transmission power of a time slot is

[0039] Constraint C2 in problem P2 represents the power of the mth mobile relay in time period n. Less than the maximum transmit power P max ;

[0040] Constraint C3 in problem P2 means that for each time period t n The length of is greater than or equal to 0;

[0041] S3.2. Convert problem P2 into problem P3. Problem P3 is expressed as:

[0042]

[0043] The necessary condition for the optimal solution of problem P3 is that the equality of constraint C1 holds.

[0044] Furthermore, in S4, the time planning sub-problem is split from problem P3 and expressed as:

[0045]

[0046] Use the interior point method to solve problem P4.

[0047] Furthermore, in S4, the power control optimization subproblem is solved using successive convex approximation. In each iteration, the problem is transformed into a convex function and solved iteratively to obtain the final result of the problem. In each iteration, the power control optimization subproblem is decomposed from problem P3 and expressed as:

[0048]

[0049] in, It represents the data transmission rate of the mth mobile relay in the nth time period, and its expression is:

[0050]

[0051] in, The value of For all The vector composed of The value of is used to obtain the power P of the mobile relay, that is, the power allocation strategy is obtained; and Both are expressed as the conversion coefficient of the mth mobile relay in the nth time period; express

[0052] In each iteration, and The values ​​are as follows:

[0053]

[0054] represents the signal-to-interference-and-noise ratio of the mobile relay in the nth time slot;

[0055] Through continuous iterative solution, we finally get P, and The convergent solution of .

[0056] Furthermore, in S4, the alternating optimization and solution process for the two sub-problems includes:

[0057] For the time planning sub-problem, fix the power P and schedule the time t to minimize the objective;

[0058] For the power control optimization sub-problem, with fixed time t, the objective function is maximized by optimizing the power P of different mobile relays in different time periods;

[0059] Add a forcing step. When the time planning subproblem and the power control optimization subproblem cannot be optimized further, determine whether the necessary conditions of the problem are met. If not, update P through the forcing step and bring the updated results into the time planning subproblem and the power control optimization subproblem for further iterative loop solution until the two subproblems converge and cannot be optimized further and the necessary conditions are met.

[0060] The process of integrating the forcing step into the alternating optimization for joint optimization includes:

[0061] use Denotes the data transmission rate matrix before the forced equalization step, which is represented by composition; adoption Denotes the data transmission rate matrix after the forced equalization step, which is given by composition;

[0062] Before conversion, define a reference data transmission rate matrix For reference conversion, each element is calculated as follows:

[0063]

[0064] Construct the necessary conditions for sub-problem P6:

[0065]

[0066] Using feasibility check and n Take the bisection method to find the solution of problem P6 and get the time interval t n The multiplication factor ξ n , and get each updated time period t n Corresponding power Finally, the updated power P of the mobile relay is obtained through the algorithm.

[0067] The updated results are brought into the time planning subproblem and the power control optimization subproblem for further iterative loop solution to make the necessary conditions meet.

[0068] Furthermore, in S5, the expression of the green communication problem P7 is:

[0069]

[0070] Among them, the objective function in the green communication problem is to minimize the energy required to transmit all data; the constraint C1 in problem P7 means that the number of time slots required for transmission cannot exceed the specified number, and the (S max -N)*L slot Guarantee Issues Convert to the original problem When the solution is obtained, the number of time slots used is still less than S max and satisfy the constraints;

[0071] The problem Substitute the solved values ​​P and t into problem P7, and solve the problem Continue to split into problem-based The power control optimization subproblem and time planning subproblem are solved iteratively. The solution method and problem of power control optimization sub-problem and time planning sub-problem The solution method of the power control optimization sub-problem and the time planning sub-problem is the same, and the specified time slot S is finally obtained. max Resource allocation strategy within quantity.

[0072] It can be seen from the above technical solutions that compared with the prior art, the present invention has the following beneficial effects:

[0073] This paper addresses the resource constraints and inter-link interference issues of high-speed rail downlinks by proposing a system that leverages millimeter waves and an efficient space-time resource optimization algorithm. By combining time and power control, this approach optimizes resource allocation based on the Space Time Division Multiple Access (STDMA) protocol, thereby avoiding inefficient resource waste and ultimately improving both the real-time nature and efficiency of data transmission.

[0074] Aiming at the problem of high computational complexity of the system, equivalent transformation is used to reduce the complexity of the algorithm. At the same time, in order to ensure the effectiveness of the algorithm, a forcing step is introduced in the alternating optimization to ensure that the necessary conditions for the optimal solution of the problem are met.

[0075] Furthermore, considering the requirements of future green communications, the solution to the original problem is incorporated into the constructed green communication model without considering the infeasibility of the initial point, thus minimizing energy consumption within the specified delay effect. This invention can be applied not only to downlink communication scenarios of high-speed rail, but also to communication scenarios similar to D2D. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are merely embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0077] Figure 1 Flowchart of the data demand-driven resource joint optimization method for high-speed rail train-to-ground communication provided by the present invention;

[0078] Figure 2 A schematic diagram of a system model for millimeter-wave vehicle underground downlink multi-link communication under the coverage of multiple base stations provided by the present invention;

[0079] Figure 3 A schematic diagram of the spatial time division multiple access protocol provided by the present invention;

[0080] Figure 4 This is a flow chart of the joint optimization algorithm for solving problem P3 provided by the present invention. DETAILED DESCRIPTION

[0081] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0082] like Figure 1 As shown, an embodiment of the present invention discloses a resource joint optimization method based on data demand driven in high-speed rail train-to-ground communication, comprising the following steps:

[0083] S1. Build a high-speed rail train-to-ground communication model based on the data needs of the equipment and users within the train. In this model, each train has a mobile relay on top, and the base station is equipped with multiple millimeter-wave directional antennas to exchange data with each mobile relay.

[0084] S2. Build a millimeter wave transmission channel model, analyze the impact of link mutual interference on transmission performance, and establish an overall optimization problem with the goal of minimizing the transmission time slot;

[0085] S3. Perform equivalent transformation on the overall optimization problem to reduce the dimension of the problem;

[0086] S4. Decompose the problem after equivalent transformation into a time planning subproblem and a power control optimization subproblem. Alternately optimize and solve the two subproblems. In the alternating optimization process, introduce a forcing step to ensure that the necessary conditions for the optimal solution of the problem are met after the alternating optimization is completed, and obtain the time slot planning strategy and power allocation strategy.

[0087] S5. Expand the consideration of the green communication problem, substitute the solution obtained in S4 as the initial value into the green communication problem, perform iterative solution, and obtain the resource allocation strategy that meets the green communication requirements.

[0088] The above steps are further explained below.

[0089] S1. Based on the high-speed train-to-ground downlink communication scenario, a high-speed train-to-ground communication model with millimeter-wave directional antennas is constructed according to the data requirements of the equipment and users in the carriage to achieve the minimum latency requirement under the data requirements; the high-speed train-to-ground communication model is as follows: Figure 2 As shown, it includes a moving high-speed train with M carriages. Each carriage has a mobile relay MR on top. The set of mobile relays is represented as Each mobile relay is represented as MR 1, MR 2, ..., MR M. The base station is deployed with multiple millimeter wave directional antennas for transmitting data and receiving information with the MRs, and uses a superframe data transmission protocol to improve space-time utilization efficiency.

[0090] Considering the communication model in the scenario of millimeter wave base station coverage area, each car is equipped with an MR on the top to facilitate the connection of devices to the base station through it. The devices in any car access the mobile relay on the top of the car through the access points (AP) distributed inside the car, and inform the corresponding mobile relay of the required data requirements. Each mobile relay communicates with the base station on behalf of the devices in its own car and obtains data resources from the base station through the superframe transmission protocol.

[0091] Specifically, when the present invention adopts the superframe transmission protocol for resource allocation, based on the spatial time division multiple access protocol (STDMA), the entire superframe includes three stages: beacon stage, request stage and transmission stage, such as Figure 3 As shown;

[0092] During the beacon phase, each mobile relay synchronizes with the base station to receive control information from the base station;

[0093] In the request phase, each mobile relay sends a request for the required data to the base station;

[0094] In the transmission phase, the set of time slots is represented as A superframe consists of hundreds or thousands of time slots. In each time slot, multiple mobile relays can simultaneously access the base station for spatial multiplexing, and each mobile relay can select multiple time slots for data transmission;

[0095] During the request phase, the base station collects data requests from each mobile relay and allocates time slots and power resources based on meeting the data needs of each mobile relay. In the next beacon phase, the base station notifies the corresponding mobile relay of the control message and resource allocation results. Therefore, the data request from the previous superframe is transmitted in the subsequent superframe.

[0096] The number of time slots in the transmission phase cannot exceed S max At the same time, it cannot be less than S min On the one hand, too many time slots will increase the transmission delay; on the other hand, too few time slots may not provide the base station with enough time to calculate the subsequent resource allocation strategy.

[0097] S2. Construct a millimeter wave transmission channel model, analyze the impact of mutual interference of links on transmission performance, and establish an overall optimization problem with the goal of minimizing the transmission time slot.

[0098] The calculation formula of the antenna's transmit and receive gain is as follows:

[0099]

[0100] Among them, G0 represents the maximum antenna gain, G sl represents the sidelobe gain, θ -3dB represents the half-power beamwidth, θ ml =2.6·θ -3db It represents the main lobe width.

[0101] The link transmitted by the base station to the jth mobile relay is link j, and the channel gain G caused by the jth link to the mth mobile relay is mj , the gain is determined by the transmit antenna gain, the receive antenna gain and the path loss distance, which can be expressed as:

[0102]

[0103] Where k represents the constant fading factor, which is related to the wavelength of the millimeter wave; G t (m,j) represents the transmit antenna gain of link j to the mth mobile relay, G r (m,j) represents the receiving antenna gain of the mth mobile relay for link j, represents the distance between the base station and the mth mobile relay, and γ represents the exponential decay factor;

[0104] According to Shannon's formula, the rate of the mth mobile relay in the sth time slot is Expressed as:

[0105]

[0106] in, represents the signal to interference and noise ratio of the mth mobile relay in the sth time slot; G mm represents the channel gain caused by the m-th link to the m-th mobile relay; represents the transmission power of the base station on link m to the mth mobile relay in the sth time slot; represents the transmission power of the base station's link j to the jth mobile relay in the sth time slot; It represents the multiple of the difference from the Shannon formula due to the actual situation, W represents the channel bandwidth, and N0 represents the power spectral density of Gaussian white noise.

[0107] Finally, based on the above formula, an optimization problem with the minimum transmission delay as the goal is formed, that is, minimizing the number of transmission time slots within a superframe to improve the real-time performance of transmission.

[0108] The final overall optimization problem P1 is expressed as:

[0109]

[0110] Among them, L slot Indicates the time length of each time slot; S represents the number of time slots used in the superframe, which is a natural number;

[0111] represents the data demand of the mth mobile relay; the constraint C1 in problem P1 means that the data transmitted to each mobile relay is greater than To meet the data requirements of the carriage represented by the mobile relay; M represents the total number of carriages;

[0112] Constraint C2 in problem P1 represents the transmission power of link m corresponding to the mth mobile relay in the sth time slot. Less than the maximum transmit power P max ; S represents the time slot set of the entire superframe; when When it is equal to 0, that is, no transmit power, it means that the link is not planned in this time slot;

[0113] Constraint C3 in problem P1 guarantees that the number of time slots does not exceed the maximum number of time slots S. max , in C3 It is expressed as a natural number, indicating that the number of time slots S is a natural number.

[0114] S3. Perform a two-step equivalent transformation on the overall optimization problem P1 to reduce the dimension of the problem so that it is easier to solve and obtain better calculation results.

[0115] Based on equivalent transformation, problem P1 is converted into the following form P2. Solving problem P1 is equivalent to solving problem P2. This transmission method does not cause much loss of accuracy, but can achieve dimensionality reduction of the problem and improve the problem solving speed. Specifically, it includes:

[0116] S3.1. Convert the overall optimization problem P1 into problem P2. Problem P2 can be expressed as:

[0117]

[0118] Where N means that the time is divided into N time periods, and the value of N is equal to M; t n Indicates the nth time period; represents the data transmission rate of the mth mobile relay corresponding to time period n; Represents the nth time period t n When , the power of the mth mobile relay;

[0119] Constraint C1 in problem P2 guarantees that the sum of data obtained by the mth mobile relay in all time periods is greater than its target data demand; where the time vector is represented by t n Composition, expressed as t=[t1,t n ,…,t N ] T , when transformed into the original problem At each time period t n The number of time slots that need to be converted into corresponding duration S n To meet the transmission requirements, the calculation formula is symbol It is rounded up, and the mth mobile relay has S n The transmission power of a time slot is

[0120] Constraint C2 in problem P2 represents the power of the mth mobile relay in time period n. Less than the maximum transmit power P max ;

[0121] Constraint C3 in problem P2 means that for each time period t n The length of is greater than or equal to 0;

[0122] S3.2, convert problem P2 into problem P3, which provides support for the subsequent alternating optimization solution. Equal to solving the problem Further equivalent to solving the problem Problem P3 is expressed as:

[0123]

[0124] in, Constraints and Problems The constraints of are consistent, and only the objective function is modified. The necessary condition for the optimal solution of problem P3 is that the equality of constraint C1 holds.

[0125] S4, based on the joint optimization algorithm, Solve the problem The problem is split into time planning sub-problem and power control optimization sub-problem for alternating optimization solution. Specifically, based on the traditional alternating optimization solution algorithm, the forcing step is introduced to make the problem The constraint C1 is equal to the problem, which satisfies the necessary conditions for the optimal solution. The specific processing flow is as follows Figure 4The joint optimization algorithm includes the following steps and iterates until the problem converges and cannot be optimized further and the necessary conditions are met. Specifically:

[0126] (1) The time planning sub-problem is decomposed from problem P3 into the following representation:

[0127]

[0128] The problem is modeled as a linear programming problem, so mature algorithms such as the interior point method can be used to solve it.

[0129] (2) The power control subproblem is decomposed from problem P3 into the following representation. The power control optimization subproblem is solved using successive convex approximation. In each iteration, the problem is transformed into a convex function and solved iteratively to obtain the final result. In each iteration, the power control optimization subproblem is expressed as:

[0130]

[0131] in, It represents the data transmission rate of the mth mobile relay in the nth time period, and its expression is:

[0132]

[0133] in, The value of For all The vector composed of The value of is used to obtain the power P of the mobile relay, that is, the power allocation strategy is obtained; and Both are expressed as the conversion coefficient of the mth mobile relay in the nth time period; express

[0134] In each iteration, and The values ​​are as follows:

[0135]

[0136] represents the signal-to-interference-and-noise ratio of the mobile relay in the nth time slot;

[0137] Through continuous iterative solution, we finally get P, and The convergent solution of .

[0138] Through the above transformation, the original non-convex problem is converted into a convex problem, and the convex problem is solved iteratively in each iteration using successive convex approximation.

[0139] (3) Necessary conditions for the establishment of sub-problems:

[0140] For the time planning sub-problem, fix the power P and schedule the time t to minimize the objective;

[0141] For the power control optimization sub-problem, with fixed time t, the objective function is maximized by optimizing the power P of different mobile relays in different time periods;

[0142] Since the problem cannot be guaranteed by relying solely on the alternating optimization of the time planning subproblem and the power control The necessary conditions for the optimal solution are met, so a forcing step is added to solve this type of problem. The optimization process using the forcing step includes:

[0143] Adding a forcing step: When the time planning subproblem and the power control optimization subproblem cannot be optimized further, determine whether the necessary conditions of the problem are met. If not, update P through the forcing step. Then, bring the updated results into the time planning subproblem and the power control optimization subproblem for further iterative loop solving until the two subproblems converge and cannot be optimized further and the necessary conditions are met.

[0144] Specifically, the process of integrating the forcing step into the alternating optimization for joint optimization includes:

[0145] use Denotes the data transmission rate matrix before the forced equalization step, which is represented by composition; adoption Denotes the data transmission rate matrix after the forced equalization step, which is given by composition;

[0146] Before conversion, define a reference data transmission rate matrix For reference conversion, each element is calculated as follows:

[0147]

[0148] Construct the necessary conditions for sub-problem P6:

[0149]

[0150] Using feasibility check and n Take the bisection method to find the solution of problem P6, so as to obtain the n The multiplication factor ξ n, we can further get each updated time period t n Corresponding power Finally, the updated power P of the mobile relay is obtained through the algorithm.

[0151] The updated results are brought into the time planning subproblem and the power control optimization subproblem for further iterative loop solution to make the necessary conditions meet.

[0152] S5. The number of time slots of the solution to the problem may be less than S max , in order to achieve the purpose of low delay, but in some cases where real-time performance is not pursued, considering green communication, the target is changed to the number of time slots S max Within the constraints of the transmission, the energy consumed to complete the task is minimized.

[0153] The solution of problem P3 is used as the initial value to substitute into problem P7 as the initial value to solve, thus avoiding the problem of infeasible initial point of the problem. The expression of green communication problem P7 is:

[0154]

[0155] Among them, the objective function in the green communication problem is to minimize the energy required to transmit all data; the constraint C1 in problem P7 means that the number of time slots required for transmission cannot exceed the specified number, and the (S max -N)*L slot Guarantee Issues Convert to the original problem When the solution is obtained, the number of time slots used is still less than S max and satisfy the constraints;

[0156] The problem Substitute the obtained values ​​P and t into the problem The problem Continue to split into problem-based The power control optimization subproblem and time planning subproblem are solved iteratively. The solution method and problem of power control optimization sub-problem and time planning sub-problem The solution method of the power control optimization sub-problem and the time planning sub-problem is the same, and the specified time slot S is finally obtained. max Resource allocation strategy within quantity.

[0157] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. Reference can be made to the common and similar parts between the various embodiments. For the devices disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple, and the relevant parts can be referred to the method description.

[0158] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not limited to the embodiments shown herein but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A data demand driven resource joint optimization method for high-speed rail train-to-ground communication, characterized in that: The following steps are involved: S1. Build a high-speed rail vehicle-to-ground communication model based on the data needs of the equipment and users in the carriage. In the high-speed rail vehicle-to-ground communication model, each carriage has a mobile relay on top, and the base station is deployed with multiple millimeter-wave directional antennas to exchange data with each mobile relay. S2. Build a millimeter wave transmission channel model, analyze the impact of link mutual interference on transmission performance, and establish an overall optimization problem with the goal of minimizing the transmission time slot; S3, performing equivalent transformation on the overall optimization problem to reduce the dimension of the problem; S4. Decompose the problem after equivalent transformation into a time planning subproblem and a power control optimization subproblem. Alternately optimize and solve the two subproblems. Introduce a forcing step in the alternating optimization process to ensure that the necessary conditions for the optimal solution of the problem are met after the alternating optimization is completed, and obtain the time slot planning strategy and power allocation strategy. S5. Establish a green communication problem, substitute the solution obtained in S4 as an initial value into the green communication problem, perform iterative solution, and obtain a resource allocation strategy that meets the green communication requirements.

2. The resource joint optimization method based on data demand driven in high-speed railway vehicle-ground communication according to claim 1 is characterized in that: In S1, in the high-speed rail train-to-ground communication model, the equipment in any car is connected to the mobile relay on the top of the car through the access points distributed inside the car. Each mobile relay communicates with the base station on behalf of the equipment in its own car and obtains data resources from the base station through the superframe transmission protocol.

3. The resource joint optimization method based on data demand driven in high-speed railway vehicle-ground communication according to claim 2 is characterized in that: When the superframe transmission protocol is used for resource allocation, the entire superframe includes three phases: a beacon phase, a request phase, and a transmission phase; During the beacon phase, each mobile relay synchronizes with the base station to receive control information from the base station; In the request phase, each mobile relay sends a request for the required data to the base station; During the transmission phase, a superframe consists of hundreds or thousands of time slots. Within each time slot, multiple mobile relays can simultaneously access the base station for spatial multiplexing, and each mobile relay can select multiple time slots for data transmission. During the request phase, the base station collects data requests from each mobile relay and allocates time slots and power resources based on meeting the data needs of each mobile relay. In the next beacon phase, the base station notifies the corresponding mobile relay of the control message and resource allocation results, so that the data request from the previous superframe is transmitted in the subsequent superframe.

4. The resource joint optimization method based on data demand driven in high-speed rail vehicle-ground communication according to claim 1 is characterized in that: In S2, the millimeter wave transmission channel model is expressed as: The link transmitted by the base station to the jth mobile relay is link j, and the channel gain G caused by the jth link to the mth mobile relay is mj Expressed as: Where k represents the constant fading factor, which is related to the wavelength of the millimeter wave; G t (m,j) represents the transmit antenna gain of link j to the mth mobile relay, G r (m,j) represents the receiving antenna gain of the mth mobile relay for link j, represents the distance between the base station and the mth mobile relay, and γ represents the exponential decay factor; According to Shannon's formula, the rate of the mth mobile relay in the sth time slot is Expressed as: in, represents the signal to interference and noise ratio of the mth mobile relay in the sth time slot; G mm represents the channel gain caused by the m-th link to the m-th mobile relay; P represents the transmission power of link m of the base station to the mth mobile relay in the sth time slot; j s represents the transmission power of the base station's link j to the jth mobile relay in the sth time slot; It represents the multiple of the difference from the Shannon formula due to the actual situation, W represents the channel bandwidth, and N0 represents the power spectral density of Gaussian white noise.

5. The resource joint optimization method based on data demand driven in high-speed railway vehicle-ground communication according to claim 4 is characterized in that: The overall optimization problem P1 established in S2 is expressed as: Among them, L slot Indicates the time length of each time slot; S represents the number of time slots used in the superframe, which is a natural number; represents the data demand of the mth mobile relay; the constraint C1 in problem P1 means that the data transmitted to each mobile relay is greater than To meet the data requirements of the carriage represented by the mobile relay; M represents the total number of carriages; Constraint C2 in problem P1 represents the transmission power of link m corresponding to the mth mobile relay in the sth time slot. Less than the maximum transmit power P max ; S represents the time slot set of the entire superframe; when When it is equal to 0, that is, no transmit power, it means that the link is not planned in this time slot; Constraint C3 in problem P1 guarantees that the number of time slots does not exceed the maximum number of time slots S. max , Expressed as a natural number.

6. The resource joint optimization method based on data demand driven in high-speed railway vehicle-ground communication according to claim 1 is characterized in that S3 include: S3.

1. Convert the overall optimization problem P1 into problem P2. Problem P2 is expressed as: Where N means that the time is divided into N time periods, and the value of N is equal to M; t n Indicates the nth time period; represents the data transmission rate of the mth mobile relay corresponding to time period n; Represents the nth time period t n When , the power of the mth mobile relay; Constraint C1 in problem P2 guarantees that the sum of data obtained by the mth mobile relay in all time periods is greater than its target data demand; where the time vector is represented by t n Composition, expressed as t=[t1,t n ,…,t N ] T , when transformed into the original problem P1, each time period t n The number of time slots that need to be converted into corresponding duration S n To meet the transmission requirements, the calculation formula is symbol It is rounded up, and the mth mobile relay has S n The transmission power of a time slot is Constraint C2 in problem P2 represents the power of the mth mobile relay in time period n. Less than the maximum transmit power P max ; Constraint C3 in problem P2 means that for each time period t n The length of is greater than or equal to 0; S3.

2. Convert problem P2 into problem P3. Problem P3 is expressed as: The necessary condition for the optimal solution of problem P3 is that the equality of constraint C1 holds.

7. The resource joint optimization method based on data demand driven in high-speed railway vehicle-ground communication according to claim 6 is characterized in that: In S4, the time planning sub-problem is split from problem P3 and is expressed as: Use the interior point method to solve problem P4.

8. The data demand driven resource joint optimization method for high-speed rail train-to-ground communication according to claim 6 is characterized in that: In S4, the power control optimization subproblem is solved by successive convex approximation. In each iteration, the problem is transformed into a convex function and solved iteratively to obtain the final result of the problem. In each iteration, the power control optimization subproblem is decomposed from problem P3 and expressed as: in, It represents the data transmission rate of the mth mobile relay in the nth time period, and its expression is: in, The value of For all The vector composed of The value of is used to obtain the power P of the mobile relay, that is, the power allocation strategy is obtained; and Both are expressed as the conversion coefficient of the mth mobile relay in the nth time period; express In each iteration, and The values ​​are as follows: represents the signal to interference and noise ratio of the mth mobile relay in the nth time slot; Through continuous iterative solution, we finally get P, and The convergent solution of .

9. The data demand driven resource joint optimization method for high-speed railway vehicle-ground communication according to claim 6 is characterized in that: In S4, the alternating optimization process for solving the two sub-problems includes: For the time planning sub-problem, fix the power P and schedule the time t to minimize the objective; For the power control optimization sub-problem, with fixed time t, the objective function is maximized by optimizing the power P of different mobile relays in different time periods; Add a forcing step. When the time planning subproblem and the power control optimization subproblem cannot be optimized any further, determine whether the necessary conditions of the problem are met. If not, use the forcing step to update P. The updated result is brought into the time planning subproblem and the power control optimization subproblem for further iterative loop solution until the two subproblems converge and the necessary conditions are met. The process of integrating the forcing step into the alternating optimization for joint optimization includes: use Denotes the data transmission rate matrix before the forced equalization step, which is represented by composition; adoption Denotes the data transmission rate matrix after the forced equalization step, which is given by composition; Before conversion, define a reference data transmission rate matrix For reference conversion, each element is calculated as follows: Construct the necessary conditions for sub-problem P6: Using feasibility check and n Take the bisection method to find the solution of problem P6 and get the time interval t n The multiplication factor ξ n , and get each updated time period t n Corresponding power Finally, the updated power P of the mobile relay is obtained through the algorithm; The updated results are brought into the time planning subproblem and the power control optimization subproblem for further iterative loop solution to make the necessary conditions meet.

10. The resource joint optimization method based on data demand driven in high-speed railway vehicle-ground communication according to claim 6 is characterized in that: In S5, the expression of the green communication problem P7 is: Among them, the objective function in the green communication problem is to minimize the energy required to transmit all data; the constraint C1 in problem P7 means that the number of time slots required for transmission cannot exceed the specified number, and the (S max -N)*L slot Guarantee Issues Convert to the original problem When the solution is obtained, the number of time slots used is still less than S max and satisfy the constraints; The problem Substitute the solved values ​​P and t into problem P7, and solve the problem Continue to split into problem-based The power control optimization subproblem and time planning subproblem are solved iteratively. The solution method and problem of power control optimization sub-problem and time planning sub-problem The solution method of the power control optimization sub-problem and the time planning sub-problem is the same, and the specified time slot S is finally obtained. max Resource allocation strategy within quantity.

Images