A communication cell switching method for 5G AeroMACS

Abstract

The present invention belongs to the field of wireless communication technology, and proposes a 5G AeroMACS communication cell switching method. The switching decision of the target communication cell is formulated according to the three factors of received signal strength, network load and remaining service time, and the deep reinforcement learning algorithm is used to realize the intelligent switching of the target communication cell, thereby improving the communication cell switching efficiency and communication quality of the airport surface equipment and meeting the real-time communication requirements. The method includes the following steps: (1) constructing a system model for airport surface communication cell switching to describe the communication channel state and spatial state of the airport surface equipment and the communication cell base station; (2) constructing an optimization problem of communication cell selection that maximizes the service quality based on the three factors of received signal strength, network load and remaining service time; (3) using a deep reinforcement learning algorithm to solve the optimization problem, thereby realizing the intelligent communication cell switching of the airport surface and ensuring the communication quality.

Description

Technical Field

[0001] The present invention relates to the field of wireless communication technologies, and in particular to a communication cell switching method for 5G AeroMACS. Background Art

[0002] As an aviation broadband mobile communication system designed for airport surface areas, the 5G Aeronautical Mobile Airport Communications System (AeroMACS) provides wireless network connections for airport surfaces and bridge areas, that is, communication services for aircraft, various ground vehicles and equipment, and is mainly used to safely and reliably transmit information related to airport operations. However, the airport surface environment is complex and dynamic, including aircraft, various ground vehicles, towers and other mobile units and fixed facilities, which move at high speed in a limited space, forming dense traffic and information flows. At the same time, under the influence of various factors such as weather changes, aircraft takeoffs and landings, flight scheduling and safety supervision, the 5G airport aviation mobile communication system must meet the real-time communication needs of the airport surface, provide stable and efficient communication services, and support the safe operation and efficient management of the airport surface.

[0003] The airport surface area is covered by multiple communication cells. When the communication quality of a certain communication cell does not meet the requirements, the equipment can switch to other communication cells to maintain the continuity and stability of communication. The switching process includes switching triggering, switching judgment and switching execution, ensuring that the equipment does not interrupt communication when moving between terminals, runways or parking spaces, thereby ensuring communication efficiency and flight safety. However, existing research usually focuses on single factors such as signal strength or network load, and fails to fully consider the complexity and dynamics of switching. In addition, existing methods often rely on preset rules and find it difficult to respond to environmental changes in a timely manner, resulting in insufficient real-time switching decisions. Therefore, in a complex electromagnetic environment, realizing intelligent communication switching of airport surface communication cells that considers multiple factors has become the key to meeting the communication needs of airport surface equipment. Summary of the invention

[0004] In order to meet the challenges of high reliability and high real-time requirements for airport surface communication cell switching in complex electromagnetic environments, this paper proposes a 5G AeroMACS communication cell switching method. An airport surface communication cell switching model based on a deep reinforcement learning algorithm is constructed, and a weighted service quality function that considers received signal strength, network load, and remaining service time is used to achieve efficient and intelligent target communication cell selection, thereby improving communication switching efficiency and communication quality.

[0005] The technical solution of the present invention is as follows:

[0006] A 5G AeroMACS communication cell switching method comprises the following steps:

[0007] Step S1: construct a system model for airport surface communication cell switching to describe the communication channel state and space state of airport surface equipment and communication cell base stations;

[0008] Step S2: constructing an optimization problem for communication cell selection that maximizes service quality based on three factors: received signal strength, network load, and remaining service time;

[0009] Step S3: A deep reinforcement learning algorithm is used to solve the optimization problem, thereby realizing intelligent communication cell switching at the airport and ensuring communication quality.

[0010] Furthermore, the step S1 specifically includes: assuming that the airport surface equipment moves along a predetermined path, at a time Enter the current communication cell when , the current communication cell base station location is , the airport equipment continues to move in the current communication cell, When the current communication cell base station and the next communication cell base station enter the overlapping area, the airport surface equipment position is , the current communication cell is C 1 , the next communication cell is C 2 ; and The distance between is expressed as:

[0011] (1)

[0012] Rice fading channel model h ( t ) is expressed as:

[0013] (2)

[0014] in, is the Rice coefficient, and They represent the line-of-sight transmission part and the non-line-of-sight transmission part, respectively, and are expressed as:

[0015] (3)

[0016] (4)

[0017] in, Indicates the reference distance The path loss when Indicates the distance between the airport surface equipment and the base station. and They represent the line-of-sight transmission coefficient and the non-line-of-sight transmission coefficient, respectively. represents a cyclically symmetric complex Gaussian variable.

[0018] Furthermore, the step S2 specifically includes:

[0019] Step S2-1: Establish a received signal strength model:

[0020] (5)

[0021] in, Indicates the received signal strength. Indicates the communication cell, is the transmitted signal power, is the transmit antenna gain, is the receiving antenna gain, ;

[0022] Normalized received signal strength It is expressed as:

[0023] (6)

[0024] Step S2-2: Establishing a network load model:

[0025] (7)

[0026] in, Indicates the network load, , , and Respectively represent the communication cells The number of airport access devices, average data transmission volume, average data transmission rate and bandwidth, It is a mapping function related to the actual situation of the airport scene and the specific configuration of the network;

[0027] Normalized network load It is expressed as:

[0028] (8)

[0029] Step S2-3: Establishing the remaining service time model:

[0030] (9)

[0031] (10)

[0032] in, Indicates the remaining service time of the current communication cell, Indicates the remaining service time of the next communication cell, When the receiving signal strength of the airport surface equipment in the current communication cell drops to the preset critical value, express to At any time between Indicates the time interval for the airport surface equipment to move from the position where the received signal strength of the current communication cell drops to the preset critical value to the position where the signal of the current communication cell is completely disconnected. When the receiving signal strength of the airport surface equipment in the next communication cell drops to the preset critical value, Indicates the time interval for the airport surface equipment to move from the position where the received signal strength of the next communication cell decreases to a preset critical value to the position where the signal of the next communication cell is completely disconnected;

[0033] Normalized remaining service time It is expressed as:

[0034] (11)

[0035] Step S2-4: Establishing a service quality model:

[0036] (12)

[0037] (13)

[0038] in, Indicates the quality of service. They represent the received signal strength weight, network load weight and remaining service time weight respectively. Indicates the target communication cell.

[0039] Furthermore, the TD3 algorithm is used in step S3 to obtain the target communication cell for communication switching, which specifically includes:

[0040] Step S3-1: Define the state space S is a set of received signal strength, network load and remaining service time of the current communication cell and the next communication cell, where the state s Expressed as ; Actions in action space A ɑ is the switching strategy for selecting a communication cell, including switching to the next communication cell and maintaining the current communication cell; the reward function Defined as: , Indicates status Service quality and status under The difference in service quality under

[0041] Step S3-2: Design the network structure of the TD3 algorithm, which consists of two target critic networks , a target Actor network , Two Current Critic Networks and a current Actor network It consists of 6 networks in total;

[0042] Step S3-3: training the TD3 algorithm;

[0043] Step S3-4: Apply the trained TD3 algorithm to the switching decision of the communication cells at the airport.

[0044] Furthermore, the airport surface equipment includes aircraft or ground vehicles.

[0045] Compared with the prior art, the present invention has the following beneficial effects:

[0046] 1. The present invention fully considers the dynamic changes of the location of airport surface equipment, constructs a motion model of airport surface equipment and a distribution model of communication cells, which are used to describe the relationship between the spatial state information of equipment and base stations and the signal strength and service time.

[0047] 2. The present invention fully considers the complex electromagnetic environment faced by airport field communications and constructs a switching strategy based on three key factors: the received signal strength of the communication cell, the network load capacity and the remaining service time.

[0048] 3. The present invention fully considers the high real-time requirements of airport surface communications, adopts deep reinforcement learning methods to achieve efficient decision-making on switching problems, and realizes efficient and accurate communication cell switching while maintaining low computational complexity, ensuring the continuity and real-time performance of communication. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. The features and advantages of the present invention can be more clearly understood by referring to the drawings. The drawings are schematic and should not be understood as limiting the present invention in any way. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0050] Figure 1 It is a flow chart of the communication cell switching method of 5G AeroMACS of the present invention.

[0051] Figure 2It is a schematic diagram of the system model for airport field communication cell switching constructed in the present invention.

[0052] Figure 3 The present invention is a flow chart of intelligent communication cell switching operation for airport scenes. DETAILED DESCRIPTION

[0053] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present invention and the features in the embodiments can be combined with each other without conflict.

[0054] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0055] Aiming at the complexity and dynamic application scenarios of airport field communication cell communication, the present invention provides a 5G AeroMACS communication cell switching method. By comprehensively considering the three key factors of the cell's received signal strength, network load capacity, and remaining service time, the TD3 algorithm in deep reinforcement learning is used as an example to achieve intelligent target communication cell selection, and adaptively select the switching target communication cell according to the constantly changing airport field environment, thereby improving the switching efficiency and communication quality of the airport field cell in a complex electromagnetic environment. Figure 1 As shown in the figure, firstly, a system model of airport surface communication cell switching is constructed, and the spatial state information of airport surface equipment and communication cells when switching communication cells is explained. Secondly, a mathematical model of three influencing factors, namely the received signal strength, network load capacity and remaining service time of the cell, is designed. Different weights are assigned to the three factors according to the actual needs of airport surface communication to form a service quality function. Finally, the service quality function is solved based on the TD3 algorithm in deep reinforcement learning, and the cell with the best service quality that meets the switching requirements is obtained for switching, realizing intelligent target communication cell selection, thereby improving the efficiency and communication quality of airport surface communication cell switching and meeting real-time requirements.

[0056] Example 1

[0057] Step S1: Construct a system model for airport field communication cell switching.

[0058] Airport surface communication is a complex and multi-layered system, involving air traffic control communication, communication between aircraft and ground service vehicles, airport operation and management measures, communication between airlines and airports, and other supporting service communications. A system model for airport surface communication cell switching is constructed to describe the spatial status information of airport surface equipment and base stations. Figure 2 As shown in the figure, in a typical airport scene, assume that a large passenger plane is moving slowly along the designated taxiway, preparing to park at the designated parking position. Enter the current communication cell when , the current communication cell base station location with which it maintains communication contact is , the aircraft glides along the predetermined path in the current communication cell, When the aircraft enters the overlapping coverage area of ​​the current communication cell base station and the next communication cell base station, the position of the aircraft is , the current communication cell is C 1 , the next communication cell is C 2 ; and The distance between is expressed as:

[0059] (1)

[0060] The aircraft can receive and analyze the broadcast signaling or control channel messages from the base station, obtain the network load of the communication cell in real time, and provide decision support for the communication cell switching. The airport environment is complex, and the signal received by the aircraft from the base station includes the direct part and the reflected part of the obstacle. Therefore, the Rice fading channel model used in the present invention h ( t ) is expressed as:

[0061] (2)

[0062] in, is the Rice coefficient, and They represent the line-of-sight transmission part and the non-line-of-sight transmission part, respectively, and are expressed as:

[0063] (3)

[0064] (4)

[0065] in, Indicates the reference distance The path loss when Indicates the distance between the aircraft and the base station. and They represent the line-of-sight transmission coefficient and the non-line-of-sight transmission coefficient, respectively. represents a cyclically symmetric complex Gaussian variable.

[0066] In the process of deciding whether to perform a communication cell handover, the received signal strength, network load, and remaining service time are comprehensively analyzed. These three parameters represent the strength of the received signal, the utilization of network resources, and the continuity of communication services. To ensure the safety and continuity of the aircraft communication system, when the aircraft is located in the overlapping area covered by the signals of two communication cells, the system will periodically perform handover decisions at fixed time intervals. The problem of communication cell selection can be abstracted as an optimization problem, the core goal of which is to select the option that best meets the airport surface communication service quality requirements from the two candidate communication cells.

[0067] Step S2: Construct an optimization problem for communication cell selection that maximizes service quality.

[0068] When considering the switching of communication cells at the airport, three factors need to be considered: received signal strength, network load, and remaining service time. Received signal strength directly reflects signal quality and affects communication stability; network load capacity indicates the available resources of the current and next communication cells, affecting the experience of airport equipment; remaining service time refers to the time interval from the current moment to the end of the communication cell service, affecting service continuity. Combining these factors, the intelligent agent can more comprehensively evaluate the potential impact of the switching decision, thereby making a more optimized choice in the complex environment of the airport.

[0069] The received signal strength is an indicator to measure the strength of wireless signals, which refers to the power level of wireless signals received at the receiving end. The received signal strength is usually affected by factors such as the signal propagation environment, the transmit signal strength and the antenna gain, and is expressed as:

[0070] (5)

[0071] in, Indicates the received signal strength. Indicates the communication cell, is the transmitted signal power, is the transmit antenna gain, is the receiving antenna gain, .

[0072] When the airport surface equipment is switched, the aircraft will receive signals from multiple communication cells during the entire taxiing process. Each switch occurs between two communication cells. In order to facilitate subsequent processing, the received signal strength is normalized to obtain the normalized received signal strength. It is expressed as:

[0073] (6)

[0074] The network load of the airport scene refers to the amount of data processed by the wireless communication network within the airport within a specific time and the usage of network resources. In the airport scene, the network load can be expressed as a function of the selected communication cell. It can vary according to factors such as the number of airport scene access devices in the communication cell, data transmission rate, communication cell coverage, and communication cell equipment processing capacity. This invention assumes that at time t , communication cell Number of airport access devices within , average data transfer volume , average data transfer rate and bandwidth Composing network load The function is expressed as:

[0075] (7)

[0076] in, It is a mapping function based on the actual situation of the airport scene and the specific configuration of the network.

[0077] Furthermore, the network load is normalized to obtain the normalized network load , expressed as:

[0078] (8)

[0079] In the airport surface communication environment, the remaining service time of the communication cell refers to the expected length of time that the aircraft can maintain effective communication in the current communication cell. It determines whether the aircraft needs to perform a handover operation in the overlapping area of ​​the communication cell coverage to avoid communication interruption due to signal weakening or network resource exhaustion. Access the current communication cell at the same time, glide along the predetermined path in the current communication cell and enter the coverage overlap area of ​​the current communication cell base station and the next communication cell base station, When the aircraft's received signal strength in the current communication cell drops to a preset critical value, and The time interval from the position where the received signal strength of the current communication cell drops to the preset critical value to the position where the signal of the current communication cell is completely disconnected is The time interval from the position where the received signal strength of the next communication cell decreases to the preset critical value to the position where the signal of the next communication cell is completely disconnected is When the aircraft enters the overlapping coverage area of ​​two base stations Time, the remaining service time of the current communication cell and the remaining service time of the next communication cell They are:

[0080] (9)

[0081] (10)

[0082] The remaining service time of the two communication cells is normalized and expressed as:

[0083] (11)

[0084] When considering the communication switching of the airport surface communication cell, different weights are assigned to the three factors according to actual needs. They are the received signal strength weight, network load weight and remaining service time weight. Taking these three factors into consideration, the service quality expressed in formula (12) is formed. , and calculate the most suitable target communication cell for switching according to formula (13) , expressed as:

[0085] (12)

[0086] (13)

[0087] Step S3: Use the TD3 algorithm to obtain the target communication cell for communication switching.

[0088] The standard reinforcement learning algorithm is an intelligent agent that interacts with the learning environment in a discrete time interval. The present invention proposes to apply the reinforcement learning algorithm to solve the intelligent communication cell switching in the airport scene, and elaborates on it in detail by taking the TD3 algorithm as an example.

[0089] Step S3-1: When switching the airport scene intelligent communication cell, the state space S It is defined as a set of received signal strength, network load and remaining service time of the current communication cell and the next communication cell, where the state s Expressed as The actions in the action space A ɑ is the handover strategy for selecting a communication cell, including switching to the next communication cell and maintaining the current communication cell. It can be defined as: , Indicates status Service quality and status under The difference in service quality under , reflects the effect of the switching decision. The goal of the agent is to find the optimal switching strategy based on the reward function so that the long-term cumulative reward is maximized, that is, switching to the communication cell with the best service quality.

[0090] Step S3-2: When designing the network structure, the TD3 algorithm is improved from the Deep Deterministic Policy Gradient (DDPG) algorithm, which consists of two target critic networks. , a target Actor network , Two Current Critic Networks and a current Actor network It consists of 6 networks in total.

[0091] Step S3-3: training the TD3 algorithm. The training process of the TD3 algorithm is as follows:

[0092] (1) At the initial moment, randomly initialize the current Actor network weight , Target Actor Network Weight , Current Critic network weight and target critic network weight , clear the experience replay buffer and initialize the state space S .

[0093] (2) The state composed of three factors: the received signal strength, network load and remaining service time in the current communication cell s Next, use the Actor network to select actions ɑ , that is, select the communication cell switching strategy to be adopted in the current state and execute the action ɑ And observe the reward composed of the difference between the service quality under the target network state and the service quality under the current state r , reflecting the effect of the switching decision, and recording the characteristics of the communication cell where the aircraft continues to communicate, forming a new state , the experience < >Stored in the experience replay buffer, through multiple repetitions, multiple samples in the experience replay buffer are obtained.

[0094] (3) Randomly extract a batch of experience from the experience replay buffer < >, the state s and actions ɑ Input two current Critic networks and output their predicted returns Q Value. > Add random noise to the output of the target Actor network to get a noisy action ,Will and Input two targets and the Critic network outputs its predicted return value.

[0095] (4) Update the current Critic network: use the target Actor network and the target Critic network to calculate the target value To prevent the network from overestimating, double Q The smaller value in the target network is calculated. The target value represents the maximum expected return that the system can obtain after the communication cell switching. The loss L is calculated and the Critic network is updated.

[0096] (5) Use gradient ascent to update the current Actor network to maximize the expected reward, that is, tend to choose actions with higher rewards.

[0097] (6) Regularly copy the parameters of the current Critic network and the current Actor network to the target Critic network and the target Actor network to achieve updates.

[0098] (7) Finally, we can get the maximum expected return Q The switching target communication cell selected by the value completes the switching decision.

[0099] Step S3-4: Apply the trained TD3 algorithm to the handover decision of the communication cells in the airport scene, for example Figure 3 As shown:

[0100] When the airport equipment is detected to enter the overlapping coverage area of ​​two communication cells, the handover decision is triggered. The service quality of the next communication cell and the current communication cell is evaluated by solving the service quality through the trained TD3 algorithm. If the service quality of the next communication cell is better than that of the current communication cell, the airport equipment will take the next communication cell as the target communication cell and switch; if the service quality of the current communication cell is better than that of the next communication cell, it will continue to communicate with the current communication cell, and periodically make handover decisions while maintaining communication with the current communication cell until the airport equipment leaves the overlapping coverage area of ​​the two communication cells.

[0101] In the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0102] In the present invention, unless otherwise clearly specified and limited, a first feature being "above" or "below" a second feature may include that the first and second features are in direct contact, or may include that the first and second features are not in direct contact but are in contact through another feature between them. Moreover, a first feature being "above", "above" and "above" a second feature includes that the first feature is directly above and obliquely above the second feature, or simply indicates that the first feature is higher in level than the second feature. A first feature being "below", "below" and "below" a second feature includes that the first feature is directly below and obliquely below the second feature, or simply indicates that the first feature is lower in level than the second feature.

[0103] In the present invention, the terms "first", "second", "third", "fourth" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance. The term "plurality" refers to two or more, unless otherwise clearly defined.

[0104] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A 5G AeroMACS communication cell switching method, characterized in that: The following steps are involved: Step S1: construct a system model for airport surface communication cell switching to describe the communication channel state and space state of airport surface equipment and communication cell base stations; Step S2: constructing an optimization problem for communication cell selection that maximizes service quality based on three factors: received signal strength, network load, and remaining service time; Step S3: using a deep reinforcement learning algorithm to solve the optimization problem, thereby realizing intelligent communication cell switching at the airport and ensuring communication quality; The step S1 specifically includes: assuming that the airport surface equipment moves along a predetermined path at a certain time. Enter the current communication cell when , the current communication cell base station location is , the airport equipment continues to move in the current communication cell, When the current communication cell base station and the next communication cell base station enter the overlapping area, the airport surface equipment position is , the current communication cell is C 1, the next communication cell is C 2; and The distance between is expressed as: (1) Rice fading channel model h ( t ) is expressed as: (2) in, is the Rice coefficient, and They represent the line-of-sight transmission part and the non-line-of-sight transmission part, respectively, and are expressed as: (3) (4) in, Indicates the reference distance The path loss when Indicates the distance between the airport surface equipment and the base station. and They represent the line-of-sight transmission coefficient and the non-line-of-sight transmission coefficient, respectively. represents a cyclically symmetric complex Gaussian variable; The step S2 specifically includes: Step S2-1: Establish a received signal strength model: (5) in, Indicates the received signal strength. Indicates the communication cell, is the transmitted signal power, is the transmit antenna gain, is the receiving antenna gain, ; Normalized received signal strength It is expressed as: (6) Step S2-2: Establishing a network load model: (7) in, Indicates the network load, , , and Respectively represent the communication cells The number of airport access devices, average data transmission volume, average data transmission rate and bandwidth, It is a mapping function related to the actual situation of the airport scene and the specific configuration of the network; Normalized network load It is expressed as: (8) Step S2-3: Establishing the remaining service time model: (9) (10) in, Indicates the remaining service time of the current communication cell, Indicates the remaining service time of the next communication cell. When the receiving signal strength of the airport surface equipment in the current communication cell drops to the preset critical value, express to At any time between Indicates the time interval for the airport surface equipment to move from the position where the received signal strength of the current communication cell drops to the preset critical value to the position where the signal of the current communication cell is completely disconnected. When the receiving signal strength of the airport surface equipment in the next communication cell drops to the preset critical value, Indicates the time interval for the airport surface equipment to move from the position where the received signal strength of the next communication cell decreases to a preset critical value to the position where the signal of the next communication cell is completely disconnected; Normalized remaining service time It is expressed as: (11) Step S2-4: Establishing a service quality model: (12) (13) in, Indicates the quality of service. They represent the received signal strength weight, network load weight and remaining service time weight respectively. Indicates the target communication cell; The step S3 uses the TD3 algorithm to solve and obtain the target communication cell for communication switching, specifically including: Step S3-1: Define the state space S is a set of received signal strength, network load and remaining service time of the current communication cell and the next communication cell, where the state s Expressed as ; Actions in action space A ɑ is the switching strategy for selecting a communication cell, including switching to the next communication cell and maintaining the current communication cell; the reward function Defined as: , Indicates status Service quality and status under The difference in service quality under Step S3-2: Design the network structure of the TD3 algorithm, which consists of two target critic networks , a target Actor network , Two Current Critic Networks and a current Actor network It consists of 6 networks in total; among them, Indicates the current Actor network weight, Represents the target Actor network weight, Represents the current Critic network weight and Represents the target Critic network weight; Step S3-3: training the TD3 algorithm; Step S3-4: Apply the trained TD3 algorithm to the switching decision of the communication cells at the airport.

2. The 5G AeroMACS communication cell switching method according to claim 1, characterized in that: The airport surface equipment includes aircraft or ground vehicles.

Images