Communication perception resource collaborative optimization method based on relaxation quantization

Abstract

The invention discloses a communication sensing resource collaborative optimization method based on relaxation quantization. The method comprises the following steps: a communication sensing integrated base station determines network parameters of resource scheduling; designing a working mode and evaluating resource overhead by the sensing integrated base station; the base station completes common sensing resource collaborative optimization problem modeling; the base station optimizes the communication sensing resources to obtain an intermediate solution; according to the method, under the condition that the communication service is guaranteed preferentially, the sensing performance of the low-altitude unmanned aerial vehicle is maximized by cooperatively scheduling the time resources of the multiple base stations, and the time resource utilization rate and the inter-station cooperation capability of the base stations are improved.

Description

Technical Field

[0001] The present invention relates to a communication-sensing integrated base station resource scheduling method for urban low-altitude sensing, and particularly to a communication-sensing resource collaborative optimization method based on relaxation quantization. Background Art

[0002] The global low-altitude economy is in the stage of application popularization and has huge potential. The global low-altitude economy market size exceeded 30 billion US dollars in 2023. According to the prediction of Roland Berger Research, by 2050, the global low-altitude economy market size will exceed 60 trillion yuan. The low-altitude economy industry is evolving from being driven by leading players to a comprehensive industrial chain structure, as Figure 1 shown. In the new tracks represented by drones, eVTOL, and UAM, China is advancing side by side with advanced countries, and the market continues to expand. According to the data of the Civil Aviation Administration, as of the end of 2023, there were approximately 2,000 drone design and manufacturing units and nearly 20,000 operating enterprises in China. Data from the CCID Research Institute shows that the scale of China's low-altitude economy reached 505.95 billion yuan in 2023, with a year-on-year growth rate of 33.8% (among which the year-on-year increase of civilian drones was 32%, and the year-on-year increase of eVTOL was 77.3%). It is expected that the scale will exceed the trillion-yuan mark in 2026.

[0003] With the booming development of the low-altitude economy, it is necessary to sense and control a large number of drones in urban low altitudes. Existing 5G base stations naturally have the ability to form a network. Only by simply transforming them can they form a communication-sensing integrated base station in future 6G to achieve full-domain coverage sensing of low-altitude drones. The existing communication-sensing integrated base station solutions use a shared antenna front-end and base station platform for communication and sensing, and achieve communication-sensing integration by allocating 10% of the communication resources to the sensing function. However, the fixed time resource scheduling scheme is difficult to adapt to the increasingly complex urban environment and cannot efficiently meet the user communication and drone sensing needs at the same time. Therefore, it is necessary to carry out research on the adaptive resource scheduling scheme of the communication-sensing integrated base station, and through the network collaboration between stations, while giving priority to ensuring communication services, dynamically optimize the time resources to maximize the overall network's sensing performance for drones.

[0004] Currently, there is no research on the communication-sensing resource collaborative scheduling scheme of the communication-sensing integrated base station under the condition of giving priority to ensuring communication services and taking into account the low-altitude drone sensing scenario. Summary of the Invention

[0005] Aiming at the problems existing in the prior art and for future urban low-altitude sensing, the present invention provides a communication-sensing resource collaborative optimization method based on relaxation quantization. Under the condition of giving priority to ensuring communication services, by collaboratively scheduling the time resources of multiple base stations, the sensing performance for low-altitude drones is maximized, and the utilization rate of the time resources of the base stations and the inter-station collaboration ability are improved.

[0006] The object of the present invention is achieved by the following technical solutions.

[0007] A communication and sensing resource collaborative optimization method based on relaxed quantization, comprising the following steps:

[0008] (10) The integrated communication and sensing base station determines the network parameters for resource scheduling: The integrated communication and sensing base station receives the communication requests of users and the sensing requests of unmanned aerial vehicles (UAVs), and evaluates the task priorities and the location service relationships between the tasks and the base station.

[0009] (20) The integrated communication and sensing base station designs the working modes and evaluates the resource overheads: Based on the time division method, the integrated communication and sensing base station designs different transmitted signals and working modes to perform communication and sensing tasks respectively, and evaluates the time resource overheads for performing different tasks based on the differences in the transmitted signals and working modes.

[0010] (30) The base station completes the modeling of the communication and sensing resource collaborative optimization problem: By representing the time resources and designing the constraints of the working modes, the base station models the communication and sensing resource collaborative optimization problem, and maximizes the completion degree of the sensing tasks under the constraint of first satisfying the communication tasks.

[0011] (40) The base station obtains an intermediate solution for the optimization of communication and sensing resources: First, the discrete variables are relaxed into continuous variables to obtain an initial solution for the communication and sensing task strategy, and then the initial solution is quantized into discrete values based on the nearest neighbor to obtain an intermediate solution.

[0012] (50) The base station performs collaborative optimization and solution for communication and sensing resources: Based on the greedy correction strategy, the quantized intermediate solution is corrected to satisfy the time resource constraints, and the final communication and sensing task strategy is generated.

[0013] The step (10) of the integrated communication and sensing base station determining the network parameters for resource scheduling includes:

[0014] (11) There are integrated communication and sensing base stations , UAV sensing and user terminal communication tasks , where the number of UAV sensing tasks is the same as the number of UAVs, and the number of communication tasks is the same as the number of user terminals.

[0015] (12) The base station evaluates the priorities of communication and sensing tasks. Communication tasks need to be guaranteed first, and their priority is infinity; the priority of the sensing tasks is , which depends on the speed, position, and size information of the UAVs.

[0016] (13) Each base station has a limited sensing range , that is, it can only perform sensing tasks within the sensing range; define the set of sensing tasks that a base station can execute, where is the physical distance between the base station and the sensing task; similarly, define the set of base stations that can execute the sensing task ;

[0017] (14) Each base station has a limited communication range , that is, it can only perform communication tasks within the communication range; define the set of communication tasks that a base station can execute, and the set of base stations that can execute the sensing task .

[0018] The above-mentioned (20) integrated communication and sensing base station design working mode and evaluation of resource overhead steps include:

[0019] (21) For the UAV sensing task, the transmitted signal of the base station is a frequency-modulated continuous wave to sense the UAV; to meet the sensing requirements of the base station for the UAV, the sensing echo signal-to-noise ratio must exceed a given signal-to-noise ratio threshold ; therefore, the time resource overhead for the base station to sense the UAV is:

[0020] ,

[0021] where is the Boltzmann constant, is the antenna temperature, is the noise figure, is the radar cross-sectional area of the UAV, is the sensing transmit power, is the transmit antenna gain, is the receive antenna gain, is the sensing signal wavelength;

[0022] (22) For the UAV sensing task, the working mode of the base station is single-base station independent sensing, that is, a single UAV is sensed by a single base station. Define the strategy matrix of the base station for the sensing task, where the element represents whether the base station executes the UAV sensing task ;

[0023] (23) For the communication tasks of user terminals, the base station uses 5G signals to transmit the information required by users to the user terminals; communication tasks The amount of data to be transmitted is Then, by the base station The time resource overhead of the service is:

[0024]

[0025] Where is the communication bandwidth, is the communication transmit power, is the small-scale fading of the communication channel, is the path fading coefficient, is the noise spectral density;

[0026] (24) For the communication tasks of user terminals, the working mode of the base station is multi-base station collaborative service, that is, multiple base stations collaborate to transmit the encoded information to the user terminals. Define the policy matrix of the base station for communication tasks Among them, the element represents the base station The proportion of the execution of communication task ;

[0027] The steps for the base station in (30) to complete the co-sensing resource collaborative optimization problem modeling include:

[0028] (31) The perception task completion degree is defined as the weighted sum of task completions based on priorities. When the perception task policy is, the perception task completion degree is :

[0029] ,

[0030] That is, if at least one base station executes the perception task, it can be determined that the perception task is completed;

[0031] (32) Communication tasks must be guaranteed first. Therefore, the communication task policy needs to satisfy:

[0032]

[0033] That is, for all communication tasks, multiple base stations collaborate to transmit the encoded data they need. When the amount of encoded data received by the user terminal is not less than the amount of data it needs, it decodes and recovers the information it needs;

[0034] (33) The time resources of each base station are limited, and the time overhead of the tasks executed cannot exceed the duration of a scheduling period , that is

[0035] ;

[0036] (34) The base station jointly optimizes the sensing task policy and the communication task policy under the constraints of limited time resources and communication task guarantee, to maximize the sensing task completion degree . This communication and sensing resource collaborative optimization problem is modeled as:

[0037] .

[0038] The steps for the base station in (40) to optimize the communication and sensing resources to obtain an intermediate solution include:

[0039] (41) Relax each element in the optimization variable in the optimization problem P1 into a continuous variable between 0 and 1, prove that the new problem is a convex optimization problem, and obtain the initial solution and through the interior point method or the dual method. This initial solution can be regarded as the upper bound of the performance of the solution to the original problem P1;

[0040] (42) Adopt the nearest neighbor method to quantize the initial solution into a discretized strategy with the shortest Euclidean distance to itself, that is ;

[0041] (43) Since multiple base stations executing the same sensing task will not increase the task completion degree, but will instead increase the resource overhead, that is, the optimal strategy needs to satisfy ; Therefore, the elements in the optimal quantization intermediate solution based on the nearest neighbor are expressed as:

[0042] .

[0043] The steps for the base station in (50) to perform collaborative optimization of communication and sensing resources include:

[0044] (51) Update the set of base stations that do not satisfy the time resource constraint as:

[0045] ;

[0046] (52) If , then jump to step (55), otherwise, update the feasible base station - sensing task pair ;

[0047] (53) Among all feasible base station - sensing task pairs, greedily remove the base station - sensing task pair that causes the smallest reduction in the task completion degree under the unit time resource consumption, that is

[0048]

[0049] Among them In order to the elements therein , and keep other elements unchanged;

[0050] (54) Update the perception task policy , and return to step (51);

[0051] (55) Output the final perception task policy and the communication task policy for the integrated communication and sensing base station to execute.

[0052] Compared with the prior art, the advantages of the present invention are as follows: For the urban low-altitude perception scenario using an integrated communication and sensing base station, aiming at the problem of time resource competition between communication services and sensing services, by designing a communication and sensing resource collaborative optimization method based on relaxed quantization, the refined allocation of communication and sensing resources is realized. On the premise of giving priority to ensuring communication services, the perception performance for low-altitude drones is maximized, filling the technical gap in related fields and meeting the service resource scheduling requirements in the integrated communication and sensing network. Description of the Drawings

[0053] Figure 1 is a low-altitude economic ecosystem diagram.

[0054] Figure 2 is the overall workflow diagram of the present invention.

[0055] Figure 3 is the network model diagram of the present invention.

[0056] Figure 4 is the performance comparison diagram of the method of the present invention and other methods under different numbers of user terminals. Detailed Embodiments

[0057] The present invention will be described in detail below in conjunction with the drawings in the specification and specific embodiments.

[0058] As Figure 2 shown, the present invention relates to a communication and sensing resource collaborative optimization method based on relaxed quantization, including the following steps:

[0059] (10) The integrated communication and sensing base station determines the network parameters for resource scheduling: The integrated communication and sensing base station receives the communication requests of users and the sensing requests of drones, and evaluates the task priorities and the location service relationships between the tasks and the base station. Its system model is as Figure 3 shown.

[0060] (11) There are integrated communication and sensing base stations , drone perceptions and user terminal communication tasks . The number of UAV sensing tasks is the same as the number of UAVs, and the number of communication tasks is the same as the number of user terminals.

[0061] (12) The base station evaluates the priorities of communication and sensing tasks. Communication tasks need to be guaranteed first, and their priority is infinite. For sensing tasks the priority is , which depends on information such as the speed, position, and size of the UAV.

[0062] (13) Each base station has a limited sensing distance , that is, it can only execute sensing tasks within the sensing distance. Define the set of sensing tasks that the base station can execute, where is the base station and the sensing task The physical distance between. Similarly, define the set of base stations that can execute the sensing task .

[0063] (14) Each base station has a limited communication distance , that is, it can only execute communication tasks within the communication distance. Define the set of communication tasks that the base station can execute, and the set of base stations that can execute the sensing task .

[0064] (20) Design the working mode of the integrated communication and sensing base station and evaluate the resource overhead: The integrated communication and sensing base station is based on the time-division method, and different transmission signals and working modes are designed to execute communication and sensing tasks respectively. Based on the differences in transmission signals and working modes, the time resource overhead of executing different tasks is evaluated.

[0065] (21) For UAV sensing tasks, the transmission signal of the base station is a frequency-modulated continuous wave to sense the UAV. To meet the sensing requirements of the base station for the UAV, the signal-to-noise ratio of the sensing echo must exceed a given signal ratio threshold . Therefore, the time resource overhead for the base station to sense the UAV is:

[0066] ,

[0067] where is the Boltzmann constant, is the antenna temperature, is the noise figure, For the unmanned aerial vehicle (UAV) the radar cross - section area is the sensing transmission power is the transmitting antenna gain is the receiving antenna gain is the sensing signal wavelength

[0068] (22) For the UAV sensing task, the working mode of the base station is single - base - station independent sensing, that is, one UAV is sensed by only a single base station. Define the policy matrix of the base station for the sensing task, where the element represents whether the base station executes the UAV sensing task .

[0069] (23) For the user terminal communication task, the base station uses 5G signals to transmit the information required by the user to the user terminal. If the amount of data to be transmitted for the communication task is , then the time - resource overhead for the base station to serve is:

[0070]

[0071] where is the communication bandwidth is the communication transmission power is the small - scale fading of the communication channel is the path - fading coefficient is the noise spectral density

[0072] (24) For the user terminal communication task, the working mode of the base station is multi - base - station collaborative service, that is, multiple base stations collaborate to transmit the encoded information to the user terminal. Define the policy matrix of the base station for the communication task, where the element represents the proportion of the base station executing the communication task .

[0073] (30) The base station completes the modeling of the communication and sensing resource collaborative optimization problem: The base station models the communication and sensing resource collaborative optimization problem through the characterization of time resources and the design constraints of the working mode, and maximizes the completion degree of the sensing task under the constraint of giving priority to satisfying the communication task

[0074] (31) The completion degree of the sensing task is defined as the weighted sum of task completions based on priorities. When the sensing task policy is , the completion degree of the sensing task is

[0075] ,

[0076] That is, if at least one base station performs the sensing task, it can be determined that the sensing task is completed.

[0077] (32) Communication tasks must be guaranteed with priority. Therefore, the communication task strategy shall satisfy:

[0078]

[0079] That is, for all communication tasks, multiple base stations cooperate to transmit the encoded data they need. When the amount of encoded data received by the user terminal is not less than the amount of data it needs, it can decode and recover the information it needs.

[0080] (33) The time resources of each base station are limited, and the time overhead of the tasks performed cannot exceed the duration of one scheduling period , that is

[0081] .

[0082] (34) Under the constraints of limited time resources and communication tasks guarantee, the base station jointly optimizes the sensing task strategy and the communication task strategy to maximize the completion degree of the sensing task . This communication and sensing resource collaborative optimization problem can be modeled as:

[0083]

[0084] (40) The base station obtains an intermediate solution for the optimization of communication and sensing resources: First, relax the discrete variable into a continuous variable to obtain an initial solution of the communication and sensing task strategy, and then quantize the initial solution into a discrete one based on the nearest neighbor to obtain an intermediate solution.

[0085] (41) Relax each element in the optimization variable in the optimization problem P1 into a continuous variable between 0 and 1. It can be proved that the new problem is a convex optimization problem, and the initial solution and can be obtained through the interior point method or the dual method. This initial solution can be regarded as the upper bound of the performance of the solution to the original problem P1.

[0086] (42) Adopt the nearest neighbor method to quantize the initial solution into a discretized strategy with the shortest Euclidean distance to itself, that is .

[0087] (43) Since multiple base stations performing the same sensing task will not increase the task completion degree, but will instead increase the resource overhead additionally, that is, the optimal strategy shall satisfy Therefore, the optimal quantization intermediate solution based on the nearest neighbor The elements in it can be expressed as:

[0088]

[0089] (50) The base station performs collaborative optimization and solution on communication sensing resources: Based on the greedy correction strategy, correct the quantized intermediate solution to make it meet the time resource constraint and generate the final communication and sensing task strategy.

[0090] (51) Update the set of base stations that do not meet the time resource constraint as:

[0091]

[0092] (52) If , then jump to step (55). Otherwise, update the feasible base station-sensing task pair

[0093] (53) Among all feasible base station-sensing task pairs, greedily remove the base station-sensing task pair with the smallest reduction in task completion degree under the unit time resource consumption, that is

[0094]

[0095] where is to The elements in it , and the other elements remain unchanged.

[0096] (54) Update the sensing task strategy , and return to step (51).

[0097] (55) Output the final sensing task strategy and the communication task strategy , for the integrated communication and sensing base station to execute.

[0098] The communication sensing resource collaborative optimization method proposed by the present invention can significantly improve the sensing task completion degree compared with the communication-sensing sequential optimization method. At the same time, the performance of the method proposed by the present invention approaches the performance upper bound. Among them, the communication-sensing sequential greedy optimization method means first using the convex optimization method to complete the allocation of all communication tasks, and then based on the remaining resources of the base station, using the greedy algorithm to complete the optimization of the sensing tasks. The performance upper bound refers to relaxing the discrete optimization variables in P1 to continuous variables between 0 and 1, and using the convex optimization method to obtain the optimal solution, which is the performance upper bound. This performance upper bound is theoretically unreachable.

[0099] Figure 4The performance comparison of the communication-sensing resource collaborative optimization method proposed in the present invention with two algorithms under different numbers of user terminals is compared. It can be seen that the method proposed in the present invention performs fine joint optimization on communication and sensing resources, and its performance is significantly better than that of the communication-sensing sequential optimization method. At the same time, the performance of this solution approaches the performance upper bound infinitely. Therefore, this method can be applied to the optimization problem of communication and sensing resources in the integrated communication and sensing network.

Claims

1. A communication-aware resource collaborative optimization method based on relaxed quantization, characterized in that Including the following steps: (10) The integrated communication and sensing base station determines the network parameters for resource scheduling: The integrated communication and sensing base station receives the communication requests of users and the sensing requests of unmanned aerial vehicles, and evaluates the task priorities and the location service relationships between the tasks and the base station; (20) The integrated communication and sensing base station designs the working modes and evaluates the resource consumption: The integrated communication and sensing base station is based on the time division method, and designs different transmission signals and working modes to execute communication and sensing tasks respectively. Based on the differences in the transmission signals and working modes, it evaluates the time resource consumption for executing different tasks; (30) The base station completes the modeling of the collaborative optimization problem of communication and sensing resources: The base station models the collaborative optimization problem of communication and sensing resources through the characterization of time resources and the design constraints of working modes, and maximizes the completion degree of sensing tasks under the constraint of giving priority to satisfying communication tasks; (40)The base station optimizes the communication sensing resources to obtain an intermediate solution: First, relax the discrete variables to continuous variables, obtain the initial solution of the communication sensing task strategy, and then quantize the initial solution into discrete values based on the nearest neighbor to obtain an intermediate solution; (50) The base station performs collaborative optimization and solution for communication and sensing resources: Based on the greedy correction strategy, the quantified intermediate solution is corrected to make it meet the time resource constraint, and the final communication and sensing task strategy is generated.

2. The communication-aware resource collaborative optimization method based on relaxed quantization according to claim 1, wherein The step (10) of the integrated communication and sensing base station determining the network parameters for resource scheduling includes: There are integrated communication and sensing base stations , UAV sensing and user terminal communication tasks , where the number of UAV sensing tasks is the same as the number of UAVs, and the number of communication tasks is the same as the number of user terminals; (12) The base station evaluates the priorities of communication and sensing tasks. Communication tasks need to be guaranteed with the highest priority, and its priority is infinity; the priority of sensing tasks is , which depends on the speed, position, and size information of the UAV; (13) Each base station has a limited sensing distance , that is, it can only perform sensing tasks within the sensing distance; define the set of sensing tasks that the base station can execute, where is the physical distance between the base station and the sensing task ; similarly, define the set of base stations that can execute the sensing task ; (14) Each base station has a limited communication range , that is, it can only perform communication tasks within the communication range; define the set of communication tasks that a base station can perform, and the set of base stations that can perform sensing tasks .

3. A communication-aware resource collaborative optimization method based on relaxed quantization according to claim 1, characterized in that, The step (20) of the integrated communication and sensing base station designing the working modes and evaluating the resource consumption includes: For the UAV sensing task, the transmission signal of the base station is a frequency-modulated continuous wave to sense the UAV. To meet the sensing requirements of the base station for the UAV, the signal-to-noise ratio of the sensing echo must exceed a given signal ratio threshold ; Therefore, the base station senses the UAV The time resource overhead is as follows: ， where is the Boltzmann constant, is the antenna temperature, is the noise figure, is the radar cross section of the UAV , is the sensing transmit power, is the transmit antenna gain, is the receive antenna gain, is the sensing signal wavelength; (22) For the UAV sensing task, the working mode of the base station is single-base station independent sensing, that is, one UAV is sensed by only a single base station, and the policy matrix of the base station for the sensing task is defined , where the element represents the base station whether to execute the UAV sensing task ; For the communication tasks of the user terminal, the base station uses 5G signals to transmit the information required by the user to the user terminal; communication tasks The amount of data to be transmitted is , then by the base station The time resource overhead of the service is: ； wherein is the communication bandwidth, is the communication transmit power, is the small-scale fading of the communication channel, is the path fading coefficient, is the noise spectral density; (24) For the communication tasks of user terminals, the working mode of the base stations is multi-base-station collaborative service, that is, multiple base stations collaborate to transmit the encoded information to the user terminals. Define the policy matrix of the base stations for communication tasks , where the element represents that the base station executes the communication task in a certain proportion.

4. A communication-aware resource collaborative optimization method based on relaxed quantization according to claim 1, characterized in that The step (30) of the base station completing the modeling of the collaborative optimization problem of communication and sensing resources includes: The completion degree of the perception task is defined as the weighted sum of task completions based on priorities. When the perception task strategy is the completion degree of the perception task is : ， That is, if at least one base station executes the sensing task, it can be determined that the sensing task is completed; (32)Communication tasks must be guaranteed with priority, so the communication task strategy shall meet the following requirements: ； That is, for all communication tasks, multiple base stations cooperate to transmit the encoded data they need. When the amount of encoded data received by the user terminal is not less than the amount of data it needs, it decodes and recovers the information it needs; (33) The time resources of each base station are limited, and the time overhead of the executed tasks cannot exceed the duration of a scheduling period , that is ； (34)The base station jointly optimizes the sensing task policy and the communication task policy under the constraints of limited guarantee of time resources and communication tasks, to maximize the sensing task completion degree . The co - optimization problem of communication and sensing resources is modeled as: 。 5. A communication-aware resource collaborative optimization method based on relaxed quantization according to claim 1, characterized in that The step (40) of the base station obtaining an intermediate solution by optimizing communication and sensing resources includes: (41) Relax each element in the optimization variables in the optimization problem P1 to a continuous variable between 0 and 1, prove that the new problem is a convex optimization problem, and obtain an initial solution by the interior point method or the dual method and , and this initial solution can be regarded as an upper bound on the performance of the solution to the original problem P1; (42) Using the nearest neighbor method, the initial solution is quantized into a discretization strategy with the shortest Euclidean distance to itself, that is ; (43) Since multiple base stations performing the same sensing task will not increase the task completion rate but will instead incur additional resource overhead, the optimal strategy needs to satisfy ; Therefore, the elements in the optimal quantization intermediate solution are represented as: 。 6. A communication-aware resource collaborative optimization method based on relaxation quantization according to claim 1, characterized in that The step (50) of the base station performing collaborative optimization and solution for communication and sensing resources includes: (51) Update the set of base stations that do not meet the time resource constraint as: ； If , jump to step (55); otherwise, update the feasible base station-sensing task pairs . (53) Among all feasible base station - sensing task pairs, greedily remove the base station - sensing task pair with the smallest reduction in the task completion degree under the unit time resource consumption, that is ； Among them In order to the elements therein , and keep other elements unchanged; (54) Update the perception task strategy , and return to step (51); (55) Output the final sensing task strategy and communication task strategy for the integrated communication and sensing base station to execute.

Images