A Data Compression-Based Method for Optimizing the Freshness of UAV Information Acquisition Data

Abstract

This invention discloses a method for optimizing the freshness of UAV information collection data based on data compression, belonging to the field of wireless communication technology. The method includes: constructing a problem to minimize the average information age of sensor data; decomposing the average information age minimization problem into three optimization sub-problems based on UAV coverage constraints: sensor relay relationship, UAV hovering position, and UAV trajectory; using a clustering algorithm to cluster sensor nodes; employing a greedy algorithm to solve the sensor relay relationship optimization sub-problem; using a simulated annealing algorithm to solve the UAV hovering position optimization sub-problem; alternately iterating the sensor relay relationship and UAV hovering position sub-problems; and using an ant colony algorithm to solve the UAV trajectory optimization sub-problem, finally obtaining the average information age of the sensors, thereby effectively reducing the average information age of data collected by the UAV.

Description

Technical Field

[0001] This invention belongs to the field of wireless communication technology, specifically relating to a method for optimizing the freshness of UAV information collection data based on data compression. Background Technology

[0002] In recent years, unmanned aerial vehicle (UAV)-assisted wireless sensor networks have been widely used in environmental monitoring, disaster relief, and smart agriculture due to their advantages such as flexible deployment and wide-area coverage. To ensure the timeliness of information, the Age of Information (AoI) is used as a standard to measure the freshness of information.

[0003] However, existing research methods mostly focus on the allocation and scheduling of communication resources, neglecting the inherent redundancy and spatiotemporal correlation of data. Data compression is a key technology for reducing the resources required for data storage and transmission. By reducing redundant information in the data, it reduces file size, thereby saving storage space, accelerating data transmission speed, and reducing bandwidth consumption. However, research on applying data compression technology to timeliness indicators in UAV information collection scenarios is still limited. Summary of the Invention

[0004] To address the shortcomings of existing technologies, the present invention aims to provide a method for optimizing the freshness of UAV information collection data based on data compression, thereby solving the problems in existing technologies.

[0005] The objective of this invention can be achieved through the following technical solutions: A method for optimizing the freshness of UAV information collection data based on data compression includes the following steps: The problem of minimizing the average information age of sensor data is constructed. Based on the coverage constraint of UAV, the problem of minimizing the average information age is decomposed into three optimization sub-problems: sensor relay relationship, UAV hovering position, and UAV trajectory. Clustering algorithms are used to cluster the nodes of the sensors, and a greedy algorithm is used to solve the sensor relay relationship optimization subproblem. Simulated annealing algorithm is used to solve the UAV hovering position optimization subproblem. The sensor relay relationship and UAV hovering position subproblems are alternately iterated, and the UAV trajectory optimization subproblem is solved using ant colony algorithm. Finally, the average information age of the sensors is obtained.

[0006] Furthermore, the method is applied to a data acquisition system, including: One sensor, one drone, and one base station; sensor nodes are randomly distributed within the target area and support direct device-to-device communication; sensor nodes ,sensor The coordinates are as The drone departs from the base station, collects data generated by all sensors in the area, and returns to the base station to unload the data. Each sensor can directly upload data to the drone or send data to other sensors. A receiving sensor can act as a relay to send received data back to the drone. If a sensor acts as a relay, it can compress the data based on the correlation between sensor data. Binary variables are used. Indicates the sensor relay relationship, if pass Relay, then When the drone hovers above the cluster, the sensors within the cluster upload data using time-division multiple access.

[0007] Furthermore, the average information age minimization problem is defined as P1: In the formula, the optimization variable is the sensor relay relationship. Drone hovering position and drone trajectories The first constraint C1 represents each sensor Only within the cluster Select one sensor It can act as a relay, but cannot choose itself as a relay; the second constraint C2 represents the sensor. With sensors Relay indication The value can only be 0 or 1; the third constraint C3 indicates that the drone is in the cluster hovering time above It depends on the sum of the time it takes for nodes within the cluster to transmit data to the drone. ,in Indicates sensor Time required to upload data For the first The total number of sensors in the cluster; the fourth constraint C4 represents the number of drones from the cluster. arrive Flight time Spacing between clusters Directly related The fifth constraint, C5, represents the flight speed of the drone; the fifth constraint, C5, represents the sensor speed within any cluster. and hovering point maximum distance Smaller than the drone's coverage radius .

[0008] Furthermore, the sensor relay relationship optimization subproblem is defined as P2: in, The revenue gained from all sensor relays. For sensors The benefits gained from relaying For sensors The time required to directly upload data to the drone, For sensors Through sensors The time required for relay upload, This represents the total number of sensors; The sub-problem of optimizing the drone's hovering position is defined as P3: The UAV trajectory optimization subproblem is defined as P4: .

[0009] Furthermore, the process of using a greedy algorithm to solve the sensor relay relationship optimization subproblem includes: using the K-means clustering algorithm to group the sensors, obtaining the cluster to which the sensors belong, using the greedy algorithm to find the nearest node of each sensor within the cluster, and evaluating whether to establish a relay relationship based on the relay benefit.

[0010] Furthermore, the process of solving the UAV hovering position optimization subproblem using the simulated annealing algorithm includes: based on the subproblem... The relay relationship uses the geometric center of the coordinates of the subset of sensors that directly upload data within the cluster as the initial solution. Within its neighborhood, new positions are continuously explored through controlled random perturbations, and the optimal solution is output.

[0011] Furthermore, the process of solving the UAV trajectory optimization subproblem using the ant colony algorithm includes: Alternating Iterative Subproblem and By obtaining stable drone hovering positions, the base station and drone hovering positions are abstracted into city nodes with hovering time and flight time. The ant colony algorithm is then used to solve the traveling salesman problem of the drone trajectory.

[0012] A device for optimizing the freshness of UAV information collection data based on data compression, comprising the following steps: Problem construction and decomposition module: Construct the problem of minimizing the average information age of sensor data. Based on the UAV coverage constraint, the problem of minimizing the average information age is decomposed into three optimization sub-problems: sensor relay relationship, UAV hovering position, and UAV trajectory. Problem-solving module: Clustering algorithm is used to cluster the nodes of the sensor, and greedy algorithm is used to solve the sensor relay relationship optimization subproblem; simulated annealing algorithm is used to solve the UAV hovering position optimization subproblem; the sensor relay relationship and UAV hovering position subproblems are alternately iterated, and ant colony algorithm is used to solve the UAV trajectory optimization subproblem, and finally the average information age of the sensor is obtained.

[0013] A computer storage medium storing a readable program, which, when executed, instructs a computing device to perform a data compression-based method for optimizing the freshness of UAV information collection data, as described above.

[0014] An electronic device includes: a processor, a memory, a communication interface, and a communication bus, wherein the processor, the memory, and the communication interface communicate with each other through the communication bus; The memory is used to store at least one executable instruction, which causes the processor to perform an operation corresponding to the above-described method for optimizing the freshness of UAV information collection data based on data compression.

[0015] The beneficial effects of this invention are: 1. This invention constructs a system model for UAVs to collect ground sensor data and unload it to a base station. The UAV departs from the base station, collects data generated by all sensors within the area, and returns to the base station to unload the data. A data compression-based method for optimizing the freshness of UAV information collection data is proposed. A hierarchical optimization framework is adopted, dividing the original problem into three sub-problems: sensor relay relationship optimization, UAV hovering point optimization, and UAV trajectory optimization. Data compression technology reduces redundant data, saves communication resources, and improves data timeliness.

[0016] 2. This invention considers the spatial correlation of information collected by sensors in close proximity. It employs K-means clustering to group sensors based on the drone's coverage area. Each cluster member node uploads its collected data to the drone using TDMA (Time-of-Flight Distributed Mapping) during the time the drone hovers above its cluster. Sensors are relayed through other sensors, and data compression is performed based on spatial correlation. A greedy relay selection algorithm is used to determine relay relationships. From these relationships, a subset of sensors directly uploading data to the drone is obtained. Since the hovering position is only affected by directly uploading sensors, simulated annealing is used to find the position with the minimum upload time within a cluster, which is the drone's hovering position. The sensor relay relationship subproblem and the drone hovering position optimization subproblem iterate until a stable relay relationship and hovering position are obtained. Finally, based on the optimal hovering positions of all clusters, an ant colony algorithm is used to plan the drone's global flight trajectory. This invention can effectively reduce the average information age of sensor data. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a schematic diagram of the optimized method flow of the present invention; Figure 2 This is a schematic diagram of the system model corresponding to the UAV information collection data freshness optimization method based on data compression of the present invention; Figure 3 This is a schematic diagram of the relay relationship and flight trajectory corresponding to the UAV information collection data freshness optimization method based on data compression of the present invention; Figure 4 This is a graph showing the variation of average AoI with the number of sensors under the present invention's method for optimizing the freshness of UAV information collection data based on data compression and comparison algorithm; Figure 5 This is a graph showing the average AoI as a function of the number of sensors under different uploaded data volumes, based on the data compression-based UAV information collection data freshness optimization method of this invention. Figure 6 This is a graph showing the variation of average AoI with the number of sensors under different data correlation levels under the data compression-based UAV information acquisition data freshness optimization method of this invention; Figure 7 This is a graph showing the variation of average AoI with the number of sensors at different flight altitudes under the UAV information collection data freshness optimization method based on data compression according to the present invention. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] Example 1 The data freshness optimization method for UAV information collection based on data compression in this invention is applied to, for example... Figure 2 The data acquisition system shown includes: The system consists of one sensor, one drone, and one base station; sensor nodes are randomly distributed within the target area and support direct device-to-device communication. Sensor nodes ,sensor The coordinates are as The drone departs from the base station, collects data generated by all sensors within the area, and returns to the base station to unload the data. Each sensor can directly upload data to the drone or send data to other sensors; receiving sensors can act as relays to send received data to the drone. If a sensor acts as a relay sensor, it can compress the data based on the correlation between sensor data. Binary variables are used. Indicates the sensor relay relationship, if pass Relay, then When the drone hovers above the cluster, the sensors within the cluster upload data using time-division multiple access.

[0021] like Figure 1 As shown, a method for optimizing the freshness of UAV information collection data based on data compression includes the following steps: S1, construct the problem of minimizing the average information age of sensor data. Based on the drone coverage constraint, the original optimization problem is decomposed into three sub-problems: sensor relay relationship, drone hovering position, and drone trajectory.

[0022] Specifically as follows: Communication between sensors within the same cluster adopts a ground-based point-to-point communication model. and Transmission rate It can be represented as: in, For system bandwidth, Indicates the sensor's transmission power. The channel gain is given at a reference distance of 1m. For sensors and The distance between, For environmentally related constants, This represents the power of Gaussian white noise.

[0023] sensor Towards The amount of data transmitted is Time spent on byte data packets for: We model the channel between the sensor and the UAV by jointly considering line-of-sight (LAS) and non-LAS effects and their probabilities. The transmission channel between the UAV and the sensor is related to both LAS and non-LAS links, and we use... Indicates the drone's flight altitude; the drone is at the [missing information - likely a typo or incomplete point]. The cluster's hovering position is ,sensor Distance between UAVs and drones for: drones Cluster collection When data is collected, drones and The probability of a line-of-sight connection existing between them can be expressed as: in, and It is a constant that depends on the environment. Angle of elevation of the drone Drones and The large-scale path loss effect between them can be calculated as follows: In the formula, and Let represent the average path loss of line-of-sight and non-line-of-sight links, respectively, which can be calculated as follows: in, For the carrier frequency of the drone, At the speed of light, and These are the additional path losses for line-of-sight and non-line-of-sight links, respectively.

[0024] The data transmission rate to the drone is: in, For the system bandwidth between the sensor and the drone, Indicates the sensor's transmission power. This represents the power of Gaussian white noise. The amount of data transmitted to the drone is Time spent on byte data packets for: The drone offloads all collected data to the base station via a line-of-sight link, ignoring the base station's altitude and the channel gain between the drone and the base station. ,in The channel gain at a reference distance of 1m corresponds to the transmission rate between the UAV and the base station. for: in, Indicates the sensor's transmission power. This refers to the system bandwidth between the sensor and the drone.

[0025] If sensors are located close to each other, the collected data will be correlated. Consider compressing the data from different sensors based on their spatial correlation to remove redundant data. and Spatial correlation can be expressed as: in, This is the spatial correlation coefficient. for and The distance between, This represents the degree of spatial correlation and is a constant. Assuming the data generated by the sensors conforms to a Gaussian distribution, for a joint Gaussian distribution... and Mutual information per byte of data Correlation coefficient The relationship is: It can be seen that for , The amount of redundant data is After removing redundant information, Data volume It can be regarded as: like pass A relay can be considered as First Bytes of data sent , Then compress Data Upload to the drone. (Set) pass Relaying the drone's data requires no waiting. The data upload time via relay is: in, for Towards The time it takes to send data. for Will The time it takes to upload to the drone.

[0026] The time it takes to upload data to the drone can be recorded as: The drone eventually received The amount of data is: The time spent by the drone after completing all sensor data collection and unloading it to the base station can be calculated as follows: AoI mainly includes the flight time of drones, Data upload time and the time it takes for the drone to arrive at the base station and complete data unloading. Each drone departs from the base station, visits each hovering point, and then returns to the base station to complete data unloading. It is assumed that when the drone returns to the data center, it immediately unloads the data. The sensors are grouped using a clustering algorithm. The cluster, the first The total number of sensors in the cluster is used Indicates. When It is the first drone The first cluster When uploading data from a sensor, The AoI can be represented as: The first term in the formula is The time of data upload, the second item is the [number]. Intra-cluster acquisition order arrive The sum of sensor upload times, the third item being all uncollected clusters arrive The fourth item is the time required for all sensors within the cluster to upload data. For drones from cluster to cluster The fifth item is the total flight time required for data collection, and the fifth item is the time when the drone ends its flight. Clusters are collected and then sent to the base station. Flight time, the last item is for drones Offload all data to the base station The time.

[0027] The average information age of all sensors is taken as the optimization objective in this embodiment. The average information age can be expressed as: By optimizing sensor relay relationships Drone hovering point location and drone trajectories Minimize the average information age (AoI) of the sensor data. The problem of minimizing the average information age of the sensor data is defined as P1: In the formula, the optimization variable is the sensor relay relationship. Drone hovering position and drone trajectories The first constraint C1 represents each sensor Only within the cluster Select one sensor It can act as a relay, but cannot choose itself as a relay; the second constraint C2 represents the sensor. With sensors Relay indication The value can only be 0 or 1; the third constraint C3 indicates that the drone is in the cluster hovering time above It depends on the sum of the time it takes for nodes within the cluster to transmit data to the drone. ,in Indicates sensor Time required to upload data For the first The total number of sensors in the cluster; the fourth constraint C4 represents the number of drones from the cluster. arrive Flight time Spacing between clusters Directly related The fifth constraint, C5, represents the flight speed of the drone; the fifth constraint, C5, represents the sensor speed within any cluster. and hovering point maximum distance Smaller than the drone's coverage radius .

[0028] Will The time required to directly upload drone data and pass The difference in time required for the relay to re-upload the data from the drone is considered the relay revenue. , can be represented as: Will The benefit obtained by all sensor relays is defined as P2. The sensor relay relationship optimization subproblem is defined as P2: The drone hovering position optimization sub-problem is defined as P3: The UAV trajectory optimization sub-problem is defined as P4: S2 uses a clustering algorithm to cluster the sensor nodes, a greedy algorithm to solve the sensor relay relationship optimization subproblem, a simulated annealing algorithm to solve the drone hovering position optimization subproblem, and alternately iterates the sensor relay relationship and drone hovering position subproblems, and uses an ant colony algorithm to solve the drone trajectory optimization subproblem, finally obtaining the average information age of the sensors.

[0029] S21, the process of using clustering algorithms to create clusters and using a greedy algorithm to find the optimal solution for sensor relay relationships: Let the coverage angle of the drone be The flight altitude is In this case, the coverage radius of the drone can be expressed as To ensure that each sensor is covered by the drone during communication, the drone's coverage radius is used as the maximum clustering radius, and the K-means algorithm is employed to cluster the ground sensors. The K-means algorithm is a centroid-based unsupervised clustering method designed to divide data samples into K non-overlapping clusters. First, K cluster centers are randomly initialized. For each cluster center, the distance between the cluster center and the data point is calculated, and then each point is assigned to the nearest cluster center. When the cluster assignment of any point changes, the mean of the data points within the cluster is recalculated as the new center, until the cluster centers converge or the maximum number of iterations is reached.

[0030] After completing sensor clustering, it is necessary to determine the data relay relationships within each cluster. The time required to directly upload drone data and pass The difference in time required for the relay to re-upload the data from the drone is considered the relay revenue. , can be represented as: like ,show pass Relays can reduce the amount of data uploaded. As The relay sensor.

[0031] Considering that data transmission efficiency is negatively correlated with node spacing, and that data from neighboring nodes typically has higher correlation, an algorithm based on a greedy approach is adopted to jointly optimize communication efficiency and data value, thus solving for the optimal relay relationship. The algorithm achieves optimal relay selection through two stages: First, the node... Search for neighbors within the cluster, calculate Get the nearest node As a candidate relay node; subsequently, the node is based on relay revenue. The algorithm evaluates whether to establish a relay relationship. For multi-hop relay sensors, to avoid accumulated latency in relay transmission and maximize the data reduction from single relay compression, the benefit of each relay segment is compared, and the segment with the greater benefit is selected as the final relay relationship. The final algorithm outputs the sensor relay relationships. .

[0032] S22, the process of using simulated annealing algorithm to find the optimal solution for the drone's hovering position: According to the subproblem The relay relationship allows us to obtain the subset of sensors actually responsible for uploading data. The time it takes for data to be uploaded to the UAV is determined only by the nodes within the cluster that directly communicate with the UAV. The geometric center of the subset of sensors directly uploading data within the cluster is used as the initial solution, and new locations are continuously explored within its neighborhood through controlled random perturbations. By introducing an annealing temperature scheduling mechanism, the algorithm can accept slightly inferior solutions with a higher probability in the early stages of the search to escape local optima, and gradually converge to an approximately optimal hovering point as iterations progress. First, the algorithm randomly generates new solutions near the initial hovering point. According to the Metropolis criterion, the probability of accepting a new solution is: in, The current temperature. The total upload time obtained in this iteration This is the retained solution from the previous iteration. The simulated annealing algorithm accepts inferior solutions with a certain probability to avoid getting trapped in local optima, thus obtaining the optimal hovering position for the UAV. When the algorithm reaches the maximum number of iterations or meets the termination condition, it outputs the current solution as the optimal solution.

[0033] In this model, sensors within the cluster sequentially upload data to the UAV using TDMA, and the AoI (Aspect-Oriented Intelligence) of the data begins to be calculated when it leaves the sensor node. This means that the AoI of each sensor's data begins to accumulate while waiting for data from other sensor nodes to upload. To minimize the system's average AoI, this embodiment adopts the "Longest Processing Time First" (LPT) principle, based on the upload time of all sensors within the cluster. Upload tasks are sorted in descending order and uploaded sequentially. This strategy ensures that the data with the longest upload time is served first, thereby suppressing the overall growth of AoI (Aspect-Oriented Income).

[0034] S23, the process of using the ant colony algorithm to find the optimal solution for the drone trajectory: Subproblems AND problem Through iterative processes, stable relay relationships and hovering positions are obtained. Each hovering node in the drone network is abstracted as a "city" node with hovering time and flight time, representing the target location that the ants need to explore. First, parameters such as ant colony size, initial pheromone distribution, pheromone evaporation rate, and maximum number of iterations are initialized. ants were randomly placed Given a hovering position and a base station, set the first element of each ant's taboo list to its current position. Use probability. To indicate the first An ant is hovering from the point Transferred to The probability of: In the formula, For the first wheel arrive The concentration of pheromones along the path, This is a heuristic function whose value is equal to... and The reciprocal of the path length ,parameter and These are the pheromone factor and the heuristic function factor, both constants, which determine the influence weights of pheromone concentration and path distance when ants choose the next node, respectively. Used to record the Nodes not visited by ants. After completing one round of trajectory planning, pheromone updates are performed. The path will contain pheromones left by ants. and The pheromone representation on the inter-path is as follows: in, As a pheromone volatile factor, Indicates the first Only one ant remained on the side in this round. The amount of pheromones on the surface.

[0035] After all ants have traversed all nodes, the objective function value corresponding to the path constructed by each ant is calculated, and the optimal solution in the current iteration is recorded. Then, the pheromone matrix is ​​updated based on path quality, and a new round of search begins. When the preset maximum number of iterations is reached, the algorithm outputs the globally optimal path (drone trajectory). .

[0036] Based on a similar inventive concept, embodiments of the present invention also provide a computer storage medium storing a readable program, which, when run by a processor, can execute the above-described method for optimizing the freshness of UAV information collection data based on data compression.

[0037] Based on a similar inventive concept, this invention provides an electronic device, including: a processor, a memory, a communication interface, and a communication bus, wherein the processor, the memory, and the communication interface communicate with each other through the communication bus; The memory is used to store at least one executable instruction, which causes the processor to perform the operation corresponding to the above-described method for optimizing the freshness of UAV information collection data based on data compression.

[0038] Based on a similar inventive concept, embodiments of the present invention also provide a computer program product, including computer instructions, which instruct a computing device to perform the operations corresponding to the above-described method for optimizing the freshness of UAV information collection data based on data compression.

[0039] Example 2 Based on the data compression-based method for optimizing the freshness of UAV information collection data proposed in Example 1, this example proposes a data compression-based device for optimizing the freshness of UAV information collection data, specifically including: Problem construction and decomposition module: Construct the problem of minimizing the average information age of sensor data. Based on the UAV coverage constraint, the problem of minimizing the average information age is decomposed into three optimization sub-problems: sensor relay relationship, UAV hovering position, and UAV trajectory.

[0040] Problem-solving module: Clustering algorithm is used to cluster the nodes of the sensor, and greedy algorithm is used to solve the sensor relay relationship optimization subproblem; simulated annealing algorithm is used to solve the UAV hovering position optimization subproblem; the sensor relay relationship and UAV hovering position subproblems are alternately iterated, and ant colony algorithm is used to solve the UAV trajectory optimization subproblem, and finally the average information age of the sensor is obtained.

[0041] Example 3 Based on the data compression-based method for optimizing the freshness of UAV information acquisition data proposed in Example 1, this example presents the experimental process and results of the data compression-based method for optimizing the freshness of UAV information acquisition data, specifically including: The method proposed in this invention was simulated using MATLAB software. The simulation environment was constructed as follows: several wireless sensor nodes were randomly deployed within a 300 m × 300 m square area, and the coordinates of the data center were […]. A drone is responsible for collecting sensing data from each node. In the simulation, the drone maintains a constant flight altitude of 100m and a uniform speed of 10m / s. Each sensor uploads the same amount of data, 2560 bytes.

[0042] Figure 3 shows the sensor relay relationships and the UAV flight trajectory. Sensors in different clusters are represented by dots of different colors, and relay relationships are indicated by gray arrows. The sensor being pointed to is the relay sensor of the pointing sensor. The UAV uses [missing information - likely a specific indicator] at each cluster's hovering point. Identifier. Base station coordinates are... The drone's flight path is drawn with a solid blue line, showing that the drone starts from the base station, follows the path to hover above each cluster to collect information, and then flies back to the base station.

[0043] The performance of the method of this invention was compared with two benchmark algorithms. Algorithm 1 was configured such that the sensors did not relay data, meaning data was directly uploaded to the UAV without compression, and the UAV only hovered at the cluster center to collect information for each cluster. Algorithm 2 was configured such that the UAV did not optimize the hovering position within the cluster, only calculated the sensor relay relationship, and the UAV hovered at the initial cluster center point to collect data.

[0044] Figure 4 The relationship between the average AoI of the method of this invention and comparative algorithms 1 and 2 as the number of sensors increases is presented. It can be seen that the average AoI of the method of this invention is the lowest for all different numbers of sensors, indicating that the method of this invention has a certain advantage in reducing information age. Furthermore, the average AoI of all algorithms increases with the increase of the number of sensors. This is because as the number of sensors increases, the number of sensors within each cluster also increases, leading to a longer hovering time for the UAV to collect data, and simultaneously increasing the time for the UAV to unload data from the base station, ultimately causing the average AoI to rise.

[0045] Figure 5 The paper presents the trend of average AoI (Aspect of Interest) as the number of sensors increases under different uploaded data volumes. With other parameters remaining constant, the average AoI increases with the increase in uploaded data volume. This is mainly because an increase in uploaded data volume leads to a longer sensor data transmission time, which in turn increases the drone's hovering time, ultimately resulting in an increase in the average AoI.

[0046] Figure 6The impact of data relevance on the average AoI is presented. It can be seen that, for the same number of sensors, a higher degree of data relevance results in a lower average AoI. Since the correlation coefficient is directly proportional to the degree of data relevance, the higher the degree of data relevance, the closer the correlation coefficient is to 1. A higher degree of data relevance means more redundant data can be compressed, resulting in higher benefits from sensor relay and a corresponding reduction in data upload time, thus lowering the average AoI.

[0047] Figure 7 The average AoI (Aspect-of-Identity) varies with the number of sensors at different flight altitudes. It can be seen that regardless of the number of sensors, the lower the flight altitude, the lower the average AoI. At lower flight altitudes, the communication rate between the UAV and ground sensors is higher, and the data upload time is shorter. Furthermore, the UAV's coverage radius... As can be seen, the coverage angle of a drone is fixed, and a lower flight altitude will reduce its coverage area. Therefore, it is necessary to divide the sensor into more clusters, which increases the flight time to some extent. Under the combined effect of the above two factors, the communication time gain from reducing the flight altitude exceeds the flight time cost caused by the increase in clustering. Therefore, the average AoI shows the downward trend shown in the figure.

[0048] The methods of the present invention can be implemented in hardware, firmware, or as software or computer code that can be stored in a recording medium (such as a CD-ROM, RAM, floppy disk, hard disk, or magneto-optical disk), or as computer code originally stored on a remote recording medium or a non-transitory machine-readable medium and subsequently stored on a local recording medium, downloaded via a network. Thus, the methods described herein can be processed by software stored on a recording medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware (such as an ASIC or FPGA). It is understood that the computer, processor, microprocessor controller, or programmable hardware includes storage components (e.g., RAM, ROM, flash memory, etc.) capable of storing or receiving software or computer code that, when accessed and executed by the computer, processor, or hardware, implements the methods described herein. Furthermore, when a general-purpose computer accesses the code used to implement the methods shown herein, the execution of the code transforms the general-purpose computer into a dedicated computer for performing the methods shown herein.

[0049] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention.

Claims

1. A method for optimizing the freshness of UAV information collection data based on data compression, characterized in that, Includes the following steps: The problem of minimizing the average information age of sensor data is constructed. Based on the coverage constraint of UAV, the problem of minimizing the average information age is decomposed into three optimization sub-problems: sensor relay relationship, UAV hovering position, and UAV trajectory. Clustering algorithms are used to cluster the nodes of the sensors, and a greedy algorithm is used to solve the sensor relay relationship optimization subproblem. The simulated annealing algorithm is used to solve the subproblem of drone hovering position optimization. The sensor relay relationship and the UAV hovering position subproblem are solved alternately and iteratively, and the UAV trajectory optimization subproblem is solved using the ant colony algorithm. Finally, the average information age of the sensors is obtained.

2. The method for optimizing the freshness of UAV information collection data based on data compression according to claim 1, characterized in that, Applied to data acquisition systems, including: One sensor, one drone, and one base station; sensor nodes are randomly distributed within the target area and support direct device-to-device communication; sensor nodes ,sensor The coordinates are as The drone departs from the base station, collects data generated by all sensors in the area, and returns to the base station to unload the data. Each sensor can directly upload data to the drone or send data to other sensors. A receiving sensor can act as a relay to send received data back to the drone. If a sensor acts as a relay, it can compress the data based on the correlation between sensor data. Binary variables are used. Indicates the sensor relay relationship, if pass Relay, then When the drone hovers above the cluster, the sensors within the cluster upload data using time-division multiple access.

3. The method for optimizing the freshness of UAV information collection data based on data compression according to claim 1, characterized in that, The average information age minimization problem is defined as P1: In the formula, the optimization variable is the sensor relay relationship. Drone hovering position and drone trajectories The first constraint C1 represents each sensor Only within the cluster Select one sensor It can act as a relay, but cannot choose itself as a relay; the second constraint C2 represents the sensor. With sensors Relay indication The value can only be 0 or 1; the third constraint C3 indicates that the drone is in the cluster hovering time above It depends on the sum of the time it takes for nodes within the cluster to transmit data to the drone. ,in Indicates sensor Time required to upload data For the first The total number of sensors in the cluster; the fourth constraint C4 represents the number of drones from the cluster. arrive Flight time Spacing between clusters Directly related The fifth constraint, C5, represents the flight speed of the drone; the fifth constraint, C5, represents the sensor speed within any cluster. and hovering point maximum distance Smaller than the drone's coverage radius .

4. The method for optimizing the freshness of UAV information collection data based on data compression according to claim 3, characterized in that, The sensor relay relationship optimization subproblem is defined as P2: in, The revenue gained from all sensor relays. For sensors The benefits gained from relaying For sensors The time required to directly upload data to the drone, For sensors Through sensors The time required for relay upload, This represents the total number of sensors; The sub-problem of optimizing the drone's hovering position is defined as P3: The UAV trajectory optimization subproblem is defined as P4: 。 5. The method for optimizing the freshness of UAV information collection data based on data compression according to claim 4, characterized in that, The process of solving the sensor relay relationship optimization subproblem using a greedy algorithm includes: using the K-means clustering algorithm to group the sensors, obtaining the cluster to which the sensors belong, using the greedy algorithm to find the nearest node of each sensor in the cluster, and evaluating whether to establish a relay relationship based on the relay benefits.

6. The method for optimizing the freshness of UAV information collection data based on data compression according to claim 4, characterized in that, The process of solving the UAV hovering position optimization subproblem using the simulated annealing algorithm includes: based on the subproblem... The relay relationship uses the geometric center of the coordinates of the subset of sensors that directly upload data within the cluster as the initial solution. Within its neighborhood, new positions are continuously explored through controlled random perturbations, and the optimal solution is output.

7. The method for optimizing the freshness of UAV information collection data based on data compression according to claim 4, characterized in that, The process of solving the drone trajectory optimization subproblem using the ant colony algorithm includes: Alternating Iterative Subproblem and By obtaining stable drone hovering positions, the base station and drone hovering positions are abstracted into city nodes with hovering time and flight time. The ant colony algorithm is then used to solve the traveling salesman problem of the drone trajectory.

8. A device for optimizing the freshness of UAV information collection data based on data compression, comprising the method described in any one of claims 1-7, characterized in that, include: Problem construction and decomposition module: Construct the problem of minimizing the average information age of sensor data. Based on the UAV coverage constraint, the problem of minimizing the average information age is decomposed into three optimization sub-problems: sensor relay relationship, UAV hovering position, and UAV trajectory. Problem-solving module: Clustering algorithm is used to cluster the nodes of the sensor, and greedy algorithm is used to solve the sensor relay relationship optimization subproblem; simulated annealing algorithm is used to solve the UAV hovering position optimization subproblem; the sensor relay relationship and UAV hovering position subproblems are alternately iterated, and ant colony algorithm is used to solve the UAV trajectory optimization subproblem, and finally the average information age of the sensor is obtained.

9. A computer storage medium storing a readable program, characterized in that, When the program runs, it can instruct the computing device to execute a method for optimizing the freshness of UAV information collection data based on data compression as described in any one of claims 1-7.

10. An electronic device, characterized in that, include: The processor, memory, communication interface, and communication bus are provided, wherein the processor, memory, and communication interface communicate with each other via the communication bus. The memory is used to store at least one executable instruction, which causes the processor to perform the operation corresponding to the data freshness optimization method for UAV information collection based on data compression as described in any one of claims 1-7.

Images