An adaptive resource scheduling method based on improved ant colony algorithm

Abstract

The present invention relates to an adaptive resource scheduling method based on an improved ant colony algorithm. In view of the current situation that the passive wideband phased array radar system has wide frequency domain and airspace coverage, large resource consumption, and great difficulty in resource scheduling, the present invention is based on combining the characteristics of the ant cycle model and the ant density model in the ant colony algorithm, using the residence time for the frequency domain and airspace in different scheduling strategies as the ant colony route, and using the scheduling time as the pheromone concentration judgment standard to perform adaptive adjustment of the route, and combining the feedback mechanism of multi-level resource scheduling to realize adaptive scheduling of the radar system. The present invention uses a multi-iteration ant colony system as an improvement on the basic ant colony algorithm, and applies it to a passive wideband phased array radar system that lacks prior knowledge, overcoming the shortcoming of low flexibility of the fixed template scheduling algorithm in conventional scheduling, and improving the resource utilization rate of the passive wideband phased array radar system. The invention has good versatility and practicality.

Description

Technical Field

[0001] The invention belongs to the technical field of radars and relates to a resource scheduling method for a broadband phased array system. Background Art

[0002] In modern radar technology, broadband phased array systems have superior performance and powerful detection capabilities. The resource scheduling method of the system has a great impact on the detection results and equipment performance. In the current multifunctional phased array radar resource scheduling technology, the main scheduling design is resource scheduling based on comprehensive priority. In view of the wide frequency domain coverage capability of broadband phased array systems, the scheduling strategy pursues wide airspace coverage and rapid scheduling. The scheduling method of related literature generally uses the inherent task model to perform comprehensive priority calculation and control the scheduling of scheduled tasks by changing the task template to achieve the purpose of adaptive scheduling. The application of multi-level cache in resource scheduling is to partition the task scheduling process and realize the multi-level processing process through partitioning.

[0003] In "Phase-controlled Array Radar Data Processing and Resource Scheduling Simulation", the author gives a resource scheduling method for phased array radars, and "A Comprehensive Design Method for Task Priority" gives a priority design method based on task attributes, and "High Probability of Intercept Time-Frequency Joint Search Scheduling Method" gives a resource scheduling method that uses the frequency domain as the weight to adjust the scheduling strategy. The existing resource scheduling methods for passive phased array radar systems have the following shortcomings: First, the adaptive scheduling process is single, which is determined by the control command received before the task scheduling. This is determined by the detection nature of the active radar and does not have sufficient adaptability in passive detection. It is impossible to feedback and update the actual information of the target based on the scheduling results, resulting in the system running in one direction, lacking a feedback mechanism, and insufficient flexibility in task scheduling. Secondly, in the process of adaptive scheduling, several fixed templates are often used for switching, which is not adaptable enough to the characteristics of the wide frequency domain and large airspace of the passive phased array radar. These two shortcomings lead to the low degree of adaptability and poor real-time performance of the system in the process of resource scheduling. Therefore, the present invention adopts a modified ant colony algorithm for feedback, uses the interception probability in different spatial and frequency domains as pheromones, calculates the pheromone concentration after scheduling, and sorts the paths to change the content of task scheduling and the order of cache areas, so as to achieve real-time adaptability of resource scheduling. Summary of the invention

[0004] The present invention proposes an adaptive resource scheduling method based on an improved ant colony algorithm, which solves the problems that the adaptive degree of the resource scheduling algorithm of the existing passive phased array radar system is not high and the feedback process has more manual intervention.

[0005] The present invention proposes an adaptive resource scheduling method based on an improved ant colony algorithm, which pre-processes received tasks in a broadband passive phased array system, performs ant colony pathfinding in the entire airspace and frequency domain during the first execution, calculates pheromone concentrations using dimensions such as the number of interceptions, interception time, and system resource utilization as parameters during task execution, finds the optimal path in the task execution process based on the pheromone concentration calculation results in combination with the characteristics of the ant cycle model and the ant density model, selects specific parameters and substitutes them into the ant colony algorithm formula, changes the task content in the second-level cache area based on the calculated pheromone concentration table, and re-arranges the tasks in the cache area.

[0006] The present invention is achieved through the following technical solutions:

[0007] Step 1: Establishment of multi-level cache and pheromone concentration table: Multiple cache areas are established to classify the tasks to be processed; the tasks to be processed enter the primary scheduling task area, and the tasks are scheduled according to the fixed template. After the scheduling is completed, the tasks are sent to the sending waiting area; at the same time, the initial pheromone concentration table is established. When the initial values ​​are the same, the scheduling with equal priority is adopted;

[0008] Step 2: Pheromone calculation for sending tasks: Calculate the pheromone concentration based on the system resource utilization, the number of interceptions in the target frequency domain and airspace, and the interception time as parameters; use the ant cycle model to calculate the pheromone concentration, and calculate the pheromone increment through the frequency domain and airspace numerical values; and use the ant density model to update the pheromone concentration table immediately after each single task is executed;

[0009] Step 3: Ant colony feedback adjusts the sending buffer: Considering that the pheromone concentration is inversely proportional to the path length, tasks with high pheromone concentrations have high detection and interception probabilities, and thus have higher task priorities. The content and order of tasks to be sent are adjusted through the updated pheromone concentration table to complete adaptive resource scheduling.

[0010] Furthermore, based on the target interception probability as the basis for selecting the shortest path, in the ant colony algorithm formula, the information heuristic factor is selected as 1, the expectation heuristic factor is selected as 4, the number of ants is 100, the information volatility factor is 0.8, and the residual factor is 0.2. ij (t+n)=(1-ρ)·τ ij (t)+△τ ij (t), After one cycle, the pheromone release with the shortest path is

[0011] The present invention utilizes the multi-result feedback feature of the ant colony system algorithm and the strong real-time and flexible adjustment characteristics of multi-level cache and multi-level processing to achieve real-time adaptive processing of external new instructions or lower-level task execution result analysis, and has good versatility and practicality. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 This is a schematic diagram of the present invention. DETAILED DESCRIPTION

[0013] The technical solution of the present invention is further explained below in conjunction with the accompanying drawings.

[0014] The adaptive resource scheduling method based on the improved ant colony algorithm proposed in the present invention sets a hierarchical cache area for task storage and processing scheduling. After each task is executed, the pheromone concentration under the path is calculated and a pheromone concentration table is formed. After the full-frequency and airspace tasks are completed, the tasks are rescheduled by analyzing the pheromone concentration table, and the optimal path for task scheduling is obtained through ant colony feedback to achieve adaptive scheduling of tasks.

[0015] The implementation flow chart of the present invention is as follows Figure 1 As shown, the specific implementation steps are as follows:

[0016] (1) Multi-level cache establishment and task preprocessing scheduling

[0017] In the resource scheduling software, firstly, a multi-level cache area is established which is independent of each other. Each cache area is recursive, and the content of the previous cache area determines the next cache arrangement. In the passive broadband phased array radar system, two cache areas are established: the first arrangement task area and the sending waiting area. For the specific airspace and frequency domain tasks that are executed for the first time, the task arrangement of the fixed module is performed. The arranged tasks are sent to the sending waiting area and sent in sequence.

[0018] (2) Using ant colony algorithm to calculate pheromones

[0019] The principle of the ant colony algorithm is: 1. Ants release pheromones on the path; 2. Randomly select a path at the intersection and release pheromones related to the path length when passing through the path; 3. The pheromone concentration is inversely proportional to the path length. When the ants encounter the intersection later, they choose the path with high pheromone concentration; 4. The pheromone concentration of the optimal path increases, and the advantage accumulates; 5. Finally, the optimal path is found. It can be seen that at time t+n, the amount of pheromone on the path (i, j) can be calculated according to the formula τ ij (t+n)=(1-ρ)·τ ij (t)+△τ ij (t) is adjusted, where There are two models for updating pheromones, namely the ant-period model and the ant-density model. Combining the advantages of the two models, the ant-period model is used for pheromone increment, in which the increment is only related to the search route and has nothing to do with the specific path. The ant-density model is used for updating the pheromone on the path after each step of movement, which becomes an improved ant colony algorithm. In terms of the selection of main parameters, the larger the information heuristic factor α, the greater the probability that the ant will choose the path it has walked before, and the smaller the randomness of the search path, the easier it is to fall into the local optimal solution. Therefore, in the process of wide-band radar search, in order to achieve a high intercept probability under real-time performance, α is set to 1. The expected heuristic factor β determines the convergence speed of the algorithm. Also considering obtaining the local optimal solution as soon as possible, β is set to 4. The more ant colonies m, the more accurate the optimal solution can be obtained, but it will greatly consume system resources and increase time complexity, so the value is set to 100. If the information volatility factor ρ is too small, there will be too many residual pheromones on the path, and the number of invalid paths searched will increase, affecting the convergence speed of the algorithm, so ρ is set to 0.8. The residual factor 1-ρ can thus be determined to be 0.2. Set the initial information amount τ ij (0) = τ max , combined with the basic formula of the ant colony system (ACS), after one cycle, the pheromone release that finds the shortest path is where τ ij (t) is limited to [τ min ,τ max ]between.

[0020] (3) Formation of pheromone concentration table and adaptive scheduling of tasks

[0021] According to the pheromone concentrations under each path formed after one cycle, the paths are arranged from high to low according to the concentration, and a ranking table of paths from good to bad can be obtained. According to this table, the tasks in the sending waiting area are rescheduled to increase the residence time of the tasks under the path with high pheromone concentration to increase the interception probability and realize a virtuous cycle of positive feedback. And the key frequency domain airspace with high pheromone concentration is prioritized to realize the priority sorting and scheduling of the secondary tasks. Each subsequent task scheduling will update the pheromone concentration table, and the tasks will be rescheduled according to the latest concentration table to achieve priority adaptive scheduling of wide frequency domain and large airspace. And as the number of scheduling times accumulates, the pheromone concentration table updates are reduced, the optimal path is determined, and the resource utilization rate and task scheduling success rate of the radar system are improved.

Claims

1. An adaptive resource scheduling method based on improved ant colony algorithm, characterized in that: Step 1: Establishment of multi-level cache and pheromone concentration table: Multiple cache areas are established to classify the tasks to be processed; the tasks to be processed enter the primary scheduling task area, and the tasks are scheduled according to the fixed template. After the scheduling is completed, the tasks are sent to the sending waiting area; at the same time, the initial pheromone concentration table is established. When the initial values ​​are the same, the scheduling with equal priority is adopted; Step 2: Calculate the pheromone concentration based on the shortest path selection based on the target interception probability: In the ant colony algorithm formula, the information heuristic factor is selected as 1, the expected heuristic factor is selected as 4, the number of ants is 100, the information volatility factor is 0.8, and the residual factor is 0.2; in the information quantity calculation formula τ ij (t+n)=(1-ρ)·τ ij (t)+Δτ ij (t), After one cycle, the pheromone release with the shortest path is When calculating the pheromone concentration, the ant-period model is used to calculate the pheromone increment through the frequency domain and spatial domain numerical calculations; at the same time, the ant-density model is used to update the pheromone concentration table immediately after each single task is executed; Step 3: Ant colony feedback adjusts the sending buffer area: Considering that the pheromone concentration is inversely proportional to the path length, the task with a high pheromone concentration has a high detection and interception probability, and the task priority is also higher; The content and order of tasks to be sent are adjusted by updating the pheromone concentration table to complete adaptive resource scheduling.

Images