CN109581342B

CN109581342B - Full-period time resource allocation method for rotary phased array radar

Info

Publication number: CN109581342B
Application number: CN201811495674.8A
Authority: CN
Inventors: 李纪三; 耿利祥
Original assignee: 724th Research Institute of CSIC
Current assignee: 724th Research Institute of CSIC
Priority date: 2018-12-07
Filing date: 2018-12-07
Publication date: 2022-12-27
Anticipated expiration: 2038-12-07
Also published as: CN109581342A

Abstract

The invention provides a method for distributing time resources of a rotating phased array radar in a full period, which belongs to the technology of resource planning in the full period. In another aspect, the newly increased tracking time of the current period compared to the previous period determines the scan table of the next circle, which is a delay compensation strategy.

Description

Method for allocating time resources of full period of rotary phased array radar

Technical Field

The invention belongs to the field of sea surveillance radar resource scheduling based on a phased array system.

Background

The large radar resource management and scheduling part is the core of a phased array radar system and is used for determining a scheduling strategy to schedule radar tasks according to a certain optimal criterion so as to effectively distribute radar resources, achieve the aim of optimizing the overall performance of the radar and mainly complete the task generation, task arrangement and beam scheduling of radar events. The radar task residency, once scheduled to execute, cannot be interrupted by other residences, i.e., the scheduling is non-preemptive. Scheduling algorithms based on time budgets and adaptive scheduling algorithms based on time windows and task priorities are common.

1. A time budget scheduling method: the time budget scheduling method designs an upper limit of resource usage for each task in advance, when the resource usage of the task or thread reaches the limit, the request of the task is not responded any more, and then other tasks are executed, and the flow is as follows: at the starting time t1 of the scheduling interval, counting tasks which can be executed at the moment, selecting the task with the highest priority to execute, wherein the residence time of the task is d, the system is idle at the moment t1+ d, counting the tasks which can be executed at the moment, meanwhile, calculating the proportion of time resources consumed by various tasks, selecting the task with the highest priority and the proportion of time resources which do not exceed the preset proportion to execute, if the task which can not be executed at the moment is not executed, then, the time is shifted backwards by 1 unit, counting the tasks at the moment t1+ d +1, and circulating the steps until the scheduling interval is full.

2. The self-adaptive scheduling algorithm based on the time window and the task priority comprises the following steps: huizing proposed the concept of radar task time windows when multi-phased array radar resource simulation was done in 1996. With this time window constraint, the resource allocation can be flexibly arranged when designing the resource scheduler. The concept of the time window is based on a radar tracking working mode, the specific meaning of the time window is that the actual emission time of a radar event can move within an effective range around the expected emission time, and if the radar event is not executed beyond the range of the time window, and even if the radar event is called again, the radar event is abandoned to be scheduled. In this way, many events that are discarded due to time conflicts can be scheduled by scheduling the time window, thereby improving the time utilization.

And scheduling the sector tasks, wherein the tasks are sequenced according to the priority in the scheduling interval, the tasks with high priority are executed first, and the tasks with low priority are abandoned under the condition that time resources are not enough. For the scheduling of the radar of the fixed area array, in a task time window, a task with high priority is executed, and a task with low priority is delayed. The realization process is as follows: determining a scheduling interval according to the time required by the radar closing operation, taking out tasks to be executed in the interval from a scanning list, a tracking list and a delay list, arranging beams according to the expected emission time from high to low according to the priority; recording the gap time interval in the interval when one wave beam is arranged; taking out the next task from the task list, and judging whether the next task can be executed in a certain idle time period under the constraint of a time window; after all tasks which can be in the interval are arranged, a sequence list of the transmitted beams is obtained, then the transmitting time of each beam is reset, the introduction time of the previous beam is used as the execution time of the next beam, and time fragments are squeezed out; the last task ending time in the interval is used as the starting time of the next scheduling interval.

The two algorithms compare:

the time pre-algorithm (TBS) is a task planning and time resource allocation focusing on the macro, while the adaptive scheduling algorithm based on time windows and priorities is biased on the micro-camping scheduling or called beam-scheduling.

The time pre-algorithm (TBS) emphasizes the allocation of time resources among tasks, and can ensure that tasks with low priority have corresponding time to execute. Particularly when the radar works and encounters false target interference, the priority of the target tracking task is higher than that of the searching task, so that the task saturation and overload of the radar can be caused. The time pre-algorithm can guarantee that a certain time is reserved for searching. The method has the disadvantages that when a plurality of high-priority tasks burst in a certain time period are processed, the high-priority tasks are discarded, and because the high-priority tasks in the time period are obviously increased, a plurality of high-priority tasks are proportionally stuck in scheduling.

The adaptive scheduling algorithm based on the time window and the priority is just opposite to the time pre-algorithm, and is characterized in that the high-priority task in a short period can be executed, but the task planning and the time resource allocation on the period are deficient. The latter algorithm is proposed in laboratory simulation, and many scholars have conducted tracking research later, but the algorithm is not suitable for actual use on surveillance radar equipment. Because, in general, the alert search can be performed as a low priority task without the limitation of a time window, when tracking a high priority task in multiple, the frame period of the search can be increased instead of discarding the task.

The disadvantage of sector planning is that the utilization rate of time resources is not high, because when the targets are concentrated in one sector, the time resources of the search of the sector can be greatly compressed, resulting in small distance reduction, and the sectors beside the sector have no targets and are far away. The resources of other sectors cannot be allocated to the sector.

Disclosure of Invention

The invention belongs to a full-period resource planning technology, and aims to solve the problem of low time resource utilization rate in a rotary phased array radar and the problem of deletion of low-priority tasks caused by sudden increase of tasks in a certain direction.

The specific scheme of the invention is as follows: in the task execution of the period, the total time of the tracking task execution of the task of the period is recorded, and when the antenna passes through the north, a scanning table of the period is generated according to the time used for tracking the previous period. In another aspect, the newly increased tracking time of the current period compared to the previous period determines the scan table of the next circle, which is a delay compensation strategy. The effect of the conventional algorithm and the cross-cycle resource scheduling algorithm is shown in fig. 1. In the conventional algorithm, the task "remote search 2" time resource is preempted by two higher priority trace tasks in scheduling interval 1 and is therefore deleted. In the improved algorithm, the full period of idle time can be taken for tracking, so that after two tracking tasks are scheduled, "remote search 2" can still be performed.

Through full verification of an external field test, the full-period resource planning technology provided by the invention is compatible with a full-circle planning technology of sector scheduling, can quickly respond to sector tasks issued by other subsystem systems, and greatly improves the time utilization rate.

Drawings

Fig. 1 shows the effect of the full-period planning compared with the traditional adaptive algorithm.

FIG. 2 is a flow chart of periodic search parameter generation.

FIG. 3 is a flow of task orchestration for a scheduling interval.

Detailed Description

The invention provides a method for allocating time resources of a rotating phased array radar in a full period, which comprises the following concrete steps:

1. the method for planning the cross-period resources of the rotary phased array radar comprises two functions, namely determining the search parameters of the whole period when the period starts (when the normal direction of the area array is over positive north), and arranging the beam execution linked list of each scheduling interval when the interruption of each scheduling interval comes.

2. Determining parameters of a full period, determining a scanning parameter table of the period according to a working mode and other control commands, wherein the parameters generally comprise: azimuth number, elevation number, pulse number, multi-beam number, pulse processing mode, pulse width, pulse repetition period, signal form, signal bandwidth and signal frequency point. Usually, the scan table is designed in advance, then the time allocated to the search in this period is calculated, and the corresponding scan table is selected according to the time, and the specific flow is shown in fig. 2.

3. The beam scheduling of each scheduling interval refers to scheduling of a search task and a tracking task of the scheduling interval. The search tasks are executed from the search chain table in the present cycle according to the swept-out execution in the previous scheduling interval, for example, 1000 search tasks are shared in the scan table in the present cycle, the first scheduling interval sweeps 50 tasks, and the second scheduling interval is executed from the 51 st task. The tracking task refers to a task falling in the scheduling interval, each scheduling interval generally corresponds to a scanning azimuth interval, if the azimuth of the tracked target falls in the interval, the tracking task is executed in the scheduling interval, but the retrace confirmation, high data rate tracking in one circle, and the scheduling interval needs to be recalculated by the advance and the lag of the tracking and scanning. If the target is 70 degrees in azimuth in geodetic coordinates and it is desired to scan the target 30 degrees in advance, the task should be scheduled when the antenna turns around 40 degrees, as shown in fig. 3.

Claims

1. A method for allocating time resources of a rotating phased array radar in a full period is characterized by comprising the following steps: the method comprises the steps of utilizing a watchdog to generate scheduling interruption of 100ms, reading the servo direction of an antenna when the interruption arrives, judging that the servo direction of the current time is smaller than the last servo direction value, indicating that the antenna is over north, when the antenna is over north, according to the total time spent by the tracking type tasks of the previous circle, budgeting the search task time of the period, according to the time budget of the search task and the working mode of a radar, determining the search parameter of each direction of the period, generating a task linked list for search residence and adding the search parameter into the search, when each interruption arrives, searching whether a task in the tracking task linked list has a task with the direction falling in the later 18 degrees of the servo direction value read by the scheduling interval, if yes, adding the tracking task into an execution list of the interval, simultaneously recording the time spent by the tracking task, taking out the task from the head of the search task linked list, adding the task to the back of the tracking task, and finishing the task arrangement of the scheduling interval after the task of 100ms is accumulated.