Real-time adaptive task scheduling method for multi-channel interference system

Abstract

The invention belongs to the field of electronic system resource management, and particularly relates to a real-time adaptive task scheduling method for a multi-channel interference system. The interference system provided by the invention has a plurality of channels, the interference frequency band of each channel can be flexibly controlled, and the system can implement simultaneous interference by the same interference beam for multiple tasks distributed to the same channel, or adopts a digital beam forming (DBF) technology to generate a plurality of interference beams at the same time. Or a simultaneous multi-interference beam is generated by adopting an aperture subarray division mode. The algorithm provided by the invention can enable the system to achieve the purpose of maximizing the execution benefit of the system by effectively controlling the multi-channel interference frequency band, the tasks allocated and executed on each channel and the specific execution mode of each task.

Description

Technical Field

[0001] The present invention belongs to the field of electronic system resource management, and particularly relates to a real-time multi-channel interference system adaptive task scheduling method. Background Art

[0002] The battlefield environment is becoming increasingly complex, and resources such as time, space, frequency, energy, and aperture of the interference system are limited. Especially in the case of intensive interference tasks, the pressure of insufficient resources faced by the system is more serious. At the same time, the confrontation between the enemy and us is uncertain and time-varying, which makes the prior decision unable to cope with the real-time battlefield situation. Therefore, the allocation strategy of the limited resources of our system needs to be adaptively changed according to the number, importance, and urgency of the actual tasks to be executed. When facing multiple interference tasks and the interference resources of the system are subject to various constraints, how to form an adaptive task scheduling strategy for our interference system is particularly crucial.

[0003] The scheduling method of the multi-task system is closely related to the system regime. For a single-beam system, multiple tasks to be executed are carried out in a time-division multiplexing manner, and only the time resource constraint of the system is considered during the scheduling process. There are several relatively common optimization and solution methods, such as: the task scheduling algorithm based on the heuristic method determines which task occupies the system resources according to the priority level of the task, and algorithms such as the particle swarm algorithm based on entropy theory, the hybrid particle swarm and genetic algorithm, and the reinforcement learning method based on transfer learning use intelligent methods to determine the execution timing of multiple tasks. However, due to the high computational complexity, the biggest weakness of the scheduling algorithm based on intelligent methods is that it is difficult to meet the real-time requirement of calculation.

[0004] As the number of tasks gradually increases and modern electronic technology develops step by step, the jammer system becomes gradually complex and diversified. The single-beam system begins to introduce aperture division technology and digital beamforming technology and develops into a multi-beam system. For the multi-beam system, the literature (Qi Wenchao, Yang Ruijuan, Li Xiaobo, Chen Xinyong, Cheng Wei. Research on the task scheduling algorithm of multi-functional integrated radar [J]. Radar Science and Technology, 2012, 10(02): 150-155.) presents a source management method for the multi-functional integrated radar system based on aperture segmentation, which uniformly allocates time and aperture resources. The literature (Guo Xiaoyi, Yuan Weiwei, Huang Jincai. A method for one-to-many allocation of radar jamming resources [J]. Fire Control & Command Control, 2008, 33(12): 22-25+29.) simultaneously considers time resource constraints, frequency resource constraints, and airspace resource constraints, and judges the radar set that can be simultaneously jammed by one jammer from three dimensions of time domain, airspace, and frequency domain, so as to group the target radars, that is, task integration. The literature (Siyu H, Ting C, Zishu H, et al. Adaptive dwell scheduling for simultaneous multi-beam radar system based on array element selection with different polarization characteristics [J]. Digital Signal Processing, 2023, 140.) refines the task scheduling problem of the multi-beam system based on a shared aperture into the task scheduling problem of the multi-beam system based on selecting array elements with different polarization characteristics. Its goal is to achieve the maximum scheduling gain of the system under the constraint conditions of the system's time, array surface, frequency, and electromagnetic compatibility.

[0005] The above research has achieved certain results in the multi-task scheduling of the jamming system, but only considered a single way of multi-task merging or aperture division during the simultaneous execution of multiple tasks, without comprehensively considering various ways. Based on the above problems, the present invention proposes a real-time adaptive task scheduling method for a jamming system with multiple channels. The jamming system in the present invention has multiple channels, and the jamming frequency band of each channel can be flexibly controlled. For multiple tasks allocated to the same channel, according to the spatial position relationship of the jammed tasks, they can be simultaneously jammed by the same jamming beam, or digital beamforming (DBF) technology can be used to generate multiple jamming beams simultaneously, or the aperture sub-array division method can be used to generate multiple jamming beams simultaneously. By effectively controlling the multi-channel jamming frequency band, the tasks allocated and executed on each channel, and the specific execution methods of each task, this system can achieve the goal of maximizing the system execution benefit. Summary of the Invention

[0006] The present invention provides a real-time multi-channel interference system adaptive task scheduling method.

[0007] The task model is T i,k ={w i,k , Ts i , Te i , P i , G i , L i , K i , R i,k , v i,k , θ i,k , α i,k , f i , B i}, where the task model is considered as a radar jamming task. Among them, w i,k is the threat level of task i at time k. The higher the threat level of the task at time k, that is, the more urgent the requirement for the task to be scheduled at time k. Ts i and Te i are the start and end times of the task interference request time respectively. P i , G i and L i are the transmit power, antenna gain and combined loss of the target corresponding to task i respectively. K i is the suppression coefficient of the target corresponding to the task. R i,k , θ i,k and α i,k reflect the position relationship between the target corresponding to task i and the jammer system at time k, including distance, azimuth angle and elevation angle information. v i,k represents the moving speed of the target corresponding to task i, usually represented by . f i and B i reflect the desired operating frequency and bandwidth of task i, that is, the operating frequency of the target.

[0008] Assume that the current analysis interval is [t0, t0 + Δt], where t0 is the start time of the current analysis interval and t is the duration of one analysis. According to the interference request time of the task, select the set of tasks to be executed at this analysis time, and represent this set as T appear . The number of multi-channels that the system can be divided into is M cha , and initialize the current number of idle channels as M fre = M cha。And set the task execution queue Ta to be empty, the task deletion queue Td to be empty, the final channel - frequency band range set Fre to be empty, the final channel - task combination matrix B to be empty, the final channel - task - aperture resource quantity matrix Aperture to be empty, and the final task execution mode set T on all channels classify to be empty, and initialize the analysis total count variable μ of all channels in the system to 0. Then a radar adaptive scheduling method for multi - interference tasks and high - dimensional system interference resource constraints includes the following steps:

[0009] Step 1:

[0010] Judge whether the set of tasks T to be executed in the current analysis interval appear is non - empty and whether there are idle channels, that is, n task > 0 and M fre > 0, where n task represents the number of elements in T appear . If it is satisfied, go to Step 2, otherwise go to Step 8.

[0011] Step 2:

[0012] Define K as the bandwidth of the interference frequency band of each channel, and its value is equal on each channel. The starting point s of each frequency band j can be freely set, and the frequency band [s j , s j + K] is regarded as a channel. Considering that the expected execution frequency band of task i is [f i - B i / 2, f i + B i / 2], if then it is considered that task i can be executed on channel j. Take the maximum and minimum values of the starting points of the expected execution frequency bands of all tasks and record them as [f min , f max , divide it into M = [(f max - f min ) / K] + 1 non - overlapping frequency bands, and remove the empty frequency bands in the divided frequency bands where there are no tasks to be executed. The range of each frequency band is denoted as [s j , s j + K], and the obtained division result is the independent frequency band range information Fre0 required by the task set, that is Record the number of frequency bands after removing the empty frequency bands as M total . Initialize the channel - task combination matrix B0 to an n task × M total zero matrix, where each column represents the frequency band [s j , s jThe set of tasks executed on +K. That is, for the set of tasks that satisfy let B0 i,j = 1, so that the task number information applied to be executed in this frequency band is stored in each column of B0. Update μ = μ + 1. Judge the relationship between the number M total of required frequency bands in Fre0 and the number M fre of idle channels. If M total > M fre , then go to step 3, otherwise go to step 4.

[0013] Step 3:

[0014] Define the set of frequency band numbers corresponding to the M total frequency bands selected from the M fre frequency bands as index cha , record the number M = M fre of the selected frequency bands to be analyzed. Set index cha , the total number N target of interference targets in the channel combination corresponding to this group of serial numbers, and the maximum total number N unique of non - overlapping interference tasks to 0. Calculate the number of tasks in each column of B0. For all possible channel combinations, if the currently analyzed frequency band number is the combination of index cha ', and its total number N target ' of channel interference targets and the maximum number N unique ' of non - overlapping interference targets in this group of channels satisfy: N target '> N target ; or satisfy: N target ' = N target , N unique '> N unique , then update index cha = index cha ', N unique = N unique ', N target = N target '. After completing the analysis of combinations, update the channel - task combination matrix channel - frequency band range set and enter step 5.

[0015] Step 4:

[0016] Select all the required frequency bands in Fre0 as the frequency band combination to be executed, update B' = B0, record the number M = M total of the selected frequency bands to be analyzed, update M fre = M fre - Mtotal , proceed to Step 5.

[0017] Step 5:

[0018] Set the channel analysis count variable to m = 1. Determine whether m ≤ M holds. If it holds, assume that the set of tasks to be executed in the m-th channel is Arbitrarily select one of the tasks, such as T i as the reference task set Task refer , and set all the remaining tasks except the reference task as the task queue to be analyzed Analyze Task refer and each task in the task queue to be analyzed to check whether the azimuth angle and elevation angle are close. If they are, it is considered that Task refer and this task (assumed to be T j ) can be combined and executed using the same beam. At this time, add T j to the reference task set, i.e., Task refer = {T i , T j}, and remove T j from the task queue to be analyzed . When there are multiple tasks in the reference task queue, the task to be analyzed needs to be close in both phase angle and elevation angle to all the tasks in the reference task set in order to be considered combinable with this set of reference tasks for execution. After analyzing each task in , record the current reference task set as a beam result Select any task in as a new reference task for analysis, and repeat the above steps until Record all the beam result sets at this time, in the specific form as m . If the total number of beams n_beam m in this channel is greater than the total number of beams that can be transmitted n_beam max , record the beam numbers corresponding to the top n_beam m items with the highest values in the threat degree sum array w_beam max as index_w_beam m , and update the task beam result in the m-th channel to Update m = m + 1. If m ≤ M, return to Step 5 and repeat; if m > M, output T beam , proceed to Step 6.

[0019] Step 6:

[0020] Analyze the beam result sets on each channel in sequence. Take the results of the m-th channel as an example for analysis. Set the set at this time, and assume there are n beams in it. Analyze whether there are multiple beams in the n beam beams that can be jointly emitted in the form of DBF. For example, beam first, it is necessary to analyze whether beams {T1, T2, T3} and {T4, T5}, etc. can be jointly emitted in the form of DBF, that is, it is necessary to analyze whether the following formula is satisfied between the beams:

[0021]

[0022] where is the maximum elevation angle corresponding to the task set executed by the multi-beam, is the minimum elevation angle corresponding to the task set executed by the multi-beam, calculated according to the elevation angle information of the task beam. Calculate the result of the multi-task beam executed in the form of DBF in . At this time, update to as shown in the following formula:

[0023]

[0024] Update

[0025] Group each set of beams executed in the form of DBF into a separate beam set, and each individually executed beam is also grouped into a separate beam set. Then, at most n_beam max sets can be divided. Divide the channel array surface according to the number of beam sets on this channel, and calculate the array surface resources occupied by each beam set. In the present invention, the maximum number of sub-array divisions is set to 4. Assume the number of beam categories on this channel is n classify , and according to n classify ∈{1, 2, 3, 4}, the specific array surface division method is as Figure 1 shown.

[0027] When n classify ≠3, update the array surface resources occupied by the task T corresponding to the beam in i to Update Aperture′.

[0028] When n classify =3, that is, the beams in are divided into 3 sets. There are two possibilities for the normalized array surface resource amount occupied by the task, 0.25 and 0.5. For the purpose of maximizing resource utilization, The beam set with the largest number of medium beams is denoted as Assume T i belongs to Then T i The amount of array surface resources occupied is 0.5. And for the task T that does not belong to , the amount of array surface resources it occupies is 0.25, and update Aperture = Aperture′. j For

[0029] Finally, assume the task T i belongs to the k-th beam set, and there are n_beam k beams in this set, then update the task T i The effective interference power that is expected to be obtained when successfully executed on this channel is Go to step 7.

[0030] Step 7:

[0031] Given that the set of channel - operating frequency band ranges at this time is Fre′, the channel - task combination matrix is B′, and the channel - task - array surface resource amount matrix is Aperture′. If the number of channel analysis times μ > 1 at this time, then when calculating the total task execution benefit, it is necessary to combine the resource allocation results of the previous μ - 1 analysis stages, that is, update Fre′ = Fre′ ∪ Fre, B′ = B′ ∪ B, and T classify ' = T classify ∪ T classify '.

[0032] Calculate the sum of the execution benefits that each task is expected to obtain on all channels according to the following formula

[0033]

[0034] where is the comprehensive benefit of the interference power and interference frequency band obtained by the task T i on channel m. (P i G i ) des is the expected interference power of the task, is its comprehensive loss, K iΣ is the radar suppression coefficient, and σ i is the radar cross - sectional area. indicates the ratio of the actual interference power to the expected interference power of the task. And if the interference power emitted by the jammer to the specified target reaches the target interference suppression effect, the excess system resources will not be able to continue to enhance the interference effect. is the coverage rate of the frequency band range of channel m to the expected execution frequency band range of the task T i .

[0035] If task T i satisfies ∑ m eff(i,m) > 0, then add T i to the execution list Ta′, and set Otherwise, Σ m eff(i,m) = 0, and task T i fails to execute. Add such tasks to the deletion list Td′.

[0036] Update Ta = Ta′, Td = Td′, Fre = Fre′, B = B′, Aperture = Aperture′, T classify = T classify ′. Proceed to step 1.

[0037] Step 8:

[0038] The current scheduling analysis process ends. Output the task execution queue Ta, the task deletion queue Td, the channel - frequency band range set Fre, the channel - task combination matrix B, the channel - task - array surface resource quantity matrix Aperture, and the set T of the final task execution methods on all channels classify .

[0039] Principle of the Invention

[0040] The present invention provides a real - time multi - channel interference system adaptive task scheduling method. This method is aimed at a multi - channel interference system and can comprehensively consider the characteristics of simultaneously executing multiple interference tasks in the ways of beam combination, aperture division, and DBF, and realizes multi - task adaptive scheduling under the system resource constraint conditions. The principle is elaborated below.

[0041] Suppose there are M independent channels in the multi - channel interference system, the execution frequency band width of each channel is K, and the starting working frequency of this frequency band can be flexibly configured. Each channel corresponds to an aperture, and each aperture can generate at most B interference beams. Suppose the duration of the confrontation between the friendly and the enemy is L, and it is discretized into multiple analysis intervals with Δt as the time step for scheduling analysis of tasks. Suppose at the k - th analysis moment, there are a total of N tasks applying for scheduling execution. Here, introduce u i,k = {0, 1} to represent whether task i is scheduled to execute at the k - th moment. If u i,k = 1, then task i is executed; otherwise, u i,k = 0, and task i is not executed. Then, the execution benefit of the system at the k - th moment can be expressed as:

[0042]

[0043] It can be seen that the interference task execution benefit at the k - th moment comprehensively considers the number of executed tasks and the benefits of each executed task.

[0044] The scheduling process has the following constraints:

[0045] First, for the interfering tasks in the scheduling execution, the selected interference time must be within its interference effective time window, that is:

[0046]

[0047] Second, the covered bandwidth of the channel resources occupied by multiple interfering tasks executed at the same time does not exceed the upper limit of the covered bandwidth of the system channel resources:

[0048]

[0049] Among them, MK represents the upper limit of the covered bandwidth of the system channel resources, which is a controllable parameter, and F i,k represents the execution frequency range of task T at time k i , where f i,k and B i,k are the main execution frequency and bandwidth of task T respectively i . The combined execution frequency ranges of all tasks in the T set can be further simplified into several continuous execution frequency bands [s j , e j , j = 1, 2,..., J, where J represents the number of continuous frequency bands in the set. The sum of the J continuous frequency bands can describe the total bandwidth amount covered by multiple tasks executed simultaneously at time k, which is the total frequency band resource demand

[0050] Third, the total number of beams emitted on each channel does not exceed the upper limit of the number of schedulable beams on that channel, that is:

[0051]

[0052] B k (:, m) is the m-th column vector in matrix B k , which describes the task occupancy of channel m, and B k (i, m) = {1, 0}, when it represents that task i occupies channel m at time k; B is the upper limit of the total number of scheduling task beams in channel m, which is a controllable parameter

[0053] Fourth, the execution mode of the task beam (whether to merge and execute with other beams, whether to apply DBF technology to execute) is determined by its positional relationship with the remaining tasks. That is:

[0054]

[0055] η i,k (m) represents the amount of array resources allocated to task i by the system on channel m, and the maximum amount of array resources is expressed as 1; Bk (i, m) represents whether task i occupies channel m. For task beams i and j that are executed simultaneously in the same channel, the threshold θ is used simultaneously th and α th are used to represent the positional relationship between the azimuth and elevation angles of tasks i and j respectively. If the difference between their azimuth or elevation angles is less than the threshold, the azimuth or elevation angles of the two tasks are defined as close; otherwise, the azimuth or elevation angles of the two tasks are defined as far apart. According to this threshold, it can be judged whether several tasks can be combined for execution, or whether several beams can be emitted in the form of DBF. In addition, this constraint also introduces the matrix sη i,k (m) to represent whether the DBF technology is applied when the system emits a beam to irradiate target i in channel m at time k; according to the above three situations, reflecting situation (1), two task beams are combined into one beam, reflecting situation (2), indicating that task i beam and task j beam are formed by the DBF technology, reflecting situation (3), task i beam and task j beam are each emitted separately by the divided sub-arrays.

[0056] Fifth, the transmission power of a task shall not exceed the upper limit of its execution power, that is:

[0057]

[0058] P total (m) represents the upper limit of the transmission power of channel m. Among them, if the DBF technology is applied when task i is executed in channel m, the upper limit of the execution power of task i is [η i,k (m)] 2 ·P total (m) / (n + 1), where n is the total number of other tasks that are transmitted in the form of DBF together with task i; the second inequality represents that when task i is executed alone on the entire array or sub-array in channel m, the upper limit of the execution power of the task is [η i,k (m)] 2 ·P total (m).

[0059] Therefore, the design of the scheduling optimization model of the present invention can be expressed as the following formula:

[0060]

[0061] It can be seen that the objective function of this optimization problem model is to increase the system execution benefit obtained at time k as much as possible. The controllable parameters include the tasks executed at time k, the array resources occupied by the executed tasks, and the array multiplexing method of multiple tasks in the same channel The channel resources occupied by the execution of tasks. The constraints comprehensively consider the time resource constraints, array surface resource constraints, channel resource constraints, and airspace resource constraints for scheduling and executing interfering tasks. To maximize the objective function, the number of tasks to be executed should be maximized as much as possible and tasks with higher execution benefits should be preferentially selected for execution, that is: select the channels with a larger number of tasks for execution, as shown in step 3; try to combine the beams of tasks with similar azimuth and elevation angles for execution, as shown in step 5, or emit the beams with similar elevation angles in the form of DBF to execute more tasks with the same amount of array surface resources, as shown in step 6; when the total number of beams exceeds the schedulable upper limit, preferentially select the beams with a higher total threat level, as shown in step 5. Brief Description of the Drawings

[0062] Figure 1 It is a display diagram of the division results of dividing the entire array surface into different numbers of sub-arrays;

[0063] Figure 2 It is a comparison diagram of three task scheduling algorithms TEP;

[0064] Figure 3 It is a display diagram of ADRT of the algorithm proposed in the present invention;

[0065] Figure 4 It is a display diagram of TCRR of the algorithm proposed in the present invention;

[0066] Figure 5 It is an explanatory description of the array surface division results of channels 1 to 4 of the algorithm proposed in the present invention; Figure 6 It is an explanatory description of the beam scheduling results of the algorithm proposed in the present invention; Detailed Embodiment

[0067] Consider analyzing at 1s intervals within the duration of 1 - 50s in the simulation, considering several interfering tasks with random occurrence times to simulate different numbers of tasks at different analysis times. A total of 10 radar targets are set in the scenario, and the 10 radar targets move with time, making the targets to be interfered with random at each analysis time. The specific parameters are shown in Table 2.

[0068]

[0069]

[0070] To comprehensively evaluate the performance of the algorithm, the Task Execution Profit (TEP), the Analysis Duration Running Time (ADRT), and the Task Conflict Resolution Rate (TCRR) are used to evaluate the algorithm performance, and their specific definitions are as follows:

[0071] Task Execution Profit (TEP): It is defined as the system execution profit at the k-th analysis moment. The specific meaning has been given in the previous text and can be briefly expressed as:

[0072]

[0073] Analysis Duration Running Time (ADRT): It is defined as the program running time required to complete the k-th analysis moment and can be expressed as:

[0074] ADRT k = t k,2 - t k,1

[0075] where, t k,1 and t k,2 represent the start time and end time of the i-th scheduling interval respectively

[0076] Task Conflict Resolution Rate (TCRR): It is defined as the ratio of the number of tasks executed at the k-th analysis moment to the total number of tasks applied for scheduling. It can be expressed as:

[0077] TCRR k = N k,exe / N k

[0078] where, N k,exe represents the total number of tasks executed at the k-th analysis moment, and N k represents the total number of tasks applied for scheduling at the k-th analysis moment.

[0079] To better demonstrate the algorithm proposed in the present invention (the heuristic algorithm with the highest profit), two comparison algorithms: the beam combining priority algorithm and the threat degree priority algorithm are used for comparison and explanation here. Figure 2 The execution profits obtained by the three algorithms at each analysis moment are given. It can be seen that the proposed method can obtain the highest system execution profit, and the system execution profit of the beam combining priority method is the lowest. Figure 3 The conflict resolution rate of the algorithm proposed in the present invention is shown. It can be seen that the proposed method can still maintain a high conflict resolution rate. Figure 4The running time of the algorithm proposed in this project is given. It can be seen that the algorithm proposed in this paper has a great advantage in running time, which is sufficient to meet the requirements of real-time performance. To sum up, the algorithm proposed in this paper has the best comprehensive performance compared with the other comparative algorithms.

[0080] Figures 5 to 6 Taking the 6th analysis moment in the scenario as an example, an illustration of the scheduling result of this algorithm is given. At this moment, a total of ten task applications are to be executed, and their numbers are 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, and their positions and the position of the local platform are all Figure 5 shown in. Figure 5 (a - d) The four rectangles represent four available channels. Each channel shows its sub-array division with a black solid line, and triangles of different colors are used to represent the beams facing different targets. The beams emitted from the same sub-array are formed through the DBF technology, and the task numbers corresponding to the executed beams are explained in words. The emission positions of all beams are correspondingly shown in Figure 5 the xoy plane projection diagram of. Taking channel three as an example, this channel is divided into two half-array surfaces. One half-array surface executes beam 9 and beam 10 in the form of DBF. Among them, beam 9 irradiates task 3, and beam 10 irradiates task 1; on the other half-array surface, beam 11 is executed, and beam 11 irradiates task 9 and task 4 at the same time.

Claims

1. A real-time multi-channel interference system adaptive task scheduling method, and the specific technical solution is as follows: The task model is T i,k ={w i,k , Ts i , Te i , P i , G i , L i , K i , R i,k , v i,k , θ i,k , α i,k , f i , B i},where the task model is considered as a radar jamming task. Among them, w i,k is the threat level of task i at time k. The higher the threat level of the task at time k, that is, the more urgent the requirement for the task to be scheduled at time k, Ts i and Te i are the start and end times of the task interference request time respectively, P i , G i and L i are the transmit power, antenna gain and combined loss of the target corresponding to task i respectively, K i is the suppression coefficient of the target corresponding to the task, R i,k , θ i,k and α i,k reflect the positional relationship between the target corresponding to task i and the jammer system at time k, including distance, azimuth angle and elevation angle information, v i,k represents the moving speed of the target corresponding to task i, usually represented by f i and B i reflect the desired operating frequency and bandwidth of task i, that is, the operating frequency of the target. Suppose the current analysis interval is [t0, t0 + Δt], where t0 is the start time of the current analysis interval and t is the duration of one analysis. According to the interference request time of the tasks, select the set of tasks to be executed at this analysis time, and denote this set as T appear The number of multi-channels that the system can be divided into is M cha , and initialize the current number of idle channels to M fre = M cha . And set the task execution queue Ta to be empty, the task deletion queue Td to be empty, the final channel-frequency band range set Fre to be empty, the final channel-task combination matrix B to be empty, the final channel-task-aperture resource matrix Aperture to be empty, and the set T classify of the final task execution methods on all channels to be empty. Initialize the total number of analysis times counting variable μ of all channels in the system to 0. Then a radar adaptive scheduling method for multi-interference tasks and high-dimensional system interference resource constraints includes the following steps: Step 1: Determine the set of tasks T to be executed in the current analysis interval appear Whether it is non-empty and whether there is an idle channel, that is, n task > 0 and M fre > 0, where n task represents the number of elements in T appear If satisfied, go to step 2; otherwise, go to step 8. Step 2: Define \(K\) as the bandwidth of the interference frequency band for each channel, which has the same value on each channel. The starting point \(s\) of each frequency band j can be freely set. The frequency band \([s j , s j + K]\) is regarded as a channel. Considering that the expected execution frequency band of task \(i\) is \([f i - B i / 2, f i + B i / 2]\), if then it is considered that task \(i\) can be executed on channel \(j\). Take the maximum and minimum values of the starting points of the expected execution frequency bands of all tasks and denote them as \([f min , f max \). Divide it into \(M = [(f max - f min ) / K]+1\) non - overlapping frequency bands, and remove the empty frequency bands where there are no tasks to be executed in the divided frequency bands. The range of each frequency band is denoted as \([s j , s j + K]\). The obtained division result is the independent frequency band range information Fre0 required by the task set, that is Denote the number of frequency bands after removing the empty frequency bands as \(M total Initialize the channel - task combination matrix \(B0\) as an \(n task ×M total zero matrix. Among them, each column represents the task set expected to be executed on the frequency band \([s j , s j + K]\). That is, for the task set that satisfies , let \(B0 i,j = 1\), so that the task number information applied to be executed on this frequency band is stored in each column of \(B0\). Update \(\mu=\mu + 1\). Judge the relationship between the number \(M total of the required frequency bands in Fre0 and the number \(M fre of the idle channels. If \(M total >M fre \), then go to step 3, otherwise go to step 4. Step 3: Defined in M total The set of frequency band numbers corresponding to the selected M fre frequency bands among the frequency bands is index cha , record the number of frequency bands M to be analyzed selected as M = M fre . Set index cha , the total number of interference targets N in the channel combination corresponding to this group of serial numbers target , the total number of non - overlapping maximum interference tasks N unique to 0. Calculate the number of tasks in each column of B0. For all possible channel combinations, if the frequency band number currently analyzed is the combination of index cha ', and the total number of channel interference targets N target ' and the number of non - overlapping maximum interference targets N unique ' in this group of channels satisfy: N target ' > N target ; or satisfy: N target ' = N target , N unique ' > N unique , then update index cha = index cha ', N unique = N unique ', N target = N target '. After completing the analysis of combinations, update the channel - task combination matrix channel - frequency band range set Enter step 5. Step 4: Select all the required frequency bands in Fre0 as the frequency band combinations to be executed, update B′ = B0, record the number M of the frequency bands to be analyzed selected = M total , update M fre = M fre - M total , and proceed to step 5. Step 5: Set the channel analysis count variable to m = 1. Determine whether m ≤ M holds. If it holds, assume that the set of tasks to be executed in the m-th channel is Arbitrarily select one of the tasks, such as T i as the reference task set Task refer , and set all the remaining tasks except the reference task as the task queue to be analyzed Analyze Task refer and each task in the task queue to be analyzed to determine whether the azimuth angle and elevation angle are close. If they are, it is considered that Task refer and this task (assumed to be T j ) can be combined and executed using the same beam. At this time, add T j to the reference task set, that is, Task refer = {T i , T j}, and remove T j from the task queue to be analyzed . When there are multiple tasks in the reference task queue, the task to be analyzed needs to be close in phase angle and elevation angle to all tasks in the reference task to be considered combinable with this group of reference tasks for execution. After analyzing each task in , record the current reference task set as a beam result Arbitrarily select one task in as the new round of reference task for analysis, and repeat the above steps until record all the beam result sets at this time, and its specific form is as m shown, and update and calculate the threat degree sum array w_beam m corresponding to each beam in this channel max . If the total number of beams n_beam m in this channel at this time is greater than the total number of beams that can be transmitted n_beam max in the channel, record the beam numbers corresponding to the top n_beam m items with the highest values in the threat degree sum array w_beam Update If m ≤ M, return to step 5 and repeat; if m > M, output T beam , and enter step 6. Step 6: Analyze the beam result sets on each channel in sequence. Take the result of the m-th channel as an example for analysis. Set the set to have n beam beams at this time. Analyze whether there are multiple beams among the n beam beams that can be jointly emitted in the form of DBF. For example, first, it is necessary to analyze whether beams {T1, T2, T3} and {T4, T5}, etc. can be jointly emitted in the form of DBF, that is, it is necessary to analyze whether the following formula is satisfied between the beams: Among them, is the corresponding maximum pitch angle in the task set executed by the multi-beam, is the corresponding minimum pitch angle in the task set executed by the multi-beam, calculated according to the pitch angle information of the task beam the result of the multi-task beam executed in the form of DBF in is updated to as shown in the following formula: Update Each group of beams executed in DBF format is classified as a separate beam set, and each individually executed beam is also classified as a separate beam set. At this time, a maximum of n_beam max sets can be divided. The channel array surface is divided according to the number of beam sets on this channel, and the array surface resources occupied by each beam set are calculated. In the present invention, the maximum number of sub-array divisions is set to 4. Assume that the number of beam categories on this channel is n classify , according to n classify ∈{1,2,3,4}. When n classify ≠ 3, update the array resources occupied by the task T corresponding to the middle beam i as Update Aperture′. When n classify = 3, that is the middle beam is divided into 3 sets. There are two possibilities for the normalized array surface resources occupied by the task, which are 0.25 and 0.

5. For the purpose of maximizing the resource utilization rate, one beam set with the largest number of middle beams is denoted as Assume T i belongs to Then T i occupies 0.5 of the array surface resources. And for the task T that does not belong to j , the array surface resources it occupies are 0.25, and update Aperture = Aperture'. Finally, assume that task T i belongs to the k-th beam set, and there are n_beam k beams in this set, then update task T i The effective interference power that is successfully executed and expected to be obtained on this channel is Go to step 7. Step 7: Given that the set of channel - operating frequency band range at this time is Fre′, the channel - task combination matrix is B′, and the channel - task - aperture resource quantity matrix is Aperture′. If the number of channel analysis times μ > 1 at this time, then when calculating the total benefit of task execution, it is necessary to combine the resource allocation results of the previous μ - 1 analysis stages, that is, update Fre′ = Fre′ ∪ Fre, B′ = B′ ∪ B, and T classify ' = T classify ∪T classify '. Calculate the sum of the execution benefits that each task expects to obtain on all channels according to the following formula. Among them, is the interference power and the comprehensive benefit of the interference frequency band obtained on channel m for task T. (P i i G i ) des is the expected interference power of the task, is its comprehensive loss, K iΣ is the radar suppression coefficient, σ i is the radar cross-sectional area. indicates the ratio of the actual interference power to the expected interference power of the task. And if the interference power emitted by the jammer to the specified target reaches the target interference suppression effect, the excess system resources will not be able to continue to enhance the interference effect. is the coverage rate of the frequency band range of channel m for the expected execution frequency band range of task T i ​​ If task T i satisfies ∑ m eff(i,m) > 0, then add T i to the execution list Ta′, and set Otherwise, ∑ m eff(i,m) = 0, and task T i fails to execute. Add such tasks to the deletion list Td′. Update Ta = Ta′, Td = Td′, Fre = Fre′, B = B′, Aperture = Aperture′, T classify = T classify ′. Proceed to Step 1. Step 8: The process of this scheduling analysis ends. The task execution queue Ta, the task deletion queue Td, the channel-frequency band range set Fre, the channel-task combination matrix B, the channel-task-aperture resource matrix Aperture, and the set T of the final task execution methods on all channels are output classify .

Images