CN107070534A - The dynamic preemptive type method for scheduling task and system of a kind of repeater satellite load balancing - Google Patents
The dynamic preemptive type method for scheduling task and system of a kind of repeater satellite load balancing Download PDFInfo
- Publication number
- CN107070534A CN107070534A CN201710057263.XA CN201710057263A CN107070534A CN 107070534 A CN107070534 A CN 107070534A CN 201710057263 A CN201710057263 A CN 201710057263A CN 107070534 A CN107070534 A CN 107070534A
- Authority
- CN
- China
- Prior art keywords
- task
- scheduling
- mrow
- msub
- dynamic
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 40
- 230000006870 function Effects 0.000 claims abstract description 42
- 238000003780 insertion Methods 0.000 claims abstract description 34
- 230000037431 insertion Effects 0.000 claims abstract description 34
- 230000011218 segmentation Effects 0.000 claims abstract description 16
- 230000003068 static effect Effects 0.000 claims description 18
- 238000011156 evaluation Methods 0.000 claims description 16
- 230000008569 process Effects 0.000 claims description 12
- 230000008859 change Effects 0.000 claims description 11
- 238000000354 decomposition reaction Methods 0.000 claims description 10
- 239000011159 matrix material Substances 0.000 claims description 9
- 238000012545 processing Methods 0.000 claims description 5
- 238000012854 evaluation process Methods 0.000 claims description 3
- 230000007246 mechanism Effects 0.000 claims description 3
- 238000013459 approach Methods 0.000 abstract description 2
- 230000009897 systematic effect Effects 0.000 abstract 1
- 230000005540 biological transmission Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 5
- 238000005265 energy consumption Methods 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 238000013500 data storage Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B7/00—Radio transmission systems, i.e. using radiation field
- H04B7/14—Relay systems
- H04B7/15—Active relay systems
- H04B7/185—Space-based or airborne stations; Stations for satellite systems
- H04B7/1851—Systems using a satellite or space-based relay
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W28/00—Network traffic management; Network resource management
- H04W28/02—Traffic management, e.g. flow control or congestion control
- H04W28/08—Load balancing or load distribution
- H04W28/082—Load balancing or load distribution among bearers or channels
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W72/00—Local resource management
- H04W72/12—Wireless traffic scheduling
- H04W72/1263—Mapping of traffic onto schedule, e.g. scheduled allocation or multiplexing of flows
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- Astronomy & Astrophysics (AREA)
- Aviation & Aerospace Engineering (AREA)
- General Physics & Mathematics (AREA)
- Radio Relay Systems (AREA)
Abstract
The present invention relates to the dynamic preemptive type method for scheduling task and system of a kind of repeater satellite load balancing, it comprises the following steps：If 1) antenna has sufficiently long free time, it is inserted directly into task in the time and is directly dispatched；If can not be inserted directly into, carry out the switching of preemptive type task and subtask segmentation is scheduled with insertion, next step is entered if it can not still dispatch；2) to using the new task not being performed after direct scheduler module, scheduling is completed by dispatching indirectly；3) selection of three object functions realizations to optimal dynamic scheduling scheme is set up.The present invention further optimizes repeater satellite task scheduling approach under the conditions of dynamic disturbances using direct scheduler module and indirect scheduler module, to provide than complete readjustment degree and the more excellent systematic function of conventional dynamic dispatching method.The present invention can be applied in Information Network repeater satellite scheduling of resource field extensively.
Description
Technical Field
The invention relates to the field of relay satellite resource scheduling of a spatial information network, in particular to a dynamic preemptive task scheduling method and system for relay satellite load balancing.
Background
With the rapid development of the spatial information network, the spatial traffic gradually shows an exponential growth trend. The relay satellite system is used as an important component of a spatial information network, can effectively relieve the transmission pressure of satellite services, realizes real-time return of global-range data, and makes up the practical problem that the ground station is difficult to establish in the open sea in China. In a relay satellite system, in order to realize reasonable arrangement of on-satellite tasks, resource scheduling becomes a main technical problem to be solved urgently.
In the related research of satellite task scheduling, the research focus is mathematical modeling and algorithm solving. By establishing a mathematical model highly equivalent to the relay satellite scheduling problem, the constraint conditions borne by the satellite control system and the specific targets to be realized by the system are accurately described, and the method is the basis for solving the problem. To date, modeling and solving of relay satellite scheduling problems can be largely divided into two categories: static mission planning and dynamic mission planning.
1. Static task planning: the relay satellite resource scheduling is influenced by various factors such as a visible time window, data storage capacity, equipment energy consumption, task effectiveness, priority and the like, and is an NP-hard problem which is satisfied by multiple constraints. At present, a static task planning mode is mainly adopted for the problem, namely, on the premise that tasks are periodically generated or known in advance, a task scheduling scheme is solved. The proposed problem model mainly comprises a linear programming model, a graph theory model and a multi-constraint satisfaction problem model. The mathematical programming model and the graph theory model have the characteristics of mature basic theory, simple solving algorithm and the like, but have no advantages for solving the scheduling problem of large-scale tasks and complex constraint conditions. The constraint satisfaction problem model can accurately describe variables and constraint conditions in the problem by using a logic language, can be solved by adopting a group intelligent optimization algorithm, obtains the optimal or better solution of the scheduling problem in a determined time by designing an appropriate search strategy, and fully exerts the offline processing capacity of the relay satellite.
2. Dynamic task planning: in fact, the relay satellite scheduling process is affected by various disturbance factors, including burst tasks, equipment faults, platform vibration, poor error code performance and the like, and dynamic changes of on-satellite tasks and terminal equipment are directly caused. When the above uncertain factors occur, if a static planning algorithm is used to rearrange tasks, not only the computation overhead and energy consumption of the satellite are increased, but also a large number of tasks cannot be transmitted in real time, and even the scheduling scheme may fail in a serious case. The scheduling research of the relay satellite needs to continuously adjust the original scheme along with the change of resources on the basis of static task planning, and realize efficient dynamic task planning on the premise of ensuring the minimum adjustment degree. The existing dynamic scheduling method adopts simpler replacement insertion and mobile insertion strategies to adjust tasks by analyzing service characteristics. This approach can result in large changes to the original scheme, resulting in the loss of more tasks in cases where the task scheduling is more intensive. In addition, a complete re-scheduling algorithm is adopted to re-calculate the scheduling scheme, so that a large amount of computing resources are wasted. Therefore, there is a need to further optimize the scheduling performance of relay satellites in dynamic mission planning.
Currently, there are many studies in academic circles and industrial circles on the problem of relay satellite resource scheduling in the first case, and there are relatively limited academic studies and design schemes related to the second case.
Disclosure of Invention
In view of the above problems, an object of the present invention is to provide a method and a system for dynamically preemptively scheduling a task in relay satellite load balancing, wherein a direct scheduling module and an indirect scheduling module are used to further optimize a task scheduling scheme for a relay satellite under a dynamic disturbance condition, so as to provide a system performance better than that of a complete rescheduling and a conventional dynamic scheduling method.
In order to achieve the purpose, the invention adopts the following technical scheme: a dynamic preemptive task scheduling method for relay satellite load balancing is characterized by comprising the following steps: 1) if the antenna has enough idle time, directly inserting a task at the time to carry out direct scheduling; if the direct insertion cannot be carried out, carrying out preemptive task switching and subtask segmentation and insertion for scheduling, and if the direct insertion cannot be carried out, entering the next step; 2) the new tasks which are not executed after the direct scheduling module is adopted are scheduled through indirect scheduling; 3) and establishing three objective functions to realize the selection of the optimal dynamic scheduling scheme.
Further, in step 1), the preemptive task switching is to pre-arrange a new task by adjusting a scheduling order of the tasks when the new task is generated and all antenna terminals are occupied during task scheduling.
Further, in the step 1), the dividing and inserting of the subtasks is to divide the new task or the original task which conflicts with the new task into a plurality of subtasks for scheduling through a subtask division strategy.
Further, in the step 1), the direct scheduling process is as follows: 1.1) obtaining a new task set TdSorting the tasks from high to low according to the weight to obtain Td＝{Td_1,Td_2,…,Td_NtIn which N istRepresenting the number of new tasks, and determining the current scheduling task as Td_i(ii) a 1.2) to task Td_iDetermining a set of antenna resources Md＝{Md_1,Md_2,…,Md_NmIn which N ismRepresenting the number of available days, selectingCurrent antenna is Md_j(ii) a 1.3) to task Td_iSelecting an antenna Md_jUpper visible time window WdObtaining the selection time window and the remaining time, and defining a task insertion point set Id＝{stw,et1,et2,…,etn_old, etw }; wherein stw and etw represent the start and end times of the current time window; n _ old represents the number of original tasks scheduled in the current time window in the static planning scheme; et alkIndicating the end time of the task scheduled by the static planning scheme in the current time window, k being 1,2, …; 1.4) sequentially judging the tasks Td_iWhether or not to insert directly in each time window; if feasible, scheduling the new task Td_iOtherwise, entering the next step; 1.5) adopting a preemptive task switching strategy to sequentially judge the tasks Td_iWhether execution can be performed in each time window; if feasible, scheduling the new task Td_iOtherwise, entering the next step; 1.6) adopting a subtask segmentation strategy to collect the insertion points I of each time windowdAccording to the time length sequence of the idle time interval, judging the top Ni dec_dynamicWhether the segmented task T can be realized in each idle periodd_iScheduling of (2); if feasible, scheduling the new task Td_iOtherwise, adopting an indirect scheduling method; wherein N isi dec_dynamicIndicating the number of subtasks of the new task.
Further, in the step 2), the indirect scheduling process is as follows: 2.1) backtracking to a new task T of the taskd_iInserting node IdAnd a set of conflicting tasks; 2.2) selecting insertion points and conflicting task sequences, comparing the new task Td_iWeights associated with conflicting tasks; if the new task Td_iIf the size is larger, the conflict task is replaced and executed preferentially; otherwise, entering step 2.4); 2.3) adopting a subtask segmentation strategy to schedule the replaced conflict task; if the processing operation is successful, the conflict task is rescheduled; otherwise, deleting the conflict task; 2.4) searching and judging whether the next inserted node can realize the new task T or not in sequenced_iScheduling of (2); if all task insertion nodes are traversed, step 3) is entered.
Further, in the step 3), the evaluation process of the dynamic scheduling scheme is as follows: 3.1) defining three objective functions as a maximum task scheduling weight, a minimum scheme change rate and a minimum subtask decomposition degree respectively; 3.2) determining a multi-target decision mechanism according to the target function.
Further, in the step 3.1), the maximum task scheduling weight f1(Ad)：
In the formula, AdIs a static mission planning scheme; y isiFor scheduling completed task i, if task i is scheduled, yi1 is ═ 1; otherwise, yi＝0；piRepresenting a weight of the corresponding task;
minimum rate of change of scheme f2(A,Ad)：
In the formula,movfor the penalty factor of a mobile task,dela penalty factor for deleting tasks; n ismovFor moving the number of tasks, ndelThe number of times of deleting the task; p is a radical ofiMoving the weight of a task for a task, pjThe weight of the task is deleted for the task.
Further, in the step 3.1), the minimum subtask decomposition degree f3(Ad)：
In the formula, Ni dec_dynamicIndicating the number of subtasks of the new task; n is a radical ofi dec_originalRepresenting the number of subtasks of the original task.
Further, in the step 3.2), the specific determination process is as follows: 3.2.1) assuming that n dynamic scheduling schemes are obtained after passing through a scheduling module, and defining a dynamic scheme matrix D; 3.2.2) normalizing and weighting each target function under different schemes and obtaining an improved dynamic scheme matrix Dw(ii) a 3.2.3) selecting the dynamic scheme matrix D of each scheme separatelywFor the optimal value and the worst value of the three objective functions, and constructing an optimal vectorWorst vector
Wherein,andrespectively representing the maximum value and the minimum value of the ith objective function; 3.2.4) calculating Euclidean distance d from each scheme to the optimal vector and the worst vectori +And di -According to the optimal vector Euclidean distance di +Euclidean distance d from the worst vectori -The ratio of (a) to (b) is obtained as an evaluation function E of the ith schemeiThe smaller the value of the evaluation function is, the better the performance of the scheme corresponding to the evaluation function is.
A dynamic preemptive task scheduling system for relay satellite load balancing is characterized by comprising the following steps: the system comprises a direct scheduling module, an indirect scheduling module and a scheduling scheme evaluation module; the direct scheduling module is used for directly inserting a task for scheduling at the time when the antenna has enough idle time; if the direct insertion cannot be carried out, carrying out preemptive task switching and sub-task segmentation and insertion for scheduling; the indirect scheduling module is used for finishing scheduling on new tasks which are not executed after the direct scheduling module is adopted through indirect scheduling; the scheduling scheme evaluation module is used for establishing three objective functions to realize selection of the optimal dynamic scheduling scheme.
Due to the adoption of the technical scheme, the invention has the following advantages: 1. according to the invention, the system performance of relay satellite resource scheduling under dynamic disturbance is fully optimized by adopting direct scheduling and indirect scheduling and establishing three objective functions to select the optimal dynamic scheduling scheme. 2. Compared with the existing relay satellite scheduling method, the method provided by the invention can be used for more effectively solving the problem of multiple types of dynamic disturbance actually faced by the resource scheduling problem, and the overall performance of the system under the condition of dynamic change of tasks is improved. Meanwhile, the preemptive dynamic scheduling algorithm has a high processing speed, and efficient dynamic task planning is realized on the premise of ensuring the minimum adjustment degree of the original scheduling scheme. 3. The invention adopts the preemptive task to switch in the direct scheduling, and then the new task is arranged in advance, thereby avoiding the conflict with the original task and effectively improving the scheduling efficiency of the system. 4. The invention adopts the division and insertion of the subtasks in the direct scheduling, can preferentially transmit the new task, and can divide the original task into a plurality of subtasks, thereby not only realizing the maximization of the scheduling task weight, but also improving the flexibility of task scheduling. In conclusion, the invention can be widely applied to the field of relay satellite resource scheduling of the spatial information network.
Drawings
FIG. 1 is a schematic diagram of a dynamic preemptive task scheduling method according to the present invention;
FIG. 2 is a schematic diagram of preemptive task switching according to the present invention;
FIG. 3 is a subtask segmentation diagram of the present invention;
FIG. 4 is a schematic diagram illustrating task scheduling of a relay satellite system according to an embodiment of the present invention;
FIG. 5 is a Gantt diagram of a WRA in an embodiment of the invention;
FIG. 6 is a Gantt chart of TDSA in an embodiment of the present invention;
fig. 7 is a gantt chart of PDSA in an embodiment of the invention.
Detailed Description
The relay satellite dynamic preemptive scheduling system takes a static task planning scheme and a dynamic scheduling strategy as heuristic information, takes three objective functions of maximum task scheduling weight, minimum scheme change rate and minimum subtask decomposition degree as a scheduling principle, and realizes efficient dynamic task planning under the condition of ensuring the minimum scheme adjustment degree. The invention provides a dynamic preemptive task scheduling method and system for relay satellite load balancing, which is mainly characterized in that in the effective time of a new task, if an antenna has enough idle time, a task can be directly inserted into the time for scheduling; and if the direct insertion cannot be carried out, scheduling by adopting a preemptive task switching and subtask splitting strategy. If the effective time of the new task is too short and the task weight is larger, resources need to be immediately preempted and the current task is replaced for execution; and if the replaced conflict task cannot be scheduled by adopting a subtask segmentation method, discarding the conflict task. The invention is described in detail below with reference to the figures and examples.
As shown in fig. 1, the method for scheduling a dynamic preemptive task for relay satellite load balancing according to the present invention includes the following steps:
1) in the process of task scheduling, if the antenna has enough idle time, the task can be directly inserted into the antenna at the time for direct scheduling; if the direct insertion cannot be carried out, carrying out preemptive task switching and subtask segmentation and insertion for scheduling, and if the direct insertion cannot be carried out, entering the next step;
2) the new tasks which are not executed after the direct scheduling module is adopted are scheduled through indirect scheduling;
3) after direct and indirect scheduling is adopted, a series of dynamic scheduling schemes are obtained, and three objective functions are established to realize selection of the optimal dynamic scheduling scheme for determining the quality degree of the schemes.
In the step 1), the preemptive task switching is to pre-arrange a new task by adjusting the scheduling order of the tasks when the new task is generated and all antenna terminals are occupied during the task scheduling process. As shown in fig. 2, the light part is the effective time of the task, and the black part is the time when the task is scheduled. Prior to the adjustment, the original task has been scheduled in the original solution and conflicts with a new task that needs to be scheduled immediately, resulting in the new task not being able to be executed. After the preemptive task is switched, the new task is arranged in advance, so that the conflict with the original task is avoided, and the scheduling efficiency of the system can be effectively improved.
In the step 1), since the dynamic task scheduling is implemented based on the initial scheme of static task planning, the scheduling of the original tasks occupies most of the time window, and the tasks are divided into a plurality of small time slots. Thus, there is rarely enough time in each time window to schedule a new task, directly resulting in increased difficulty in task scheduling. If only the strategy of preemptive task scheduling is adopted, the original task will exceed the effective time range of the original task. The division and insertion of the subtasks adopted by the invention are to divide the new task or the original task which conflicts with the new task into a plurality of subtasks for scheduling through a subtask division strategy: as shown in fig. 3, if the original task has a sufficient effective time for task transmission, when a new task is accessed, the new task can be transmitted preferentially, and the original task can be divided into a plurality of subtasks, so that not only is the maximization of the task scheduling weight achieved, but also the flexibility of task scheduling is improved.
In the step 1), the direct scheduling process is as follows:
1.1) obtaining a new task set TdTasks are weighted fromHigh to low ordering to obtain Td＝{Td_1,Td_2,…,Td_NtIn which N istRepresenting the number of new tasks, and determining the current scheduling task as Td_i；
1.2) to task Td_iDetermining a set of antenna resources MdObtaining Md＝{Md_1,Md_2,…,Md_NmIn which N ismRepresenting the number of available space, selecting the current antenna as Md_j；
1.3) to task Td_iSelecting an antenna Md_jUpper visible time window WdObtaining the selection time window and the remaining time, and defining a task insertion point set Id＝{stw,et1,et2,…,etn_oldEtw }; wherein stw and etw represent the start and end times of the current time window; n _ old represents the number of original tasks scheduled in the current time window in the static planning scheme; et alkIndicating the end time of the task scheduled by the static planning scheme in the current time window, k being 1,2, …;
1.4) sequentially judging the tasks Td_iWhether or not to insert directly in each time window; if feasible, scheduling the new task Td_iOtherwise, entering the next step;
1.5) adopting a preemptive task switching strategy to sequentially judge the tasks Td_iWhether execution can be performed in each time window; if feasible, scheduling the new task Td_iOtherwise, entering the next step;
1.6) adopting a subtask segmentation strategy to collect the insertion points I of each time windowdAccording to the time length sequence of the idle time interval, judging the top Ni dec_dynamicWhether the segmented task T can be realized in each idle periodd_iScheduling of (2); if feasible, scheduling the new task Td_iOtherwise, adopting an indirect scheduling method; wherein N isi dec_dynamicIndicating the number of subtasks of the new task.
In the step 2), the indirect scheduling process is as follows:
2.1) backtracking to a new task T of the taskd_iInserting node IdAnd a set of conflicting tasks;
2.2) selecting insertion points and conflicting task sequences, comparing the new task Td_iWeights associated with conflicting tasks; if the new task Td_iIf the size is larger, the conflict task is replaced and executed preferentially; otherwise, entering step 2.4);
2.3) adopting a subtask segmentation strategy to schedule the replaced conflict task; if the processing operation is successful, the conflict task is rescheduled; otherwise, the conflicting task is deleted.
2.4) searching and judging whether the next inserted node can realize the new task T or not in sequenced_iThe scheduling of (2). If all task insertion nodes are traversed, step 3) is entered.
In the step 3), the evaluation process of the dynamic scheduling scheme is as follows:
3.1) defining three objective functions as a maximum task scheduling weight, a minimum scheme change rate and a minimum subtask decomposition degree respectively:
① maximum task scheduling weight f1(Ad)：
The main goal of dynamic mission planning is to improve the scheduling capability of the relay satellite, i.e. the relay satellite system needs to transmit more missions. Thus, a dynamic scheduling objective function f is established1(Ad) For the maximum scheduling weight:
in the formula, AdIs a static mission planning scheme; y isiFor scheduling completed task i, if task i is scheduled, yi1 is ═ 1; otherwise, yi＝0；piRepresenting the weight of the corresponding task.
② minimal solution changeRate f2(A,Ad)：
After the initial scheduling plan is determined, each user satellite typically performs a data transmission plan according to the plan. Therefore, the dynamic adjustment of the initial scheme not only affects the task of current adjustment, but also may cause the change of subsequent decision of each user. In addition, after the task is adjusted, a new scheduling scheme needs to be issued to each user satellite again, and the overhead of a relay satellite system is increased. Thus, a second objective function f of the dynamic scheduling model is established2(A,Ad) Is the minimum rate of change of the protocol.
Because the initial scheme task is adjusted in two modes of moving tasks and deleting tasks, the initial scheme is affected to different degrees by different adjustment modes, and the influence degree is related to the importance degree of the adjusted task, so the objective function f2(A,Ad) The definition is as follows:
in the formula,movfor the penalty factor of a mobile task,dela penalty factor for deleting tasks; n ismovFor moving the number of tasks, ndelThe number of times of deleting the task; p is a radical ofiMoving the weight of a task for a task, pjThe weight of the task is deleted for the task.
③ minimum subtask decomposition degree f3(Ad)：
The scheduling efficiency of the system can be effectively improved by adopting a subtask decomposition strategy, but too many times of subtask decomposition can result in the increase of switching time and system energy consumption, which can affect the overall performance of the system, so that the number of task segmentation is reduced while the task scheduling weight is ensured to be higher. Thus, the objective function minimum subtask decomposition degree f3(Ad) The definition is as follows:
in the formula, Ni dec_dynamicIndicating the number of subtasks of the new task; n is a radical ofi dec_originalRepresenting the number of subtasks of the original task.
In a preferred embodiment, to ensure the switching power consumption of the satellite antenna, N is seti dec_dynamicAnd Ni dec _originalIs not more than 3.
3.2) determining a multi-target decision mechanism according to the target function:
3.2.1) assuming that n dynamic scheduling schemes are obtained after passing through the scheduling module, a dynamic scheme matrix D can be defined as:
wherein f isi jThe ith objective function value for the jth scenario is shown.
3.2.2) normalizing and weighting each target function under different schemes and obtaining an improved dynamic scheme matrix Dw：
Wherein, FiWhich represents the ith objective function value after the normalization weighting operation. I fi||2Representing the 2-norm of the ith objective function. w ═ w1,w2,w3]Represents the degree of importance of the objective function, and w1+w2+w3＝1。
3.2.3) selecting the dynamic scheme matrix D of each scheme separatelywFor the optimal and worst values of the three objective functionsAnd constructing an optimal vectorWorst vector
Wherein,andrespectively representing the maximum and minimum of the ith objective function.
3.2.4) calculating Euclidean distance d from each scheme to the optimal vector and the worst vectori +And di -According to the optimal vector Euclidean distance di +The worst Euclidean distance di -Obtaining an evaluation function E of the ith schemei：
The smaller the value of the evaluation function is, the better the performance of the scheme corresponding to the evaluation function is.
The invention also provides a dynamic preemptive task scheduling system for relay satellite load balancing, which comprises a direct scheduling module, an indirect scheduling module and a scheduling scheme evaluation module. Wherein:
the direct scheduling module is used for directly inserting a task for scheduling at the time when the antenna has enough idle time; if the direct insertion cannot be carried out, carrying out preemptive task switching and sub-task segmentation and insertion for scheduling;
the indirect scheduling module is used for finishing scheduling on the new tasks which are not executed after the direct scheduling module is adopted through indirect scheduling;
the scheduling scheme evaluation module is used for establishing three objective functions to realize the selection of the optimal dynamic scheduling scheme.
Example (b):
as shown in fig. 4, the scene setting: 3 relay satellites and 8 user satellites, and the simulation time is 00: 00: 00-06: 00: 00. three relay satellites are respectively positioned at 0 DEG E, 162.324 DEG W and 162.324 DEG E, and each relay satellite carries 2 single-address antenna terminals comprising a Ka-band antenna and a laser antenna. The data rate of the Ka antenna is 600MB/s, and the switching time is 2 s; the data rate of the laser antenna was 1500MB/s and the switching time was 20 s. Assuming that the initial task number is 64, and the number of burst tasks caused by various disturbance factors is 8, the detailed definition of the tasks is shown in table 1. The relay satellite system preferentially carries out off-line static task planning during scheduling, and system scheduling performance of a complete re-scheduling algorithm (WRA) and a Preemptive Dynamic Scheduling Algorithm (PDSA) is optimized on the basis.
TABLE 1 parameter settings for bursty tasks
| Task numbering | Task weight | User satellite | Task size/Gbit |
| T65 | 5 | LEO 01 | 220 |
| T66 | 10 | LEO 02 | 200 |
| T67 | 8 | LEO 03 | 150 |
| T68 | 6 | LEO 04 | 270 |
| T69 | 7 | LEO 05 | 150 |
| T70 | 7 | LEO 06 | 160 |
| T71 | 8 | LEO 07 | 100 |
| T72 | 9 | LEO 08 | 400 |
Fig. 5 is a gantt chart of the WRA, where the lighter colored part (i.e., the lower half of each antenna corresponding to a transmission task) represents the static mission planning scheme and the darker colored part (i.e., the upper half of each antenna corresponding to a transmission task) represents the scheduling scheme of the WRA. The length of the color block corresponds to the task execution time on the selected antenna. Obviously, the WRA schedules all 8 new tasks, but discards the tasks in the original scheme. In addition, only 8 tasks are allocated after dynamic scheduling is carried out by using WRA, and the re-scheduling rate is up to 87.50 percent. Although the WRA algorithm meets the requirement of task scheduling, the rescheduling rate of the scheme and the waste of resources caused by rescheduling have a large influence on the system performance.
FIG. 6 shows a Gantt chart of TDSA. The new task is preferentially scheduled to be executed in idle time, and if the idle time is insufficient, the new task replaces the task in the original plan. In the figure, two original tasks are replaced by new tasks, and the task completion rate is 97.22%. Compared with static mission planning, the probability of a change of the scenario is 3.13%. TDSA is preferred to ensure that new tasks are completed, and therefore, if a large number of real-time tasks are generated, system scheduling performance is greatly affected.
Fig. 7 is a gantt chart of PDSA, in which an oval part represents that a new task T5 is replaced by a T69 and is moved to a free time slot for scheduling. The diamond-shaped portion indicates that the new tasks T65 and T66 are divided into subtasks for execution due to the limited time available. As can be seen from the figure, the PDSA realizes the scheduling of all new task tasks, and the rescheduling rate is only 1.56%. Therefore, compared with WRA, PDSA is optimized by 1.39% and 85.94% in terms of task scheduling amount and rescheduling rate respectively.
In conclusion, the invention carries out evaluation through direct scheduling, indirect scheduling and three objective functions, and compared with a complete re-scheduling algorithm (WRA) and a Traditional Dynamic Scheduling Algorithm (TDSA), the performance of the invention is better.
The above embodiments are only for illustrating the present invention, and the steps may be changed, and on the basis of the technical solution of the present invention, the modification and equivalent changes of the individual steps according to the principle of the present invention should not be excluded from the protection scope of the present invention.
Claims (10)
1. A dynamic preemptive task scheduling method for relay satellite load balancing is characterized by comprising the following steps:
1) if the antenna has enough idle time, directly inserting a task at the time to carry out direct scheduling; if the direct insertion cannot be carried out, carrying out preemptive task switching and subtask segmentation and insertion for scheduling, and if the direct insertion cannot be carried out, entering the next step;
2) the new tasks which are not executed after the direct scheduling module is adopted are scheduled through indirect scheduling;
3) and establishing three objective functions to realize the selection of the optimal dynamic scheduling scheme.
2. The method for dynamically preemptively scheduling a relay satellite load balancing according to claim 1, wherein: in the step 1), the preemptive task switching is to pre-arrange a new task by adjusting the scheduling sequence of the tasks when the new task is generated and all antenna terminals are occupied in the task scheduling process.
3. The method for dynamically preemptively scheduling a relay satellite load balancing according to claim 1, wherein: in the step 1), the sub-task division and insertion is to divide the new task or the original task which conflicts with the new task into a plurality of sub-tasks for scheduling through a sub-task division strategy.
4. A dynamic preemptive task scheduling method for relay satellite load balancing according to any of claims 1 to 3, characterized by: in the step 1), the direct scheduling process is as follows:
1.1) obtaining a new task set TdSorting the tasks from high to low according to the weight to obtain Td＝{Td_1,Td_2,…,Td_NtIn which N istRepresenting the number of new tasks, and determining the current scheduling task as Td_i；
1.2) to task Td_iDetermining a set of antenna resources Md＝{Md_1,Md_2,…,Md_NmIn which N ismRepresenting the number of available space, selecting the current antenna as Md_j；
1.3) to task Td_iSelecting an antenna Md_jUpper visible time window WdObtaining the selection time window and the remaining time, and defining a task insertion point set Id＝{stw,et1,et2,…,etn_oldEtw }; wherein stw and etw represent the start and end times of the current time window; n _ old represents silenceThe number of original tasks scheduled in the current time window in the dynamic planning scheme; et alkIndicating the end time of the task scheduled by the static planning scheme in the current time window, k being 1,2, …;
1.4) sequentially judging the tasks Td_iWhether or not to insert directly in each time window; if feasible, scheduling the new task Td_iOtherwise, entering the next step;
1.5) adopting a preemptive task switching strategy to sequentially judge the tasks Td_iWhether execution can be performed in each time window; if feasible, scheduling the new task Td_iOtherwise, entering the next step;
1.6) adopting a subtask segmentation strategy to collect the insertion points I of each time windowdAccording to the time length sequence of the idle time interval, judging the top Ni dec_dynamicWhether the segmented task T can be realized in each idle periodd_iScheduling of (2); if feasible, scheduling the new task Td_iOtherwise, adopting an indirect scheduling method; wherein N isi dec_dynamicIndicating the number of subtasks of the new task.
5. The method for dynamically preemptively scheduling a relay satellite load balancing according to claim 1, wherein: in the step 2), the indirect scheduling process is as follows:
2.1) backtracking to a new task T of the taskd_iInserting node IdAnd a set of conflicting tasks;
2.2) selecting insertion points and conflicting task sequences, comparing the new task Td_iWeights associated with conflicting tasks; if the new task Td_iIf the size is larger, the conflict task is replaced and executed preferentially; otherwise, entering step 2.4);
2.3) adopting a subtask segmentation strategy to schedule the replaced conflict task; if the processing operation is successful, the conflict task is rescheduled; otherwise, deleting the conflict task;
2.4) searching and judging whether the next inserted node can realize the new task T or not in sequenced_iScheduling of (2); if all task insertion nodes are traversed, step 3) is entered.
6. The method for dynamically preemptively scheduling a relay satellite load balancing according to claim 1, wherein: in the step 3), the evaluation process of the dynamic scheduling scheme is as follows:
3.1) defining three objective functions as a maximum task scheduling weight, a minimum scheme change rate and a minimum subtask decomposition degree respectively;
3.2) determining a multi-target decision mechanism according to the target function.
7. The method for dynamically preemptively scheduling a relay satellite load balancing according to claim 6, wherein: in the step 3.1), the maximum task scheduling weight f1(Ad)：
<mrow> <mi>M</mi> <mi>a</mi> <mi>x</mi> <mi> </mi> <msub> <mi>f</mi> <mn>1</mn> </msub> <mrow> <mo>(</mo> <msub> <mi>A</mi> <mi>d</mi> </msub> <mo>)</mo> </mrow> <mo>=</mo> <munder> <mo>&Sigma;</mo> <mrow> <mi>i</mi> <mo>&Element;</mo> <mo>{</mo> <mi>T</mi> <mo>,</mo> <msub> <mi>T</mi> <mi>d</mi> </msub> <mo>}</mo> </mrow> </munder> <msub> <mi>y</mi> <mi>i</mi> </msub> <msub> <mi>p</mi> <mi>i</mi> </msub> <mo>,</mo> </mrow>
In the formula, AdIs a static mission planning scheme; y isiFor scheduling completed task i, if task i is scheduled, yi1 is ═ 1; otherwise, yi＝0；piRepresenting a weight of the corresponding task;
minimum rate of change of scheme f2(A,Ad)：
<mrow> <mi>M</mi> <mi>i</mi> <mi>n</mi> <mi> </mi> <msub> <mi>f</mi> <mn>2</mn> </msub> <mrow> <mo>(</mo> <mi>A</mi> <mo>,</mo> <msub> <mi>A</mi> <mi>d</mi> </msub> <mo>)</mo> </mrow> <mo>=</mo> <munderover> <mo>&Sigma;</mo> <mrow> <mi>i</mi> <mo>=</mo> <mn>1</mn> </mrow> <msub> <mi>n</mi> <mrow> <mi>m</mi> <mi>o</mi> <mi>v</mi> </mrow> </msub> </munderover> <msub> <mi>&delta;</mi> <mrow> <mi>m</mi> <mi>o</mi> <mi>v</mi> </mrow> </msub> <msub> <mi>p</mi> <mi>i</mi> </msub> <mo>+</mo> <munderover> <mo>&Sigma;</mo> <mrow> <mi>j</mi> <mo>=</mo> <mn>1</mn> </mrow> <msub> <mi>n</mi> <mrow> <mi>d</mi> <mi>e</mi> <mi>l</mi> </mrow> </msub> </munderover> <msub> <mi>&delta;</mi> <mrow> <mi>d</mi> <mi>e</mi> <mi>l</mi> </mrow> </msub> <msub> <mi>p</mi> <mi>j</mi> </msub> <mo>,</mo> </mrow>
In the formula,movfor the penalty factor of a mobile task,dela penalty factor for deleting tasks; n ismovFor moving the number of tasks, ndelThe number of times of deleting the task; p is a radical ofiMoving the weight of a task for a task, pjThe weight of the task is deleted for the task.
8. The method for dynamically preemptively scheduling a relay satellite load balancing according to claim 6, wherein: in said step 3.1), the minimum subtask decomposition degree f3(Ad)：
<mrow> <mi>M</mi> <mi>i</mi> <mi>n</mi> <mi> </mi> <msub> <mi>f</mi> <mn>3</mn> </msub> <mrow> <mo>(</mo> <msub> <mi>A</mi> <mi>d</mi> </msub> <mo>)</mo> </mrow> <mo>=</mo> <munder> <mo>&Sigma;</mo> <mrow> <mi>i</mi> <mo>&Element;</mo> <mo>{</mo> <msub> <mi>T</mi> <mi>d</mi> </msub> <mo>}</mo> </mrow> </munder> <msubsup> <mi>N</mi> <mi>i</mi> <mrow> <mi>d</mi> <mi>e</mi> <mi>c</mi> <mo>_</mo> <mi>d</mi> <mi>y</mi> <mi>n</mi> <mi>a</mi> <mi>m</mi> <mi>i</mi> <mi>c</mi> </mrow> </msubsup> <msub> <mi>p</mi> <mi>i</mi> </msub> <mo>+</mo> <munder> <mo>&Sigma;</mo> <mrow> <mi>j</mi> <mo>&Element;</mo> <mo>{</mo> <mi>T</mi> <mo>}</mo> </mrow> </munder> <msubsup> <mi>N</mi> <mi>j</mi> <mrow> <mi>d</mi> <mi>e</mi> <mi>c</mi> <mo>_</mo> <mi>o</mi> <mi>r</mi> <mi>i</mi> <mi>g</mi> <mi>i</mi> <mi>n</mi> <mi>a</mi> <mi>l</mi> </mrow> </msubsup> <msub> <mi>p</mi> <mi>j</mi> </msub> <mo>,</mo> </mrow>
In the formula, Ni dec_dynamicIndicating the number of subtasks of the new task; n is a radical ofi dec_originalRepresenting the number of subtasks of the original task.
9. The method for dynamically preemptively scheduling a relay satellite load balancing according to claim 6, wherein: in the step 3.2), the specific determination process is as follows:
3.2.1) assuming that n dynamic scheduling schemes are obtained after passing through a scheduling module, and defining a dynamic scheme matrix D;
3.2.2) normalizing and weighting each target function under different schemes and obtaining an improved dynamic scheme matrix Dw；
3.2.3) selecting the dynamic scheme matrix D of each scheme separatelywFor the optimal value and the worst value of the three objective functions, and constructing an optimal vectorWorst vector
<mrow> <mfenced open = "{" close = ""> <mtable> <mtr> <mtd> <mrow> <msubsup> <mi>D</mi> <mi>w</mi> <mo>+</mo> </msubsup> <mo>=</mo> <mfenced open = "[" close = "]"> <mtable> <mtr> <mtd> <msubsup> <mi>F</mi> <mn>1</mn> <mi>max</mi> </msubsup> </mtd> <mtd> <msubsup> <mi>F</mi> <mn>2</mn> <mi>min</mi> </msubsup> </mtd> <mtd> <msubsup> <mi>F</mi> <mn>3</mn> <mi>min</mi> </msubsup> </mtd> </mtr> </mtable> </mfenced> </mrow> </mtd> </mtr> <mtr> <mtd> <mrow> <msubsup> <mi>D</mi> <mi>w</mi> <mo>-</mo> </msubsup> <mo>=</mo> <mfenced open = "[" close = "]"> <mtable> <mtr> <mtd> <msubsup> <mi>F</mi> <mn>1</mn> <mi>min</mi> </msubsup> </mtd> <mtd> <msubsup> <mi>F</mi> <mn>2</mn> <mi>max</mi> </msubsup> </mtd> <mtd> <msubsup> <mi>F</mi> <mn>3</mn> <mi>max</mi> </msubsup> </mtd> </mtr> </mtable> </mfenced> </mrow> </mtd> </mtr> </mtable> </mfenced> <mo>,</mo> </mrow>
Wherein,andrespectively representing the maximum value and the minimum value of the ith objective function;
3.2.4) calculating Euclidean distance d from each scheme to the optimal vector and the worst vectori +And di -According to the optimal vector Euclidean distance di +Euclidean distance d from the worst vectori -The ratio of (a) to (b) is obtained as an evaluation function E of the ith schemeiThe smaller the value of the evaluation function is, the better the performance of the scheme corresponding to the evaluation function is.
10. A dynamic preemptive task scheduling system for relay satellite load balancing is characterized by comprising the following steps: the system comprises a direct scheduling module, an indirect scheduling module and a scheduling scheme evaluation module;
the direct scheduling module is used for directly inserting a task for scheduling at the time when the antenna has enough idle time; if the direct insertion cannot be carried out, carrying out preemptive task switching and sub-task segmentation and insertion for scheduling;
the indirect scheduling module is used for finishing scheduling on new tasks which are not executed after the direct scheduling module is adopted through indirect scheduling;
the scheduling scheme evaluation module is used for establishing three objective functions to realize selection of the optimal dynamic scheduling scheme.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710057263.XA CN107070534B (en) | 2017-01-26 | 2017-01-26 | A kind of the dynamic preemptive type method for scheduling task and system of repeater satellite load balancing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710057263.XA CN107070534B (en) | 2017-01-26 | 2017-01-26 | A kind of the dynamic preemptive type method for scheduling task and system of repeater satellite load balancing |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107070534A true CN107070534A (en) | 2017-08-18 |
| CN107070534B CN107070534B (en) | 2018-11-30 |
Family
ID=59599265
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710057263.XA Active CN107070534B (en) | 2017-01-26 | 2017-01-26 | A kind of the dynamic preemptive type method for scheduling task and system of repeater satellite load balancing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107070534B (en) |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107608793A (en) * | 2017-09-13 | 2018-01-19 | 航天恒星科技有限公司 | The Agile satellite task planning method of fast search |
| CN107678848A (en) * | 2017-10-17 | 2018-02-09 | 合肥工业大学 | Repeater satellite task scheduling constrains the construction method and device of plan model |
| CN107909207A (en) * | 2017-11-16 | 2018-04-13 | 北京卫星信息工程研究所 | Remote Sensing Ground Station Multiobjective Optimal Operation method and system |
| CN108053051A (en) * | 2017-11-16 | 2018-05-18 | 西安电子科技大学 | Task based access control splits the data relay satellite system mission planning method of polymerization |
| CN108199763A (en) * | 2017-12-29 | 2018-06-22 | 清华大学 | Repeater satellite method for scheduling task and system |
| CN108322249A (en) * | 2018-01-22 | 2018-07-24 | 新疆联海创智信息科技有限公司 | The restructural constellation satellite networking decision-making technique and system of oriented mission driving |
| CN108494470A (en) * | 2018-02-05 | 2018-09-04 | 西安电子科技大学 | Information Network repeater satellite antenna scheduling method based on optimization weight |
| CN109190938A (en) * | 2018-08-17 | 2019-01-11 | 中南大学 | The repeater satellite single access antenna dynamic dispatching method reached for temporary duty |
| CN110113212A (en) * | 2019-05-16 | 2019-08-09 | 中南大学 | A kind of repeater satellite single access antenna dispatching method suitable for breakpoint transmission |
| CN113393146A (en) * | 2021-06-25 | 2021-09-14 | 重庆大学 | Multi-stage flexible test system reliability evaluation method based on Bayesian network |
| CN113852406A (en) * | 2021-08-24 | 2021-12-28 | 合肥工业大学 | Multi-beam relay satellite task scheduling method and device |
| CN115293451A (en) * | 2022-08-24 | 2022-11-04 | 中国西安卫星测控中心 | Resource dynamic scheduling method based on deep reinforcement learning |
| CN115314500A (en) * | 2022-07-28 | 2022-11-08 | 三峡大学 | Dynamic load balancing method based on improved TOPSIS model |
| CN116151039A (en) * | 2023-04-19 | 2023-05-23 | 中国西安卫星测控中心 | Distributed collaborative scheduling method and system based on random access task |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109934259B (en) * | 2019-01-30 | 2021-09-28 | 上海卫星工程研究所 | Satellite-borne working mode design method based on dual-mode control |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1445897A2 (en) * | 2002-12-23 | 2004-08-11 | Alcatel | Scheduling system and method for a burst switch |
| CN101739293A (en) * | 2009-12-24 | 2010-06-16 | 航天恒星科技有限公司 | Method for scheduling satellite data product production tasks in parallel based on multithread |
| CN102402461A (en) * | 2011-12-08 | 2012-04-04 | 曙光信息产业(北京)有限公司 | Balanced scheduling method based on operation scale |
| CN102594433A (en) * | 2011-12-14 | 2012-07-18 | 大连大学 | Dynamic priority dispatching method of satellite network many-to-one communication conflicts |
| CN105717940A (en) * | 2016-01-26 | 2016-06-29 | 中国空间技术研究院 | Autonomous task planning method of relay satellite |
-
2017
- 2017-01-26 CN CN201710057263.XA patent/CN107070534B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1445897A2 (en) * | 2002-12-23 | 2004-08-11 | Alcatel | Scheduling system and method for a burst switch |
| CN101739293A (en) * | 2009-12-24 | 2010-06-16 | 航天恒星科技有限公司 | Method for scheduling satellite data product production tasks in parallel based on multithread |
| CN102402461A (en) * | 2011-12-08 | 2012-04-04 | 曙光信息产业(北京)有限公司 | Balanced scheduling method based on operation scale |
| CN102594433A (en) * | 2011-12-14 | 2012-07-18 | 大连大学 | Dynamic priority dispatching method of satellite network many-to-one communication conflicts |
| CN105717940A (en) * | 2016-01-26 | 2016-06-29 | 中国空间技术研究院 | Autonomous task planning method of relay satellite |
Non-Patent Citations (1)
| Title |
|---|
| 张彦等: "中继卫星动态调度问题研究", 《系统仿真学报》 * |
Cited By (25)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107608793A (en) * | 2017-09-13 | 2018-01-19 | 航天恒星科技有限公司 | The Agile satellite task planning method of fast search |
| CN107608793B (en) * | 2017-09-13 | 2021-08-13 | 航天恒星科技有限公司 | Quick search agile satellite task planning method |
| CN107678848B (en) * | 2017-10-17 | 2020-07-07 | 合肥工业大学 | Method and device for constructing relay satellite task scheduling constraint planning model |
| CN107678848A (en) * | 2017-10-17 | 2018-02-09 | 合肥工业大学 | Repeater satellite task scheduling constrains the construction method and device of plan model |
| CN108053051A (en) * | 2017-11-16 | 2018-05-18 | 西安电子科技大学 | Task based access control splits the data relay satellite system mission planning method of polymerization |
| CN107909207A (en) * | 2017-11-16 | 2018-04-13 | 北京卫星信息工程研究所 | Remote Sensing Ground Station Multiobjective Optimal Operation method and system |
| CN108053051B (en) * | 2017-11-16 | 2021-08-10 | 西安电子科技大学 | Task planning method of data relay satellite system based on task splitting and aggregation |
| CN108199763A (en) * | 2017-12-29 | 2018-06-22 | 清华大学 | Repeater satellite method for scheduling task and system |
| CN108322249A (en) * | 2018-01-22 | 2018-07-24 | 新疆联海创智信息科技有限公司 | The restructural constellation satellite networking decision-making technique and system of oriented mission driving |
| CN108322249B (en) * | 2018-01-22 | 2020-09-08 | 新疆联海创智信息科技有限公司 | Task-driven reconfigurable constellation satellite network access decision method and system |
| CN108494470A (en) * | 2018-02-05 | 2018-09-04 | 西安电子科技大学 | Information Network repeater satellite antenna scheduling method based on optimization weight |
| CN108494470B (en) * | 2018-02-05 | 2020-08-18 | 西安电子科技大学 | Space information network relay satellite antenna scheduling method based on optimized weight |
| CN109190938A (en) * | 2018-08-17 | 2019-01-11 | 中南大学 | The repeater satellite single access antenna dynamic dispatching method reached for temporary duty |
| CN109190938B (en) * | 2018-08-17 | 2022-04-19 | 中南大学 | Relay satellite single-address antenna dynamic scheduling method for temporary task arrival |
| CN110113212B (en) * | 2019-05-16 | 2020-10-13 | 中南大学 | Relay satellite single-address antenna scheduling method suitable for breakpoint continuous transmission |
| CN110113212A (en) * | 2019-05-16 | 2019-08-09 | 中南大学 | A kind of repeater satellite single access antenna dispatching method suitable for breakpoint transmission |
| CN113393146A (en) * | 2021-06-25 | 2021-09-14 | 重庆大学 | Multi-stage flexible test system reliability evaluation method based on Bayesian network |
| CN113393146B (en) * | 2021-06-25 | 2022-12-20 | 重庆大学 | Multi-stage flexible test system reliability evaluation method based on Bayesian network |
| CN113852406A (en) * | 2021-08-24 | 2021-12-28 | 合肥工业大学 | Multi-beam relay satellite task scheduling method and device |
| CN113852406B (en) * | 2021-08-24 | 2024-04-30 | 合肥工业大学 | Multi-beam relay satellite task scheduling method and device |
| CN115314500A (en) * | 2022-07-28 | 2022-11-08 | 三峡大学 | Dynamic load balancing method based on improved TOPSIS model |
| CN115314500B (en) * | 2022-07-28 | 2024-01-16 | 三峡大学 | Dynamic load balancing method based on improved TOPSIS model |
| CN115293451A (en) * | 2022-08-24 | 2022-11-04 | 中国西安卫星测控中心 | Resource dynamic scheduling method based on deep reinforcement learning |
| CN115293451B (en) * | 2022-08-24 | 2023-06-16 | 中国西安卫星测控中心 | Resource dynamic scheduling method based on deep reinforcement learning |
| CN116151039A (en) * | 2023-04-19 | 2023-05-23 | 中国西安卫星测控中心 | Distributed collaborative scheduling method and system based on random access task |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107070534B (en) | 2018-11-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107070534B (en) | A kind of the dynamic preemptive type method for scheduling task and system of repeater satellite load balancing | |
| CN109388484B (en) | Multi-resource cloud job scheduling method based on Deep Q-network algorithm | |
| Nakasuka et al. | Dynamic scheduling system utilizing machine learning as a knowledge acquisition tool | |
| CN112766813A (en) | Air-space cooperative observation complex task scheduling method and system | |
| CN111010294A (en) | Electric power communication network routing method based on deep reinforcement learning | |
| CN107045667B (en) | Measurement and control data transmission integrated station network resource scheduling method | |
| CN113190351B (en) | Efficient resource distribution system for distributed deep learning training task | |
| CN110113212A (en) | A kind of repeater satellite single access antenna dispatching method suitable for breakpoint transmission | |
| CN116629511A (en) | Multi-star dynamic task planning method and device based on two-stage hybrid scheduling in uncertain environment | |
| CN117055619A (en) | Unmanned aerial vehicle scheduling method based on multi-agent reinforcement learning | |
| CN116939866A (en) | Wireless federal learning efficiency improving method based on collaborative computing and resource allocation joint optimization | |
| Wei et al. | Dynamic beam scheduling of multibeam low earth orbit satellites based on an enhanced artificial bee colony algorithm | |
| CN114154685A (en) | Electric energy data scheduling method in smart power grid | |
| CN106936905B (en) | Openstack-based scheduling method and scheduling system of Nova component virtual machine | |
| CN113891466B (en) | Online scheduling system and method for UDL tasks in edge wireless network | |
| Lin et al. | Hypergraph-based autonomous networks: adaptive resource management and dynamic resource scheduling | |
| CN115454585A (en) | Adaptive batch processing and parallel scheduling system for deep learning model inference of edge equipment | |
| Chen et al. | New two-stage deep reinforcement learning for task admission and channel allocation of wireless-powered mobile edge computing | |
| CN113946440A (en) | Resource scheduling method in green cloud environment | |
| Wang et al. | A queueing framework for channel allocation protocol in multi-user multi-channel cognitive radio network | |
| Meng et al. | Clustering based distributed multi-robot task allocation algorithm in large-scale systems | |
| CN116909717B (en) | Task scheduling method | |
| CN116307241B (en) | Distributed job shop scheduling method based on reinforcement learning with constraint multiple agents | |
| CN113347593B (en) | Relay selection method of unmanned aerial vehicle | |
| Shaojun et al. | Research on resource scheduling algorithms for satellite earth stations |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |