Service dynamic matching time-frequency resource scheduling method based on adaptive control

Abstract

The invention discloses a service dynamic matching space-time-frequency resource scheduling method based on adaptive control, which belongs to the technical field of space-time-frequency resource scheduling, and comprises the following steps of: constructing a resource scheduling evaluation function, acquiring a service queuing queue, and coding by adopting a space-time-frequency resource to be scheduled on the basis of the service queuing queue; according to the method, multiple resource scheduling parameter codes are obtained, local cosine optimization, local joint optimization, local rapid optimization and global jump optimization are carried out on the resource scheduling parameter codes for multiple times to obtain the optimal resource scheduling parameter code, and space-time-frequency resource scheduling can be finally carried out according to the optimal resource scheduling parameter code. The resource arrangement utilization rate is improved, and network congestion can be effectively avoided.

Description

Technical Field

[0001] The present invention belongs to the technical field of spatio-temporal-frequency resource scheduling, and particularly relates to a spatio-temporal-frequency resource scheduling method for dynamic matching of services based on adaptive control. Background Art

[0002] Spatio-temporal-frequency resources refer to the general term for frequency, time, and spatial location resources used for transmitting information in a communication network. These resources are the basic elements in a wireless communication system and are crucial for ensuring the reliability, effectiveness, and quality of service of communication. With the rapid development of communication technologies, the demand for spatio-temporal-frequency resources by various services is increasing day by day. Existing resource scheduling methods are difficult to meet the requirements of dynamic service changes, resulting in low resource utilization and serious network congestion. Summary of the Invention

[0003] The present invention provides a spatio-temporal-frequency resource scheduling method for dynamic matching of services based on adaptive control to solve the problems of low resource utilization and serious network congestion in the prior art.

[0004] A spatio-temporal-frequency resource scheduling method for dynamic matching of services based on adaptive control includes:

[0005] Constructing a resource scheduling evaluation function and obtaining a service queue, and based on the service queue, encoding with the spatio-temporal-frequency resources to be scheduled to obtain multiple resource scheduling parameter encodings;

[0006] Performing local cosine optimization, local joint optimization, local fast optimization, and global jump optimization on the resource scheduling parameter encodings multiple times to obtain the optimal resource scheduling parameter encoding;

[0007] Decoding the optimal resource scheduling parameter encoding to obtain the actual scheduling value of the spatio-temporal-frequency resources to be scheduled, and scheduling the spatio-temporal-frequency resources of the service queue according to the actual scheduling value of the spatio-temporal-frequency resources to be scheduled.

[0008] Further, the constructed resource scheduling evaluation function is:

[0009]

[0010] where Q(t + 1) represents the length of the communication service queue of all users at scheduling time t + 1, Q(t) represents the length of the communication service queue of all users at scheduling time t, C(t) represents the transmission rate allocation, γ(t) represents the beam scheduling allocation, T(t) represents the time allocation, S n (C(t), γ(t), T(t)) represents the queue service rate under the transmission rate allocation, beam scheduling allocation, and time allocation, a n (t) represents the service queue of the nth user, and N represents the total number of users.

[0011] Further, based on the service queuing queue, encoding is performed using the space-time-frequency resources to be scheduled, and multiple resource scheduling parameter encodings are obtained, including:

[0012] For the service queuing queue corresponding to any user, randomly allocate space-time-frequency resource values within the upper and lower limits of the space-time-frequency resources to be scheduled;

[0013] Encode the space-time-frequency resource values corresponding to the space-time-frequency resources to be scheduled into vectors to obtain resource scheduling parameter encodings, and obtain multiple different resource scheduling parameter encodings.

[0014] Further, perform multi-information fusion optimization, local cooperation optimization, local cosine optimization, and global mutation optimization on the resource scheduling parameter encodings multiple times to obtain the optimal resource scheduling parameter encoding, including:

[0015] Set the optimization times counter k = 1 and the maximum optimization times as K;

[0016] According to the resource scheduling evaluation function, obtain the evaluation function values corresponding to each resource scheduling parameter encoding, and according to the evaluation function values corresponding to each resource scheduling parameter encoding, obtain the optimal resource scheduling parameter encoding, the first relatively optimal parameter encoding, and the second relatively optimal parameter encoding;

[0017] According to the optimal resource scheduling parameter encoding, the first relatively optimal parameter encoding, and the second relatively optimal parameter encoding, perform multi-information fusion optimization on the resource scheduling parameter encoding to obtain the resource scheduling parameter encoding after multi-information fusion optimization;

[0018] Determine the historical optimal value of each resource scheduling parameter encoding during the k -th optimization process. For any resource scheduling parameter encoding after multi-information fusion optimization, randomly determine another historical optimal value, and perform local cooperation optimization on the resource scheduling parameter encoding according to the determined other historical optimal value to obtain the resource scheduling parameter encoding after local cooperation optimization;

[0019] For the resource scheduling parameter encoding after local cooperation optimization, randomly match another resource scheduling parameter encoding, and perform local cosine optimization on the resource scheduling parameter encoding according to the other resource scheduling parameter encoding using the pre - oscillation search strategy to obtain the resource scheduling parameter encoding after local cosine optimization;

[0020] For the resource scheduling parameter encoding after local cosine optimization, perform global mutation optimization on the resource scheduling parameter encoding using the multi - historical optimal value fusion mutation strategy to obtain the resource scheduling parameter encoding after global mutation optimization;

[0021] Determine whether the count value of the optimization times counter k is greater than or equal to the maximum optimization times K. If so, re-determine the optimal resource scheduling parameter encoding based on the resource scheduling parameter encoding after global mutation optimization, and output the optimal resource scheduling parameter encoding. Otherwise, increment the count value of the optimization times counter k by one, and return to the step of obtaining the evaluation function value.

[0022] Furthermore, according to the resource scheduling evaluation function, obtain the evaluation function value corresponding to each resource scheduling parameter encoding, and based on the evaluation function value corresponding to each resource scheduling parameter encoding, obtain the optimal resource scheduling parameter encoding, the first relatively optimal parameter encoding, and the second relatively optimal parameter encoding, including:

[0023] According to the resource scheduling evaluation function, obtain the evaluation function value corresponding to each resource scheduling parameter encoding;

[0024] Determine the resource scheduling parameter encoding with the smallest evaluation function value as the optimal resource scheduling parameter encoding, determine the resource scheduling parameter encoding with the second smallest evaluation function value as the first relatively optimal parameter encoding, and determine the resource scheduling parameter encoding with the third smallest evaluation function value as the second relatively optimal parameter encoding.

[0025] Furthermore, according to the optimal resource scheduling parameter encoding, the first relatively optimal parameter encoding, and the second relatively optimal parameter encoding, perform multi-information fusion optimization on the resource scheduling parameter encoding to obtain the resource scheduling parameter encoding after multi-information fusion optimization, including:

[0026]

[0027] Among them, α 1 represents the first random resource scheduling parameter encoding, α 2 represents the second random resource scheduling parameter encoding, α 3 represents the third random resource scheduling parameter encoding. The dimensions of the first random resource scheduling parameter encoding, the second random resource scheduling parameter encoding, and the third random resource scheduling parameter encoding are the same as the dimension of the resource scheduling parameter encoding, and each dimension is randomly generated. k represents the current optimization times, K represents the preset maximum optimization times, r represents the first random number between (0, 1), β 1 represents the first coefficient, β 1 represents the second coefficient, β 2 represents the third coefficient, and the first coefficient, the second coefficient, and the third coefficient are all generated by 2r 3 r 2 represents the second random number between (0, 1), 2 represents the optimal resource scheduling parameter encoding, represents the first relatively optimal resource scheduling parameter encoding, Denote the second relatively optimal resource scheduling parameter encoding, Denote the historical optimal value corresponding to the i-th resource scheduling parameter encoding, Denote the i-th resource scheduling parameter encoding after guided update, where i = 1, 2, …, NP, NP represents the total number of resource scheduling parameter encodings, and |*| represents the modulo operation.

[0028] Furthermore, determine the historical optimal value of each resource scheduling parameter encoding during k optimization processes. For any resource scheduling parameter encoding after multi-information fusion optimization, randomly determine another historical optimal value, and perform local collaborative optimization on the resource scheduling parameter encoding according to the determined other historical optimal value to obtain the resource scheduling parameter encoding after local collaborative optimization, including:

[0029] Determine the historical optimal value of each resource scheduling parameter encoding during k optimization processes;

[0030] For any resource scheduling parameter encoding after multi-information fusion optimization, randomly determine another historical optimal value

[0031]

[0032] Perform local collaborative optimization on the resource scheduling parameter encoding according to the determined other historical optimal value, and the resource scheduling parameter encoding after local collaborative optimization is:

[0033]

[0034] Among them, Denote the historical optimal value corresponding to the h-th resource scheduling parameter encoding after multi-information fusion optimization, Denote the other historical optimal value, Denote the resource scheduling parameter encoding after local collaborative optimization α 4 Denote the fourth random vector, and it has the same dimension as the resource scheduling parameter encoding, and each dimension is randomly generated by r 3 Denote a random number between (0, 1); β 4 Denote the fourth coefficient, and it is generated by 2r 4 r 4 Denote a random number between (0, 1).

[0035] Furthermore, for the resource scheduling parameter encoding after local collaborative optimization, randomly match another resource scheduling parameter encoding, and perform local cosine optimization on the resource scheduling parameter encoding according to the other resource scheduling parameter encoding using the pre-shock search strategy to obtain the resource scheduling parameter encoding after local cosine optimization, including:

[0036] Encoding of resource scheduling parameters after local collaboration optimization Randomly match an encoding of other resource scheduling parameters

[0037]

[0038] According to the encoding of other resource scheduling parameters Adopt a pre - oscillation search strategy to perform local cosine optimization on the encoding of resource scheduling parameters, and the encoding of resource scheduling parameters after local cosine optimization is:

[0039]

[0040] wherein, represents the d - th dimension parameter of the n - th encoding of resource scheduling parameters after local collaboration optimization in the k - th optimization process, n = 1, 2, …, NP, d = 1, 2, …, D, and D represents the total dimension of the encoding of resource scheduling parameters represents the d - th dimension parameter of the n - th encoding of resource scheduling parameters after local cosine optimization, π represents the pi, r 5 represents a random number between (0, 1), r 6 represents a random number between (0, 1), and cos represents the cosine function

[0041] Furthermore, for the encoding of resource scheduling parameters after local cosine optimization, adopt a multi - historical optimal value fusion mutation strategy to perform global mutation optimization on the encoding of resource scheduling parameters, and the encoding of resource scheduling parameters after global mutation optimization includes:

[0042] For the encoding of resource scheduling parameters after local cosine optimization, determine L target resource scheduling parameter encodings with the smallest evaluation function values; where L is less than NP

[0043] For any one of the target resource scheduling parameter encodings, perform global mutation optimization on the encoding of resource scheduling parameters as:

[0044]

[0045] wherein, represents the m - th target resource scheduling parameter encoding after global mutation optimization represents the optimal resource scheduling parameter encoding represents the first relatively optimal resource scheduling parameter encoding represents the second relatively optimal resource scheduling parameter encoding represents the first random value among the historical optimal values of all resource scheduling parameter encodings represents the second random value among the historical optimal values of all resource scheduling parameter encodings, c1 represents the first scaling factor, c 2 represents the second scaling factor, r 7 Represents a random number between (0,1), r 8 Represents a random number between (0,1);

[0046] For the target resource scheduling parameter code, when the evaluation function value corresponding to the target resource scheduling parameter code after global variation optimization decreases, the target resource scheduling parameter code after global variation optimization is used as the resource scheduling parameter code after global variation optimization, otherwise the original target resource scheduling parameter code is used as the resource scheduling parameter code after global variation optimization;

[0047] For non-target resource scheduling parameter coding, the original target resource scheduling parameter coding is used as the resource scheduling parameter coding after global variation optimization.

[0048] Furthermore, after the multi-information fusion optimization, local collaborative optimization, local cosine optimization and global variation optimization, it also includes: performing out-of-bounds processing on the resource scheduling parameter encoding.

[0049] The present invention provides a method for scheduling space-time and frequency resources for dynamic matching of services based on adaptive control. By constructing a resource scheduling evaluation function and obtaining a service queue, the space-time and frequency resources to be scheduled are encoded based on the service queue, and multiple resource scheduling parameter codes are obtained. The resource scheduling parameter codes are subjected to local cosine optimization, local joint optimization, local fast optimization and global jump optimization for multiple times to obtain the optimal resource scheduling parameter code. Finally, the space-time and frequency resources can be scheduled according to the optimal resource scheduling parameter code, thereby improving the utilization rate of the sorting resources and effectively avoiding network congestion. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0051] Figure 1 A flowchart of a method for dynamically matching time, space and frequency resources for services based on adaptive control is provided in an embodiment of the present invention.

[0052] The above drawings have shown clear embodiments of the present invention, which will be described in more detail below. These drawings and text descriptions are not intended to limit the scope of the present invention in any way, but to illustrate the concept of the present invention for those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0053] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Instead, they are merely examples of devices and methods consistent with some aspects of the present invention as detailed in the appended claims.

[0054] The embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0055] like Figure 1 As shown, an embodiment of the present invention provides a method for scheduling time, space and frequency resources for dynamic matching of services based on adaptive control, including:

[0056] S101, constructing a resource scheduling evaluation function and obtaining a service queue, and based on the service queue, using the time-space-frequency resources to be scheduled for encoding to obtain multiple resource scheduling parameter codes;

[0057] The resource scheduling parameter encoding can be evaluated through the resource scheduling evaluation function, so that the time-space-frequency resource scheduling can be optimized based on this. For the service queue sequence corresponding to any user, the time-space-frequency resources can be allocated within a preset interval. Therefore, based on this, the time-space-frequency resource scheduling can be optimized in combination with the intelligent optimization algorithm.

[0058] S102, performing local cosine optimization, local joint optimization, local fast optimization, and global jump optimization on the resource scheduling parameter code multiple times to obtain the optimal resource scheduling parameter code;

[0059] Based on multiple resource scheduling parameter codes, an intelligent optimization algorithm can be used to perform coding optimization, so that an optimal resource scheduling solution (i.e., optimal resource scheduling parameter code) can be determined. An embodiment of the present invention provides an intelligent optimization algorithm, including optimal resource scheduling parameter code, which can effectively find a global optimal solution in the solution space, thereby achieving optimal time-space-frequency resource scheduling.

[0060] S103, decoding the optimal resource scheduling parameter code to obtain the actual scheduling value of the time-space-frequency resources to be scheduled, and scheduling the time-space-frequency resources of the service queue according to the actual scheduling value of the time-space-frequency resources to be scheduled.

[0061] The embodiments of the present invention combine intelligent optimization algorithms to schedule time, space and frequency resources, which can effectively improve the utilization efficiency of time, space and frequency resources, thereby improving network smoothness.

[0062] In the embodiment of the present invention, the resource scheduling evaluation function is constructed as follows:

[0063]

[0064] Where Q(t+1) represents the length of the communication service queue of all users at the scheduling time t+1, Q(t) represents the length of the communication service queue of all users at the scheduling time t, C(t) represents the transmission rate allocation, γ(t) represents the beam scheduling allocation, T(t) represents the time allocation, S n (C(t),γ(t),T(t)) represents the queue service rate under transmission rate allocation, beam scheduling allocation and time allocation, a n (t) represents the service queue of the nth user, and N represents the total number of users.

[0065] The queue service rate can be obtained through deep learning algorithms or existing technologies. Therefore, after specifying the space-time and frequency resource allocation (such as transmission rate allocation, beam scheduling allocation, and time allocation), the remaining communication service queue length can be determined when the next service cycle arrives. Therefore, this can be used as an optimization condition to maximize resource utilization and improve overall network utilization efficiency.

[0066] It is worth noting that, in addition to the above functions, the evaluation function of time-space-frequency resources can also use some existing technologies for evaluation, and the types of time-space-frequency resources are not limited to the example types in the embodiments of the present invention.

[0067] In the embodiment of the present invention, based on the service queue, the time-space-frequency resources to be scheduled are used for encoding to obtain multiple resource scheduling parameter codes, including:

[0068] For the service queue corresponding to any user, randomly allocate the time-space-frequency resource value within the upper and lower limits of the time-space-frequency resources to be scheduled;

[0069] For example, the space-time and frequency resource set allocated to user A is [transmission rate allocation a1, beam scheduling allocation a2, time allocation a3], and the space-time and frequency resource set allocated to user B is [transmission rate allocation b1, beam scheduling allocation b2, time allocation b3], then a resource scheduling parameter encoding can be [a1, a2, a3, b1, b2, b3].

[0070] The time-space-frequency resource values ​​corresponding to the time-space-frequency resources to be scheduled are encoded into a vector to obtain a resource scheduling parameter code, and multiple different resource scheduling parameter codes are obtained.

[0071] In an embodiment of the present invention, the resource scheduling parameter encoding is subjected to multiple information fusion optimization, local collaborative optimization, local cosine optimization, and global variation optimization to obtain the optimal resource scheduling parameter encoding, including:

[0072] Set the optimization times counter k=1 and the maximum optimization times to K;

[0073] According to the resource scheduling evaluation function, obtaining the evaluation function value corresponding to each resource scheduling parameter code, and according to the evaluation function value corresponding to each resource scheduling parameter code, obtaining the optimal resource scheduling parameter code, the first relatively optimal parameter code and the second relatively optimal parameter code;

[0074] According to the optimal resource scheduling parameter code, the first relatively optimal parameter code and the second relatively optimal parameter code, the resource scheduling parameter code is optimized by multi-information fusion to obtain the resource scheduling parameter code after multi-information fusion optimization;

[0075] Determine the historical optimal value of each resource scheduling parameter code in the k-times optimization process, randomly determine another historical optimal value for any resource scheduling parameter code after multi-information fusion optimization, and perform local collaborative optimization on the resource scheduling parameter code according to the other determined historical optimal value to obtain the resource scheduling parameter code after local collaborative optimization;

[0076] For the resource scheduling parameter code after local collaborative optimization, randomly match another resource scheduling parameter code, and according to the other resource scheduling parameter code, use the pre-oscillation search strategy to perform local cosine optimization on the resource scheduling parameter code to obtain the resource scheduling parameter code after local cosine optimization;

[0077] For the resource scheduling parameter coding after local cosine optimization, a multi-historical optimal value fusion mutation strategy is used to perform global mutation optimization on the resource scheduling parameter coding to obtain the resource scheduling parameter coding after global mutation optimization;

[0078] Determine whether the count value of the optimization times counter k is greater than or equal to the maximum optimization times K. If so, redetermine the optimal resource scheduling parameter code based on the resource scheduling parameter code after global variation optimization, and output the optimal resource scheduling parameter code. Otherwise, add one to the count value of the optimization times counter k and return to the step of obtaining the evaluation function value.

[0079] The intelligent optimization algorithm provided by the embodiment of the present invention can find the best time-space-frequency resource scheduling solution in the solution space, thereby improving the scheduling capability of time-space-frequency resources and ensuring smooth network operation.

[0080] In an embodiment of the present invention, according to the resource scheduling evaluation function, obtaining the evaluation function value corresponding to each resource scheduling parameter code, and according to the evaluation function value corresponding to each resource scheduling parameter code, obtaining the optimal resource scheduling parameter code, the first preferred parameter code and the second preferred parameter code, including:

[0081] According to the resource scheduling evaluation function, obtain the evaluation function value corresponding to each resource scheduling parameter code;

[0082] The resource scheduling parameter code with the smallest evaluation function value is determined as the optimal resource scheduling parameter code, the resource scheduling parameter code with the second smallest evaluation function value is determined as the first preferred parameter code, and the resource scheduling parameter code with the third smallest evaluation function value is determined as the second preferred parameter code.

[0083] In an embodiment of the present invention, according to the optimal resource scheduling parameter code, the first preferred parameter code and the second preferred parameter code, the resource scheduling parameter code is optimized by multi-information fusion to obtain the resource scheduling parameter code after multi-information fusion optimization, including:

[0084]

[0085] Among them, α 1 represents the first random resource scheduling parameter encoding, α 2 represents the second random resource scheduling parameter encoding, α 3 represents the third random resource scheduling parameter code, the dimensions of the first random resource scheduling parameter code, the second random resource scheduling parameter code and the third random resource scheduling parameter code are the same as the dimension of the resource scheduling parameter code, and each dimension is represented by Randomly generated, k represents the current optimization times, K represents the preset maximum optimization times, r 1 represents the first random number between (0,1), β 1 represents the first coefficient, β 2 represents the second coefficient, β 3 represents the third coefficient, and the first coefficient, the second coefficient and the third coefficient are all through 2r 2 Produce, r 2 represents the second random number between (0,1), represents the optimal resource scheduling parameter encoding, represents the first optimal resource scheduling parameter encoding, represents the second best resource scheduling parameter encoding, represents the historical optimal value corresponding to the i-th resource scheduling parameter encoding, It represents the i-th resource scheduling parameter code after the boot update, i=1,2,…,NP, NP represents the total number of resource scheduling parameter codes, and |*| represents a modulo operation.

[0086] The embodiment of the present invention can enable resource scheduling parameter encoding to be quickly optimized according to a better position, which helps to improve algorithm accuracy and convergence speed.

[0087] In an embodiment of the present invention, a historical optimal value of each resource scheduling parameter code in the k-times optimization process is determined, and for any resource scheduling parameter code after multi-information fusion optimization, another historical optimal value is randomly determined, and local collaborative optimization is performed on the resource scheduling parameter code according to the other determined historical optimal value to obtain the resource scheduling parameter code after local collaborative optimization, including:

[0088] Determine the historical optimal value of each resource scheduling parameter encoding in the k optimization processes;

[0089] For any resource scheduling parameter encoding after multi-information fusion optimization, randomly determine another historical optimal value

[0090]

[0091] The resource scheduling parameter encoding is locally collaboratively optimized according to other determined historical optimal values, and the resource scheduling parameter encoding after local collaborative optimization is obtained as follows:

[0092]

[0093] in, represents the historical optimal value corresponding to the resource scheduling parameter encoding after the h-th multi-information fusion optimization, represents other historical optimal values, Represents the resource scheduling parameter encoding after local collaborative optimization α 4 represents the fourth random vector, and its dimension is the same as the resource scheduling parameter encoding, and each dimension is represented by Randomly generated, r 3 Represents a random number between (0,1); β 4 represents the fourth coefficient, and it is expressed by 2r 4 Produce, r 4 Represents a random number between (0,1).

[0094] The embodiment of the present invention performs collaborative optimization through historical optimal values, which can greatly increase the local optimization capability and optimization speed of the algorithm.

[0095] In an embodiment of the present invention, for a resource scheduling parameter code after local collaborative optimization, a random match is made to another resource scheduling parameter code, and according to the other resource scheduling parameter code, a pre-oscillation search strategy is used to perform local cosine optimization on the resource scheduling parameter code to obtain the resource scheduling parameter code after local cosine optimization, including:

[0096] Resource scheduling parameter encoding after local collaboration optimization Randomly match another resource scheduling parameter encoding

[0097]

[0098] Encoding based on other resource scheduling parameters The resource scheduling parameter encoding is locally cosine optimized using the pre-oscillation search strategy, and the resource scheduling parameter encoding after local cosine optimization is obtained as follows:

[0099]

[0100] in, represents the d-th dimension parameter of the resource scheduling parameter encoding after the n-th local collaborative optimization in the k-th optimization process, n = 1, 2, ..., NP, d = 1, 2, ..., D, D represents the total dimension of the resource scheduling parameter encoding, represents the d-th dimension parameter of the resource scheduling parameter encoding after the n-th local cosine optimization, π represents the circumference of a circle, and r 5 Represents a random number between (0,1), r 6 Represents a random number between (0,1), and cos represents the cosine function.

[0101] The local cosine optimization provided by the embodiment of the present invention can enable the algorithm to have more local exploration capabilities, which helps the algorithm to escape from the local optimum.

[0102] In the embodiment of the present invention, for the resource scheduling parameter coding after local cosine optimization, a multi-historical optimal value fusion mutation strategy is adopted to perform global mutation optimization on the resource scheduling parameter coding to obtain the resource scheduling parameter coding after global mutation optimization, including:

[0103] For the resource scheduling parameter codes after local cosine optimization, determine L target resource scheduling parameter codes with the smallest evaluation function value; where L is less than NP;

[0104] For any target resource scheduling parameter encoding, the resource scheduling parameter encoding is globally mutated and optimized as follows:

[0105]

[0106] in, represents the target resource scheduling parameter encoding after the mth global mutation optimization, represents the optimal resource scheduling parameter encoding, represents the first optimal resource scheduling parameter encoding, represents the second best resource scheduling parameter encoding, represents the first random value among the historical optimal values ​​of all resource scheduling parameter encodings, represents the second random value among the historical optimal values ​​of all resource scheduling parameter codes, c 1represents the first scaling factor, c 2 represents the second scaling factor, r 7 Represents a random number between (0,1), r 8 Represents a random number between (0,1);

[0107] The global variation optimization provided by the embodiment of the present invention can realize global search with information encoded by different resource scheduling parameters, which can effectively prevent the algorithm from falling into local optimum and ultimately improve the time-space-frequency resource scheduling capability.

[0108] For the target resource scheduling parameter code, when the evaluation function value corresponding to the target resource scheduling parameter code after global variation optimization decreases, the target resource scheduling parameter code after global variation optimization is used as the resource scheduling parameter code after global variation optimization, otherwise the original target resource scheduling parameter code is used as the resource scheduling parameter code after global variation optimization;

[0109] For non-target resource scheduling parameter coding, the original target resource scheduling parameter coding is used as the resource scheduling parameter coding after global variation optimization.

[0110] In the embodiment of the present invention, after the multi-information fusion optimization, local collaborative optimization, local cosine optimization and global variation optimization, it also includes: performing out-of-bounds processing on the resource scheduling parameter encoding.

[0111] The present invention provides a method for scheduling space-time and frequency resources for dynamic matching of services based on adaptive control. By constructing a resource scheduling evaluation function and obtaining a service queue, the space-time and frequency resources to be scheduled are encoded based on the service queue, and multiple resource scheduling parameter codes are obtained. The resource scheduling parameter codes are subjected to local cosine optimization, local joint optimization, local fast optimization and global jump optimization for multiple times to obtain the optimal resource scheduling parameter code. Finally, the space-time and frequency resources can be scheduled according to the optimal resource scheduling parameter code, thereby improving the utilization rate of the sorting resources and effectively avoiding network congestion.

[0112] Those skilled in the art will readily appreciate other embodiments of the present invention after considering the specification and practicing the invention disclosed herein. The present invention is intended to cover any variations, uses or adaptations of the present invention that follow the general principles of the present invention and include common knowledge or customary techniques in the art that are not disclosed by the present invention. It should be understood that the present invention is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from the scope thereof. The scope of the present invention is limited only by the appended claims.

Claims

1. A method for scheduling time, space and frequency resources based on adaptive control and dynamic matching of services, characterized in that: include: Constructing a resource scheduling evaluation function and obtaining a service queue, and based on the service queue, using the time-space-frequency resources to be scheduled for encoding to obtain multiple resource scheduling parameter codes; The resource scheduling parameter encoding is repeatedly subjected to local cosine optimization, local joint optimization, local fast optimization, and global jump optimization to obtain the optimal resource scheduling parameter encoding; The optimal resource scheduling parameter code is decoded to obtain the actual scheduling value of the time-space-frequency resources to be scheduled, and the time-space-frequency resources of the service queuing queue are scheduled according to the actual scheduling value of the time-space-frequency resources to be scheduled.

2. The method for scheduling time, space and frequency resources based on adaptive control of dynamic matching of services according to claim 1, characterized in that: The resource scheduling evaluation function is constructed as follows: Where Q(t+1) represents the length of the communication service queue of all users at the scheduling time t+1, Q(t) represents the length of the communication service queue of all users at the scheduling time t, C(t) represents the transmission rate allocation, γ(t) represents the beam scheduling allocation, T(t) represents the time allocation, S n (C(t),γ(t),T(t)) represents the queue service rate under transmission rate allocation, beam scheduling allocation and time allocation, a n (t) represents the service queue of the nth user, and N represents the total number of users.

3. The method for scheduling time, space and frequency resources based on adaptive control of dynamic matching of services according to claim 2 is characterized in that: Based on the service queue, the time, space and frequency resources to be scheduled are used for encoding to obtain multiple resource scheduling parameter codes, including: For the service queue corresponding to any user, randomly allocate the time-space-frequency resource value within the upper and lower limits of the time-space-frequency resources to be scheduled; The time-space-frequency resource values ​​corresponding to the time-space-frequency resources to be scheduled are encoded into a vector to obtain a resource scheduling parameter code, and multiple different resource scheduling parameter codes are obtained.

4. The method for scheduling time, space and frequency resources based on adaptive control of dynamic matching of services according to claim 3 is characterized in that: The resource scheduling parameter encoding is optimized multiple times by multi-information fusion optimization, local collaborative optimization, local cosine optimization, and global mutation optimization to obtain the optimal resource scheduling parameter encoding, including: Set the optimization times counter k=1 and the maximum optimization times to K; According to the resource scheduling evaluation function, obtaining the evaluation function value corresponding to each resource scheduling parameter code, and according to the evaluation function value corresponding to each resource scheduling parameter code, obtaining the optimal resource scheduling parameter code, the first relatively optimal parameter code and the second relatively optimal parameter code; According to the optimal resource scheduling parameter code, the first relatively optimal parameter code and the second relatively optimal parameter code, the resource scheduling parameter code is optimized by multi-information fusion to obtain the resource scheduling parameter code after multi-information fusion optimization; Determine the historical optimal value of each resource scheduling parameter code in the k-times optimization process, randomly determine another historical optimal value for any resource scheduling parameter code after multi-information fusion optimization, and perform local collaborative optimization on the resource scheduling parameter code according to the other determined historical optimal value to obtain the resource scheduling parameter code after local collaborative optimization; For the resource scheduling parameter code after local collaborative optimization, randomly match another resource scheduling parameter code, and according to the other resource scheduling parameter code, use the pre-oscillation search strategy to perform local cosine optimization on the resource scheduling parameter code to obtain the resource scheduling parameter code after local cosine optimization; For the resource scheduling parameter coding after local cosine optimization, a multi-historical optimal value fusion mutation strategy is used to perform global mutation optimization on the resource scheduling parameter coding to obtain the resource scheduling parameter coding after global mutation optimization; Determine whether the count value of the optimization times counter k is greater than or equal to the maximum optimization times K. If so, redetermine the optimal resource scheduling parameter code based on the resource scheduling parameter code after global variation optimization, and output the optimal resource scheduling parameter code. Otherwise, add one to the count value of the optimization times counter k and return to the step of obtaining the evaluation function value.

5. The method for scheduling time, space and frequency resources based on adaptive control of dynamic matching of services according to claim 4, characterized in that: According to the resource scheduling evaluation function, an evaluation function value corresponding to each resource scheduling parameter code is obtained, and according to the evaluation function value corresponding to each resource scheduling parameter code, an optimal resource scheduling parameter code, a first relatively optimal parameter code, and a second relatively optimal parameter code are obtained, including: According to the resource scheduling evaluation function, obtain the evaluation function value corresponding to each resource scheduling parameter code; The resource scheduling parameter code with the smallest evaluation function value is determined as the optimal resource scheduling parameter code, the resource scheduling parameter code with the second smallest evaluation function value is determined as the first preferred parameter code, and the resource scheduling parameter code with the third smallest evaluation function value is determined as the second preferred parameter code.

6. The method for scheduling time, space and frequency resources based on adaptive control for dynamic matching of services according to claim 5, characterized in that: According to the optimal resource scheduling parameter code, the first relatively optimal parameter code and the second relatively optimal parameter code, the resource scheduling parameter code is optimized by multi-information fusion to obtain the resource scheduling parameter code after the multi-information fusion optimization, including: Among them, α1 represents the first random resource scheduling parameter code, α2 represents the second random resource scheduling parameter code, α3 represents the third random resource scheduling parameter code, the dimensions of the first random resource scheduling parameter code, the second random resource scheduling parameter code and the third random resource scheduling parameter code are the same as the dimension of the resource scheduling parameter code, and each dimension is represented by Randomly generated, k represents the current optimization times, K represents the preset maximum optimization times, r1 represents the first random number between (0,1), β1 represents the first coefficient, β2 represents the second coefficient, β3 represents the third coefficient, and the first coefficient, the second coefficient and the third coefficient are all generated by 2r2, r2 represents the second random number between (0,1), represents the optimal resource scheduling parameter encoding, represents the first optimal resource scheduling parameter encoding, represents the second best resource scheduling parameter encoding, represents the historical optimal value corresponding to the i-th resource scheduling parameter encoding, It represents the i-th resource scheduling parameter code after the boot update, i=1,2,…,NP, NP represents the total number of resource scheduling parameter codes, and |*| represents a modulo operation.

7. The method for scheduling time, space and frequency resources based on adaptive control for dynamic matching of services according to claim 6, characterized in that: Determine the historical optimal value of each resource scheduling parameter code in the k-times optimization process, randomly determine another historical optimal value for any resource scheduling parameter code after multi-information fusion optimization, and perform local collaborative optimization on the resource scheduling parameter code according to the determined other historical optimal value, and obtain the resource scheduling parameter code after local collaborative optimization, including: Determine the historical optimal value of each resource scheduling parameter encoding in the k optimization processes; For any resource scheduling parameter encoding after multi-information fusion optimization, randomly determine another historical optimal value The resource scheduling parameter encoding is locally collaboratively optimized according to other determined historical optimal values, and the resource scheduling parameter encoding after local collaborative optimization is obtained as follows: in, represents the historical optimal value corresponding to the resource scheduling parameter encoding after the h-th multi-information fusion optimization, represents other historical optimal values, Represents the resource scheduling parameter encoding after local collaborative optimization h=1,2,…,NP;α4 represents the fourth random vector, and its dimension is the same as the resource scheduling parameter encoding, and each dimension is Randomly generated, r3 represents a random number between (0,1); β4 represents the fourth coefficient, and it is generated by 2r4, r4 represents a random number between (0,1).

8. The method for scheduling time, space and frequency resources based on adaptive control of dynamic matching of services according to claim 7, characterized in that: For the resource scheduling parameter code after local collaborative optimization, randomly match another resource scheduling parameter code, and perform local cosine optimization on the resource scheduling parameter code using a pre-oscillation search strategy based on the other resource scheduling parameter code, to obtain the resource scheduling parameter code after local cosine optimization, including: Resource scheduling parameter encoding after local collaboration optimization Randomly match another resource scheduling parameter code Encoding based on other resource scheduling parameters The resource scheduling parameter encoding is locally cosine optimized using the pre-oscillation search strategy, and the resource scheduling parameter encoding after local cosine optimization is obtained as follows: in, represents the d-th dimension parameter of the resource scheduling parameter encoding after the n-th local collaborative optimization in the k-th optimization process, n = 1, 2, ..., NP, d = 1, 2, ..., D, D represents the total dimension of the resource scheduling parameter encoding, It represents the d-th dimension parameter of the resource scheduling parameter encoding after the n-th local cosine optimization, π represents the pi, r5 represents a random number between (0,1), r6 represents a random number between (0,1), and cos represents the cosine function.

9. The method for scheduling time, space and frequency resources based on adaptive control of dynamic matching of services according to claim 8, characterized in that: For the resource scheduling parameter encoding after local cosine optimization, a multi-historical optimal value fusion mutation strategy is used to perform global mutation optimization on the resource scheduling parameter encoding, and the resource scheduling parameter encoding after global mutation optimization is obtained, including: For the resource scheduling parameter codes after local cosine optimization, determine L target resource scheduling parameter codes with the smallest evaluation function value; where L is less than NP; For any target resource scheduling parameter encoding, the resource scheduling parameter encoding is globally mutated and optimized as follows: in, represents the target resource scheduling parameter encoding after the mth global mutation optimization, represents the optimal resource scheduling parameter encoding, represents the first optimal resource scheduling parameter encoding, represents the second best resource scheduling parameter encoding, represents the first random value among the historical optimal values ​​of all resource scheduling parameter encodings, represents the second random value among the historical optimal values ​​of all resource scheduling parameter codes, c1 represents the first scaling factor, c2 represents the second scaling factor, r7 represents a random number between (0, 1), and r8 represents a random number between (0, 1); For the target resource scheduling parameter code, when the evaluation function value corresponding to the target resource scheduling parameter code after global variation optimization decreases, the target resource scheduling parameter code after global variation optimization is used as the resource scheduling parameter code after global variation optimization, otherwise the original target resource scheduling parameter code is used as the resource scheduling parameter code after global variation optimization; For non-target resource scheduling parameter coding, the original target resource scheduling parameter coding is used as the resource scheduling parameter coding after global variation optimization.

10. The method for scheduling time, space and frequency resources based on adaptive control of dynamic matching of services according to claim 9, characterized in that: After multi-information fusion optimization, local collaborative optimization, local cosine optimization and global variation optimization, it also includes: out-of-bounds processing of resource scheduling parameter encoding.

Images