Space-time-frequency combined adaptive resource management method

Abstract

The invention discloses a space-time-frequency joint adaptive resource management method, which belongs to the technical field of space-time-frequency resource management, codes space-time-frequency resource allocation by adopting a coding strategy of an intelligent optimization algorithm to obtain a plurality of resource allocation codes, and performs resource allocation on the basis of the plurality of resource allocation codes and a target optimization function. According to the method, resource allocation optimization is carried out by adopting an intelligent optimization algorithm to obtain an optimal space-time-frequency resource scheduling scheme, and finally, dynamic scheduling of space-time-frequency resources is carried out based on the optimal space-time-frequency resource scheduling scheme, a space-time-frequency resource proportioning relation is optimized, and the cross-functional-domain joint performance of communication radars in an integrated system is comprehensively improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of spatio-temporal-frequency resource management, and particularly relates to a spatio-temporal-frequency joint adaptive resource management method. Background Art

[0002] With the continuous emergence of emerging scenarios such as vehicle-to-everything (V2X) networks and unmanned aerial vehicle (UAV) networks, communication-sensing integrated systems have received extensive attention in recent years. Through integrated design, a communication-sensing integrated system realizes dual functions of data transmission and target detection based on the same hardware and spectrum resources, reducing hardware costs and improving system integration and spectrum resource utilization performance. During the operation of a vehicle network, there are often radar services and communication services that need to occupy spatio-temporal-frequency resources to work, so as to achieve communication-sensing integration. However, in the prior art, one or more spatio-temporal-frequency resources are often preset for simple adjustment, making it difficult to achieve dynamic adjustment of spatio-temporal-frequency resources, and thus difficult to achieve performance balance between radar services and communication services. Summary of the Invention

[0003] The present invention provides a spatio-temporal-frequency joint adaptive resource management method to solve the problem that in the prior art, one or more spatio-temporal-frequency resources are often preset for simple adjustment, making it difficult to achieve dynamic adjustment of spatio-temporal-frequency resources, and thus difficult to achieve performance balance between radar services and communication services.

[0004] A spatio-temporal-frequency joint adaptive resource management method includes:

[0005] For radar services and communication services, a joint optimization strategy is adopted to construct a spatio-temporal-frequency joint optimization function, and an objective optimization function for optimizing spatio-temporal-frequency resource allocation is obtained;

[0006] For the spatio-temporal-frequency resources of radar services and communication services, an encoding strategy of an intelligent optimization algorithm is adopted to encode the spatio-temporal-frequency resource allocation to obtain multiple resource allocation encodings;

[0007] Based on the multiple resource allocation encodings and the objective optimization function, an intelligent optimization algorithm is adopted to optimize resource allocation to obtain an optimal spatio-temporal-frequency resource scheduling scheme;

[0008] Based on the optimal spatio-temporal-frequency resource scheduling scheme, spatio-temporal-frequency resources are allocated to radar services and communication services to complete spatio-temporal-frequency joint adaptive resource management.

[0009] Further, the objective optimization function for optimizing spatio-temporal-frequency resource allocation is:

[0010]

[0011] Wherein, f represents the objective optimization function, Q n$(t)$ represents the length of the communication service queue of the $n$-th user at time $t$, and $a$ n $(t)$ represents the admitted traffic volume of the $n$-th user at time $t$, and $S$ n $(C$ n $(t), B$ n $(t), T$ n $(t))$ represents the communication queue service rate, and $S$ n $(C$ n $(t), B$ n $(t), T$ n $(t))$ is jointly determined by spatio-temporal-frequency resources, and the spatio-temporal-frequency resources include the transmission rate $C$ n $(t)$, beam scheduling $B$ n $(t)$ and time allocation $T$ n $(t)$, and $Y$ m $(t)$ represents the length of the detection virtual queue obtained by the $m$-th radar wave position at time $t$, and $\beta$ m $(t)$ represents the number of detections of the $m$-th radar wave position at time $t$, and $\varphi$ m $(B$ m $(t), T$ m $(t))$ represents the radar queue service rate, and $\varphi$ m $(B$ m $(t), T$ m $(t))$ is jointly determined by beam scheduling $B$ n $(t)$ and time allocation $T$ n $(t)$.

[0012] Furthermore, for the spatio-temporal-frequency resources of radar services and communication services, an encoding strategy of an intelligent optimization algorithm is adopted to encode the spatio-temporal-frequency resource allocation to obtain multiple resource allocation encodings, including:

[0013] For any radar service or communication service, perform random initialization within the upper and lower limits of spatio-temporal-frequency resources to obtain the spatio-temporal-frequency resource values corresponding to each radar service or communication service;

[0014] Encode the spatio-temporal-frequency resource values corresponding to all radar services and communication services into vectors to obtain resource allocation encodings, and repeat to obtain multiple different resource allocation encodings.

[0015] Furthermore, based on multiple said resource allocation encodings and a target optimization function, an intelligent optimization algorithm is used to optimize resource allocation to obtain an optimal spatio-temporal-frequency resource scheduling scheme, including:

[0016] Based on the target optimization function, obtain the target optimization function value corresponding to each resource allocation encoding, and determine the optimal resource allocation encoding according to the target optimization function value corresponding to each resource allocation encoding;

[0017] According to the optimal resource allocation coding, an information fusion strategy is adopted to perform information fusion on each resource allocation coding to obtain the resource allocation coding after information fusion;

[0018] An unknown region search is performed on the resource allocation coding after information fusion by adopting a coding position weighted guiding strategy to obtain the resource allocation coding after unknown region search;

[0019] According to the optimal resource allocation coding, a dual-optimal guiding strategy is adopted to perform an optimal search on the resource allocation coding after unknown region search to obtain the resource allocation coding after optimal search;

[0020] A global search is performed on the resource allocation coding after optimal search by adopting a mutation fusion strategy to obtain the resource allocation coding after global search;

[0021] Repeat information fusion, unknown region search, optimal search, and global search until the optimization end condition is satisfied, re-obtain the optimal resource allocation coding, and decode the optimal resource allocation coding to obtain the optimal spatio-temporal-frequency resource scheduling scheme.

[0022] Furthermore, based on the target optimization function, obtain the target optimization function value corresponding to each resource allocation coding, and determine the optimal target optimization function value according to the target optimization function value corresponding to each resource allocation coding, including:

[0023] Based on the target optimization function, obtain the target optimization function value corresponding to each resource allocation coding;

[0024] Determine that the resource allocation coding with the smallest target optimization function value is the optimal resource allocation coding.

[0025] Furthermore, according to the optimal resource allocation coding, an information fusion strategy is adopted to perform information fusion on each resource allocation coding to obtain the resource allocation coding after information fusion, including:

[0026] Randomly determine a first random coding among all resource allocation codings A second random coding And a third random coding

[0028] Fuse the optimal resource allocation coding The first random coding The second random coding And the third random coding to obtain the coding fusion information as And after perturbing the coding fusion information, obtain the perturbed coding fusion information as where rand1 represents a random number between (0, 1).

[0029] Add the resource allocation code and the encoded fusion information after perturbation to obtain the resource allocation code after information fusion as wherein represents the i-th resource allocation code in the k-th optimization process, i = 1, 2, …, L, and L represents the total number of resource allocation codes.

[0030] Furthermore, adopt a coding position weighted guiding strategy to search for the unknown area of the resource allocation code after information fusion, and obtain the resource allocation code after unknown area search, including:[[]]

[0031] Based on the resource allocation code after information fusion, determine the weight value corresponding to each resource allocation code wherein represents the position quality degree corresponding to the h-th resource allocation code after information fusion in the k-th optimization process,[[]] represents the position quality degree corresponding to the j-th resource allocation code after information fusion in the k-th optimization process,[[]] represents the weight value corresponding to the j-th resource allocation code after information fusion in the k-th optimization process; the position quality degree = 1 / (objective optimization function value + 0.001);[[]]

[0032] Perform weighted processing based on the obtained weight values to obtain the reference resource allocation code as:[[]] wherein represents the reference resource allocation code;[[]]

[0033] For any resource allocation code after information fusion, exchange information between the resource allocation code and other resource allocation codes to obtain the information exchange code as where λ represents the convergence factor, α1 represents the first exchange coefficient and is set to 0.6; α2 represents the second exchange coefficient and is set to 1.5; e represents the natural constant, d jg represents the j-th resource allocation code[[]] and the g-th resource allocation code[[]] the Euclidean distance between them;[[]]

[0034] Add the reference resource allocation code and the information exchange code to obtain the resource allocation code after unknown area search wherein represents the resource allocation code after unknown area search[[]]

[0035] Furthermore, according to the optimal resource allocation code, adopt a double-optimal guiding strategy to perform an optimal search on the resource allocation code after unknown area search, and obtain the resource allocation code after optimal search, including:[[]]

[0036] The resource allocation code after searching the unknown area is guided by the optimal resource allocation code to obtain the first guiding code as where c1 represents the first learning factor, e represents the natural constant, and χ represents the learning range control coefficient, represents the resource allocation code after searching the u-th unknown area, u = 1, 2,..., L, d ubest represents the Euclidean distance between the resource allocation code after searching the u-th unknown area and the optimal resource allocation code;

[0037] The resource allocation code corresponding to the historical optimal value after searching the unknown area is used to guide it to obtain the second guiding code as where, represents the resource allocation code after searching the u-th unknown area corresponding historical optimal value, d uup represents the Euclidean distance between the resource allocation code after searching the u-th unknown area and its historical optimal value, and c2 represents the second learning factor;

[0038] Remember the previous update speed, and fuse the first guiding code and the second guiding code to obtain the current update speed as: where, represents the optimization speed corresponding to the resource allocation code after searching the u-th unknown area in the k-th optimization process, represents the optimization speed corresponding to the resource allocation code after searching the u-th unknown area in the (k + 1)-th optimization process;

[0039] The resource allocation code after searching the unknown area is added to the current update speed to obtain the resource allocation code after the optimal search

[0040] Furthermore, a mutation fusion strategy is used to globally search the resource allocation code after the optimal search to obtain the resource allocation code after the global search, including:

[0041] The optimal resource allocation code is weighted to obtain the first weighted term as: where rand2 represents a random number between (0, 1);

[0042] The resource allocation code after the optimal search is weighted to obtain the second weighted term as: where rand3 represents a random number between (0, 1), represents the v-th resource allocation code after the optimal search;

[0043] After fusing the resource allocation encoding after the optimal search with a randomly selected different resource allocation encoding, a random flight step size is used for weighting to obtain the third weighted term. Among them, represents the fourth random encoding, levy represents the Levy flight step size, γ represents the global search control factor, and γ = 1 - k / K, where K represents the maximum number of optimizations.

[0044] Fusing the first weighted term, the second weighted term, and the third weighted term, the resource allocation encoding after global search is obtained as:

[0045] Furthermore, the optimization end condition is set as: when the number of optimizations k is greater than or equal to the preset maximum number of optimizations K, it is determined that the optimization ends.

[0046] A spatio-temporal-frequency joint adaptive resource management method provided by the present invention encodes the spatio-temporal-frequency resource allocation by adopting the encoding strategy of an intelligent optimization algorithm to obtain multiple resource allocation encodings. Based on the multiple resource allocation encodings and the target optimization function, an intelligent optimization algorithm is used for resource allocation optimization to obtain an optimal spatio-temporal-frequency resource scheduling scheme. Finally, based on the optimal spatio-temporal-frequency resource scheduling scheme, dynamic scheduling of spatio-temporal-frequency resources is performed to optimize the spatio-temporal-frequency resource ratio relationship and comprehensively improve the joint performance of communication radar across functional domains in the integrated system. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] The accompanying drawings herein are incorporated into the specification and form a part of the specification, showing embodiments consistent with the present invention and, together with the specification, are used to explain the principles of the present invention.

[0048] Figure 1 It is a flowchart of a spatio-temporal-frequency joint adaptive resource management method provided by an embodiment of the present invention.

[0049] Through the above accompanying drawings, specific embodiments of the present invention have been shown, and there will be more detailed descriptions hereinafter. These drawings and textual descriptions are not intended to limit the scope of the inventive concept in any way, but to illustrate the concept of the present invention to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0050] Here, exemplary embodiments will be described in detail, and the examples are shown in the accompanying drawings. When the following description refers to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. On the contrary, they are merely examples of devices and methods consistent with some aspects of the present invention as detailed in the appended claims.

[0051] The embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0052] As Figure 1 shown, a spatio-temporal-frequency joint adaptive resource management method provided by an embodiment of the present invention includes:

[0053] S101. For radar services and communication services, a joint optimization strategy is adopted to construct a spatio-temporal-frequency joint optimization function, and an objective optimization function for optimizing spatio-temporal-frequency resource allocation is obtained;

[0054] By adopting a joint optimization strategy to construct a spatio-temporal-frequency joint optimization function, the radar service and the communication service can be self-balanced. When the amount of one service increases, its influence on the objective optimization function will be greater. Therefore, more spatio-temporal-frequency resources can be obtained, thereby achieving optimization.

[0055] S102. For the spatio-temporal-frequency resources of radar services and communication services, an encoding strategy of an intelligent optimization algorithm is adopted to encode the spatio-temporal-frequency resource allocation to obtain a plurality of resource allocation encodings;

[0056] Although constructing the objective optimization function can effectively evaluate the effect of spatio-temporal-frequency resource allocation, it is still necessary to allocate the spatio-temporal-frequency resources to minimize the objective optimization function. Therefore, the embodiment of the present invention adopts an encoding strategy of an intelligent optimization algorithm for encoding and optimizes with an intelligent optimization algorithm, and finally can achieve resource adaptive management.

[0057] S103. Based on the plurality of resource allocation encodings and the objective optimization function, an intelligent optimization algorithm is adopted to optimize the resource allocation to obtain an optimal spatio-temporal-frequency resource scheduling scheme;

[0058] The intelligent optimization algorithm is a class of computational methods that simulate biological evolution, physical phenomena, or human intelligent behavior in nature and are used to solve complex optimization problems. Therefore, the embodiment of the present invention adaptively solves the allocation of spatio-temporal-frequency resources with continuous numerical segments, thereby achieving adaptive management.

[0059] Optionally, for spatio-temporal-frequency resources with discontinuous numerical segments, the type of spatio-temporal-frequency resources can be mapped to a numerical segment between (0, 1). For all types of a spatio-temporal-frequency resource, the combined numerical segments of all types are (0, 1), so continuous optimization can be achieved.

[0060] S104. Based on the optimal spatio-temporal-frequency resource scheduling scheme, spatio-temporal-frequency resources are allocated to radar services and communication services to complete spatio-temporal-frequency joint adaptive resource management.

[0061] Combined with the dynamic characteristics of services in practical scenarios such as drones and vehicle-to-everything (V2X) networks, a spatio-temporal-frequency adaptive resource scheduling algorithm that matches the dynamic characteristics of services is designed. When the service of one side of communication / radar is low, opportunistic scheduling is implemented. When the services of both communication / radar sides are heavy, the performance of communication / radar is balanced. By utilizing the reciprocity between spatio-temporal-frequency resources and aiming at the differences in the impacts of spatio-temporal-frequency resources on different service performances of communication / radar, the ratio relationship of spatio-temporal-frequency resources is optimized to comprehensively improve the joint performance of communication and radar across functional domains in the integrated system.

[0062] In the embodiment of the present invention, the objective optimization function for optimizing the spatio-temporal-frequency resource allocation is as follows:

[0063]

[0064] where f represents the objective optimization function, Q n (t) represents the length of the communication service queue of the nth user at time t, a n (t) represents the admitted traffic volume of the nth user at time t, S n (C n (t), B n (t), T n (t)) represents the communication queue service rate, and S n (C n (t), B n (t), T n (t)) is jointly determined by spatio-temporal-frequency resources. The spatio-temporal-frequency resources include the transmission rate C n (t), the beam scheduling B n (t), and the time allocation T n (t). Y m (t) represents the length of the detection virtual queue obtained by the mth radar wave position at time t, β m (t) represents the number of detections of the mth radar wave position at time t, φ m (B m (t), T m (t)) represents the radar queue service rate, and φ m (B m (t), T m (t)) is jointly determined by the beam scheduling B n (t) and the time allocation T n (t).

[0065] It should be noted that the above target optimization function for optimizing spatio-temporal frequency resource allocation is only a preferred implementation manner of the embodiments of the present invention. Other existing optimization functions can also be used for optimization, and adaptive dynamic resource management can also be achieved. The communication queue service rate and the radar queue service rate can be obtained through deep learning technology or existing technologies. For example, some transmission rate C n (t), beam scheduling B n (t), and time allocation T n (t) sample values can be collected. After actual execution, the communication queue service rate S n (C n (t), B n (t), T n (t)) actual value is obtained. Taking the sample value as training data and the actual value as the expected output data, the deep learning model can be trained, and the resource communication queue service rate S n (C n (t), B n (t), T n (t)) can be quickly predicted in the subsequent process.

[0066] In the embodiments of the present invention, for the spatio-temporal frequency resources of radar services and communication services, an encoding strategy of an intelligent optimization algorithm is used to encode the spatio-temporal frequency resource allocation to obtain multiple resource allocation encodings, including:

[0067] For any radar service or communication service, random initialization is performed within the upper and lower limits of the spatio-temporal frequency resources to obtain the spatio-temporal frequency resource values corresponding to each radar service or communication service;

[0068] Encode the spatio-temporal frequency resource values corresponding to all radar services and communication services into vectors to obtain resource allocation encodings, and repeat to obtain multiple different resource allocation encodings.

[0069] Optionally, in order to simplify the optimization complexity, after obtaining the resource allocation encoding, the resource allocation encoding can be normalized. Then, after obtaining the optimal resource allocation encoding, denormalization is performed, so as to ensure the correctness of the algorithm and also ensure the simplicity of the algorithm.

[0070] In the embodiments of the present invention, based on the multiple resource allocation encodings and the target optimization function, an intelligent optimization algorithm is used to optimize the resource allocation to obtain an optimal spatio-temporal frequency resource scheduling scheme, including:

[0071] Based on the target optimization function, obtain the target optimization function value corresponding to each resource allocation encoding, and determine the optimal resource allocation encoding according to the target optimization function value corresponding to each resource allocation encoding;

[0072] According to the optimal resource allocation coding, an information fusion strategy is adopted to perform information fusion on each resource allocation coding to obtain the resource allocation coding after information fusion;

[0073] An unknown region search is performed on the resource allocation coding after information fusion by adopting a coding position weighted guiding strategy to obtain the resource allocation coding after unknown region search;

[0074] According to the optimal resource allocation coding, a double-optimal guiding strategy is adopted to perform an optimal search on the resource allocation coding after unknown region search to obtain the resource allocation coding after optimal search;

[0075] A mutation fusion strategy is adopted to perform a global search on the resource allocation coding after optimal search to obtain the resource allocation coding after global search;

[0076] Repeat information fusion, unknown region search, optimal search, and global search until the optimization end condition is met, re-obtain the optimal resource allocation coding, and decode the optimal resource allocation coding to obtain the optimal spatio-temporal-frequency resource scheduling scheme.

[0077] In the existing intelligent optimization algorithms, during the continuous solution process of multi-dimensional data, there is often a problem of being easily trapped in local optimum, which ultimately leads to poor solution effects of multi-dimensional data and it is difficult to obtain the optimal spatio-temporal-frequency resource scheduling scheme. Therefore, the embodiments of the present invention provide a new intelligent optimization algorithm to improve the global search ability and search speed of the algorithm, so that the algorithm can successfully jump out of the local optimum and find the optimal spatio-temporal-frequency resource scheduling scheme.

[0078] In the embodiments of the present invention, based on the target optimization function, the target optimization function value corresponding to each resource allocation coding is obtained, and the optimal target optimization function value is determined according to the target optimization function value corresponding to each resource allocation coding, including:

[0079] Based on the target optimization function, the target optimization function value corresponding to each resource allocation coding is obtained;

[0080] The resource allocation coding with the smallest target optimization function value is determined as the optimal resource allocation coding.

[0081] In the embodiments of the present invention, according to the optimal resource allocation coding, an information fusion strategy is adopted to perform information fusion on each resource allocation coding to obtain the resource allocation coding after information fusion, including:

[0082] Randomly determine a first random coding a second random coding and a third random coding

[0084] Encode the optimal resource allocation The first random encoding The second random encoding And the third random encoding Perform fusion to obtain the encoded fusion information as And after perturbing the encoded fusion information, the perturbed encoded fusion information is Where rand1 represents a random number between (0, 1).

[0085] Add the resource allocation encoding and the perturbed encoded fusion information to obtain the resource allocation encoding after information fusion as Where represents the i-th resource allocation encoding in the k-th optimization process, i = 1, 2,..., L, and L represents the total number of resource allocation encodings.

[0086] The information fusion provided by the embodiments of the present invention can effectively fuse the encoded information, so that during the process of moving the resource allocation encoding towards the optimal position, it is deviated by the influence of other encodings, thereby effectively searching the local area, increasing the possibility of finding the optimal solution, and also increasing the diversity to avoid falling into the local optimum.

[0087] In the embodiments of the present invention, an unknown area search is performed on the resource allocation encoding after information fusion by using an encoding position weighted guidance strategy to obtain the resource allocation encoding after unknown area search, including:

[0088] Based on the resource allocation encoding after information fusion, determine the weight value corresponding to each resource allocation encoding Where represents the degree of position superiority corresponding to the h-th resource allocation encoding after information fusion in the k-th optimization process, represents the degree of position superiority corresponding to the j-th resource allocation encoding after information fusion in the k-th optimization process, represents the weight value corresponding to the j-th resource allocation encoding after information fusion in the k-th optimization process; the degree of position superiority = 1 / (objective optimization function value + 0.001);

[0089] Perform weighted processing based on the obtained weight values to obtain the reference resource allocation encoding as:[[]] Where represents the reference resource allocation encoding;

[0090] For any one of the resource allocation encodings after information fusion, exchange information between the resource allocation encoding and other resource allocation encodings to obtain the information exchange encoding as Among them, λ represents the convergence factor, α1 represents the first exchange coefficient and is set to 0.6; α2 represents the second exchange coefficient and is set to 1.5; e represents the natural constant, and d jg represents the j-th resource allocation code and the g-th resource allocation code The Euclidean distance between them;

[0091] The resource allocation code after the search of the unknown area is obtained by adding the reference resource allocation code and the information exchange code Among them, represents the resource allocation code after the search of the unknown area

[0092] The search for the unknown area provided by the embodiment of the present invention can effectively fuse the information of all codes, thereby realizing the search of the strange area and improving the spatial search efficiency.

[0093] In the embodiment of the present invention, according to the optimal resource allocation code, a double-optimal guidance strategy is used to perform an optimal search on the resource allocation code after the search of the unknown area, and the resource allocation code after the optimal search is obtained, including:

[0094] The resource allocation code after the search of the unknown area is guided by the optimal resource allocation code, and the first guidance code is obtained as Among them, c1 represents the first learning factor, e represents the natural constant, χ represents the learning range control coefficient, represents the resource allocation code after the search of the u-th unknown area, u = 1, 2,..., L, d ubest represents the Euclidean distance between the resource allocation code after the search of the u-th unknown area and the optimal resource allocation code;

[0095] The resource allocation code after the search of the unknown area is guided by the corresponding historical optimal value, and the second guidance code is obtained as Among them, represents the resource allocation code after the search of the u-th unknown area The corresponding historical optimal value, d uup represents the Euclidean distance between the resource allocation code after the search of the u-th unknown area and its historical optimal value, and c2 represents the second learning factor;

[0096] Remember the update speed of the previous time, and fuse the first guidance code and the second guidance code to obtain the current update speed as: Among them, represents the optimization speed corresponding to the resource allocation code after the search of the u-th unknown area in the k-th optimization process, Denote the optimization speed corresponding to the resource allocation code after the search of the $u$-th unknown region in the $(k + 1)$-th optimization process;

[0097] Add the resource allocation code after the unknown region search to the current update speed to obtain the resource allocation code after the optimal search

[0098] The optimal search provided by the embodiments of the present invention can make the resource allocation code quickly move forward in a better direction, thereby improving the convergence speed and convergence accuracy of the algorithm.

[0099] In the embodiments of the present invention, a mutation fusion strategy is adopted to perform a global search on the resource allocation code after the optimal search to obtain the resource allocation code after the global search, including:

[0100] Weight the optimal resource allocation code to obtain the first weighted term as: where rand2 represents a random number between (0, 1);

[0101] Weight the resource allocation code after the optimal search to obtain the second weighted term as: where rand3 represents a random number between (0, 1), Denote the resource allocation code after the $v$-th optimal search;

[0102] After fusing the resource allocation code after the optimal search with a randomly different resource allocation code, perform weighting with a random flight step length to obtain the third weighted term where, Denote the fourth random code, levy represents the Levy flight step length, γ represents the global search control factor, and γ = 1 - k / K, where K represents the maximum number of optimizations.

[0103] Fuse the first weighted term, the second weighted term, and the third weighted term to obtain the resource allocation code after the global search as:

[0104] Optionally, a greedy algorithm can also be used to control the global search process during the global search, thereby further improving the convergence speed of the algorithm.

[0105] With the completion of the previous three stages, the convergence speed and spatial traversal degree of the algorithm are satisfied. Therefore, the embodiments of the present invention provide a global search strategy, such that during the update of the resource allocation code, there will be a situation of jumping search, thereby helping the algorithm escape from the local optimum.

[0106] The resource allocation code can be processed for out-of-bounds after each search, thereby ensuring the effectiveness of the optimization process.

[0107] In an embodiment of the present invention, the optimization end condition is set as follows: when the number of optimization times k is greater than or equal to a preset maximum number of optimization times K, it is determined that the optimization ends.

[0108] After considering the specification and practicing the invention disclosed herein, those skilled in the art will readily conceive of other embodiments of the present invention. The present invention is intended to cover any variations, uses, or adaptations of the present invention, which follow the general principles of the present invention and include known common general knowledge or conventional technical means in the technical field not disclosed by the present invention. It should be understood that the present invention is not limited to the exact structures described above and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the present invention is only limited by the appended claims.

Claims

1. A time-space-frequency joint adaptive resource management method, characterized in that: include: For radar business and communication business, a joint optimization strategy is adopted to construct a joint optimization function of time, space and frequency, and the target optimization function for optimizing the allocation of time, space and frequency resources is obtained; Aiming at the space-time and frequency resources of radar business and communication business, the coding strategy of intelligent optimization algorithm is adopted to encode the space-time and frequency resource allocation to obtain multiple resource allocation codes; Based on the plurality of resource allocation codes and the target optimization function, an intelligent optimization algorithm is used to optimize resource allocation to obtain an optimal time-space-frequency resource scheduling solution; Based on the optimal time-space-frequency resource scheduling scheme, time-space-frequency resources are allocated to radar services and communication services to complete time-space-frequency joint adaptive resource management.

2. The time-space-frequency joint adaptive resource management method according to claim 1, characterized in that: The objective optimization function for optimizing the allocation of time, space and frequency resources is: Where f represents the target optimization function, Q n (t) represents the length of the communication service queue of the nth user at time t, a n (t) represents the admitted traffic volume of the nth user at time t, S n (C n (t),B n (t),T n (t)) represents the communication queue service rate, and S n (C n (t),B n (t),T n (t)) is determined by the combination of time, space and frequency resources, which include the transmission rate C n (t), beam scheduling B n (t) and time allocation T n (t), Y m (t) represents the length of the virtual queue of the mth radar wave position at time t, β m (t) represents the number of detections of the mth radar wave at time t, φ m (B m (t),T m (t)) represents the radar queue service rate, and φ m (B m (t),T m (t)) is scheduled by beam B n (t) and time allocation T n (t) Joint decisions.

3. The time-space-frequency joint adaptive resource management method according to claim 1, characterized in that: For the time-space-frequency resources of radar services and communication services, the coding strategy of the intelligent optimization algorithm is used to encode the time-space-frequency resource allocation to obtain multiple resource allocation codes, including: For any radar service or communication service, random initialization is performed within the upper and lower limits of the time-space-frequency resources to obtain the time-space-frequency resource value corresponding to each radar service or communication service; The time-space-frequency resource values ​​corresponding to all radar services and communication services are encoded into vectors to obtain resource allocation codes, and multiple different resource allocation codes are repeatedly obtained.

4. The time-space-frequency joint adaptive resource management method according to claim 1, characterized in that: Based on the plurality of resource allocation codes and the target optimization function, an intelligent optimization algorithm is used to optimize resource allocation to obtain an optimal time-space-frequency resource scheduling solution, including: Based on the objective optimization function, the objective optimization function value corresponding to each resource allocation code is obtained, and the optimal resource allocation code is determined according to the objective optimization function value corresponding to each resource allocation code; According to the optimal resource allocation code, an information fusion strategy is used to fuse each resource allocation code to obtain the resource allocation code after information fusion; The resource allocation code after information fusion is searched in unknown areas by using the code position weighted guidance strategy to obtain the resource allocation code after the unknown area search; According to the optimal resource allocation code, a double optimal guidance strategy is used to optimally search the resource allocation code after the unknown area search to obtain the resource allocation code after the optimal search; A variation fusion strategy is used to perform a global search on the resource allocation code after the optimal search to obtain the resource allocation code after the global search; Repeat information fusion, unknown area search, optimal search and global search until the optimization end condition is met, re-acquire the optimal resource allocation code, decode the optimal resource allocation code, and obtain the optimal time-space-frequency resource scheduling solution.

5. The time-space-frequency joint adaptive resource management method according to claim 4, characterized in that: Based on the objective optimization function, obtaining the objective optimization function value corresponding to each resource allocation code, and determining the optimal objective optimization function value according to the objective optimization function value corresponding to each resource allocation code, including: Based on the objective optimization function, obtaining the objective optimization function value corresponding to each resource allocation code; The resource allocation code with the smallest target optimization function value is determined as the optimal resource allocation code.

6. The time-space-frequency joint adaptive resource management method according to claim 5, characterized in that: According to the optimal resource allocation code, an information fusion strategy is used to fuse each resource allocation code to obtain the resource allocation code after information fusion, including: Randomly determine the first random code among all resource allocation codes Second random code And the third random code Encoding the optimal resource allocation First random code Second random code And the third random code After fusion, the coded fusion information is obtained as After perturbing the coded fusion information, the perturbed coded fusion information is obtained as Among them, rand1 represents a random number between (0,1). Add the resource allocation code to the perturbed code fusion information to get the resource allocation code after information fusion: in, represents the i-th resource allocation code in the k-th optimization process, i = 1, 2, ..., L, and L represents the total number of resource allocation codes.

7. The time-space-frequency joint adaptive resource management method according to claim 6, characterized in that: The resource allocation code after information fusion is searched in unknown areas by using the code position weighted guidance strategy, and the resource allocation code after the unknown area search is obtained, including: Based on the resource allocation code after information fusion, determine the weight value corresponding to each resource allocation code in, Indicates the superiority of the position corresponding to the resource allocation code after the hth information fusion in the kth optimization process, Indicates the superiority of the position corresponding to the resource allocation code after the jth information fusion in the kth optimization process, Indicates the weight value corresponding to the resource allocation code after the jth information fusion in the kth optimization process; Position quality = 1 / (target optimization function value + 0.001); Based on the obtained weight values, weighted processing is performed to obtain the reference resource allocation code: in, represents the reference resource allocation code; For any resource allocation code after information fusion, the resource allocation code is made to exchange information with other resource allocation codes, and the information exchange code is obtained as Where λ represents the convergence factor, α1 represents the first exchange coefficient and is set to 0.6; α2 represents the second exchange coefficient and is set to 1.5; e represents the natural constant, d jg Indicates the jth resource allocation code and the g-th resource allocation code The Euclidean distance between The reference resource allocation code is added to the information exchange code to obtain the resource allocation code after the unknown area search. in, Indicates the resource allocation code after the unknown area search 8. The time-space-frequency joint adaptive resource management method according to claim 7, characterized in that: According to the optimal resource allocation code, a dual optimal guidance strategy is used to perform an optimal search on the resource allocation code after the unknown area search, and the resource allocation code after the optimal search is obtained, including: The optimal resource allocation code is used to guide the resource allocation code after the unknown area search, and the first guided code is obtained as Where c1 represents the first learning factor, e represents the natural constant, and χ represents the learning range control coefficient. represents the resource allocation code after the uth unknown area search, u=1,2,…,L,d ubest represents the Euclidean distance between the resource allocation code after the u-th unknown area search and the optimal resource allocation code; The historical optimal value corresponding to the resource allocation code after the unknown area search is used to guide it, and the second guided code is obtained as in, Represents the resource allocation code after the uth unknown area search The corresponding historical optimal value, d uup represents the Euclidean distance between the resource allocation code after the u-th unknown area search and its historical optimal value, and c2 represents the second learning factor; Remember the last update speed, and merge the first boot code and the second boot code to get the current update speed: in, represents the optimization speed corresponding to the resource allocation encoding after the u-th unknown area search in the k-th optimization process, represents the optimization speed corresponding to the resource allocation coding after the uth unknown area search in the k+1th optimization process; The resource allocation code after the unknown area search is added to the current update speed to obtain the resource allocation code after the optimal search.

9. The time-space-frequency joint adaptive resource management method according to claim 8, characterized in that: The variation fusion strategy is used to perform a global search on the resource allocation code after the optimal search, and the resource allocation code after the global search is obtained, including: The optimal resource allocation code is weighted, and the first weighted term is obtained as follows: Among them, rand2 represents a random number between (0,1); The resource allocation code after the optimal search is weighted to obtain the second weighted item: Among them, rand3 represents a random number between (0,1), represents the resource allocation code after the vth optimal search; After the optimal search resource allocation code is fused with a random different resource allocation code, the random flight step length is used for weighting to obtain the third weighted term in, represents the fourth random code, levy represents the Levy flight step, γ represents the global search control factor, and γ=1-k / K, K represents the maximum number of optimization times. The first weighted term, the second weighted term and the third weighted term are combined to obtain the resource allocation code after global search:

10. The time-space-frequency joint adaptive resource management method according to claim 9, characterized in that: The optimization end condition is set as: when the optimization times k is greater than or equal to the preset maximum optimization times K, the optimization is determined to be ended.

Images