Method, system, device and medium for allocating heterogeneous network resources for time-sensitive services

Abstract

The present invention discloses a method, system, device and medium for allocating heterogeneous network resources for time-sensitive services, which belongs to the field of energy dispatching, including: obtaining the transmission delay performance of historical time-sensitive services of distributed wind and solar power stations; constructing a communication network resource allocation model with minimization of the total data transmission delay of power Internet of Things terminals as the objective function based on the resource allocation architecture of wind and solar power stations; converting the objective function into a multi-armed bandit model; solving the multi-armed bandit model with a delay-sensitivity-based cognitive greedy algorithm to obtain the distributed wind and solar power station heterogeneous network communication resource allocation decision; using the historical time-sensitive service transmission delay performance to perform delay-sensitive cognitive learning to obtain the resource allocation decision of different time-sensitive service power Internet of Things terminals. The method learns the distributed wind and solar power station heterogeneous network communication resource allocation decision, solves the resource allocation problem under non-global information, and realizes the reasonable allocation of time-sensitive service communication resources.

Description

Technical Field

[0001] The present invention belongs to the field of wind and solar power station communication technology, and in particular relates to a method, system, equipment and medium for allocating heterogeneous network resources for time-sensitive services. Background Art

[0002] Distributed wind and solar power stations have the advantages of energy saving and environmental protection, high safety, low transmission and distribution loss, and high automation, and they occupy an important position in the new power system. Distributed wind and solar power stations generate a variety of time-sensitive services, including information collection services, control command services, and emergency communication services. Due to different subsequent applications, they often have differentiated transmission delay requirements. At the same time, distributed wind and solar power stations are often built in different scenarios such as railway slopes, industrial park roofs, and rural vacant land. There are many participating entities, using power line communication (PLC), 4G / 5G, power wireless private network, optical fiber and other communication means to form a heterogeneous communication network, which makes resource allocation more difficult. It is difficult for the power Internet of Things terminal of distributed wind and solar power stations to make reasonable choices from many heterogeneous resources, and it is impossible to meet the differentiated transmission delay requirements of different services, and the utilization rate of heterogeneous network resources decreases. Therefore, it is urgent to design a heterogeneous network resource allocation method for time-sensitive services of distributed wind and solar power stations to meet the differentiated transmission delay requirements of services and provide guarantees for the realization of subsequent services.

[0003] In order to solve the problem of heterogeneous network resource allocation, existing methods such as the maximum channel resource gain method select communication resources by identifying the channel resource gain of the communication link between power IoT terminals when forwarding at different link nodes, thereby realizing service data transmission. However, this method is based on the situation where global information is known, and it is difficult to apply it to the actual scenario where global information is unknown. At the same time, it ignores the low-latency communication requirements of different time-sensitive services of distributed wind and solar power stations.

[0004] The existing distributed wind and solar power station communication resource allocation methods still face the following problems:

[0005] The existing distributed wind and solar power station communication resource allocation method cannot realize heterogeneous communication network resource allocation when global information is unknown. The existing method uses the communication resource channel resource gain information to achieve resource allocation based on the situation where global information is known. However, in actual complex operation scenarios, the distributed wind and solar power station gateway cannot obtain the real-time channel resource status and perform communication resource allocation.

[0006] The data transmission of time-sensitive services in distributed wind and solar power stations has a low latency requirement, and different time-sensitive services have different latency sensitivities, that is, minimum latency requirements. Most of the existing network communication resource configuration methods ignore the differentiated minimum latency requirements of different time-sensitive services, and cannot achieve the optimal adaptation of the heterogeneous resource configuration of distributed wind and solar power stations and the low latency requirements of time-sensitive services. Summary of the invention

[0007] The present invention provides a method, system, device and medium for allocating heterogeneous network resources for time-sensitive services. The method of the present invention can optimize the allocation of heterogeneous communication network resources for time-sensitive services under non-global information, meet the differentiated delay sensitivity requirements of time-sensitive services of wind and solar power stations, and achieve the purpose of low-delay control of complex heterogeneous networks of distributed wind and solar power stations.

[0008] In order to achieve the above object, the present invention adopts the following technical solutions:

[0009] A method for allocating heterogeneous network resources for time-sensitive services, comprising:

[0010] Obtain the historical transmission delay performance of time-sensitive services of distributed wind and solar power stations;

[0011] Based on the resource allocation architecture of wind and solar power stations, a communication network resource allocation model with the minimization of the total data transmission delay of power Internet of Things terminals as the objective function is constructed; the objective function is converted into a multi-armed bandit model; the multi-armed bandit model is solved by a cognitive greedy algorithm based on delay sensitivity to obtain the communication resource allocation decision of the heterogeneous network of distributed wind and solar power stations;

[0012] The historical time-sensitive service transmission delay performance is used to conduct delay-sensitive cognitive learning on the communication resource allocation decision of the distributed wind and solar power stations in heterogeneous networks, and the resource allocation decisions of power Internet of Things terminals with different time-sensitive services are obtained.

[0013] As a further improvement of the present invention, the wind-solar power station resource allocation architecture includes: distributed wind power components and photovoltaic components, and M power Internet of Things terminals are deployed on the wind power components and photovoltaic components to provide data for time-sensitive services; the power Internet of Things terminals transmit the collected channel resources to the wind-solar power station gateway through power line communication or micro-power wireless;

[0014] The set of M power IoT terminals is represented by D = {d1,...,d m ,...,d M}; There are K1 power line channel resources and K2 micro-power wireless channel resources, and the set is expressed as Among them, c k , k=1,..., K1 is the power line channel resource, c k , k=K1+1,...,K1+K2 are micro-power wireless channel resources.

[0015] As a further improvement of the present invention, the construction of a communication network resource allocation model with minimization of the total data transmission delay of the power Internet of Things terminal as the objective function includes:

[0016] The time slot model is adopted, with a total of T equal-length time slots, and the set is represented by T = {11,...,t,...,T}; at the beginning of each time slot, the power Internet of Things terminal collects data, the wind and solar power station gateway allocates channel resources for it, and the power Internet of Things terminal transmits data according to the resource allocation decision; the binary variable s m,k,t ∈{0,1} is the channel resource allocation indicator variable;

[0017] Power Internet of Things Terminal m Select channel resource c in time slot t k The transmission rate R m,k,t for:

[0018]

[0019] Among them, B k Indicates the channel resource c k The bandwidth, p m,k,t and h m,k,t denote transmission power and channel resource gain, σ0 and σ EMI Represent Gaussian white noise and electromagnetic interference respectively; power Internet of Things terminal d m Select channel resource c k The transmission delay of data is expressed as:

[0020]

[0021] Among them, F m,t Indicates the power Internet of Things terminal d m The size of data transmitted in the tth time slot;

[0022] The communication network resource allocation model selects the resource allocation decision to minimize the total data transmission delay of the power Internet of Things terminal.

[0023] The constraints are:

[0024] Constraint 1: Channel resource allocation constraint, that is, only one channel resource can be allocated to each power IoT terminal in each time slot;

[0025] Constraint 2: Transmission rate constraint, that is, the transmission rate of the channel resources allocated to the power IoT terminal in each time slot must be no less than the threshold

[0026] As a further improvement of the present invention, the objective function is converted into a multi-armed bandit model, specifically:

[0027] The wind and solar power station gateway is the decision maker, and the rocker arm is the channel resource collection The reward is the inverse of the transmission delay of the power IoT terminal, that is,

[0028]

[0029] Power Internet of Things Terminal m The exploration factor κ m for Among them, α is a constant used to adjust the order of magnitude, η m For power Internet of Things terminal m delay sensitivity.

[0030] As a further improvement of the present invention, the method of using the historical time-sensitive service transmission delay performance to perform delay-sensitive cognitive learning on the distributed wind and solar power station heterogeneous network communication resource allocation decision to obtain resource allocation decisions for different time-sensitive service power Internet of Things terminals includes:

[0031] Initializes m,k,t =0 and in, Represented as power Internet of Things terminal d m Allocate channel resources k times; when t≤K1+K2, the wind-solar power station gateway selects each rocker arm in turn and obtains the initial value;

[0032] Generate a random number λ∈(0,1) and select the rocker arm according to the following formula

[0033]

[0034] When λ>κ m When , the gateway of the wind and solar power station greedily selects the rocker arm with the best historical performance;

[0035] When λ≤κ m When the wind and solar power station gateway selects an arm at random;

[0036] Wind and solar power station gateway observes latency performance and updates and in, Represents the average reward of the rocker arm ck up to the t-1th iteration;

[0037] When t>T, it ends.

[0038] A heterogeneous network resource allocation system for time-sensitive services, comprising:

[0039] Data acquisition module, used to obtain the transmission delay performance of historical time-sensitive services of distributed wind and solar power stations;

[0040] The decision-making solution module is used to construct a communication network resource allocation model based on the wind and solar power station resource allocation architecture with the minimization of the total data transmission delay of the power Internet of Things terminal as the objective function; convert the objective function into a multi-armed bandit model; use a delay-sensitive cognitive greedy algorithm to solve the multi-armed bandit model to obtain a distributed wind and solar power station heterogeneous network communication resource allocation decision;

[0041] The cognitive learning module is used to perform delay-sensitive cognitive learning on the communication resource allocation decisions of the heterogeneous network of distributed wind and solar power stations by utilizing the historical time-sensitive service transmission delay performance, and obtain the resource allocation decisions of the power Internet of Things terminals with different time-sensitive services.

[0042] As a further improvement of the present invention, the objective function of the decision-making solution module includes:

[0043] Resource allocation decision-making to minimize the total data transmission delay of power Internet of Things terminals And the following constraints are met:

[0044] Constraint 1: Channel resource allocation constraint, that is, only one channel resource can be allocated to each power IoT terminal in each time slot;

[0045] Constraint 2: Transmission rate constraint, that is, the transmission rate of the channel resources allocated to the power IoT terminal in each time slot must be no less than the threshold

[0046] As a further improvement of the present invention, the decision-solving module converts the objective function into a multi-armed bandit model, specifically:

[0047] The wind and solar power station gateway is the decision maker, and the rocker arm is the channel resource collection The reward is the inverse of the transmission delay of the power IoT terminal, that is,

[0048]

[0049] Power Internet of Things Terminal m The exploration factor κ m for Among them, α is a constant used to adjust the order of magnitude, η m For power Internet of Things terminal m delay sensitivity.

[0050] An electronic device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the method for allocating heterogeneous network resources for time-sensitive services when executing the computer program.

[0051] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the method for allocating heterogeneous network resources for time-sensitive services.

[0052] Compared with the prior art, the present invention has the following beneficial effects:

[0053] In view of the differentiated delay sensitivity requirements of time-sensitive services generated by distributed wind and solar power stations under the background of dual carbon, the present invention provides a heterogeneous network resource allocation method for time-sensitive services, which is oriented to the time-sensitive services of wind and solar power stations, comprehensively considers the differentiated delay sensitivity requirements of time-sensitive services of wind and solar power stations, and uses the historical time-sensitive service transmission delay performance to achieve differentiated time-sensitive service resource allocation decision optimization under non-global information. The present invention adopts a heterogeneous network resource allocation algorithm based on delay-sensitive cognitive learning to reduce the data transmission delay of time-sensitive services of distributed wind and solar power stations. The method uses the historical time-sensitive service transmission delay performance to learn the heterogeneous network communication resource allocation decision of distributed wind and solar power stations, solves the resource allocation problem under non-global information, and realizes the reasonable allocation of time-sensitive service communication resources. The present invention reduces the data transmission delay of time-sensitive services of distributed wind and solar power stations. The method uses the historical time-sensitive service transmission delay performance to learn the heterogeneous network communication resource allocation decision of distributed wind and solar power stations, solves the resource allocation problem under non-global information, and realizes the reasonable allocation of time-sensitive service communication resources.

[0054] Furthermore, the distributed wind and solar power station heterogeneous network resource allocation problem of the present invention is modeled as a multi-armed gambling machine model. The present invention innovates on the traditional ε-greedy algorithm and proposes a delay-sensitivity-based cognitive greedy algorithm. The exploration factor is dynamically adjusted according to the differentiated delay sensitivity of the power Internet of Things terminals of different time-sensitive services, and the relationship between exploration and utilization of the rocker arm selection is balanced, thereby specifically reducing the data transmission delay of different time-sensitive services of the distributed wind and solar power station, and realizing low-latency control of time-sensitive services of complex heterogeneous networks of wind and solar power stations. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 A flow chart of heterogeneous network resource allocation for time-sensitive services provided by the present invention;

[0056] Figure 2 This is a diagram of the resource allocation architecture for wind and solar power stations;

[0057] Figure 3 A flowchart of heterogeneous network resource allocation based on delay sensitivity cognitive greedy algorithm;

[0058] Figure 4 A heterogeneous network resource allocation system for time-sensitive services provided by the present invention;

[0059] Figure 5A schematic diagram of an electronic device provided by the present invention. DETAILED DESCRIPTION

[0060] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0061] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0062] like Figure 1 As shown, the present invention proposes a method for allocating heterogeneous network resources for time-sensitive services, comprising:

[0063] S1, obtain the historical transmission delay performance of time-sensitive services of distributed wind and solar power stations;

[0064] S2, based on the resource allocation architecture of wind and solar power stations, construct a communication network resource allocation model with the minimization of the total data transmission delay of the power Internet of Things terminal as the objective function; convert the objective function into a multi-armed bandit model; use a cognitive greedy algorithm based on delay sensitivity to solve the multi-armed bandit model to obtain the communication resource allocation decision of the distributed wind and solar power station heterogeneous network;

[0065] S3, using the historical time-sensitive service transmission delay performance to conduct delay-sensitive cognitive learning on the communication resource allocation decisions of the distributed wind and solar power stations in heterogeneous networks, obtains the resource allocation decisions of power Internet of Things terminals with different time-sensitive services.

[0066] The present invention innovates on the traditional ε-greedy algorithm and proposes a cognitive greedy algorithm based on delay sensitivity. It dynamically adjusts the exploration factor according to the differentiated delay sensitivity of the power Internet of Things terminals with different time-sensitive services, and balances the relationship between exploration and utilization of rocker arm selection. When the service delay sensitivity is high, the exploration factor will decrease, so that it is more inclined to use rockers with better delay performance to meet the higher delay requirements of the service; on the contrary, when the service delay sensitivity is low, the exploration factor will increase, so that it is more inclined to explore other rockers to avoid resource allocation decisions from falling into local optimality. The proposed algorithm can specifically reduce the data transmission delay of different time-sensitive services of distributed wind and solar power stations, and realize low-latency control of time-sensitive services in complex heterogeneous networks of wind and solar power stations.

[0067] The specific method is described in detail as follows with reference to the accompanying drawings and embodiments.

[0068] The present invention considers the wind and solar power station resource allocation architecture diagram as shown in Figure 2 As shown, wind and solar power stations deploy a large number of distributed wind power components and photovoltaic components, and a large number of power Internet of Things terminals are deployed on the components to provide data for time-sensitive services.

[0069] Suppose there are M power IoT terminals, the set is represented by D = {d1,...,d m ,...,d M There are channel resources such as power line communication and micro-power wireless in the heterogeneous network of wind and solar power stations. The power Internet of Things terminal transmits the collected data to the wind and solar power station gateway for processing through power line communication or micro-power wireless.

[0070] Assume that there are K1 power line channel resources and K2 micro-power wireless channel resources, the set is expressed as Among them, c k , k=1,..., K1 is the power line channel resource, c k , k=K1+1,...,K1+K2 are micro-power wireless channel resources.

[0071] The present invention adopts a time slot model, assuming that there are T equal-length time slots, and the set is represented by T = {11,...,t,...,T}. At the beginning of each time slot, the power Internet of Things terminal collects data, the wind and solar power station gateway allocates channel resources for it, and the power Internet of Things terminal transmits data according to the resource allocation decision. Define a binary variable s m,k,t ∈{0,1} is the channel resource allocation indicator variable, s m,k,t =1 indicates power Internet of Things terminal d m Select channel resource c in time slot t k Transmit data, otherwise s m,k,t =0.

[0072] Power Internet of Things Terminal mSelect channel resource c in time slot t k The transmission rate R m,k,t for

[0073]

[0074] Among them, B k Indicates the channel resource c k The bandwidth, p m,k,t and h m,k,t denote transmission power and channel resource gain, σ0 and σ EMI represent Gaussian white noise and electromagnetic interference respectively. Therefore, the power Internet of Things terminal d m Select channel resource c k The transmission delay of data is expressed as

[0075]

[0076] Among them, F m,t Indicates the power Internet of Things terminal d m The size of data transmitted in the tth time slot.

[0077] This invention solves the problem of heterogeneous network resource allocation for time-sensitive services in wind and solar power stations, and studies how to minimize the total data transmission delay of power Internet of Things terminals by optimizing the resource allocation decision of power Internet of Things terminals. And the following constraints are met:

[0078] Constraint 1: Channel resource allocation constraint, that is, only one channel resource can be allocated to each power Internet of Things terminal in each time slot.

[0079] Constraint 2: Transmission rate constraint, that is, the transmission rate of the channel resources allocated to the power IoT terminal in each time slot must be no less than the threshold

[0080] In order to solve the above problems, the present invention transforms the heterogeneous network resource allocation problem of time-sensitive services of wind and solar power stations into a multi-armed bandit model. The main elements of the multi-armed bandit model include decision makers, rockers, and rewards.

[0081] In the multi-armed gambling machine model, the decision maker is the wind and solar power station gateway, and the rocker arm is the channel resource set. The reward is defined as the inverse of the transmission delay of the power IoT terminal, that is, In particular, when constraint 2 is not satisfied, the reward is 0.

[0082] In order to solve the above multi-armed bandit model, the present invention proposes a delay sensitivity cognitive greedy algorithm to solve the above multi-armed bandit model, dynamically adjust the exploration factor according to the differentiated delay sensitivity of the power Internet of Things terminal for different time-sensitive services, and balance the relationship between exploration and utilization of the rocker arm selection. m The exploration factor κ m The definition is as follows Among them, α is a constant used to adjust the order of magnitude, η m For power Internet of Things terminal m When the latency sensitivity is high, the exploration factor will be reduced, which will be more inclined to "utilization" and reduce the performance degradation caused by unnecessary "exploration". On the contrary, when the latency sensitivity is low, the exploration factor will be increased, which will be more inclined to "explore" resource allocation strategies with better performance.

[0083] The proposed delay sensitivity based cognitive greedy algorithm flow chart is as follows Figure 3 As shown, the algorithm implementation mainly includes the following three stages.

[0084] 1) Initialization phase: Initialize s m,k,t =0 and in, Represented as power Internet of Things terminal d m Allocate channel resources k When t≤K1+K2, the wind and solar power station gateway selects each rocker arm in turn and obtains the initial value.

[0085] 2) Rocker arm selection stage: First, generate a random number λ∈(0,1). Then, select the rocker arm according to the following formula

[0086]

[0087] When λ>κ m When λ≤κ, the wind and solar power station gateway greedily selects the rocker arm with the best historical performance. m When the wind and solar power station gateway selects a rocker arm at random.

[0088] 3) Learning phase: The wind and solar power station gateway observes the latency performance and updates and in, Represents the average reward of the rocker arm ck up to the t-1th iteration.

[0089] Finally, when t>T, the algorithm ends.

[0090] Among them, the learning process includes: in each time slot, the wind and solar power station gateway (decision maker) selects a channel (rocker arm), obtains a resource configuration decision, transmits data according to the configured channel resources, and obtains the corresponding reward power Internet of Things terminal transmission delay performance (reward), thereby updating the historical time-sensitive service transmission delay performance of the selected channel (rocker arm). Therefore, the wind and solar power station gateway (decision maker) can continuously learn which channel (rocker arm) has better power Internet of Things terminal transmission delay performance.

[0091] The present invention also proposes a heterogeneous network resource allocation algorithm based on delay-sensitive cognitive learning to reduce the data transmission delay of time-sensitive services of distributed wind and solar power stations. This method uses the historical time-sensitive service transmission delay performance to learn the distributed wind and solar power station heterogeneous network communication resource allocation decision without obtaining the global information of the wind and solar power station heterogeneous network, solves the resource allocation problem under non-global information, and realizes the reasonable allocation of time-sensitive service communication resources.

[0092] like Figure 4 As shown, the present invention also provides a system for allocating heterogeneous network resources for time-sensitive services, comprising:

[0093] Data acquisition module, used to obtain the transmission delay performance of historical time-sensitive services of distributed wind and solar power stations;

[0094] The decision-making solution module is used to construct a communication network resource allocation model based on the wind and solar power station resource allocation architecture with the minimization of the total data transmission delay of the power Internet of Things terminal as the objective function; convert the objective function into a multi-armed bandit model; use a delay-sensitive cognitive greedy algorithm to solve the multi-armed bandit model to obtain a distributed wind and solar power station heterogeneous network communication resource allocation decision;

[0095] The cognitive learning module is used to perform delay-sensitive cognitive learning on the communication resource allocation decisions of the heterogeneous network of distributed wind and solar power stations by utilizing the historical time-sensitive service transmission delay performance, and obtain the resource allocation decisions of the power Internet of Things terminals with different time-sensitive services.

[0096] As an optional embodiment, the model building module, the objective function includes:

[0097] Resource allocation decision-making to minimize the total data transmission delay of power Internet of Things terminals And the following constraints are met:

[0098] Constraint 1: Channel resource allocation constraint, that is, only one channel resource can be allocated to each power IoT terminal in each time slot;

[0099] Constraint 2: Transmission rate constraint, that is, the transmission rate of the channel resources allocated to the power IoT terminal in each time slot must be no less than the threshold

[0100] As an optional embodiment, the decision-solving module converts the objective function into a multi-armed bandit model, specifically:

[0101] The wind and solar power station gateway is the decision maker, and the rocker arm is the channel resource collection The reward is the inverse of the transmission delay of the power IoT terminal, that is,

[0102]

[0103] Power Internet of Things Terminal m The exploration factor κ m for Among them, α is a constant used to adjust the order of magnitude, η m For power Internet of Things terminal m delay sensitivity.

[0104] like Figure 5 As shown, the present invention provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the method for allocating heterogeneous network resources for time-sensitive services when executing the computer program.

[0105] The method for allocating heterogeneous network resources for time-sensitive services comprises the following steps:

[0106] Obtain the historical transmission delay performance of time-sensitive services of distributed wind and solar power stations;

[0107] Based on the resource allocation architecture of wind and solar power stations, a communication network resource allocation model with the minimization of the total data transmission delay of power Internet of Things terminals as the objective function is constructed; the objective function is converted into a multi-armed bandit model; the multi-armed bandit model is solved by a cognitive greedy algorithm based on delay sensitivity to obtain the communication resource allocation decision of the heterogeneous network of distributed wind and solar power stations;

[0108] The historical time-sensitive service transmission delay performance is used to conduct delay-sensitive cognitive learning on the communication resource allocation decision of the distributed wind and solar power stations in heterogeneous networks, and the resource allocation decisions of power Internet of Things terminals with different time-sensitive services are obtained.

[0109] The present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method for allocating heterogeneous network resources for time-sensitive services are implemented.

[0110] The method for allocating heterogeneous network resources for time-sensitive services comprises the following steps:

[0111] Obtain the historical transmission delay performance of time-sensitive services of distributed wind and solar power stations;

[0112] Based on the resource allocation architecture of wind and solar power stations, a communication network resource allocation model with the minimization of the total data transmission delay of power Internet of Things terminals as the objective function is constructed; the objective function is converted into a multi-armed bandit model; the multi-armed bandit model is solved by a cognitive greedy algorithm based on delay sensitivity to obtain the communication resource allocation decision of the heterogeneous network of distributed wind and solar power stations;

[0113] The historical time-sensitive service transmission delay performance is used to conduct delay-sensitive cognitive learning on the communication resource allocation decision of the distributed wind and solar power stations in heterogeneous networks, and the resource allocation decisions of power Internet of Things terminals with different time-sensitive services are obtained.

[0114] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0115] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0116] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0117] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0118] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for allocating heterogeneous network resources for time-sensitive services, characterized in that: include: Obtain the historical transmission delay performance of time-sensitive services of distributed wind and solar power stations; Based on the resource allocation architecture of wind and solar power stations, a communication network resource allocation model with the minimization of the total data transmission delay of power Internet of Things terminals as the objective function is constructed; the objective function is converted into a multi-armed bandit model; the multi-armed bandit model is solved by a cognitive greedy algorithm based on delay sensitivity to obtain the communication resource allocation decision of the heterogeneous network of distributed wind and solar power stations; The transmission delay performance of historical time-sensitive services is used to conduct delay-sensitive cognitive learning on the communication resource allocation decision of the distributed wind and solar power stations in heterogeneous networks, and the resource allocation decision of the power Internet of Things terminals with different time-sensitive services is obtained; The objective function is converted into a multi-armed bandit model, specifically: The wind and solar power station gateway is the decision maker, and the rocker arm is the channel resource collection , the reward is the inverse of the transmission delay of the power Internet of Things terminal, that is, ; Power Internet of Things Terminal The exploration factor for ,in, is a constant used to adjust the order of magnitude, For power Internet of Things terminals Delay sensitivity; The method uses the historical time-sensitive service transmission delay performance to perform delay-sensitive cognitive learning on the distributed wind and solar power station heterogeneous network communication resource allocation decision to obtain resource allocation decisions for power Internet of Things terminals with different time-sensitive services, including: initialization and ,in, Represented as a power IoT terminal Allocate channel resources The number of times; when When , the wind and solar power station gateway selects each rocker arm in turn and obtains the initial value; Generate random numbers , select the rocker arm according to the following formula : （3） when When , the gateway of the wind and solar power station greedily selects the rocker arm with the best historical performance; when When the wind and solar power station gateway selects an arm at random; Wind and solar power station gateway observes latency performance and updates and ,in, Indicates that until Iteration rocker arm The average reward of when When, it ends.

2. A method for allocating heterogeneous network resources for time-sensitive services according to claim 1, characterized in that: The wind and solar power station resource allocation architecture includes: distributed wind power components and photovoltaic components, both of which are deployed with A power Internet of Things terminal provides data for time-sensitive services; the power Internet of Things terminal transmits the collected data to the wind and solar power station gateway via power line communication or micro-power wireless; The set of power IoT terminals is represented as ;exist Power line channel resources, micro-power wireless channel resources, the set is expressed as ,in, is the power line channel resource, It is a micro-power wireless channel resource.

3. A method for allocating heterogeneous network resources for time-sensitive services according to claim 1, characterized in that: The construction of a communication network resource allocation model with minimization of the total data transmission delay of the power Internet of Things terminal as the objective function includes: Using the time slot model, there are equal-length time slots, the set is represented as At the beginning of each time slot, the power Internet of Things terminal collects data, the wind and solar power station gateway allocates channel resources for it, and the power Internet of Things terminal transmits data according to the resource allocation decision; binary variables A variable indicating channel resource allocation; Power Internet of Things Terminal In the Time slot selection channel resources The transmission rate for: （1） in, Indicates channel resources bandwidth, and represent transmission power and channel resource gain respectively, and Represent Gaussian white noise and electromagnetic interference respectively; power Internet of Things terminal Select channel resources The transmission delay of data is expressed as: （2） in, Represents power Internet of Things terminal In the The size of data transmitted in the time slot; The communication network resource allocation model selects the resource allocation decision to minimize the total data transmission delay of the power Internet of Things terminal. ; The constraints are: Constraint 1: Channel resource allocation constraint, that is, only one channel resource can be allocated to each power IoT terminal in each time slot; Constraint 2: Transmission rate constraint, that is, the transmission rate of the channel resources allocated to the power IoT terminal in each time slot must be no less than the threshold .

4. A heterogeneous network resource allocation system for time-sensitive services, characterized in that: include: Data acquisition module, used to obtain the transmission delay performance of historical time-sensitive services of distributed wind and solar power stations; The decision-making solution module is used to construct a communication network resource allocation model based on the wind and solar power station resource allocation architecture with the minimization of the total data transmission delay of the power Internet of Things terminal as the objective function; convert the objective function into a multi-armed bandit model; use a delay-sensitive cognitive greedy algorithm to solve the multi-armed bandit model to obtain a distributed wind and solar power station heterogeneous network communication resource allocation decision; The cognitive learning module is used to use the historical time-sensitive service transmission delay performance to conduct delay-sensitive cognitive learning on the distributed wind and solar power station heterogeneous network communication resource allocation decisions, and obtain resource allocation decisions for power Internet of Things terminals with different time-sensitive services; The objective function is converted into a multi-armed bandit model, specifically: The wind and solar power station gateway is the decision maker, and the rocker arm is the channel resource collection , the reward is the inverse of the transmission delay of the power Internet of Things terminal, that is, ; Power Internet of Things Terminal The exploration factor for ,in, is a constant used to adjust the order of magnitude, For power Internet of Things terminals Delay sensitivity; The method uses the historical time-sensitive service transmission delay performance to perform delay-sensitive cognitive learning on the distributed wind and solar power station heterogeneous network communication resource allocation decision to obtain resource allocation decisions for power Internet of Things terminals with different time-sensitive services, including: initialization and ,in, Represented as a power IoT terminal Allocate channel resources The number of times; when When , the wind and solar power station gateway selects each rocker arm in turn and obtains the initial value; Generate random numbers , select the rocker arm according to the following formula : （3） when When , the gateway of the wind and solar power station greedily selects the rocker arm with the best historical performance; when When the wind and solar power station gateway selects an arm at random; Wind and solar power station gateway observes latency performance and updates and ,in, Indicates that until Iteration rocker arm The average reward of when When, it ends.

5. A heterogeneous network resource allocation system for time-sensitive services according to claim 4, characterized in that: The objective function of the decision-making solution module includes: Resource allocation decision-making to minimize the total data transmission delay of power Internet of Things terminals , and the following constraints are satisfied: Constraint 1: Channel resource allocation constraint, that is, only one channel resource can be allocated to each power IoT terminal in each time slot; Constraint 2: Transmission rate constraint, that is, the transmission rate of the channel resources allocated to the power IoT terminal in each time slot must be no less than the threshold .

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the method for allocating heterogeneous network resources for time-sensitive services as described in any one of claims 1 to 3 when executing the computer program.

7. A computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the steps of the method for allocating heterogeneous network resources for time-sensitive services as described in any one of claims 1 to 3.

Images