61 results about "Optimal deployment" patented technology

The invention discloses a distributed service deployment method and a distributed service deployment device based on Kubernetes. According to the method, resource object definition and service metadata are extracted by automatically analyzing a configuration file, and accurate acquisition and standardized processing of deployment information are realized; optimal deployment nodes are dynamically matched based on service characteristics and cluster real-time states, and resource scheduling rationality and service operation efficiency are improved; then automatically creating and starting a service instance at a target node according to an analysis result, and ensuring configuration consistency and standardization of a deployment process; and the service flow is controllably migrated to the new instance through the predefined version switching rule, so that smooth transition and risk controllability of version upgrading are realized. According to the method, the efficiency, the reliability and the version management capability of micro-service deployment are remarkably improved through full-process automatic closed loop from configuration analysis, intelligent scheduling to instance creation and flow switching.

The invention relates to an aggregation scene unmanned aerial vehicle cluster three-dimensional deployment method and device based on cyclic segmentation and multi-objective optimization. The method comprises the following steps: designing a cyclic segmentation optimization framework and a self-adaptive switching mechanism oriented to an aggregation scene; performing cyclic segmentation optimization on the unmanned aerial vehicle cluster based on a cyclic segmentation optimization framework and an adaptive switching mechanism, and updating a Pareto file; designing a comprehensive fitness function, screening out elite parent individuals based on the Pareto file, and performing cyclic segmentation optimization on the unmanned aerial vehicle cluster by using the elite parent individuals; and performing three-dimensional deployment of the unmanned aerial vehicle cluster in the aggregation scene according to a cyclic segmentation optimization result. According to the method, through multi-objective optimization based on Pareto and an innovative segmented optimization framework, and through dynamic switching of level and height optimization stages, astringency and global diversity of understanding are remarkably improved, optimal deployment of a multi-unmanned aerial vehicle communication system can be realized in a complex scene, and theoretical and method support is provided for practical engineering application.

A data processing system includes a processor and a memory for the processor. The memory stores executable instructions that, when executed by the processor alone or in combination with other processors, cause the data processing system to perform functions of: receive a deployment request to deploy a software change; determine factors corresponding to the deployment request that impact an optimal deployment policy for the software change; query an Artificial Intelligence (AI) trained with a dataset of optimized deployment policies based on corresponding sets of the factors, the query requesting an optimized deployment policy for the software change of the received deployment request based on the determined factors and including a ring rollout policy, ring bake time and deployment time; execute the deployment request using the optimized deployment policy returned by the AI; and update training of the AI based on the determined factors and results of the optimized deployment policy.

The invention provides a UWB anchor point deployment method and system based on multi-objective optimization and dynamic hotspot prediction. The method comprises the following steps: initially installing a UWB positioning system; constructing a dynamic hot area prediction model; predicting a candidate area of the dynamic hot area; constructing a multi-target performance function, and evaluating the candidate region of the dynamic hot region; performing anchor point position search by adopting a hybrid optimization algorithm; performing feedback adjustment based on a state monitoring result; outputting a result; according to the system, a multi-source data acquisition module is used for acquiring multi-source target data; the dynamic hot area prediction module is used for predicting a dynamic hot area candidate area; the multi-target performance evaluation module is used for screening high-value candidate regions; the optimal deployment module is used for searching an optimal UWB anchor point deployment position; the state feedback module is used for feedback adjustment; and the result output module is used for outputting the optimal UWB anchor point deployment scheme. According to the method and the system, the optimal UWB anchor point deployment scheme can be efficiently and accurately obtained, and the positioning accuracy under the underground complex working condition can be ensured.

The invention relates to the technical field of cloud computing, and discloses a multi-cloud certificate deployment intelligent arrangement engine system, which comprises a unified abstraction layer used for providing a standardized certificate deployment interface; the intelligent arrangement engine is used for analyzing a deployment task dependency relationship and generating an optimal deployment path; the distributed transaction management module is used for coordinating deployment transactions of the cross-cloud platform and ensuring the consistency of operation; and the strategy optimization and learning module is used for optimizing the deployment strategy based on the historical deployment data. According to the method, the cloud platform difference is shielded through the unified abstraction layer, the consistent deployment experience is provided, and the adaptation workload in the multi-cloud environment can be remarkably reduced; the deployment path is optimized through the intelligent arrangement engine, so that the deployment time can be remarkably shortened; through distributed transaction management, transaction consistency of cross-cloud deployment is ensured, and strong fault recovery capability is provided. Through multi-objective optimization and online learning, the system can automatically balance the cost, the performance and the reliability, so that the resource utilization rate is greatly improved.

The invention provides a fault repair and task unloading optimization method under a space-air-ground fusion vehicle-mounted network framework, and belongs to the technical field of Internet of Vehicles. According to the first stage, the optimal deployment point location of the unmanned aerial vehicle is searched globally based on the wolf pack algorithm, the flight path is planned dynamically in combination with the near-end strategy optimization algorithm, and network connectivity repair of a fault area is achieved; and in the second stage, a task unloading decision initial population is generated through multiple deep reinforcement learning agents, and a Pareto optimal solution set is obtained through optimization of a non-dominated sorting genetic algorithm II, so that multi-target balance of energy consumption, cost and time delay is realized. According to the method, the fault self-healing capability and task processing efficiency of the SAGVN can be remarkably improved, and the method adapts to the high-dynamic and high-reliability requirements of intelligent traffic.

The invention discloses a card type fast application deployment method, system and device based on a multi-kernel fusion system, and the method comprises the steps: decomposing a fast application into a plurality of card modules with independent functions, determining the dependency relationship between the card modules, and then determining a deployment sequence; configuring and determining a self-adaptive interface and a constraint value of each card module; and determining the adaptation degree of the kernel and each card module according to various indexes, further determining the minimum deployment condition value of each card module and the kernel, performing deployment, and managing the running state of the card module according to the multi-dimensional context data. According to the method, the fast application is decomposed into a plurality of card modules with independent functions, the optimal deployment combination for deploying the card modules on the kernel is determined by comprehensively considering multiple indexes, efficient operation of each kernel and the application module is ensured, and the method can be widely applied to the technical field of fast application deployment.

A method for developing a platform-independent cloud application with a flexible deployment model and performing automated deployment on a target cloud platform is fulfilled in the ongoing description by (i) obtaining user-written application source code targeted to a Logical Application Model, (ii) obtaining from the user an Application Manifest that is agnostic to specific deployment models and target clouds, (iii) combining the user-written application source code with system-generated bindings and a main file to obtain a final application source code, (iv) dynamically determining based on heuristics or user-provided preferences, the optimal deployment model for each specific component of the cloud application, (v) transforming the Application Manifest into a deployment configuration file based on the optimal deployment model and the target cloud platform, (vi) automatically generating IaC scripts based on the deployment configuration file, and (vii) automatically deploying executable components of the cloud application on the target cloud platform.

The invention discloses an intelligent metasurface auxiliary directional charger deployment method based on a heterogeneous graph neural network and deep reinforcement learning, and belongs to the field of wireless energy supplementation of the Internet of Things. The method aims at solving the problems that at present, related work is mostly one-time static geometric deployment, the coverage range is limited, flexibility is insufficient, the optimization granularity is coarse, high-quality charging full coverage is difficult to achieve under the cost constraint, and the overall charging effectiveness of a network is difficult to maximize. According to the method, a heterogeneous network graph fusing the relation of sensor nodes, directional chargers and intelligent super surfaces (RIS) is constructed, a heterogeneous graph neural network is deployed in a base station in a centralized mode to extract the network structure and energy state characteristics, and a near-end strategy optimization algorithm is further combined. And joint optimization of a directional charger deployment position and a charging direction as well as an RIS position and reflection configuration is realized. According to the method, the charging coverage rate, the charging utility and the deployment cost are taken as a comprehensive target, the optimal deployment strategy can be intelligently generated, the charging coverage is expanded, the charging utility is improved and the node failure rate is reduced on the premise of controllable cost, so that the purpose of prolonging the service life of the network is achieved.

The invention discloses a Kubernetes micro-service optimal deployment method and device based on award accumulation deep reinforcement learning and a medium, and relates to the field of electric digital data processing, the deployment method comprises the following steps: S1, constructing a multi-target micro-service deployment optimization problem; s2, constructing a micro-service resource preference function and combining with an optimization target; s3, constructing a reinforcement learning model; s4, constructing a reward function, and training the model; and S5, obtaining an optimal deployment strategy on line according to a real state through the trained model. According to the method, a multi-target micro-service deployment optimization problem is constructed, two targets of minimum service response delay and resource load balancing are considered, and a Kubernetes deployment demand is effectively converted into a computable and solvable optimal problem; meanwhile, a micro-service resource preference function is designed and combined with an optimization target, the calculation complexity during model training is reduced, the real-time performance of Kubernetes deployment is guaranteed, and therefore the development requirement of a micro-service architecture is met.

The invention discloses an enterprise resource supervision system and method based on a cloud platform, and relates to the technical field of enterprise informatization management and cloud computing. Comprising a cloud platform deployment module, a resource collection module, a security management module and a resource supervision module. By constructing a 9-dimensional capability index space, the mainstream cloud platform capability is quantitatively evaluated, and optimal deployment type selection is realized based on the Euclidean distance; the acquisition module supports multi-source data to be accessed and pushed to the processing channel; the security module realizes identity authentication and authority control; the supervision module performs verification around data integrity, time consistency, primary key uniqueness and semantic consistency, and constructs a data quality score and label system; according to the method, abnormal data are automatically classified and processed, the operations including field supplementation, time restoration, major key conflict analysis, semantic standardization and the like are included, an interactive supplementary recording process is started for serious abnormal data, authentication is carried out again, closed-loop supervision of high-credibility data is finally achieved, and the intelligence and compliance level of enterprise resource management is improved.

The invention relates to the field of network security, and particularly provides a security cooperation system and method in a hybrid cloud environment. The system comprises a unified security management module, a security solution cooperation module and a security function module. The unified security management module analyzes a self-healing algorithm through a policy graph, carries out graph modeling on a multi-cloud security policy, automatically identifies policy conflicts and drifts through graph difference analysis, and realizes continuous consistency and self-healing of a cross-cloud policy. And the security solution cooperation module is embedded into a reinforcement learning agent, dynamically decides an optimal deployment scheme of security resources according to a reward function by sensing a threat situation, a resource utilization rate and response delay, and realizes intelligent self-adaptive arrangement of protection resources. The security function module is used for micro-servitizing and containerizing security capability based on a cloud native technology, transparent injection is realized through a service grid, an automatic compliance repair and verification mechanism is built in, and a compliance closed loop for detection, repair and verification is formed. And the management efficiency and the protection accuracy of the security policy are improved.

The invention discloses a machine room resource capacity intelligent planning system and method based on multi-source data fusion, and belongs to the technical field of data center management, and the system comprises a multi-source data collection module, a resource manifold modeling module, a geodesic optimization module, a self-adaptive planning module and a verification and inspection module. The method comprises the following steps: mapping multi-dimensional resource parameters such as space, power, heat dissipation and load bearing of a machine room into a Riemannian manifold mathematical model, and constructing a resource measurement tensor representing a resource distribution density and a constraint relationship; calculating a resource allocation optimal path on the resource manifold, constructing a multi-objective function including space utilization rate, energy efficiency, heat dissipation efficiency and cost effectiveness, and generating an optimal deployment scheme of the machine room equipment; in combination with topological characteristic analysis of resource manifolds, potential resource bottlenecks are predicted, and a resource allocation scheme is dynamically adjusted; and the security and compliance of the evaluation scheme are verified through digital twinborn simulation, so that the problem of unbalanced resource allocation caused by independent planning of each system in traditional machine room planning is solved.

The invention discloses a low-voltage power line carrier communication network repeater deployment method and system, and relates to the technical field of power communication. The method comprises the following steps: (1) according to PLC network topology information, acquiring an STA set which cannot directly communicate with a CCO, and inputting model parameters; (2) calling a general optimization solver to carry out model solution to obtain a solution result; (3) if the feasible solution exists, outputting a PCO deployment mark; if not, adjusting the received signal threshold value and the transmitting power, and repeating the step (2); and (5) once a feasible solution is obtained or the model parameter reaches the set limit value, the solution is ended. According to the deployment planning model constructed by the invention, the optimal deployment position of the PCO in the network is solved by taking the minimum weighted sum of the total transmission time delay and the total cost as a target and taking the received signal strength and the single-hop relay as constraints; the problem that seamless coverage of a low-voltage power distribution network is difficult to realize by single-hop communication due to high background noise and frequency selective fading of a PLC channel is solved.

The invention provides a low-altitude unmanned system air node optimal deployment method based on communication-link-energy consumption joint optimization. The method comprises the following steps: initializing an unmanned aerial vehicle group; the system constructs a performance model according to the real-time channel information, constructs a multi-objective collaborative optimization problem, and solves the multi-objective optimization problem by adopting an MOPSO-HGD algorithm to obtain an optimal result; deploying the unmanned aerial vehicle group to the optimal position according to the optimal result to form a virtual distributed MIMO array structure; setting an unmanned aerial vehicle electric quantity threshold value, judging the size of each unmanned aerial vehicle remaining electric quantity and the unmanned aerial vehicle electric quantity threshold value, and if the unmanned aerial vehicle remaining electric quantity is smaller than the unmanned aerial vehicle electric quantity threshold value, recycling the unmanned aerial vehicle; otherwise, re-determining the position of the unmanned aerial vehicle group; through a closed-loop control mechanism of channel perception-multi-target optimization-dynamic adjustment, the unmanned aerial vehicle relay system can still keep a high signal-to-noise ratio and strong robustness in a complex low-altitude environment, is suitable for scenes such as low-altitude broadband communication, emergency rescue and urban information networks, and has a good engineering application prospect.

The invention relates to an intelligent networking planning system based on multi-objective dynamic optimization, which comprises a data input module, a data preprocessing module, a system modeling module, a multi-objective optimization module and a scheme output module, and is characterized in that the data input module is used for receiving geographic information, environmental conditions, equipment models and performance parameters input by a user; the data preprocessing module is used for carrying out standardization processing on the input geographic information and equipment parameters; the system modeling module is used for establishing a terrain and terrain model; the multi-target optimization module is used for carrying out multi-target dynamic optimization and generating an optimal deployment scheme; and the scheme output module is used for generating an equipment deployment scheme report. According to the invention, complementary coverage between devices is realized, the system cost is reduced, and the deployment efficiency is improved. The system has good expansibility and adaptability, can be flexibly adjusted according to different application scenes and requirements, improves the overall performance and adaptability of a low-altitude security system, and meets diversified low-altitude security requirements.

The invention discloses a model pruning method for a heterogeneous cloud edge-end cooperative system, and belongs to the technical field of model pruning of machine learning. The problem that in the prior art, a traditional pruning method based on a deep learning model is low in resource utilization rate is solved. The method comprises the following steps: S1, initializing a to-be-called model, analyzing the structure of the to-be-called model to form a structure set, distributing the structure set to a computing node, and generating an overhead predictor; s2, according to a pruning shape control strategy, calling an overhead predictor, and using a pruning evaluation function to obtain a used pruning evaluation function; and S3, in the pruned model set of different segmentation points, according to a time overhead evaluation function and a network condition, screening to obtain an optimal deployment model. According to the method, the operation efficiency of the pruned model on different hardware is effectively improved, the reasoning efficiency and the resource utilization rate can be controlled, different task requirements are met through dynamic weight adjustment, and the method can be applied to the pruning model.

Embodiments of the present application disclose a non-real-time resource scheduling method and device for heterogeneous computing resources. The method comprises: receiving a to-be-processed task and an optimized performance index sent by a request end; determining target computing resources corresponding to each subtask in the current available computing resources; constructing all deployment schemes according to the correspondence between each subtask and each target computing resource; determining the value of each performance index corresponding to each deployment scheme, and determining the optimal deployment scheme in each deployment scheme and returning to the request end. By applying the scheme provided by the embodiments of the present application, the characteristics of the tasks in the automatic driving platform can be targeted, that is, the tasks on the automatic driving platform have less variability, and periodic tasks are usually deployed, and the scheduling time has no effect on the execution efficiency. When the to-be-processed task is received, all possible deployment schemes are generated, and the performance index corresponding to each deployment scheme is determined, so that the optimal deployment scheme can be selected according to the user's optimized performance index.

The method and system for joint configuration of network security functions based on the maximum satisfiability modulo theory include the following steps: (1) Problem modeling process: modeling the joint configuration problem of network security functions as a maximum satisfiability modulo theory MaxSMT problem; (2) Hard constraint generation process; (3) Soft constraint generation process; (4) MaxSMT solution process; (5) Result extraction process: extracting the joint configuration results of network security functions from the model output by the solver; The method and system of the present invention can automatically determine the optimal deployment location, configuration rules and execution order of various network security functions in the network, significantly reduce the configuration complexity and improve the intelligence level of network security configuration while strictly meeting all security requirements.

The application discloses a low-altitude defense platform deployment system and method. The system comprises: a data acquisition module for acquiring defense area related data and available platform information; a capability quantification module for uniformly quantifying the capability parameters of each platform to obtain a standardized capability description; an AI agent search module comprising a plurality of parallel working AI agents and a shared memory area; each AI agent iteratively executes the generation of a candidate deployment scheme, constraint violation degree calculation and comprehensive performance score evaluation, and updates a local optimal deployment scheme based on the evaluation results, and exchanges scheme information through the shared memory area to guide subsequent search; and a scheme determination module for determining a global optimal deployment scheme when a termination condition is met. Through capability quantification and multi-agent parallel collaborative search, the application improves the comprehensive protection performance and planning efficiency of the heterogeneous defense platform deployment scheme, and realizes automatic global optimization deployment under complex constraint conditions.

The invention discloses a base station deployment method for mine space constraints and related products, which can be applied to the technical field of mine communication network deployment, and the method comprises the following steps: constructing M to-be-selected particles; wherein each particle to be selected comprises three-dimensional position information and antenna angle information of N base stations. And determining a three-dimensional coverage rate of each to-be-selected particle for the to-be-deployed mine, and taking the to-be-selected particle with the highest three-dimensional coverage rate as a first target particle. And iteratively adjusting the first target particle based on a preset updating strategy, and determining a three-dimensional coverage rate of the first target particle to the to-be-deployed mine after each iteration adjustment. And outputting the optimal deployment scheme when the obtained three-dimensional coverage rate meets a preset iteration convergence difference value through continuous K times of iteration adjustment. Therefore, the three-dimensional space information of the mine is taken as a constraint, the three-dimensional position and the antenna angle of the base station are jointly optimized by combining the particle swarm optimization algorithm, and the three-dimensional space coverage rate of deployment of the base station in the mine environment is improved.

The invention provides a MaaS platform large model reasoning parameter automatic optimization acceleration method, which comprises the following steps: constructing a standardized container environment and preparing pressure measurement data by detecting hardware configuration; traversing a multi-dimensional parameter space by adopting an intelligent search algorithm, and automatically discovering an optimal reasoning parameter of a specific hardware and model combination through a pressure measurement and performance scoring formula; the optimal parameter combinations corresponding to various hardware models and model versions are persistently stored in a structured database, and a reusable parameter knowledge base is formed; and when the model is deployed, the current hardware environment is automatically identified, and corresponding optimal parameter configuration is intelligently matched and loaded from the parameter library, so that zero-configuration optimal deployment is realized. Through a parameter combination feasibility verification mechanism, parameter combinations which can cause system errors or memory overflow can be found and eliminated in time, it is ensured that the finally selected parameters can improve the performance, and stable operation of the system can also be ensured.

The present application relates to cloud computing resource scheduling technical field, especially in a kind of real-time virtual machine deployment method and system based on buffer queue ant colony algorithm, comprising: complete computing period is divided into several sub-periods, each sub-period includes scheduling time period and deployment time period;In scheduling time period, the optimal deployment result of current sub-period is solved using buffer queue ant colony algorithm;In deployment time period, according to the optimal deployment result of current sub-period, virtual machine is deployed.The present application can significantly reduce the number of working servers in different scale scenarios and improve resource utilization, effectively solve the energy consumption optimization and resource balance problem under the real-time load of cloud computing.

The invention discloses a multi-layer satellite network control node optimization deployment method based on a GCN, relates to the technical field of satellite network management and graph neural network application, and solves the problems that in an existing large-scale network, the deployment calculation complexity of a controller is high; in order to solve the problems that an MCD model lacks research on inter-layer interactive modeling in a multi-layer control plane and the like, the cooperative work of a super controller and a controller is realized by constructing a multi-layer satellite network architecture, so that the network response capability of a cross-domain service is improved. And a mathematical model which aims at reducing the network delay and balancing the load of the controller is established, and solving is carried out through an SA algorithm. Besides, a multi-controller deployment algorithm of the GCN network is provided, a solution process of a neural network learning SA algorithm is utilized, a rapid approximate optimal deployment strategy is realized, and technical support is provided for application of a large-scale satellite network. Experimental results show that the method is superior to other related schemes in the aspects of network response time delay, load balancing and expandability.

The application discloses a Kubernetes micro-service optimal deployment method and device based on reward accumulation deep reinforcement learning and a medium, relates to the field of electric digital data processing, and comprises the following steps: S1, constructing a multi-target micro-service deployment optimization problem; S2, constructing a micro-service resource preference function combined with an optimization target; S3, constructing a reinforcement learning model; S4, constructing a reward function and training the model; and S5, obtaining an optimal deployment strategy online according to a real state through the trained model. The application considers both the lowest service response delay and resource load balancing by constructing a multi-target micro-service deployment optimization problem, effectively converts the Kubernetes deployment requirement into a computable and solvable optimal problem, designs a micro-service resource preference function, combines the function with the optimization target, reduces the calculation complexity during model training, guarantees the real-time performance of the Kubernetes deployment, and thus meets the development requirement of the micro-service architecture.

The present invention relates to the field of intelligent reflective surface deployment position optimization, and specifically to an intelligent reflective surface deployment position optimization system. The solution includes: a channel estimator cooperates with a transmitter, a receiver, and an intelligent reflective surface to carry out wireless channel estimation, and then sends the obtained wireless channel information to a deployment optimizer. After receiving the instruction signal sent by the coordinator and the wireless channel information sent by the channel estimator, the deployment optimizer obtains the optimal deployment position of the intelligent reflective surface according to the deployment optimization algorithm in a single receiver scenario, and then sends the optimal deployment position of the intelligent reflective surface to the position controller corresponding to the intelligent reflective surface. After receiving the optimal deployment position of the corresponding intelligent reflective surface, the position controller generates a control signal for the driver, and controls the driver to move the intelligent reflective surface to the corresponding optimal deployment position. The present invention is suitable for optimizing the deployment position of intelligent reflective surfaces.

The application provides an APT attack active defense method based on a four-honey system, which comprises the following steps: collecting events and external threat intelligence fed back by defense components in the four-honey system to obtain security observation data, and generating an alignment subgraph representing anchored inter-tactical behavior relationships; combining the alignment subgraph with an APT knowledge graph to implement explicit relationship reasoning and implicit relationship reasoning to generate a candidate attack intent set and a confidence distribution thereof; generating an optimal deployment strategy under the constraint of system resource state, encapsulating the optimal deployment strategy into an executable work order; calling resources for deployment and generating a deployment state reply to complete new trapping environment construction; collecting attacker behavior data, evaluating strategy effectiveness according to the attacker behavior data and the deployment strategy, and updating the strategy deployment priority. The method can be applied to real-time strategy adaptation and automatic resource scheduling in response to attack behavior evolution, and can significantly improve flexibility and continuous interference capability in a complex attack and defense environment.

The invention relates to a controller deployment method and device for a software-defined network and a medium, and the method comprises the steps: S1, obtaining a topological graph of the software-defined network, and carrying out the problem modeling of network controller deployment; wherein the network controller deployment problem comprises a sub-problem of reducing S-C average time delay, a sub-problem of reducing S-C worst time delay, a sub-problem of reducing C-C average time delay and a sub-problem of reducing global time delay; s2, performing mutual information propagation between adjacent nodes by adopting a multilayer graph convolutional network, extracting embedded features from a topological graph of the software defined network, and performing clustering operation on the embedded features to obtain segmented sub-graphs; and S3, for each segmented sub-graph, determining the optimal deployment position of the controller in each sub-graph in combination with network controller deployment problem modeling information. Compared with the prior art, the controller deployment scheme obtained through optimization effectively reduces network time delay.

The application discloses a deployment method, system and device of a service function chain based on a micro-service architecture. The method comprises the following steps: acquiring position information of each terminal and communication time consumption of each terminal to each edge server; taking the position information and the communication time consumption of each terminal as a feature vector, and dividing all terminals into several terminal clusters by using a clustering algorithm according to the feature vector; acquiring a service function chain to be deployed, distributing all micro-services of each terminal in the service function chain to edge servers corresponding to terminal clusters to which the terminals belong, and obtaining an initial deployment scheme; inputting the initial deployment scheme into a deployment optimization model, and iteratively optimizing by using a genetic algorithm until a termination condition is met, and outputting an individual with the optimal fitness in a current population as an optimal deployment scheme; and deploying the service function chain on the edge servers according to the optimal deployment scheme. Therefore, the application can find the optimal deployment position for a complex service function chain.

The embodiment of the invention provides a low-altitude communication node deployment method and system based on deep learning and reinforcement learning. The method comprises the following steps: acquiring a city three-dimensional environment map; wherein the urban three-dimensional environment map comprises the terrain, the building height and the building shielding information of the city; through a predetermined radio map prediction model, carrying out signal to interference plus noise ratio distribution prediction on the urban three-dimensional environment map under different node deployments to obtain a signal to interference plus noise ratio prediction distribution map; performing iterative optimization on the signal to interference plus noise ratio prediction distribution diagram through a predetermined reinforcement learning optimization model, and determining an optimal deployment strategy; determining the position of a target base station based on the optimal deployment strategy; wherein the target base station position is the position of the target node in the plurality of different nodes determined based on the optimal deployment strategy. According to the scheme, high-precision propagation prediction and adaptive optimization deployment are realized in a complex low-altitude environment, and the deployment efficiency and the communication performance can be improved.