Self-guiding lift-off maneuvering network workload prediction method

Abstract

The invention discloses a self-guiding lift-off maneuvering network work load prediction method, and relates to the technical field of network load prediction, the self-guiding lift-off maneuvering network work load prediction comprises a resource prediction model, a federated learning module and a local data set, the federated learning module is specifically a distributed training architecture and is used for training a deep neural network in a specific scene, and the local data set is used for predicting the work load of the self-guiding lift-off maneuvering network. Aggregating the models of the local clients through an aggregation strategy to obtain a global model; the resource prediction model comprises a non-ground mobile network module, a local prediction module and an available resource cooperation module; the non-ground mobile network module is specifically a communication computing network constructed in the maneuvering process of the unmanned aerial vehicle group; according to the self-guiding lift-off maneuvering network work load prediction method, the non-ground mobile network has a good supplement effect on the ground network due to the flexibility of the structure, the maneuverability of the nodes and the diversity of the carried resources, and meanwhile, the resource guarantee requirement of an imminent scene is met.

Description

Technical Field

[0001] The present invention relates to the technical field of network load prediction, and in particular to a method for predicting the workload of a self-guided aerial maneuvering network. Background Art

[0002] In order to cope with the computing and communication needs of resource-constrained IoT devices without preset infrastructure, non-terrestrial mobile networks (NTMNs) came into being. It refers to a flexible and efficient communication and computing network built by drones (groups) equipped with high-performance computing equipment during maneuvers, and has the following characteristics: First, drones can dynamically adjust flight strategies and plan flight trajectories according to user needs. Secondly, non-terrestrial mobile networks with a certain flight altitude can reduce geographical effects and strengthen and expand cellular network coverage. Finally, because line-of-sight communication with ground equipment can be achieved, there will be an opportunity to reduce drone channel restrictions.

[0003] Available resource prediction is an important basis for efficient scheduling of network and computing resources in non-terrestrial mobile networks, and is an important prerequisite for ensuring the reliable operation of non-terrestrial mobile networks. Continuous prediction of available resources can form a preliminary expectation of future network conditions and node performance. This can assist in the intelligent configuration of network and computing resources, enabling non-terrestrial mobile networks to automatically adjust bandwidth and allocate computing resources based on accurate prediction results and service requirements. This can avoid potential performance issues and reduce the risk of service interruptions. Ensure that the system can maintain stable operation under different loads and network conditions.

[0004] Although the research on resource prediction has been relatively sufficient and has achieved remarkable results. However, traditional methods are usually aimed at production clusters in cloud environments. The nearly unlimited computing resources mean that they do not need to consider the parameter scale of the prediction model. At the same time, due to the load limit of drones, non-ground mobile networks naturally face the practical challenge of relatively low resources. Prediction models with large parameter scales will increase the workload on the non-business side and reduce the actual resource utilization efficiency. The practical challenge of low resources in non-ground mobile networks will bring about two trade-offs.

[0005] In order to solve the above problems, we designed a workload prediction method for self-guided aerial maneuvering networks. Summary of the invention

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: a method for predicting workload of a self-guided aerial mobility network, including a resource prediction model, a federated learning module and a local data set; The federated learning module is specifically a distributed training architecture, which is used to train deep neural networks in specific scenarios, and to obtain a global model by aggregating the models of each local client through an aggregation strategy.

[0007] The resource prediction model includes a non-terrestrial mobile network module, a local prediction module and an available resource coordination module; The non-ground mobile network module is specifically a communication computing network constructed by the drone group during the maneuvering process; Non-terrestrial mobile networks provide both computing resources and network resources. Their main application direction is to provide resource guarantees for service needs in highly temporary scenarios. Such scenarios usually have no preset infrastructure, and the natural, social and information environments present stronger uncertainties. Intelligent services also show the characteristics of rapid changes in time, space and resource requirements. Due to the flexibility of their structure, the mobility of their nodes and the diversity of resources they can carry, non-terrestrial mobile networks are a good supplement to terrestrial networks, while also adapting to the resource guarantee requirements of temporary scenarios.

[0008] The local prediction module is used to predict the resource margin of a single node in a non-terrestrial mobile network; The available resource collaboration module makes full use of data distributed on different nodes by adopting a federated learning framework. The different nodes are based on local data sets and update the model parameters to the global model through an encrypted communication protocol. The global model can integrate knowledge and information from multiple nodes, thereby improving the accuracy and robustness of predictions; Preferably, the non-terrestrial mobile network is composed of: Each non-ground mobile node is carried by a medium-sized rotor drone, equipped with a mobile 5G base station and server. Compared with traditional fixed base stations, drones have more flexible coverage and service capabilities. At the cost of some flight energy consumption, drones can reach more suitable locations, which can reduce data transmission time and computing energy consumption and improve service quality.

[0009] Predicting available resources is crucial for non-terrestrial mobile networks to achieve efficient allocation of network and computing resources. It is the basis for maintaining stable operation of non-terrestrial mobile networks. By continuously predicting available resources, we can have a basic understanding of the future trends of network status and node performance. Such predictions can help achieve intelligent configuration of network and computing resources, allowing non-terrestrial mobile networks to automatically adjust bandwidth and allocate computing resources based on prediction results and service requirements. Doing so can prevent potential performance issues, reduce the possibility of service interruptions, and ensure that the system can maintain stable operation when facing different workloads and network conditions.

[0010] An improved heuristic method based on the black hole phenomenon is used to train neurons in feedforward neural networks. The prediction error trend is also captured by calculating the deviation of the most recent prediction, further improving the accuracy of the prediction. An improved heuristic method based on the black hole phenomenon is used to train neurons in feedforward neural networks. The prediction error trend is also captured by calculating the deviation of the most recent predictions, further improving the accuracy of the prediction.

[0011] The present invention provides a method for predicting the workload of a self-guided aerial maneuvering network. It has the following beneficial effects: 1. The workload prediction method of the self-guided aerial mobile network provides both computing resources and network resources through non-ground mobile networks. The main application direction is to provide resource guarantees for service needs in highly temporary scenarios. Such scenarios usually have no preset infrastructure, and the natural, social and information environments present stronger uncertainties. Intelligent services also show significant characteristics of rapid changes in time, space and resource requirements. Due to the flexibility of their structure, the mobility of their nodes and the diversity of resources they can carry, non-ground mobile networks are a good complement to ground networks, and at the same time they are adapted to the resource guarantee requirements of temporary scenarios.

[0012] Second, the workload prediction method of the self-guided launch maneuver network can integrate knowledge and information from multiple nodes through a global model, thereby improving the accuracy and robustness of the prediction.

[0013] 3. The self-guided aerial mobile network workload prediction method uses drones to have more flexible coverage and service capabilities than traditional fixed base stations. At the cost of part of the flight energy consumption, the drone can reach a more suitable location, which can reduce data transmission time and computing energy consumption and improve service quality.

[0014] Fourth, the workload prediction method of the self-guided aerial mobile network is crucial for non-ground mobile networks to achieve efficient allocation of network and computing resources by predicting available resources. It is the basis for maintaining the stable operation of non-ground mobile networks. By continuously predicting available resources, we can have a basic understanding of the future trends of network status and node performance. Such predictions can help realize the intelligent configuration of network and computing resources, so that non-ground mobile networks can automatically adjust bandwidth and allocate computing resources based on prediction results and service needs. Doing so can prevent potential performance issues, reduce the possibility of service interruptions, and ensure that the system can maintain stable operation when facing different workloads and network conditions. DETAILED DESCRIPTION

[0015] The present invention will be further described in detail below in conjunction with specific embodiments. The embodiments of the present invention are provided for the purpose of illustration and description, and are not intended to be exhaustive or to limit the present invention to the disclosed forms. Many modifications and variations will be apparent to those of ordinary skill in the art. The embodiments are selected and described to better illustrate the principles and practical applications of the present invention, and to enable those of ordinary skill in the art to understand the present invention and thereby design various embodiments with various modifications suitable for specific uses.

[0016] In a first embodiment, the present invention provides a technical solution: a method for predicting workload of a self-guided aerial mobility network, including a resource prediction model, a federated learning module, and a local data set; The federated learning module is specifically a distributed training architecture, which is used to train deep neural networks in specific scenarios, and to obtain a global model by aggregating the models of each local client through an aggregation strategy.

[0017] The resource prediction model includes a non-terrestrial mobile network module, a local prediction module and an available resource coordination module; The non-ground mobile network module is specifically a communication computing network constructed by the drone group during the maneuvering process; Non-terrestrial mobile networks provide both computing resources and network resources. Their main application direction is to provide resource guarantees for service needs in highly temporary scenarios. Such scenarios usually have no preset infrastructure, and the natural, social and information environments present stronger uncertainties. Intelligent services also show the characteristics of rapid changes in time, space and resource requirements. Due to the flexibility of their structure, the mobility of their nodes and the diversity of resources they can carry, non-terrestrial mobile networks are a good supplement to terrestrial networks, while also adapting to the resource guarantee requirements of temporary scenarios.

[0018] The local prediction module is used to predict the resource margin of a single node in a non-terrestrial mobile network; The available resource collaboration module makes full use of data distributed on different nodes by adopting a federated learning framework. The different nodes are based on local data sets and update the model parameters to the global model through an encrypted communication protocol. The global model can integrate knowledge and information from multiple nodes, thereby improving the accuracy and robustness of predictions; In the second embodiment, based on the first embodiment, the non-terrestrial mobile network is composed of the following: Each non-ground mobile node is carried by a medium-sized rotor drone, equipped with a mobile 5G base station and server. Compared with traditional fixed base stations, drones have more flexible coverage and service capabilities. At the cost of some flight energy consumption, drones can reach more suitable locations, which can reduce data transmission time and computing energy consumption and improve service quality.

[0019] Predicting available resources is crucial for non-terrestrial mobile networks to achieve efficient allocation of network and computing resources. It is the basis for maintaining stable operation of non-terrestrial mobile networks. By continuously predicting available resources, we can have a basic understanding of the future trends of network status and node performance. Such predictions can help achieve intelligent configuration of network and computing resources, allowing non-terrestrial mobile networks to automatically adjust bandwidth and allocate computing resources based on prediction results and service requirements. Doing so can prevent potential performance issues, reduce the possibility of service interruptions, and ensure that the system can maintain stable operation when facing different workloads and network conditions.

[0020] An improved heuristic method based on the black hole phenomenon is used to train neurons in feedforward neural networks. The prediction error trend is also captured by calculating the deviation of the most recent prediction, further improving the accuracy of the prediction. An improved heuristic method based on the black hole phenomenon is used to train neurons in feedforward neural networks. The prediction error trend is also captured by calculating the deviation of the most recent predictions, further improving the accuracy of the prediction.

[0021] Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field and related fields without creative work should fall within the scope of protection of the present invention. The structures, devices and operating methods not specifically described and explained in the present invention are implemented according to the conventional means in the field unless otherwise specified and limited.

Claims

1. A method for predicting workload of a self-guided aerial maneuvering network, characterized in that: Includes resource prediction models, federated learning modules, and local datasets; The federated learning module is specifically a distributed training architecture, which is used to train deep neural networks in specific scenarios, and to obtain a global model by aggregating the models of each local client through an aggregation strategy; The resource prediction model includes a non-terrestrial mobile network module, a local prediction module and an available resource coordination module; The non-ground mobile network module is specifically a communication computing network constructed by the drone group during the maneuvering process; The local prediction module is used to predict the resource margin of a single node in a non-terrestrial mobile network; The available resource collaboration module fully utilizes the data distributed on different nodes by adopting a federated learning framework. The different nodes are based on local data sets, and the different nodes aggregate the model parameter updates to the global model through an encrypted communication protocol.

2. A method for predicting workload of a self-guided aerial maneuvering network according to claim 1, characterized in that: The non-terrestrial mobile network is composed of: Each non-ground mobile node is based on a medium-sized rotorcraft drone and is equipped with a mobile 5G base station and server.