A Distributed Intelligent Routing Method for UAV Network Slicing

A machine network and distributed technology, applied in the field of distributed intelligent routing for UAV network slicing, to achieve the effect of fast path selection

Active Publication Date: 2021-06-11
UNIV OF ELECTRONICS SCI & TECH OF CHINA
View PDF7 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The purpose of the present invention is to provide a distributed intelligent routing method for UAV network slicing, to solve the problem of how to improve communication quality in the dynamic network environment of UAV network slicing scenarios

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A Distributed Intelligent Routing Method for UAV Network Slicing
  • A Distributed Intelligent Routing Method for UAV Network Slicing
  • A Distributed Intelligent Routing Method for UAV Network Slicing

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0049] Such as figure 1shown. Embodiment 1 is a distributed intelligent routing method for UAV network slicing. First, by modeling the UAV network as a network model; then setting constraints on the network model, the constraints include: delay limit, rate limit, packet loss rate limit; when facing low-latency slices, The constrained network model is built as a multi-constraint optimization model; then the multi-constraint optimization model is solved by a reinforcement learning model to obtain a solution value. During the solution process, each communication node independently saves the link status and update the link status in real time; and then perform dynamic route selection according to the solution value and the link status.

[0050] In a static network environment, the network model in Embodiment 1 is established as an undirected graph G(V, E, W) with weighted edges, where V is the set of communication nodes in the network, expressed as V={v 1 ,v 2 ,...,v n}, n is...

Embodiment 2

[0071] by image 3 The specific process of Q-learning is introduced as an example. Find a path from v1 to v7 that meets the QoS requirements. Designate each node as a state, in image 3 In the network shown, the number of states is 7. In each state, the action is defined as the optional next hop of each node. Taking v1 as an example, the size of its action set is 3, and there are three optional actions: v2, v3, and v4. Three Q values ​​are defined in each state, which are respectively used to describe the delay, packet loss rate, and bandwidth from the current node to the destination node. v1 first uses the ε-greedy strategy to send data packets, and judges which node the next hop of the data packet should be sent to according to the Q value of v2, v3, and v4. If the packet loss rate and bandwidth meet the requirements when v2 is selected (judged by Q12(e) and Q12(v)), update Q12(d), and repeat the same steps for v2 until v7 is selected, and an iterative process is complet...

Embodiment 3

[0075] Suppose a service flow in the current network initiates a routing request, its source node is s, destination node is d, Q ij (d), Q ij (e) and Q ij (v) respectively represent the cumulative link delay, cumulative packet loss rate and path rate from the current node to the destination node d, ∞ represents infinity, each node will maintain its own three types of Q value information, Indicates the neighbor node i of node j, and l indicates the experimental round.

[0076] Now known: the source node s of the service flow, the destination node j, the service type (here, the low-latency service is taken as an example), and the QoS requirement;

[0077] The purpose is to obtain: the path selection scheme π of this service.

[0078] Initialization phase:

[0079]

[0080] Qij (d)=∞,Q ij (e)=1,Q ij (v)=0 / / initialize Q-value

[0081] Online learning phase:

[0082]

[0083]

[0084] output:

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses a distributed intelligent routing method for unmanned aerial vehicle network slicing, comprising the steps of: modeling the unmanned aerial vehicle network as a network model; setting constraints on the network model, the constraint conditions include: delay limit, rate limit , packet loss rate limitation; when facing low-latency slicing, the constrained network model is established as a multi-constraint optimization model; the multi-constraint optimization model is solved by the reinforcement learning model to obtain the solution value. During the solution process, each Each communication node independently saves the link status of itself and its neighbor nodes and updates the link status in real time; performs dynamic routing selection according to the solution value and link status. The invention effectively solves the problem that the routing method in the prior art cannot adapt to the high delay requirement of the UAV network and the characteristics that the network changes dynamically at any time. The invention fills up this technical blank and creates better conditions for the network environment of the drone.

Description

technical field [0001] The invention relates to a wireless network routing method, in particular to a distributed intelligent routing method for unmanned aerial vehicle network slicing. Background technique [0002] The traditional algorithm based on the shortest path algorithm to solve the routing policy will be difficult to apply to the scenario of UAV network slicing. This is because the current network is not adaptable to the real-time changing network environment, and the path selected by the shortest path algorithm is relatively single. When the number of business flows in the network gradually increases, the possibility of network congestion will increase significantly. On the basis of the shortest path algorithm, if the degree of network congestion is expected to be reduced, the number of business flows that can be carried will be sacrificed. In addition, in actual scenarios, due to the mobility of nodes, the communication quality of links in the network will also c...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Patents(China)
IPC IPC(8): H04W16/22H04W40/12H04W40/22H04W40/24
CPCH04W16/22H04W40/12H04W40/22H04W40/248Y02T10/40
Inventor 陈博伦孙耀秦爽冯钢
Owner UNIV OF ELECTRONICS SCI & TECH OF CHINA
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap