Method for figuring out shortest path of large scale graphs based on fewest resource neural networks

A shortest path and neural network technology, applied in the direction of biological neural network models, can solve the problems that the quality and computational complexity are easily affected by other parameters, affect the total number of iterations of network convergence, and cannot be applied to large-scale problems. Effects of quality and computational complexity, minimal resources, fast convergence

Inactive Publication Date: 2012-02-22
XIDIAN UNIV
View PDF4 Cites 12 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The disadvantage of this method is that the number of neurons required by the PCNN model is proportional to the sum of the side lengths in the graph. Many neural network algorithms based on PCNN have proposed to use only the same number of neurons as the number of nodes in the graph to find the shortest path. But the neurons of the network are still based on integrate-and-fire like PCNN, so the network behaves as a nonlinear differential equation system. When the number of nodes in the graph becomes larger, the calculation of the system is very complicated; secondly, in the user Within the predetermined parameter range, the time step parameter will affect the total number of iterations of network convergence, thereby affecting the computational complexity of the algorithm; at the same time, the quality and computational complexity of the solution are also easily affected by other parameters, so these methods cannot be applied to large scale problem

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
  • Method for figuring out shortest path of large scale graphs based on fewest resource neural networks
  • Method for figuring out shortest path of large scale graphs based on fewest resource neural networks
  • Method for figuring out shortest path of large scale graphs based on fewest resource neural networks

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0039] Example 1: Comparison between the present invention and the Dijkstra method Dijkstra, the A-star method A* and the pulse-coupled neural network method PCNN in the prior art under simulated data.

[0040] Embodiment 1 has selected Dijkstra method Dijkstra of prior art, A star method A* and these three methods of pulse-coupled neural network PCNN and the present invention in the quality Quality of solution, CPU running time CPU time and number of iterations Iters are compared in three aspects. The four methods in Embodiment 1 are all coded and implemented in the vc++6.0 environment, and run on a computer with a 2.5GHz processor and 2G memory.

[0041] In the traditional Dijkstra method Dijkstra, the A star method A* and the pulse-coupled neural network method PCNN in the prior art, the Dijkstra method Dijkstra finds the single-source shortest path that does not contain negative edges in the graph, and generates a shortest path tree. The A star method A* solves the short...

Embodiment 2

[0050] Embodiment 2 is the comparison result of the present invention and the prior art Dijkstra method Dijkstra large-scale real data.

[0051]

[0052] Embodiment 2 compares the present invention with large-scale real data. Embodiment 2 selects six large-scale urban maps on the website http: / / www.dis.uniroma1.it / ~challenge9 / index.shtml, respectively New York, San Francisco Bay area, Colorado, Florida, Northwest USA and Northeast USA. The first five nodes are set as source nodes respectively, and the average performance of Dijkstra's method Dijkstra and the present invention in terms of solution quality, iterations and CPU time are compared. Embodiment 2 runs on a computer with a 3.6GHz processor and 3G internal memory.

[0053] The above table is obtained after comparing the Dijkstra method Dijkstra of the present invention and the prior art in terms of iterations, CPU running time CPU time and solution quality. It can be seen from the above table that for large-scale ...

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 method for figuring out shortest path of large scale graphs based on fewest resource neural networks. The invention is used for finding a shortest path tree of large scale data through fewest resources, and the found resolution is an optimal resolution, so that the optimal resolution can be applied on network route, transport dispatch and urban traffic planning. The method provided by the invention comprises the steps of: initializing a large scale graph network; generating an automatic wave at a network source node; calculating transmission time of the node; determining the latest arrival time of the node; figuring out the competition win node, and adding the competition win node in the shortest path of the node; adding an iteration time, repeatedly calculating the transmission times and latest arrival times of the corresponding nodes, and updating the shortest path until the network convergence; and outputting the found shortest path and the corresponding transmission time.

Description

technical field [0001] The invention belongs to the technical field of data processing, and further relates to the shortest path method for solving large-scale graphs based on the least resource neural network in the field of discrete optimization. This method can solve the shortest path tree of a large-scale graph with the least resources, and can be used in network routing, transportation scheduling and urban traffic planning. Background technique [0002] In recent years, with the popularization of the network and the rapid development of traffic, how to solve the problems of network routing, transportation scheduling and traffic planning more effectively has become more and more important, which involves how to find the shortest path. [0003] The patent "Network Protection Method Based on Shared Risk Link Group in WDM Optical Network" (publication number CN 101026482A, application number 200610007954.0) applied by Beijing University of Posts and Telecommunications disc...

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 Applications(China)
IPC IPC(8): G06N3/02
Inventor 张军英何晓涛
Owner XIDIAN UNIV
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