World expo problem and its mixed integer programming based solution

  • Hongteng Xu
  • , Dixin Luo
  • , Xiaoming Huo
  • , Xiaokang Yang

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

2 Citations (Scopus)

Abstract

In this paper, we introduce an interesting "World Expo problem", which aims to identify and track multiple targets in a sensor network, and propose a solution to this problem based on the mixed integer programming. Compared with traditional tracking problem in the sensor network, the World Expo problem has following two features. Firstly, the target in the network is not limited to single individuals. It can also be a group composed of multiple individuals with same path in the network, which implies that multiple targets can share the same path and be detected by the same sensor at the same time. Moreover, both the size and the number of groups are unknown. Secondly, differing from traditional sensor networks, the sensor network in the World Expo problem usually is sparse. These two features increase the difficulty in identification and tracking. To solve the aforementioned problem, we analyze the solvability of this problem and come up with a mixed integer programming based algorithm. The simulation result shows that our method has good performances and is robust to errors in the data.

Original languageEnglish
Title of host publicationBehavior and Social Computing - Int. Workshop on Behavior and Social Informatics, BSI 2013 and Int. Workshop on Behavior and Social Informatics and Computing, BSIC 2013, Revised Selected Papers
PublisherSpringer Verlag
Pages56-67
Number of pages12
ISBN (Print)9783319040479
DOIs
Publication statusPublished - 2013
Externally publishedYes
Event2013 International Workshop on Behavior and Social Informatics and Computing, BSIC 2013 - Beijing, China
Duration: 3 Aug 20139 Aug 2013

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume8178 LNAI
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference2013 International Workshop on Behavior and Social Informatics and Computing, BSIC 2013
Country/TerritoryChina
CityBeijing
Period3/08/139/08/13

Keywords

  • Integer programming
  • Path identification
  • Sensor networks
  • Social collective behavior

Fingerprint

Dive into the research topics of 'World expo problem and its mixed integer programming based solution'. Together they form a unique fingerprint.

Cite this