A proposal of a quasi-solution state evolution algorithm for channel assignment problems

Nobuo Funabiki, Toru Nakanishi, Tokumi Yokohira, Shigeto Tajima, Teruo Higashino

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

The channel assignment problem (CAP) in a cellular network requires finding a channel assignment to the call requests from cells such that three types of interference constraints are not only satisfied, but also the number of channels (channel span) is minimized. This paper presents a three-stage iterative algorithm, called the Quasi-solution state evolution algorithm for CAP (QCAP). QCAP evolutes quasi-solution states where a subset of call requests is assigned channels and no more request can be satisfied without violating the constraint. The first stage computes the lower bound on the channel span. After the second stage greedily generates an initial quasi-solution state, the third stage evolutes them for a feasible solution by iteratively generating best neighborhoods, with help of the dynamic state jump and the gradual span expansion for global convergence. The performance is evaluated through solving benchmark instances in literature, where QCAP always finds the optimum or near-optimum solution in very short time.

Original languageEnglish
Title of host publicationInformation Networking
Subtitle of host publicationWireless Communications Technologies and Network Applications - International Conference, ICOIN 2002, Revised Papers
EditorsIlyoung Chong
PublisherSpringer Verlag
Pages32-41
Number of pages10
ISBN (Electronic)3540442553, 9783540442554
DOIs
Publication statusPublished - 2002
EventInternational Conference on Information Networking, ICOIN 2002 - Cheju Island, Korea, Republic of
Duration: Jan 30 2002Feb 1 2002

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume2344
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

OtherInternational Conference on Information Networking, ICOIN 2002
Country/TerritoryKorea, Republic of
CityCheju Island
Period1/30/022/1/02

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Fingerprint

Dive into the research topics of 'A proposal of a quasi-solution state evolution algorithm for channel assignment problems'. Together they form a unique fingerprint.

Cite this