Skip to content

A multi-agent architecture for designing and simulating large scale wireless systems resource allocation

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

The simulation model adaptability to real network behavior is the key concept in wireless communications. In a cellular network, many procedures such as call admission, hand-off, etc take place simultaneously for every individual user. Every network procedure acts autonomously, interacts with the network environment (gathers information such as interference conditions), takes decisions (e.g. call establishment), etc. Although this is known in the literature, there is lack of suitable representations for such network procedures in the simulation systems proposed so far, thus compromising simulation model adaptability to real network behavior. To achieve such adaptability we herein propose to change the point of view in network procedure representation. Instead of viewing them as independent programming functions or even objects in a high level language, which are sequentially executed, due to their aforementioned properties it is proposed that such network procedures could be more efficiently modeled as agents. Considering this new approach, the agent cooperation and communication in terms of negotiation and agreement is a critical issue. In this paper we present a centralized cooperative multi-agent negotiation scheme applied to a multi-agent layered architecture for designing and simulating resource allocation in cellular communication systems, based on organizational modeling. Moreover, we show the way that the rules and implementation methods of agent negotiation affect the adaptation grade of simulation model to the real cellular network behavior.

Original languageEnglish
Title of host publicationAgent and Multi-Agent Systems
Subtitle of host publicationTechnologies and Applications - First KES International Symposium, KES-AMSTA 2007, Proceedings
PublisherSpringer
Pages405-415
Number of pages11
ISBN (Print)9783540728290
DOIs
Publication statusPublished - 1 Jun 2007
Event1st KES International Symposium on Agent and Multi-Agent Systems - Technologies and Applications - Wroclaw, Poland
Duration: 31 May 20071 Jun 2007

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume4496 LNAI
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference1st KES International Symposium on Agent and Multi-Agent Systems - Technologies and Applications
Abbreviated titleKES-AMSTA 2007
CountryPoland
CityWroclaw
Period31/05/071/06/07

Related information

Relations Get citation (various referencing formats)

ID: 20987240