Dynamic Agent Population in Agent-Based Distance Vector Routing Page: 1
6 p.View a full description of this paper.
- Highlighting
- Highlighting On/Off
- Color:
- Adjust Image
- Rotate Left
- Rotate Right
- Brightness, Contrast, etc. (Experimental)
- Cropping Tool
- Download Sizes
- Preview all sizes/dimensions or...
- Download Thumbnail
- Download Small
- Download Medium
- Download Large
- High Resolution Files
- IIIF Image JSON
- IIIF Image URL
- Accessibility
- View Extracted Text
Extracted Text
The following text was automatically extracted from the image on this page using optical character recognition software:
Dynamic Agent Population in Agent-Based Distance
Vector Routing
Kaizar A. Amin and Armin R. Mikler
Department of Computer Science
University of North Texas
Denton, TX 76203
(amin, mikler)@cs.unt.edu
Abstract
The Intelligent mobile agent paradigm can be applied to a wide variety of intrinsically parallel and
distributed applications. Network routing is one such application that can be mapped to an agent-based
approach. The performance of any agent-based system will depend on its agent population. Although
a lot of research has been conducted on agent-based systems, little consideration has been given to
the importance of agent population in dynamic networks. A large number of constituent agents can
increase the resource overhead of the system, thereby impeding the overall performance of the network.
Hence, it is imperative to find the optimal number of agents in the system that would maximize the
efficiency of the agent-based mechanism in the network. This optimal value cannot be determined
manually, thereby emphasizing the need for an adaptive approach that manipulates the number of
agents in the system based on its resource availability. This paper discusses an agent-based approach
to Distance Vector Routing, referred as Agent-based Distance Vector Routing and also describes an
adaptive approach controlling the number of agents in the network using pheromones and discusses
their limitations.
1. INTRODUCTION
Agents, Software Agents, and Intelligent Mobile Agents are terms, which describe the concept of mobile
computing or mobile code. The mobile agent paradigm has attracted attention from many fields of com-
puter science. The appeal of mobile agents is quite alluring - mobile agents roaming the network could
search for information, meet and interact with other agents that roam the network or remain bound to a
particular machine. An agent manifests four distinct characteristics, namely, Intelligence, Communication,
Autonomy, and Mobility [Minar et al. 1998, Di Caro & Dorigo 1997]. Intelligence is the ability of the agent
to adapt itself and/or change its environment based on the information available to it. Communication is
the property of an agent whereby it coordinates with other agents residing on the same node in exchanging
data, making decisions to merge itself [Mikler & Chokhani 2001], or planning its future strategies. Through
autonomy, the agent has the authority to control its actions and strategies without the necessity of human
control. Mobility is the property of the agents that make them conducive for distributed network applica-
tions. Mobile agents have the ability to migrate throughout the network, performing specific tasks at each
node reaching towards a global goal.
The agent-based paradigm can be applied to many intrinsically parallel network application. A large
number of applications in communication networks have been identified that can benefit from an agent-
based approach. Some examples on the system level include load balancing, network management and
network routing [Di Caro & Dorigo 1997, Minar et al. 1999]. In this paper we apply the agent paradigm
to network routing. Most of the work in agent-based network routing is based on insect colonies. It relies
on the principles that individual insects exhibit a simple behavior while collective communities exhibit
complex problem solving capabilities. For example, individual ants have limited abilities, however ant
colonies are capable of performing tasks that are remarkably complex. Considerable research has been
conducted in mapping the foraging activities of ants to routing and network management activities of
mobile agents. Real ants are represented as artificial agents that traverse the network collecting specific
information from their environment and coordinate their actions through pheromones. On the basis of this
information the agents make several decisions to adapt their behavior (Reactive Agents) and/or change
the existing environment affecting their future inputs (Proactive Agents).
This paper focusses on a new implementation of Distance Vector Routing Algorithm (DVR) using
an agent-based approach- Agent-based Distance Vector Routing (ADVR). DVR is a simple, iterative,
asynchronous and completely distributed routing algorithm [Bertsekas & Gallager 1987]. Certain imple-
mentations of DVR such as Routing Information Protocol (RIP) are used widely in many networks as
Upcoming Pages
Here’s what’s next.
Search Inside
This paper can be searched. Note: Results may vary based on the legibility of text within the document.
Tools / Downloads
Get a copy of this page or view the extracted text.
Citing and Sharing
Basic information for referencing this web page. We also provide extended guidance on usage rights, references, copying or embedding.
Reference the current page of this Paper.
Amin, Kaizar A. & Mikler, Armin R. Dynamic Agent Population in Agent-Based Distance Vector Routing, paper, August 2002; (https://digital.library.unt.edu/ark:/67531/metadc132968/m1/1/?q=%22Mikler%2C%20Armin%20R.%22: accessed May 15, 2024), University of North Texas Libraries, UNT Digital Library, https://digital.library.unt.edu; crediting UNT College of Engineering.