Harmony search based algorithms for bandwidth-delay-constrained least-cost multicast routing

  • Authors:

  • R. Forsati;A. T. Haghighat;M. Mahdavi

  • Affiliations:

  • Department of Electrical and Computer Engineering, Islamic Azad University, Qazvin Branch, Qazvin, Iran;Department of Electrical and Computer Engineering, Islamic Azad University, Qazvin Branch, Qazvin, Iran;Department of Computer Engineering, Sharif University of Technology, Tehran, Iran

  • Venue:
  • Computer Communications
  • Year:
  • 2008

Quantified Score

Hi-index 0.25

Visualization

Abstract

The advent of various real-time multimedia applications in high-speed networks creates a need for quality of service (QoS) based multicast routing. Two important QoS constraints are the bandwidth constraint and the end-to-end delay constraint. The QoS based multicast routing problem is a known NP-complete problem that depends on (1) bounded end-to-end delay and link bandwidth along the paths from the source to each destination, and (2) minimum cost of the multicast tree. In this paper, we presents novel centralized algorithms to solve the bandwidth-delay-constrained least-cost multicast routing problem based on the harmony search (HS) algorithm. Our first algorithm uses modified Prufer number as Steiner tree representation that is called HSPR. Prufer number has poor locality and heritability in evolutionary search, so, we describe a new representation, node parent index (NPI) representation, for representing trees and describe harmony operations accord to this representation. Our second algorithm is based on NPI representation that is called HSNPI, an empirical study to determine the impacts of different parameters of the HSNPI algorithm on the solution quality and convergence behavior was performed. We evaluate the performance and efficiency of our proposed methods with a GA-based algorithm and a modified version of the bounded shortest multicast algorithm (BSMA). Simulation results on randomly generated networks and real topologies indicate that HSNPI algorithm that we proposed has overcome other three algorithms on a variety of random generated networks considering average tree cost.