A Heuristic Algorithm for the Multi-constrained Multicast Tree

Abstract

Abstract.Multicasting is an important communication mechanism for implementing real-time multimedia applications, which usually require the underlying network to provide a number of quality-of-service (QoS) guarantees to users. In this paper, we study the problem concerning how to find a feasible multicast tree in which all the multicast paths satisfy the given set of QoS constraints. This problem is referred as the multiconstrained multicast tree (MCMT) problem. Based on tabu-search technique, a heuristic algorithm named MCMTS for the MCMT problem is proposed in this study. According to the experimental results, the probability of finding a feasible multicast tree by our MCMTS method is more than 99% if one exists. Furthermore, the MCMTS is also shown to be a simple and highly efficient method. As a result, it would be a practical approach for developing multicast routing protocol.

Genetic local search for multicast routing with pre-processing by logarithmic simulated annealing

“…by about 1.8% on P 12 for steinb18). The run-times of PMX-runs differ only marginally from the times for GLS-runs, which confirms (20) and the setting N PMX := 223 according to (21). As mentioned above, the run-times of sequential GLS simulations can be divided roughly by M for parallel executions of the program.…”

“…Compared to LSA-runs, all the solutions on P 7 ,…, P 12 as well as the run-times are worse for PMX-runs. Although the order of the time complexity estimation is the same for (16) and (20) if N PMX := 223 in (21), the fact that the run-times are worse for PMX-runs can be explained by auxiliary calculations related to the PMX operator (e.g., selection of M fittest solutions after each step), which are hidden in the constants of the time complexity estimation.…”

“…Numerous heuristics have been devised to find good approximations of minimum solutions efficiently [21,22,[47][48][49][50][51][52]. Since in our approach single requests may have to be rerouted many times in order to satisfy capacity constraints, we employ the simple but efficient KMB algorithm [48].…”

“…Wang et al [23] discuss the same problem as in [21,22]. The authors propose an efficient algorithm based on tabu search for delay constrained, low cost multicast trees (TSDLMRA).…”

“…Yang [21] devised a tabu search algorithm for finding a single, feasible multicast tree efficiently that satisfies a number of QoS constraints. The method is tested on randomly generated networks with 100 nodes (and on 8 × 8 meshes).…”