Two algorithms for multi‐constrained optimal multicast routing

Abstract

SUMMARYMultimedia applications, such as video-conferencing and video-on-demand, often require quality of service (QoS) guarantees from the network, typically in the form of minimum bandwidth, maximum delay, jitter and packet loss constraints, among others. The problem of multicast routing subject to various forms of QoS constraints has been studied extensively. However, most previous efforts have focused on special situations where a single or a pair of constraints is considered. In general, routing under multiple constraints, even in the unicast case is an NP-complete problem. We present in this paper two practical and efficient algorithms, called multi-constrained QoS dependent multicast routing (M QDMR) and (multicasting routing with multi-constrained optimal path selection (M MCOP)), for QoS-based multicast routing under multiple constraints with cost optimization. We provide proof in the paper that our algorithms are correct. Furthermore, through extensive simulations, we illustrate the effectiveness and efficiency of our proposals and demonstrate their significant performance improvement in creating multicast trees with lower cost and higher success probability.

Multicast routing based on data envelopment analysis with multiple Quality of Service parameters

“…However, as can be seen in the Table III, the gap of the delays and profits are loose for all different network topologies, which indicate that the resulting multicast tree from the proposed approach leads to better performance in terms of the total delay and profit. The delay gap between the third and sixth columns represented the difference between delays on the resulting multicast trees of the proposed method and Model (9). These conclude that the delay and profit of solutions obtained from the proposed method are 46% and 32% better than solutions obtained from Model (9), respectively.…”

“…Note that the total costs of solutions resulted from the proposed method are approximately 40% worse than optimal cost of Model (9), because Model (9) minimizes the cost of a multicast tree and does not attempt to improve the total delay and profit of resulting tree. The delay gap between the third and sixth columns represented the difference between delays on the resulting multicast trees of the proposed method and Model (9). However, as can be seen in the Table III, the gap of the delays and profits are loose for all different network topologies, which indicate that the resulting multicast tree from the proposed approach leads to better performance in terms of the total delay and profit.…”

“…The delay gap between the third and sixth columns represented the difference between delays on the resulting multicast trees of the proposed method and Model (9). Hence, in the second set of computational experiments, we studied an extension of Model (9) in which the total weight of a multicast tree must additionally fulfill a budget constraint. Indeed, Model (9) performs poorly with respect to the total delay and profit of a resulting multicast tree.…”

“…The second, third, and fourth columns of the table describe the cost, delay, and profit of the multicast tree obtained from our proposed method, respectively. The sixth column reports the minimum cost of a Steiner tree based on Model (9). The seventh and eighth columns measure the total delay and profit of the multicast tree obtained by Model (9), respectively.…”

“…The problem of constructing a constrained Steiner tree is also NP-complete [7]. Many heuristic algorithms for the constrained Steiner tree problem have been proposed such as heuristic algorithm, genetic algorithm, and ant algorithm [8][9][10][11][12][13][14][15][16][17][18].…”