An Effective Randomized QoS Routing Algorithm on Networks with Inaccurate Parameters

This paper develops an effective randomized on-demand QoS routingalgorithm on networks with inaccurate link-state information. Severalnew techniques are proposed in the algorithm. First, the maximum safetyrate and the minimum delay for each node in the network arepre-computed, which simplify the network complexity and provide therouting process with useful information. The routing process isdynamically directed by the safety rate and delay of the partialrouting path developed so far and by the maximum safety rate and theminimum delay of the next node. Randomness is used at the link leveland depends dynamically on the routing configuration. This providesgreat flexibility for the routing process, prevents the routing processfrom overusing certain fixed routing paths, and adequately balances thesafety rate and delay of the routing path. A network testing environmenthas been established and five parameters are introduced to measurethe performance of QoS routing algorithms. Experimental resultsdemonstrate that in terms of the proposed parameters, the algorithmoutperforms existing QoS algorithms appearing in the literature.