Error Control in Packet Switched Wide Area Networks using Bayesian Computational Model

The paper developed an improved model called the Dynamic Erasure Coding Model for controlling error in typical packet-switched Wide Area Networks. The Dynamic Erasure Coding Model was developed by designing a fluid flow model. The mathematical model was built from queuing theory and probability theory. The model of Dynamic Erasure Control Model was evaluated and characterized. The model was done by establishing a Bayesian theory model for the posterior probability of routers accepting packets using service rate of packets, and network utilization. In all these, the Dynamic Erasure Control Model was compared with the Erasure Control Model. The model produced lower posterior probability than the Erasure Coding Model with which it was compared by superimposition. The Bayesian Theoretic Model demonstrated that the higher the service rate of packets, the higher the posterior probability of several routers accepting packets. The model also showed that the higher the network utilization, the lower the posterior probability.