?
The Transport Coding Model for Delay Reduction in Communication Networks using OMNET++ toolkit
Transport coding is a promising method for reducing message delay in packet-switched networks by adding controlled redundancy at the transport layer. Classical analytical works show that encoding k original packets into n coded packets and reconstructing the message after the first k successful deliveries effectively shifts the latency metric from the maximum to the k-th order statistic of packet delays, yielding lower mean delay and variance. However, the gap between this theory and concrete implementations in realistic networks is still insufficiently documented. In this paper we focus on the modelling side of Transport coding and present a detailed discrete-event implementation in the OMNeT++ simulator. We construct a Kleinrock-type multi-hop network with FIFO queues, exponential service and link delays, and an explicit message-reconstruction logic that tracks the arrival times of coded packets and records message-level delay and deadline violations. The model is driven by a configurable pair of uncoded (n=k) and coded (n>k) configurations with the same message rate, allowing direct comparison of load, delay, and reliability. Simulation results for various code rates and input loads demonstrate consistent delay reductions and lower probability of late message delivery. The proposed modelling framework serves as a transparent bridge between analytical formulas and executable code and can be reused to tune Transport coding parameters for low-latency services in 5G and beyond networks. A references to the source code of the model is provided in the end of the article.
