On throughput-delay tradeoff of network coding for wireless communications

This paper consists of two main parts. In the first part, our aim is to introduce the reader to the fundamental issue of throughput-delay tradeoff in network coding for wireless communication systems. We discuss important classes of wireless network coding in the literature and identify their advantages and disadvantages in terms of throughput and delay. We also briefly present practical design considerations such as how to control the data rate, feedback overhead, and implementation complexity. In the second part, we present a case study of the throughput-delay tradeoff in generation-based network coding for wireless broadcast. We first show how partitioning a partially-received generation of data packets into sub-generations will impact the throughout and delay.We then introduce the concepts of local- and Pareto-optimal partitioning solutions, and derive as a special case, the Pareto-optimal partitioning solution obtained from instantly decodable network coding. This solution is then used to construct a new coding framework that provides more operating points on the throughput-delay tradeoff curve compared to existing schemes in the literature. Finally, we briefly discuss the design flexibilities this framework offers, such as reduced feedback overhead.