Communication over Finite-Field Matrix Channels

TL;DR

This paper studies error control in network coding over finite-field matrix channels, providing capacity bounds, asymptotic analysis, and a capacity-achieving coding scheme with efficient decoding and exponential error decay.

Contribution

It introduces a new model for random network coding errors, derives capacity bounds, and proposes a simple, effective coding scheme that achieves capacity asymptotically.

Findings

Capacity bounds for the matrix channel are established.

A simple coding scheme achieves capacity in large field or matrix size limits.

Decoding complexity is O(n^2 m) with exponentially decreasing error probability.

Abstract

This paper is motivated by the problem of error control in network coding when errors are introduced in a random fashion (rather than chosen by an adversary). An additive-multiplicative matrix channel is considered as a model for random network coding. The model assumes that n packets of length m are transmitted over the network, and up to t erroneous packets are randomly chosen and injected into the network. Upper and lower bounds on capacity are obtained for any channel parameters, and asymptotic expressions are provided in the limit of large field or matrix size. A simple coding scheme is presented that achieves capacity in both limiting cases. The scheme has decoding complexity O(n^2 m) and a probability of error that decreases exponentially both in the packet length and in the field size in bits. Extensions of these results for coherent network coding are also presented.