An Improved Greedy Cover Algorithm for Pliable Index Coding

We present the ImpGrCov algorithm–a novel algorithm for devising binary linear codes for pliable index coding that have lower codelengths than those devised using state-of-the-art algorithms such as the greedy cover, and random cover algorithms. The proposed algorithm employs a novel, improved decoder that uses a representative matrix to keep track of relationships between undecoded messages, along with a demand-based exponential weighting of receivers. Simulations indicate 10–19% improvement in average codelength over the best performing algorithm in the literature.