In this work we investigate bounds on throughput and delay performance of a scheduling protocol that derives its decisions from codes traditionally used to correct or detect errors in the information carried over a noisy channel. In this paper we study the particular cases in which the Reed-Solomon and Hermitian code constructions are used. It is found that Hermitian codes outperform Reed-Solomon codes in minimum throughput guarantee and delay metrics when the number of nodes is in the order of thousands. The relative minimum distance of the code used to schedule the transmissions is identified as an important property that can be used to identify codes that can enable scheduling patterns with better minimum performance guarantees. Furthermore, the terminology of error control coding is used to present a more general and constructive framework for the study of code-based scheduling protocols.
展开▼
