A Quadrature Amplitude Modulation (QAM) resource allocation algorithm for Video on Demand (VoD)traffic is presented. Based on stream encoding rates and available system capacity, the proposed on-line non-mixing algorithm will select the QAM modulator that maximizes system efficiency in a VoD service group. A discrete event simulator has been developed to compare the performance of the proposed algorithm against two commonly deployed algorithms known as the least-loaded and most-loaded algorithms. Our analysis suggests that the non-mixing algorithm performs better than the two algorithms under a range of assumptions that take into account peak concurrent usage rate and traffic mixture composition. We show that the non-mixing algorithm leads to an average of 4.39% higher allowed peak usage rate than the least-loaded and most-loaded algorithms.
Read full abstract