Abstract— A Quadrature Amplitude Modulation (QAM)resource allocation algorithm for Video on Demand (VoD)traffic is presented. Based on stream encoding rates andavailable system capacity, the proposed on-line non-mixingalgorithm will select the QAM modulator that maximizessystem efficiency in a VoD service group. A discrete eventsimulator has been developed to compare the performanceof the proposed algorithm against two commonly deployedalgorithms known as the least-loaded and most-loadedalgorithms. Our analysis suggests that the non-mixingalgorithm performs better than the two algorithms under arange of assumptions that take into account peakconcurrent usage rate and traffic mixture composition. Weshow that the non-mixing algorithm leads to an average of4.39% higher allowed peak usage rate than the least-loadedand most-loaded algorithms.

Index Terms—VoD, Broadband access, Capacity planning,Congestion control, Traffic modeling, Resource allocation,Network simulation.

Cite: Jiong Gong, David Reed, Terry Shaw, Daniel Vivanco and Jim Martin, "QAM Resource Allocation in Mixed-Format VoD Systems," Journal of Communications, vol. 2, no. 1, pp. 1-9, 2007.