Enhanced multiuser resource allocation using near-continuous loading algorithms with adaptive coding for digital subscriber loops

In this paper, we present a multi-user resource allocation technique for multi-carrier digital subscriber loop (DSL) systems using an adaptive coding/modulation scheme. Traditional bit loading based on only adaptive modulation requires the number of bits to be integer and hence causes granularity loss that can be significant in dynamic spectrum management (DSM) when each user's sub-carrier-specific power is implicitly constrained due to other user's rate demand. By adaptively varying the code rate of a programmable Reed Solomon (RS) code, a fine-granularity loading in conjunction with the purely distributed iterated water filling (IWF) is proposed to recover such granularity losses. Simulation results for various VDSL-DMT scenarios show that the proposed scheme can provide higher rates than the distributed IWF and centralized optimum spectrum management (OSM) with integer-bit loading