Joint signaling techniques and spectral optimization for symmetric bit-rate communication over self-NEXT-dominated channels

We present a framework for maximizing the capacity of symmetric bit-rate communication services dominated by Gaussian crosstalk, in particular, digital subscriber line (DSL) services. We solve for optimal transmit power spectral densities (PSDs) that maximize the joint capacity of same-service users and yield significant gains in bit rates (or performance margins) over current schemes. Our results differ from previous work in that we develop transmit spectra in the presence of self-far-end crosstalk in addition to self-near-end crosstalk, present optimal contiguous spectra for practical modulation schemes, and derive optimal spectra under an additional frequency-domain peak-power constraint. Furthermore, by design, the optimal transmit PSDs are spectrally compatible with existing services on neighboring lines.