Green x-DSL provision via Dynamic Spectrum Management

x-DSL technology has been the dominant application of communication systems in the last mile of wireline communications. Although  fiber optics will replace this technology, due to economic crisis many countries in Europe has not been prepared for this implementation and next generation will be accomplished by VSDL. To this end, x-DSL technology manufacturers and operators should refine this technology in terms of energy efficiency towards the realization of green communications as ICT power consumption has been identified a significant contributor to global warming. In order to guarantee this requirement, dynamic spectrum management (DSM) can be considered as a key technology for reducing the total power consumption. Traditional DSM design in x-DSL should be revised towards a much wider power-efficient scope of this technology.

The aim of this project is to develop such a ‘green-DSL’ technology, that we will be able to quantify the power-rate trade-off for different practical settings for several x-DSL systems. We expect that the power savings can be achieved up to 50%, while still guaranteeing 95% of full-power data rate performance. Within this project, we will focus on the following types of algorithms: (i) optimal, (ii) iterative and autonomous spectrum balancing, (iii) iterative and selective iterative water-filling, (iv) successive convex approximation for low complexity and convex functions and (v) distributed spectrum balancing. Several physical and MAC layers will be also investigated towards this end.