High-performance and parallel computation in Communication Systems
Wireless communications have evolved from the first generation system that focused on voice transmission to 4G in which data and applications will be responsible for most of the generated traffic, so hardware must also be improved to allow for higher data rates processing. Indeed, industry is considering migrating from the current massive use of ASIC devices to those that can offer more reprogrammability together with limited energy consumption. In the first steps of this migration, the use of multicore and many-core processors as Graphics Processing Units (GPU) can provide high performance at a remarkably low price. However, its architecture can be fully exploited only if its parallel programming is optimized from the very first steps of the algorithm formulation.