Hybrid Architecture for Data-Dependent Superimposed Training in Digital Receivers

Many digital communications algorithms present characteristics that make very difficult to implement them in either a software solution or as a fully custom hardware architecture. Their inherent complexity implies two challenges at the same time: to process the information as fast as possible to present the results when they are required, and to build a system that meets the power consumption and space constraints imposed by the application, while trying to maintain a low design intricacy. This work describes a hybrid hardware-software architecture designed to run a wireless communication algorithm named data-dependent superimposed training.The resulting system can be used partially or in its totality to implement many other algorithms with similar needs, and in fact it is an interesting source of information for implementing solutions for some of the most common operations encountered in the DSP field.