In this paper, we present a high-level approach for reducingthe required resources for performing multi-user detectionin a 3GPP TDD multi-user system. This approach isbased on a displacement representation of the parameters thatdescribe the transmission system, and a generalized Schur algorithmthat works on this representation. The Schur algorithmnaturally leads to a highly parallel hardware implementationusing CORDIC cells. It is shown that this hardwarearchitecture can also be used to compute the initial displacementrepresentation.
It is very beneficial to introduce incremental refinementstructures into the solution process, both at the algorithmiclevel and in the individual cells of the hardware architecture.We detail these approximations and present simulation resultsthat confirm their effectiveness.