Low-complexity vector precoding for multi-user systems

Vector precoding enables non-cooperative signal acquisition in the multi-user broadcast channel. The performance advantage with respect to the more straightforward linear precoding algorithms comes as a consequence of an added perturbation vector, which enhances the properties of the precoded signal. Nevertheless, the computation of the perturbation signal entails a search for the closest point in an infinite lattice, which is known to belong in the class of non-deterministic polynomial-time hard (NP-hard) problems. This contribution presents a novel tree search scheme that achieves an error-rate performance that is close to the optimum given by the sphere encoder, but with a significantly simpler tree-search structure that only considers the most promising nodes for expansion. With the aim of better showcasing the low-complexity and simple datapath of the proposed tree-search technique, its hardware implementation on a 65 nm ASIC target device has been performed.