Soft-Input Soft-Output List Sphere Detection with a Probabilistic Radius Tightening

In this paper, we present a low-complexity list sphere detection algorithm for achieving near-optimal a posteriori probability (APP) detection in an iterative detection and decoding (IDD). Motivated by the fact that the list sphere decoding searching a fixed number of candidates is computationally inefficient in many scenarios, we design a criterion to search lattice points with non-vanishing likelihood and then derive a hypersphere radius satisfying this condition. Further, in order to exploit the original sphere constraint as it is instead of using necessary conditioned version, we combine a probabilistic tree pruning strategy and the proposed list sphere search. Two features, tightened hypersphere radius and probabilistic tree pruning, collaborate and improve the search efficiency in a complementary fashion. Through simulations on 4x4 MIMO system, we show that the proposed method provides substantial reduction in complexity while achieving negligible performance loss over the conventional list sphere detection.