The branch-and-bound principle was applied to solve the equivalent nonlinear integer programming problem. A recursive formula for fast metric update was obtained by exploiting the intrinsic characteristic of the objective function. The impacts of reordering the data sequence and selective detection were addressed. By suggesting a preferred order and selective detection method, the detector complexity was greatly reduced while giving up little performance loss.
|Number of pages||1|
|Journal||IEEE International Symposium on Information Theory - Proceedings|
|State||Published - 20 Oct 2003|
|Event||Proceedings 2003 IEEE International Symposium on Information Theory (ISIT) - Yokohama, Japan|
Duration: 29 Jun 2003 → 4 Jul 2003