The application of semidefinite programming for detection in CDMA

摘要

A detection strategy based on a semidefinite relaxation of thenCDMA maximum-likelihood (ML) problem is investigated. Cutting planes arenintroduced to strengthen the approximation. The semidefinite programnarising from the relaxation can be solved efficiently using interiornpoint methods. These interior point methods have polynomialncomputational complexity in the number of users. The simulated bit errornrate performance demonstrates that this approach provides a goodnapproximation to the ML performance