Blind Separation of Radio Signals in Fading Channels

Part of Advances in Neural Information Processing Systems 10 (NIPS 1997)

Bibtex Metadata Paper


Kari Torkkola


We apply information maximization / maximum likelihood blind source separation [2, 6) to complex valued signals mixed with com(cid:173) plex valued nonstationary matrices. This case arises in radio com(cid:173) munications with baseband signals. We incorporate known source signal distributions in the adaptation, thus making the algorithms less "blind". This results in drastic reduction of the amount of data needed for successful convergence. Adaptation to rapidly changing signal mixing conditions, such as to fading in mobile communica(cid:173) tions, becomes now feasible as demonstrated by simulations.