Title :
A semi-blind algebraic constant modulus algorithm
Author :
Swindlehurst, A. Lee
Author_Institution :
Dept. of Electr. & Comput. Eng., Brigham Young Univ., Brigham City, UT, USA
Abstract :
A modification to the algebraic constant modulus (ACM) algorithm is proposed that exploits the presence of known pilot symbols within the transmitted data. The pilot symbols are used to place soft constraints on the subspace in which the ACM solution should lie, and relax some of the identifiability conditions required by ACM. The primary advantage of the proposed method is that sources transmitting linearly independent pilot data can be separated without resorting to a joint diagonalization procedure, and hence without computing the beamformers of signals that are not of interest.
Keywords :
blind source separation; data communication; identification; ACM algorithm; blind source separation; identifiability conditions; known pilot symbols; linearly independent pilot data; semi-blind algebraic constant modulus algorithm; soft constraints; Antenna arrays; Block codes; Closed-form solution; Equations; Interference constraints; Performance analysis; Singular value decomposition; Subspace constraints;
Conference_Titel :
Acoustics, Speech, and Signal Processing, 2004. Proceedings. (ICASSP '04). IEEE International Conference on
Print_ISBN :
0-7803-8484-9
DOI :
10.1109/ICASSP.2004.1326859