Antipodal paraunitary matrices and their application to OFDM systems

• Published in: IEEE transactions on signal processing Vol. 53; no. 4; pp. 1374 - 1386
• Main Authors: Phoong, S.-M., Kai-Yen Chang
• Format: Journal Article
• Language: English
• Published: New York, NY IEEE 01.04.2005; Institute of Electrical and Electronics Engineers; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Adaptive filters; Adaptive signal processing; Antipodal paraunitary matrices; Applied sciences; Butterflies; Construction costs; Costs; Detection, estimation, filtering, equalization, prediction; Error correction; Error correction codes; Exact sciences and technology; filterbanks; Information, signal and communications theory; Low cost; Mathematical analysis; Matrices; Matrix methods; Multiaccess communication; Multiplexing; Noise reduction; OFDM; Orthogonal Frequency Division Multiplexing; Signal and communications theory; Signal, noise; Source coding; Spread spectrum communication; Telecommunications and information theory; Performance evaluation; OFDM; Orthogonal frequency division multiplexing; Production cost; filterbanks; Filter bank; Antipodal paraunitary matrices
• ISSN: 1053-587X; 1941-0476
• DOI: 10.1109/TSP.2005.843735

Paraunitary (PU) matrices and filterbanks have played an important role in many applications. This paper studies a special class of PU matrices, called antipodal paraunitary (APU) matrices. APU matrices are PU matrices whose coefficient matrices consist of /spl plusmn/1 only. Several new methods will be introduced for the construction of APU matrices. In particular, a new method based on the butterfly structure that has a low cost implementation is proposed. Moreover, one application of APU matrices in precoded orthogonal frequency division multiplexing (OFDM) systems will be considered. By using an APU precoding matrix, the complexity will be very low, and experiments show that the use of APU matrices can greatly enhance the performance.