Title :
Selection diversity and linear equalization over frequency selective channels for single carrier filter bank-based transmissions
Author :
Harris, Fred ; Xiaofei Chen ; Venosa, Elettra ; Rao, Bo
Author_Institution :
San Diego State Univ., San Diego, CA, USA
Abstract :
This paper investigates the filter bank (FB) based selection diversity combining as well as linear equalization for single carrier (SC)transmissions over frequency selective channels. In contrast to the multicarrier carrier (MC) transmissions, e.g., OFDM, the FB based approach avoids the use of cyclic prefix (CP) or guard band and offers a number of superior properties such as synchronization, and peak to average power ratio (PAPR), etc. However, due to the lack of practical diversity techniques and cost effective equalizers, the broadband SC signal can hardly be deployed under highly dispersive channel environment. We propose a practical FB based selection diversity based on non-maximally decimated filter banks with perfect reconstruction support (PR-NMDFB) with its companion linear equalizer in the FB transformed domain. We shall give detailed minimum mean square error (MMSE) and bit error rate (BER) analysis and compare it to the optimal maximum ratio combining (MRC) solution.
Keywords :
OFDM modulation; channel bank filters; diversity reception; equalisers; error statistics; least mean squares methods; synchronisation; BER; FB based selection diversity; MMSE; OFDM; PAPR; PR-NMDFB; bit error rate analysis; broadband SC signal; cost effective equalizers; filter bank based selection diversity combining; frequency selective channels; linear equalization; minimum mean square error; multicarrier carrier transmissions; nonmaximally decimated filter banks; peak-to-average power ratio; perfect reconstruction support; single carrier filter bank-based transmissions; synchronization; Acoustics; Conferences; OFDM; Optimized production technology; Speech; Speech processing; filter bank; maximum ratio combining; polyphase; selection diversity; single carrier;
Conference_Titel :
Acoustics, Speech and Signal Processing (ICASSP), 2014 IEEE International Conference on
Conference_Location :
Florence
DOI :
10.1109/ICASSP.2014.6854711