Title :
Decision Directed Channel Estimation Employing Projection Approximation Subspace Tracking
Author :
Akhtman, Jos ; Hanzo, Lajos
Author_Institution :
Sch. of Electr. & Comput. Sci., Southampton Univ.
Abstract :
The attainable capacity and integrity of a state-of-the-art broadband multi-carrier communication system is highly sensitive to the accuracy of the information available concerning the channel conditions encountered. The majority of existing advanced channel estimation schemes assume knowledge of the channel´s power delay profile (PDP) in order to estimate the corresponding instantaneous channel impulse response (CIR). It is evident however that in realistic mobile channels, where at least one of the communicating terminals is in motion, the power delay profile will also be time-variant and thus may not be known a priori at the receiver. In this paper we propose a decision directed channel estimation scheme employing the so-called projection approximation subspace tracking (PAST) algorithm. The PAST algorithm is utilised for the sake of achieving low-complexity recursive tracking of the channel´s PDP. The achievable performance of the proposed method is documented in the context of an OFDM system communicating in realistic channel conditions characterized by a time-variant fractionally-spaced PDP.
Keywords :
broadband networks; channel estimation; tracking; wireless channels; broadband multi-carrier communication system; decision directed channel estimation; instanteneous channel impulse response; low-complexity recursive tracking; power delay profile; projection approximation subspace tracking; time-variant fractionally-spaced PDP; Approximation algorithms; Bandwidth; Broadband communication; Channel estimation; Context; Delay estimation; OFDM; Recursive estimation; Transfer functions; Wireless networks;
Conference_Titel :
Vehicular Technology Conference, 2007. VTC2007-Spring. IEEE 65th
Conference_Location :
Dublin
Print_ISBN :
1-4244-0266-2
DOI :
10.1109/VETECS.2007.626