D. Hatzinakos, C. L. Nikias, "Estimation of Multipath Channel Response in Frequency Selective Channels," IEEE Journal on selected areas in communications, vol. 7, no. 1, pp. 12-19, January 1989.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....and equalizing the channel response based solely on the channel output without the use of a training sequence, has received considerable attention recently in communications and signal processing. Earlier approaches to blind identification exploit the higher order statistics of the output; see [5,10,52,41,20,14] and the references therein. These methods, although reliable and robust in some scenarios, require a large number of data samples and a large amount of computation. In fast changing environments, such as in cellular communications, their applications may be limited. These problems are alleviated ....

....on K may sometimes be dropped when there is no possibility of confusion. Also denote by x(k) and s(k) the kth column vectors of X(K) and (K) respectively. When K = 1, they reduce to the output and input vectors in (3) 3. Blind identification using oversampling techniques Classical solutions [38, 18, 4, 20] to blind identification in digital communication systems are based upon data sampled at the baud rate, although it has been known for some time [54, 17] that fractionally spaced equalizers are more robust under timing uncertainties. Since communication channels are, in general, nonminimum phase, ....

D. Hatzinakos and C. Nikias, "Estimation of multipath channel response in frequency selective channels", IEEE J. Select. Areas Commun., Vol. S7, No. 1, January 1989, pp. 12 19.

....has also been proposed in [46] Another temporal structure is the cyclostationarity of digital communication signals which can be exploited for non time selective channels. It was widely believed that blind channel identification was only possible using higher order statistics based techniques [47, 48, 49], since the phase of the channel is lost if second order statistics of baud spaced data samples are used for identifying a non minimum phase channel. It was observed in [50] that a cyclostationary signal contains phase information that can be used to identify the channel. The work in [51] proved ....

....and the fact that the polyspectra of a stationary Gaussian random process is zero. By virtue of these properties, methods based on the HOS of the channel output can be used to identify non minimum phase channels. Various techniques for blind channel identification using HOS can be found in [47, 49, 56, 57]. To obtain reliable estimates of the HOS of the received signal, a large amount of data is needed and the computational complexity can be high. Its usefulness for estimating wireless channels is thus limited. ffl Second order statistics (SOS) methods It was shown in [50] that the SOS of ....

D. Hatzinakos and C. Nikias, "Estimation of multipath channel response in frequency selective channels," IEEE Jour. Select. Areas in Commun., vol. 7, pp. 12--19, Jan. 1989.

....have the multichannel model considered in this paper. In such a case it may be necessary to exploit higher order statistics. There is an extensive literature dealing with blind channel estimation using higher order statistics in both time and frequency domains. See, for example, 12] 30] 39] [54], 90] 125] and [126] and a tutorial in [81] For the multiuser case, see [38] 115] and [116] 2) The Bayesian Approach: In this paper, the channel is modeled by finite dimensional deterministic unknown parameters. In some applications, however, channels can be modeled as a random vector or ....

D. Hatzinakos and C. Nikias, "Estimation of multipath channel response in frequency selective channels," IEEE J. Select. Areas Commun., vol. 7, pp. 12--19, Jan. 1989.

....of data (in 1000 s of symbols) which may be unacceptable for time varying wireless channels. SISO case: These methods are motivated by the fact that most communication channels are non minimum phase, and that second order statistics of a stationary process do not contain phase information. See [7, 8, 9, 10], and the references therein for some techniques in this class. MIMO VC case: Methods in this class use HOS to separate multiple independent sources received an array of sensors by taking advantage of differing probability distributions [11] or non Gaussianity of the source signals [12] ....

D. Hatzinakos and C. Nikias. Estimation of Multipath Channel Response in Frequency Selective Channels. IEEE J. Selected Areas Comm., 7:12--19, Jan 1989.

....the transmission capability by eliminating training sequences. It is well known that nonminimum phase channels driven by wide sense stationary input sequences, cannot be identified from second order statistics. Therefore blind identification techniques, to date, use either higher order statistics [1, 2, 3, 4] or use cyclostationary input signals [5, 6, 7] to identify a possibly nonminimum phase channel. Due to their slow rate of convergence, these techniques may be impractical for mobile communications environments. In general, signals arrive at the receiver not only with different delays, but also ....

D. Hatzinakos and C. Nikias, Estimation of multipath channel response in frequency selective channels, IEEE J. Select. Areas Commun., 7(1):12--19, January 1989.

No context found.

D. Hatzinakos, C. L. Nikias, "Estimation of Multipath Channel Response in Frequency Selective Channels," IEEE Journal on selected areas in communications, vol. 7, no. 1, pp. 12-19, January 1989.

No context found.

D. Hatzinakos, C.L. Nikias, Estimation of multipath channel response in frequency selective fading channels, IEEE J. Selected Areas Commun. 7 (1989) 12-19.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC