Telecommunications – Transmitter and receiver at separate stations – Plural transmitters or receivers
Reexamination Certificate
1999-12-22
2003-01-07
Nguyen, Lee (Department: 2683)
Telecommunications
Transmitter and receiver at separate stations
Plural transmitters or receivers
C455S278100, C455S506000, C342S383000, C375S267000, C375S349000
Reexamination Certificate
active
06505053
ABSTRACT:
FIELD OF THE INVENTION
The present invention is directed to a method for predicting the multipath fading that degrades performance in wireless networks in order to correct for it.
BACKGROUND OF THE INVENTION
It is well known that a principal detrimental factor limiting the performance of wireless communication systems is signal fading, in that it is unknown and imposes multiplicative distortion on the transmitted signal. An effective technique to tackle this difficulty is spatial diversity which uses multiple receiving antennas and employs a certain scheme to combine the multiple independently-faded received signals. However, for such a scheme to be successful, it is often necessary to estimate or track the fading process to determine the appropriate combiner weights to be applied to the signal from each antenna during the combining process. Under slow fading conditions, the fading estimation is often done using a windowing and averaging technique. However, because the fading rate is proportional to the carrier frequency and the speed at which the mobile wireless unit is moving, the variances in the fading become faster as the carrier frequency and/or mobile unit speed is increased. For such a fast varying fading channel, the conventional averaging method needs to use a shorter window length, resulting in degraded performance.
The problem of fading prediction has been studied previously and other researchers have assumed that the fading process could be modeled by a small number of time varying sinusoids. (G. Giannakis, C. Tepedelenlioglu, Basis Expansion Models and Diversity Techniques for Blind Identification and Equalization of Time-Varying Channels, Proceedings of the IEEE, October 1998). In investigating the statistical properties of fading, researchers often began with a model of multiple superimposed sinusoids which correspond to many reflected rays with different Doppler frequency offsets, attenuations, and phases. This physical conception leads to a sinusoidal model with its parameters regarded as random variables. Based on this idea, Jakes proposed a widely adopted Rayleigh fading model to generate artificial fading process for system simulations. (C. Jakes, Microwave Mobile Communications, IEEE Press Reprint, 1974). In fact, spectral analysis of real-world fading data strongly supports the conjecture that the complex baseband fading process often mainly consists of a small number of time varying sinusoids.
Markov-type models have also long been used for studying the statistical properties of fading processes, such as level crossing rate, etc. For such a model to be applied to fading prediction, it is rational to consider the Kalman filtering techniques. (H. Meyr, Digital Communication Receivers, John Wiley 1998). However, since the whole state-space signal model entails an unpredictable process noise and a measurement noise, it is feasible to predict only a very few steps ahead. Furthermore, this model has no direct link to the physical origin that gives rise to the fading phenomenon which makes it inappropriate for fading prediction.
SUMMARY OF THE INVENTION
In modeling the fading as a superimposed sinusoidal process, the present invention assumes that the sinusoidal parameters can be treated as unknown constants during a short enough interval. Thus, the fading process may be viewed as a deterministic sinusoidal process with time-varying parameters. Using the method of the present invention these time-varying parameters can be determined and used to predict the fading process. If a receiver is equipped with the capability of predicting the fading process using the method of the present invention, then significant performance improvement may be possible.
The present invention assumes that, during a relatively short interval, the parameters of a sinusoidal model are constants and if those constants can be calculated then the model can predict fading behavior and be used to compensate for it at least a short time period before and after the time frame of the data that was used to construct the model. The present invention teaches a method of determining the parameters for a sinusoidal model of the multipath fading of a wireless system based on the behavior of the received signal and then using those parameters to predict fading behavior.
The present invention accomplishes this by first obtaining the noisy fading data from the discrete time sampled received signal. This can be accomplished by decoding the symbols from the received signal and removing the modulation of the signal using the decoded symbols, through the use of pilot symbols that are already known to the receiver and hence would not have to be decoded and could simply be demodulated, or other known techniques. The resulting record of noisy fading data is then low pass filtered based on known values for the carrier frequency and maximum Doppler shift in order to improve the fading power to noise ratio. The filtered fading data is then processed using the root-MUSIC algorithm in order to obtain frequency estimates for the superimposed sinusoids, and a linear least square fit is performed to obtain an amplitude and phase for each sinusoid. Once these values have been obtained, they are used to construct the sinusoidal model which can then be used to compensate for the fading behavior in incoming signals resulting in improved receiver performance.
The present invention also teaches that the model can be improved by setting a constraint on the frequency estimates based on the maximum Doppler frequency and improving the least square fit by a control point constraint based on critical data points in the filtered fading data including boundary points, level crossing points, and/or peak points.
REFERENCES:
patent: 4173759 (1979-11-01), Bakhru
patent: 4661993 (1987-04-01), Leland et al.
patent: 4939791 (1990-07-01), Bochmann et al.
patent: 5262789 (1993-11-01), Silverstein
patent: 5581580 (1996-12-01), Lindbom et al.
patent: 5890068 (1999-03-01), Fattouche et al.
Winters, Jack H., “Signal Acquisitions and Tracking with Adaptive Arrays in the Digital Mobile Radio System IS-54 with Flat Fading”,IEEE Transactions on Vehicular Technology,vol. 42, No. 4, pp. 374-384 Nov. 1993.
Dumont, L. et al., “Super-resolution of multipath channels in a spread spectrum location system”,IEE1994, Jul. 25, 1994.
Wang, Hong Shen et al., “Finite-State Markov Channel-A Useful Model for Radio Communication Channels”,IEEE Transactions on Vehicular Technology,vol. 44, No. 1, pp. 163-171, Feb. 1995.
Wang, Hong Shen et al., “On Verifying the First-Order Markovian Assumption for a Rayleigh Fading Channel Model”,IEEE Transactions on Vehicular Technology,vol. 45, No. 2, pp. 353-357, May 1996.
Cupo, Robert L. et al., “Adaptive Antenna Applique Field Test”, presented at Smart Antenna Workshop, 1997.
Klukas, Richard et al., “Cellular Telephone Positioning Using GPS Time Synchronization”,GPS World Magazine,Apr. 1998.
Wu, Wei-Chiang, et al., “Root-MUSIC Based Joint Identification and Timing Estimation of Asynchronous CDMA System over Rayleigh Fading Channel”,IEICE Trans. Fundamentals,vol. E81-A, No. 8, pp. 1550-1559, Aug. 1998.
Giannakis, Georgios B. et al., “Basic Expansion Models and Diversity Techniques for Blind Identification and Equalization of Time-Varying Channels”,Proceedings of the IEEE,vol. 86, No. 10, pp. 1969-1986, Oct. 1998.
Hwang Jeng-Kuang
Winters Jack Harriman
AT&T Corp.
Nguyen Lee
LandOfFree
Method for sinusoidal modeling and prediction of fast fading... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method for sinusoidal modeling and prediction of fast fading..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for sinusoidal modeling and prediction of fast fading... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3066319