Method of channel order selection and channel estimation in a wireless communication system

• US 6,907,092 B1
• Filed: 07/14/2000
• Issued: 06/14/2005
• Est. Priority Date: 07/14/2000
• Status: Expired due to Fees

• Current Assignee: Comsys Communication & Signal Processing Ltd. (Intel Corporation)
• Original Assignee: Comsys Communication & Signal Processing Ltd. (Intel Corporation)
• Inventors: Cohen, Ron, Yakhnich, Evgeny
• Primary Examiner(s): TSE, YOUNG TOI
• Application Number:
  US09/616,161
• Time in Patent Office: 1,796 Days
• Field of Search: 375/231, 375/232, 375/233, 375/262, 375/341, 375/346, 375/348, 375/350, 714/794, 714/795, 455/67.11, 455/67.13
• US Class Current: 375/346
• CPC Class Codes: H04L 25/0216 with estimation of channel length

Abstract

A novel and useful apparatus for and method of determining the channel order and channel estimate in a communications system such as a wireless communication system including cellular and cordless. Such channels are typically characterized by rapidly changing impulse response and their taps can be modeled as zero-mean, complex, Gaussian random processes. A sufficiently long, initial channel estimate of length is performed so as to ensure that the actual channel taps will be contained in the estimated taps thus making certain that the equalizer will effectively eliminate intersymbol interference. Channel estimation is performed during each burst using the training sequence transmitted in the middle of the burst. The tap energies are averaged so as to track slow variations in the pattern of resultant channel taps. A noise floor is calculated using the lowest averaged taps and a threshold is computed based thereon and applied to the average taps. The channel order and the tap positions are then selected in accordance those average taps that are above the threshold.

Claims

1. A method of estimating channel order of a bounded length channel having at most L non zero taps located within an M symbol time interval, said method comprising the steps of:

calculating a channel estimate over a plurality of bursts using a channel length of M taps to yield estimated channel taps;

calculating tap energies of said estimated channel taps;

averaging said tap energies over time to generate average tap energies;

selecting a threshold in accordance with a noise floor estimate calculated using the lowest M-L average tap energies;

setting said channel order to a number of average tap energies N above said threshold; and

wherein L, M and N are positive integers.

2. The method according to claim 1, wherein said step of calculating said channel estimate is performed using a least squares technique.

3. The method according to claim 1, wherein said step of calculating said channel estimate is performed using a correlation technique.

4. The method according to claim 1, wherein said estimated channel taps are represented as zero-mean, complex, Gaussian random processes.

5. The method according to claim 1, wherein said estimated channel taps are represented as non zero-mean, complex, Gaussian random processes.

6. The method according to claim 1, wherein said estimated channel taps vary over time.

7. The method according to claim 1, further comprising the step of refining said estimated channel taps by recalculating said channel estimate utilizing said channel order.

8. A method of calculating an estimate of a bounded length channel having at most L non zero taps located within M symbol time intervals, said method comprising the steps of:

calculating a channel estimate over a plurality of bursts using a channel length of M taps to yield estimated channel taps;

calculating tap energies of said estimated channel taps;

averaging said tap energies over time to generate average tap energies;

selecting a threshold in accordance with a noise floor estimate calculated using the lowest M-L average tap energies;

setting a channel order to a number of average tap energies N that are above said threshold;

refining said estimated channel taps by recalculating the channel estimate utilizing said channel order; and

wherein L, M and N are positive integers.

9. The method according to claim 8, wherein said step of calculating said channel estimate is performed using a least squares technique.

10. The method according to claim 8, wherein said step of calculating said channel estimate is performed using correlation technique.

11. The method according to claim 8, wherein said estimate channel taps are represented as zero-mean, complex, Gaussian random processes.

12. The method according to claim 8, wherein said estimated channel taps are represented as non zero-mean, complex, Gaussian random processes.

13. The method according to claim 8, wherein said estimated channel taps vary over time.

14. A cellular radio receiver for receiving and decoding a transmitted cellular signal, comprising:

a radio frequency (RF) receiver circuit for receiving and downconverting said transmitted cellular signal to a baseband signal;

a demodulator adapted to demodulate said baseband signal in accordance with the modulation scheme used to generate said transmitted cellular signal;

an equalizer comprising signal processing means programmed to;

estimate a channel order of a bounded length channel having at most L non zero taps located within an M symbol time interval;

calculate a channel estimate over a plurality of bursts using a channel length of M taps to yield estimated channel taps;

calculate tap energies of said estimated channel taps;

average said tap energies over time to generate average tap energies;

select a threshold in accordance with a noise floor estimate calculated using the lowest M-L average tap energies;

set said channel order to a number of average tap energies N that are above said threshold;

a channel decoder adapted to decode the output of said equalizer to generate a decoded output data signal; and

wherein L, M and N are positive integers.

15. The receiver according to claim 14, further comprising a speech decoder operative to convert said decoded output data signal to an audible speech signal.

16. The receiver according to claim 14, wherein said signal processing means is programmed to further refine said estimated channel taps by recalculating said channel estimate utilizing said channel order.

17. The receiver according to claim 14, wherein said equalizer is adapted to calculate said channel estimate utilizing a least squares technique.

18. The receiver according to claim 14, wherein said equalizer is adapted to calculate said channel estimate utilizing a correlation technique.

19. The receiver according to claim 14, wherein said estimated channel taps are represented as zero-means, complex, Gaussian random processes.

20. The receiver according to claim 14, wherein said estimated channel taps are represented as non zero-mean, complex, Gaussian random processes.

21. The receiver according to claim 14, wherein said estimated channel taps vary over time.

22. The receiver according to claim 14, wherein said equalizer comprises means for performing a maximum likelihood sequence estimation (MLSE) technique.

23. The receiver according to claim 14, wherein said equalizer comprises means for performing a sub-optimal complexity reduced maximum likelihood sequence estimation (MLSE) technique.

24. The receiver according to claim 14, wherein said equalizer comprises means for performing a decision feedback equalization (DFE) technique.

25. The receiver according to claim 14, wherein said receiver is adapted to receive and decode a global system for mobile communications (GSM) cellular signal.

26. In a communications receiver coupled to a bounded length channel having a plurality of non zero taps located within a plurality of symbol time intervals, a method of estimating channel order, said method comprising the steps of:

calculating a channel estimate over a plurality of bursts using a channel length comprising a first number taps to yield estimated channel taps;

averaging over time tap energies of said tap estimated channel taps to generate average tap energies;

selecting a threshold in accordance with a noise floor estimate calculated using a predetermined number of the lowest average tap energies; and

setting said channel order to a number of average tap energies above said threshold.

27. The method according to claim 26, wherein said step of calculating said channel estimate is performed using a least squares technique.

28. The method according to claim 26, wherein said step of calculating said channel estimate is performed using a correlation technique.

29. The method according to claim 26, wherein said estimated channel taps are represented as zero-mean, complex, Gaussian random processes.

30. The method according to claim 26, wherein said estimated channel taps are represented as non zero-mean, complex, Gaussian random processes.

31. The method according to claim 26, wherein said estimated channel taps vary over time.

32. A computer readable storage medium having a computer program embedded therein for causing a suitably programmed system to estimate the order of a channel having a plurality of non zero taps located within a plurality of symbol time intervals, by performing the following steps when said computer program is executed on said system:

calculating a channel estimate over a plurality of bursts using a channel length comprising a first number taps to yield estimated channel taps;

averaging tap energies over time, said tap energies determined from said channel taps;

selecting a threshold in accordance with a noise floor estimate calculated using a second number of lowest average tap energies; and

setting said channel order to a number of average tap energies above said threshold.