Channel estimation in a spread spectrum receiver

• US 20040052306A1
• Filed: 09/13/2002
• Published: 03/18/2004
• Est. Priority Date: 09/13/2002
• Status: Active Grant

• Current Assignee: Avago Technologies General IP PTE Limited (Broadcom, Inc.)
• Original Assignee: Broadcom Corporation (Broadcom, Inc.)
• Inventors: Desai, Prasanna, Ibrahim, Brima
• Granted Patent: US 7,301,993 B2
• Time in Patent Office: Days
• US Class Current: 375/152
• CPC Class Codes: H04B 1/7093 Matched filter type
• CPC Class Codes: H04L 25/0232 by interpolation between sounding signals

Abstract

A method and apparatus for channel estimation in a spread spectrum receiver include processing that begins by despreading a received spread spectrum complex baseband sample utilizing a pseudorandom spreading sequence to produce a despread baseband symbol. The processing continues by filtering the despread baseband symbol to attenuate adverse channel effects based on a channel response estimate to produce a filtered despread baseband symbol. The processing then continues by decoding the filtered despread baseband symbol to produce recaptured data. The processing continues by modulating the recaptured data to produce a remodulated symbol. The processing then continues by producing a reconstructed spread spectrum complex baseband sample from the remodulated symbol based on the pseudorandom spreading sequence. The processing then continues by generating the channel response estimate such that a convolution of the channel response estimate with the reconstructed spread spectrum complex baseband sample substantially emulates the received spread spectrum complex baseband sample.

Claims

1. A channel estimation module for use in a spread spectrum receiver, the channel estimation module comprises:

matched filter module operably coupled to produce a despread baseband symbol and delayed spread spectrum complex baseband samples from received spread spectrum complex baseband samples;

channel matched filter operably coupled to filter the despread baseband symbol based on an estimated channel response to produce an filtered symbol;

reconstruction module operably coupled to receive the filtered symbol and to produce therefrom reconstructed spread spectrum complex baseband samples; and

channel response determination module operably coupled to generate the estimated channel response based on differences between the delayed spread spectrum complex baseband samples and the reconstructed spread spectrum complex baseband samples.

2. The channel estimation module of claim 1, wherein the matched filter module further comprises:

finite impulse response filter module operably coupled to perform matched spread spectrum filtering on the received spread spectrum complex baseband samples to produce the despread baseband symbol.

3. The channel estimation module of claim 2, wherein the matched filter module further comprises:

plurality of delay elements operably coupled to produce the delayed spread spectrum complex baseband signal by delaying the received spread spectrum complex baseband symbol;

plurality of multipliers operably coupled to the plurality of delay elements, wherein the plurality of multipliers multiple a pseudorandom binary sequence with corresponding outputs of the plurality of delay elements to produce a plurality of despreading symbol components; and

summing module operably coupled to sum the plurality of despreading symbol components to produce the despread baseband symbol.

4. The channel estimation module of claim 3, wherein the pseudorandom binary sequence further comprises:

a Barker code sequence.

5. The channel estimation module of claim 1, wherein the channel matched filter further comprises:

RAKE combining finite impulse response filter module operably coupled to multiply the despread filtered symbol by a complex conjugate of channel gain that corresponds to the estimated channel response such that phase shift of a channel carrying the received spread spectrum baseband samples are compensated and multiple paths of the channel are proportionally weighted based on corresponding signal strengths.

6. The channel estimation module of claim 5, wherein the channel matched filter further comprises:

plurality of delay elements operably coupled to produce a series of delayed despread baseband symbols by delaying the despread baseband symbol;

plurality of multipliers operably coupled to the plurality of delay elements, wherein the plurality of multipliers multiply the series of delayed despread baseband symbols with a plurality of coefficients representing the complex conjugate of the channel gain to produce a plurality of filtered symbol components; and

summing module operably coupled to sum the plurality of filtered symbols to produce the filtered symbol.

7. The channel estimation module of claim 1, wherein the reconstruction module further comprises:

demodulation module operably coupled to demodulate the filtered symbol to produce recaptured data;

modulation module operably coupled to modulate the recaptured data to produce a remodulated symbol; and

spread spectrum module operably coupled to spread the remodulated symbol based on a pseudorandom spreading sequence to produce the reconstructed spread spectrum complex baseband samples.

8. The channel estimation module of claim 1, wherein the channel response determination module further comprises:

finite impulse response filter module operably coupled to perform a least mean square (LMS) algorithm on the reconstructed spread spectrum complex baseband samples in accordance with converging channel coefficients to produce to a finite impulse response (FIR) output;

summing module operably coupled to sum the FIR output with the delayed spread spectrum complex baseband samples to produce an error signal; and

channel coefficient converging module operably coupled to produce the converging channel coefficients based on the error signal.

9. The channel estimation module of claim 8, wherein the FIR filter module further comprises:

plurality of delay elements operably coupled to delay the reconstructed spread spectrum complex baseband samples to produce delayed reconstructed spread spectrum complex baseband samples;

plurality of multiplying modules operably coupled to the plurality of delay elements, wherein the plurality of multiplying modules multiple the converging channel coefficients with the delayed reconstructed spread spectrum complex baseband samples to produce a plurality of FIR output components; and

summing module operably coupled to sum the plurality of FIR output components to produce the FIR output.

10. The channel estimation module of claim 9, wherein the channel coefficient converging module further comprises:

plurality of multiply-accumulate modules operably coupled the plurality of delay elements, wherein the each of the multiply-accumulate modules multiplies a corresponding one of the delayed reconstructed spread spectrum complex baseband samples with the error signal to produce a current resultant and accumulates the current resultant with at least one previous resultant to produce a corresponding one of the converging channel coefficients.

11. The channel estimation module of claim 10, wherein the channel coefficient converging module further comprises:

step size module operably coupled to adjust step size of the error signal.

12. A method for channel estimation in a spread spectrum receiver, the method comprises:

despreading received spread spectrum complex baseband samples based on a pseudorandom spreading sequence to produce a despread baseband symbol;

filtering the despread baseband symbol to attenuate adverse channel effects based on a channel response estimate to produce a filtered despread baseband symbol;

demodulating the filtered despread baseband symbol to produce recaptured data;

modulating the recaptured data to produce a remodulated symbol;

producing reconstructed spread spectrum complex baseband samples from the remodulated symbol based on the pseudorandom spreading sequence; and

generating the channel response estimate such that a convolution of the channel response estimate with the reconstructed spread spectrum complex baseband samples substantially emulates the received spread spectrum complex baseband samples.

13. The method of claim 12, wherein the despreading the received spread spectrum complex baseband signal further comprises:

performing matched spread spectrum filtering on the received spread spectrum complex baseband samples to produce the despread baseband symbol.

14. The method of claim 13, wherein the performing the matched spread spectrum filtering further comprises:

delaying the received spread spectrum complex baseband samples to produce delayed spread spectrum complex baseband samples;

multiplying a pseudorandom binary sequence with the delayed spread spectrum complex baseband samples to produce a plurality of despreading symbol components; and

summing the plurality of despreading symbol components to produce the despread baseband symbol.

15. The method of claim 12, wherein the filtering of the despread baseband symbol further comprises:

multiplying, in accordance with a RAKE combining finite impulse response filter function, the despread filtered symbol by a complex conjugate of channel gain that corresponds to the channel response estimate such that phase shift of a channel carrying the received spread spectrum baseband samples are compensated and multiple paths of the channel are proportionally weighted based on corresponding signal strengths.

16. The method of claim 15, wherein the filtering of the despread baseband symbol further comprises:

delaying the despread baseband symbol to produce a series of delayed despread baseband symbols;

multiplying the series of delayed despread baseband symbols with a plurality of coefficients representing the complex conjugate of the channel gain to produce a plurality of filtered symbol components; and

summing the plurality of filtered symbol components to produce the filtered despread baseband symbol.

17. The method of claim 12, wherein the generating the channel response estimate module further comprises:

performing a least mean square (LMS) algorithm on the reconstructed spread spectrum complex baseband samples in accordance with converging channel coefficients to produce to an finite impulse response (FIR) output;

delaying the received spread spectrum complex baseband samples to produce delayed spread spectrum complex baseband samples;

summing the FIR output with the delayed spread spectrum complex baseband samples to produce an error signal; and

producing the converging channel coefficients based on the error signal.

18. The method of claim 17, wherein the performing the FIR algorithm further comprises:

delaying the reconstructed spread spectrum complex baseband samples to produce delayed reconstructed spread spectrum complex baseband samples;

multiplying the converging channel coefficients with the delayed reconstructed spread spectrum complex baseband samples to produce a plurality of FIR output components; and

summing the plurality of FIR output components to produce the FIR output.

19. The method of claim 18, wherein the performing the LMS algorithm further comprises:

multiplying a corresponding one of the delayed reconstructed spread spectrum complex baseband samples with the error signal to produce a current resultant; and

accumulating the current resultant with at least one previous resultant to produce a corresponding one of the converging channel coefficients.

20. The method of claim 19 further comprises:

adjusting step size of the error signal.

21. A spread spectrum receiver comprises:

processing module; and

memory operably coupled to the processing module, wherein the memory includes operational instructions that cause the processing module to;

despread received spread spectrum complex baseband samples based on a pseudorandom spreading sequence to produce a despread baseband symbol;

filter the despread baseband symbol to attenuate adverse channel effects based on a channel response estimate to produce a filtered despread baseband symbol;

demodulate the filtered despread baseband symbol to produce recaptured data;

modulate the recaptured data to produce a remodulated symbol;

produce reconstructed spread spectrum complex baseband samples from the remodulated symbol based on the pseudorandom spreading sequence; and

generate the channel response estimate such that a convolution of the channel response estimate with the reconstructed spread spectrum complex baseband samples substantially emulate the despread baseband symbol.

22. The spread spectrum receiver of claim 21, wherein the memory further comprises operational instructions that cause the processing module to despread the received spread spectrum complex baseband signal by:

performing matched spread spectrum filtering on the received spread spectrum complex baseband samples to produce the despread baseband symbol.

23. The spread spectrum receiver of claim 22, wherein the memory further comprises operational instructions that cause the processing module to perform the matched spread spectrum filtering by:

delaying the received spread spectrum complex baseband samples to produce delayed spread spectrum complex baseband samples;

multiplying a pseudorandom binary sequence with the delayed spread spectrum complex baseband samples to produce a plurality of despreading symbol components; and

summing the plurality of despreading symbol components to produce the despread baseband symbol.

24. The spread spectrum receiver of claim 21, wherein the memory further comprises operational instructions that cause the processing module to filter the despread baseband symbol by:

multiplying, in accordance with a RAKE combining finite impulse response filter function, the despread filtered symbol by a complex conjugate of channel gain that corresponds to the channel response estimate such that phase shift of a channel carrying the received spread spectrum baseband samples are compensated and multiple paths of the channel are proportionally weighted based on corresponding signal strengths.

25. The spread spectrum receiver of claim 24, wherein the memory further comprises operational instructions that cause the processing module to filter the despread baseband symbol by:

delaying the despread baseband symbol to produce a series of delayed despread baseband symbols;

multiplying the series of delayed despread baseband symbols with a plurality of coefficients representing the complex conjugate of the channel gain to produce a plurality of filtered symbol components; and

summing the plurality of filtered symbol components to produce the filtered despread baseband symbol.

26. The spread spectrum receiver of claim 21, wherein the memory further comprises operational instructions that cause the processing module to generate the channel response estimate module by:

performing a least mean square (LMS) algorithm on the reconstructed spread spectrum complex baseband sample in accordance with converging channel coefficients to produce to a finite impulse response (FIR) output;

summing the FIR output with the delayed spread spectrum complex baseband samples to produce an error signal; and

producing the converging channel coefficients based on the error signal.

27. The spread spectrum receiver of claim 26, wherein the memory further comprises operational instructions that cause the processing module to perform the LMS algorithm by:

delaying the reconstructed spread spectrum complex baseband samples to produce delayed reconstructed spread spectrum complex baseband samples;

multiplying the converging channel coefficients with the delayed reconstructed spread spectrum complex baseband samples to produce a plurality of FIR output components; and

summing the plurality of FIR output components to produce the FIR output.

28. The spread spectrum receiver of claim 27, wherein the memory further comprises operational instructions that cause the processing module to perform the LMS algorithm by:

multiplying a corresponding one of the delayed reconstructed spread spectrum complex baseband samples with the error signal to produce a current resultant; and

accumulating the current resultant with at least one previous resultant to produce a corresponding one of the converging channel coefficients.

29. The spread spectrum receiver of claim 28, wherein the memory further comprises operational instructions that cause the processing module to:

adjust step size of the error signal.

30. A spread spectrum radio comprises:

transmitter section operably coupled to convert outbound data into an outbound radio frequency (RF) signal;

receiver section operably coupled to convert an inbound RF signal into inbound data, wherein the receiver section includes;

down conversion module operably coupled to convert the inbound RF signal into inbound baseband signal;

gain control module operably coupled to increase amplitude of the inbound baseband signal to produce a gained inbound baseband signal;

analog to digital converter operably coupled to convert the gained inbound baseband signal into a digital baseband signal;

channel select filter module operably coupled to filter the digital baseband signal to produce at least one spread spectrum complex baseband sample;

matched filter module operably coupled to produce a despread baseband symbol and a delayed spread spectrum complex baseband sample from a spread spectrum complex baseband sample of the at least one spread spectrum complex baseband sample;

channel matched filter operably coupled to filter the despread baseband symbol based on an estimated channel response to produce an filtered symbol;

reconstruction module operably coupled to receive the filtered symbol and to produce therefrom a reconstructed spread spectrum complex baseband sample and the inbound data; and

channel response determination module operably coupled to generate the estimated channel response based on differences between the delayed spread spectrum complex baseband sample and the reconstructed spread spectrum complex baseband sample.

31. The spread spectrum radio of claim 30, wherein the matched filter module further comprises:

finite impulse response filter module operably coupled to perform matched spread spectrum filtering on the spread spectrum complex baseband sample to produce the despread baseband symbol.

32. The spread spectrum radio of claim 31, wherein the matched filter module further comprises:

plurality of delay elements operably coupled to produce the delayed spread spectrum complex baseband signal by delaying the spread spectrum complex baseband symbol;

plurality of multipliers operably coupled to the plurality of delay elements, wherein the plurality of multipliers multiple a pseudorandom binary sequence with corresponding outputs of the plurality of delay elements to produce a plurality of despreading symbol components; and

summing module operably coupled to sum the plurality of despreading symbol components to produce the despread baseband symbol.

33. The spread spectrum radio of claim 32, wherein the pseudorandom binary sequence further comprises:

a Barker code sequence.

34. The spread spectrum radio of claim 30, wherein the channel matched filter further comprises:

RAKE combining finite impulse response filter module operably coupled to multiply the despread filtered symbol by a complex conjugate of channel gain that corresponds to the estimated channel response such that phase shift of a channel carrying the received spread spectrum baseband sample is compensated and multiple paths of the channel are proportionally weighted based on corresponding signal strengths.

35. The spread spectrum radio of claim 34, wherein the channel matched filter further comprises:

plurality of delay elements operably coupled to produce a series of delayed despread baseband symbols by delaying the despread baseband symbol;

plurality of multipliers operably coupled to the plurality of delay elements, wherein the plurality of multipliers multiply the series of delayed despread baseband symbols with a plurality of coefficients representing the complex conjugate of the channel gain to produce a plurality of filtered symbol components; and

summing module operably coupled to sum the plurality of filtered symbols to produce the filtered symbol.

36. The spread spectrum radio of claim 30, wherein the reconstruction module further comprises:

demodulation module operably coupled to demodulate the filtered symbol to produce the inbound data;

modulation modulate operably coupled to modulate the inbound data to produce a remodulated symbol; and

spread spectrum module operably coupled to spread the remodulated symbol based on a pseudorandom spreading sequence to produce the reconstructed spread spectrum complex baseband sample.

37. The spread spectrum radio of claim 30, wherein the channel response determination module further comprises:

finite impulse response filter module operably coupled to perform a least mean square (LMS) algorithm on the reconstructed spread spectrum complex baseband sample in accordance with converging channel coefficients to produce to a finite impulse response (FIR) output;

summing module operably coupled to sum the FIR output with the delayed spread spectrum complex baseband sample to produce an error signal; and

channel coefficient converging module operably coupled to produce the converging channel coefficients based on the error signal.

38. The spread spectrum radio of claim 37, wherein the FIR filter module further comprises:

plurality of delay elements operably coupled to delay the reconstructed spread spectrum complex baseband sample to produce delayed reconstructed spread spectrum complex baseband samples;

plurality of multiplying modules operably coupled to the plurality of delay elements, wherein the plurality of multiplying modules multiple the converging channel coefficients with the delayed reconstructed spread spectrum complex baseband samples to produce a plurality of FIR output components; and

summing module operably coupled to sum the plurality of FIR output components to produce the FIR output.

39. The spread spectrum radio of claim 38, wherein the channel coefficient converging module further comprises:

plurality of multiply-accumulate modules operably coupled the plurality of delay elements, wherein the each of the multiply-accumulate modules multiplies a corresponding one of the delayed reconstructed spread spectrum complex baseband samples with the error signal to produce a current resultant and accumulates the current resultant with at least one previous resultant to produce a corresponding one of the converging channel coefficients.

40. The spread spectrum radio of claim 39, wherein the channel coefficient converging module further comprises:

step size module operably coupled to adjust step size of the error signal.