Wireless receiver

• US 10,142,145 B2
• Filed: 11/28/2016
• Issued: 11/27/2018
• Est. Priority Date: 11/27/2015
• Status: Active Grant

• Current Assignee: Cohda Wireless Pty Ltd.
• Original Assignee: Cohda Wireless Pty Ltd.
• Inventors: Alexander, Paul Dean
• Primary Examiner(s): LAM, KENNETH T
• Application Number:
  US15/362,426
• Publication Number:
  US 20170155532A1
• Time in Patent Office: 729 Days
• CPC Class Codes: H04L 1/0054 Maximum-likelihood or sequential decoding, e.g. Viterbi, Fano, ZJ algorithms
• CPC Class Codes: H04L 2025/03624 Zero-forcing
• CPC Class Codes: H04L 25/0212 of impulse response
• CPC Class Codes: H04L 25/0258 Channel estimation using zero-forcing criteria
• CPC Class Codes: H04L 25/03006 Arrangements for removing intersymbol interference
• CPC Class Codes: H04L 25/067 providing soft decisions, i.e. decisions together with an estimate of reliability H04L25/068 and H04L25/069 take precedence; sequence estimation techniques H04L25/03178
• CPC Class Codes: H04L 27/2338 using sampling H04L27/2331 - H04L27/2335 take precedence

Abstract

The present invention relates to a method and apparatus for demodulation in a wireless communications system transmitted across a wireless communications channel. The described wireless receiver includes a first antenna for receiving a wireless signal including a symbol transmitted across a wireless communications channel perceived by the first antenna, an observation modifier for generating a modified observation (y) of the symbol based on a product of the received observation (r) and the complex conjugate of a channel estimate (h*), a log-likelihood ratio (LLR) module generating log-likelihood ratios (LLRs) based on the modified observation and the channel estimate, and a maximum-likelihood-based decoder for decoding the symbol based on the LLRs.

Claims

1. A method comprising:

receiving, at a first antenna, an observation of a symbol transmitted across a wireless communications channel perceived by the first antenna;

generating a modified observation of the symbol based on a product of the received observation and the complex conjugate of a channel estimate of the channel; and

generating, based on the modified observation and the channel estimate, log-likelihood ratios (LLRs) for the symbol for a maximum-likelihood-based decoder to decode,

wherein generating log-likelihood ratios includes;

generating a LLR associated with a most significant bit of the symbol, and

generating a LLR associated with a next most significant bit of the symbol;

wherein the log-likelihood ratio (LLR) associated with the most significant bit is generated based on the ratio of the real part (y_I) or imaginary part (y_Q) of the modified observation to Gaussian-distributed noise power of the channel.

2. The method of claim 1 wherein the LLR associated with the most significant bit is estimated to be 2y_I/σ

² or 2y_Q/σ

², where σ

² represents the Gaussian-distributed noise power of the channel.

3. The method of claim 1 wherein the LLR associated with the most significant bit is estimated to be 4y_I/σ

² or 4y_Q/σ

², where σ

² represents the Gaussian-distributed noise power of the channel.

4. The method of claim 1 wherein the LLR associated with the most significant bit is estimated to be 4y_I/σ

²−

4|h|²/σ

² or 4y_Q/σ

²−

4|h|²/σ

², where σ

² represents the Gaussian-distributed noise power of the channel and |h| represents the magnitude of the channel estimate.

5. The method of claim 1 wherein the LLR associated with the most significant bit is 4y_I/σ

²+4|h|²/σ

², where σ

² represents the variance of the Gaussian-distributed noise power of the channel and |h| represents the magnitude of the channel estimate.

6. The method of claim 1 wherein the LLR associated with the next most significant bit is generated based on the ratio of the real part (y_I) or imaginary part (y_Q) of the modified observation to Gaussian-distributed noise power of the channel.

7. The method of claim 6 wherein the LLR associated with the most significant bit is 4|h|/σ

²−

2y_I/σ

² or 4|h|/σ

²−

2y_Q/σ

², where σ

² represents the variance of the Gaussian-distributed noise power of the channel and |h| represents the magnitude of the channel estimate.

8. The method of claim 1 wherein the Gaussian-distributed noise power is estimated or measured.

9. The method of claim 1 further comprising receiving, at a second antenna located separately from the first antenna, another observation of the symbol transmitted across a wireless communications channel perceived by the second antenna, and wherein generating a modified observation includes generating the modified observation also based on a product of the other received observation and the complex conjugate of a channel estimate of the channel perceived by the second antenna.

10. The method of claim 9 wherein generating the modified observation is based on a weighted sum of (a) the product of the received observation and the complex conjugate of a channel estimate of the channel and (b) the product of the other received observation and the complex conjugate of a channel estimate of the channel perceived by the second antenna.

11. The method of claim 10 wherein the weighted sum is based on weights associated with the noise power as observed by the respective antennas.

12. The method of claim 1 further comprising decoding, by a maximum-likelihood-based decoder, the LLRs.

13. A wireless receiver comprising:

a first antenna for receiving an observation including a symbol transmitted across a wireless communications channel perceived by the first antenna;

an observation modifier for generating a modified observation of the symbol based on a product of the received observation and the complex conjugate of a channel estimate of the channel;

a log-likelihood ratio (LLR) for the symbol module generating log-likelihood ratios (LLRs) based on the modified observation,

wherein generating log-likelihood ratios includes;

generating a LLR associated with a most significant bit of the symbol, and

generating a LLR associated with a next most significant bit of the symbol;

wherein the log-likelihood ratio (LLR) associated with the most significant bit is generated based on the ratio of the real part (y_I) or imaginary part (y_Q) of the modified observation to Gaussian-distributed noise power of the channel; and

a maximum-likelihood-based decoder for decoding the symbol based on the LLRs.

14. The wireless receiver of claim 13 further comprising a second antenna for receiving another observation of the symbol transmitted across a wireless communications channel by the second antenna, the first antenna and the second antenna are located to provide spatial diversity,

wherein the observation modifier is configured to generate the modified observation based on weighted sum of (a) the product of the received observation and the complex conjugate of a channel estimate of the channel and (b) the product of the other received observation and the complex conjugate of a channel estimate of the channel perceived by the second antenna.

15. The wireless receiver of claim 14 wherein the weighted sum is based on weights associated with the noise power as observed by the respective antennas.

16. A non-transitory machine-readable medium comprising instructions executable by one or more processors, the instructions including the steps of:

receiving, at a first antenna, an observation of a symbol transmitted across a wireless communications channel perceived by the first antenna;

generating a modified observation of the symbol based on a product of the received observation and the complex conjugate of a channel estimate of the channel; and

generating, based on the modified observation and the channel estimate, log-likelihood ratios (LLRs) for the symbol for a maximum-likelihood-based decoder to decode,

wherein generating log-likelihood ratios includes;

generating a LLR associated with a most significant bit of the symbol, and

generating a LLR associated with a next most significant bit of the symbol;

wherein the log-likelihood ratio (LLR) associated with the most significant bit is generated based on the ratio of the real part (y_I) or imaginary part (y_Q) of the modified observation to Gaussian-distributed noise power of the channel.