Method for the soft bit metric calculation with linear MIMO detection for LDPC codes

• US 7,751,506 B2
• Filed: 12/01/2005
• Issued: 07/06/2010
• Est. Priority Date: 12/01/2005
• Status: Expired due to Fees

• Current Assignee: Samsung Electronics Co. Ltd.
• Original Assignee: Samsung Electronics Co. Ltd.
• Inventors: Niu, Huaning, Ngo, Chiu
• Primary Examiner(s): Liu; Shuwang
• Assistant Examiner(s): MCKIE, GINA M
• Application Number:
  US11/292,853
• Publication Number:
  US 20070127603A1
• Time in Patent Office: 1,678 Days
• Field of Search: None
• US Class Current: 375/341
• CPC Class Codes: H03M 13/1102 Codes on graphs and decoding on graphs, e.g. low-density parity check [LDPC] codes
• CPC Class Codes: H04L 1/0045 Arrangements at the receiver end
• CPC Class Codes: H04L 1/0057 Block codes H04L1/0061, H04L1/0064 take precedence
• CPC Class Codes: H04L 2025/03414 Multicarrier
• CPC Class Codes: H04L 2025/03426 transmission using multiple-input and multiple-output channels
• CPC Class Codes: H04L 2025/03605 Block algorithms
• CPC Class Codes: H04L 25/03159 operating in the frequency domain H04L25/03165, H04L25/03178 take precedence

Abstract

A MIMO receiver implements a method for the soft bit metric calculation with linear MIMO detection for LDPC codes, after linear matrix inversion MIMO detection. In the receiver, a detector detects the estimated symbol and the noise variance. Further, a soft metric calculation unit computes the distance between the estimated symbol and the constellation point, and then divides the distance by the noise variance to determine the soft bit metrics.

Claims

1. A method of soft bit metric calculation for received data signals via a channel in a telecommunications receiver utilizing Low Density Parity Check (LDPC) decoding, comprising:

detecting data symbols in the received signals by applying a linear Multiple-Input Multiple-Output (MIMO) detection, wherein the linear MIMO detection comprises determining distances between the detected data symbols and constellation points;

determining a noise variance of the channel;

calculating the soft bit metrics as a function of the distances between the detected data symbols and the constellation points, and calculating a channel noise variance matrix E[vv^H], wherein the distances are divided by the noise variance, wherein calculating the soft bit metrics further includes;

selecting diagonal elements of the channel noise variance matrix E[vv^H], and

calculating the soft bit metrics as a function of the distances and σ

_v², where the diagonal elements σ

_v² is defined as;

σ

_v²=E[diag{vv^H}]

=σ

_n²diag{WW^H},

W is a linear filter and W^H is a pseudo-inverse of W, σ

_n² represents the noise variance and v is a new noise term and v^H is a pseudo-inverse of V.

2. The method of claim 1, further comprising:

LDPC decoding using the soft bit metrics to determine data values in the received signals.

3. The method of claim 1, wherein the determining the noise variance further includes performing minimum mean squared error (MMSE) criterion MIMO detection.

4. The method of claim 3 wherein the channel is defined by a NrxNt channel matrix H in a MIMO system having a receiver with Nt transmitter antennas and Nr receiver antennas, x is the Ntx1 transmitted signal vector, the received symbol is represented by a Nrx1 received symbol vector y=Hx+n, n represents a Nrx1 noise vector, the method further comprising applying a linear filter W to the received symbols y to compute estimated symbols {circumflex over (x)} as:

{circumflex over (x)}=Wy=WHx+Wn,

where W=(H^HH+σ

_n²I)<sup>−

1</sup>H^H with MMSE criterion, where H^H is a pseudo-inverse of H.

5. The method of claim 1 wherein the determining the noise variance further includes performing Zero Forcing (ZF) criterion MIMO detection.

6. The method of claim 5 wherein the channel is defined by a NrxNt channel matrix H in a MIMO system having a receiver with Nt transmitter antennas and Nr receiver antennas, x is the Ntx1 transmitted signal vector, the received symbol is represented by a Nrx1 received symbol vector y=Hx+n, n represents a Nrx1 noise vector, the method further comprising applying a linear filter W to the received symbols y to compute estimated symbols {circumflex over (x)} as:

{circumflex over (x)}=Wy=WHx+Wn,

where W=(H^HH)<sup>−

1</sup>H^H with ZF criterion and H^H is a pseudo-inverse of H.

7. The method of claim 1, wherein the calculating the soft bit metrics further includes:

determining the new noise term v=(WH−

diag(WH))x+Wn where filtering effect and cross-talk are considered,

selecting the diagonal elements σ

_v² of the noise variance matrix E[vv^H],

calculating the soft bit metrics as a function of the distance and σ

_v², where W is a linear filter, H is a NrxNt channel matrix in a MIMO system having a receiver with Nt transmitter antennas and Nr receiver antennas, x is the Ntx1 transmitted signal vector, n represents a Nrx1 noise vector and v^H is a pseudo-inverse of v.

8. The method of claim 7, wherein the determining the noise variance further includes performing Zero Forcing (ZF) criterion MIMO detection, wherein the channel is defined by a NrxNt channel matrix H in a MIMO system having a receiver with Nt transmitter antennas and Nr receiver antennas, x is the Ntx1 transmitted signal vector, the received symbol is represented by a Nrx1 received symbol vector y=Hx+n, n represents a Nrx1 noise vector, the method further comprising applying a linear filter W to the received symbols y to compute estimated symbols {circumflex over (x)} as:

{circumflex over (x)}=Wy=WHx+Wn,

where W=(H^HH)<sup>−

1</sup>H^H with ZF criterion,

such that for ZF estimation;

WH=(H^HH)<sup>−

1</sup>H^HH=I.

9. The method of claim 7 wherein the soft bit metric of a first received data stream is divided by a first diagonal element of the noise variance matrix, and the soft metric of a second received data stream is divided by a second diagonal element of the noise variance matrix.

10. The method of claim 1 wherein the receiver comprises a wireless MIMO receiver.

11. The method of claim 10 wherein the receiver receives multiple data streams signals from a transmitter with multiple antennas.

12. The method of claim 1 wherein:

the detecting data symbols further includes detecting the data symbols in the received signal using linear matrix inversion MIMO detection, and

the calculating the bit metrics further includes computing the bit metrics by determining distances between the detected data symbols and constellation points.

13. A telecommunications receiver, comprising:

a detector that detects data symbols in received signals by determining distances between received signal points and constellation points, and determines channel noise variance;

a metric calculator that calculates soft bit metrics for the received signals via a channel, as a function of the distances and the noise variance, wherein the distances are divided by the noise variance, wherein the metric calculator calculates the soft bit metrics by;

selecting diagonal elements σ

_v² of the noise variance matrix E[vv^H],

calculating the soft bit metrics as a function of the distances and σ

_v², where the diagonal elements σ

_v² is defined as;

σ

_v²=E[diag{vv^H}]

=σ

_n²diag{WW^H},

W is a linear filter and W^H is a pseudo-inverse of W, σ

_n² represents the noise variance and v is a new noise term and v^H is a pseudo-inverse of v; and

a decoder that performs LDPC decoding of the received signals using the soft bit metrics.

14. The receiver of claim 13 wherein the decoder performs LDPC decoding using the soft bit metrics to determine data values in the received signals.

15. The receiver of claim 13 wherein the detector determines the noise variance by performing minimum mean squared error (MMSE) criterion MIMO detection.

16. The receiver of claim 15 wherein the channel is defined by a NrxNt channel matrix H in a MIMO system having a receiver with Nt transmitter antennas and Nr receiver antennas, x is the Ntx1 transmitted signal vector, the received symbol is represented by a Nrx1 received symbol vector y=Hx+n, n represents a Nrx1 noise vector, such that detector applies a linear filter W to the received symbols y to compute estimated symbols {circumflex over (x)} as:

{circumflex over (x)}=Wy=WHx+Wn,

where W=(H^HH+σ

_n²I)<sup>−

1</sup>H^H with MMSE criterion and H^H is a pseudo-inverse of H.

17. The receiver of claim 13 wherein the detector determines the noise variance by performing Zero Forcing (ZF) criterion MIMO detection.

18. The receiver of claim 17 wherein the channel is defined by a NrxNt channel matrix H in a MIMO system having a receiver with Nt transmitter antennas and Nr receiver antennas, x is the Ntx1 transmitted signal vector, the received symbol is represented by a Nrx1 received symbol vector y=Hx+n, n represents a Nrx1 noise vector, such that the detector applies a linear filter W to the received symbols y to compute estimated symbols {circumflex over (x)} as:

{circumflex over (x)}=Wy=WHx+Wn,

where W=(H^HH)<sup>−

1</sup>H^H with ZF criterion and H^H is a pseudo-inverse of H.

19. The receiver of claim 13, wherein the detector determines the noise variance by performing Zero Forcing (ZF) criterion MIMO detection, wherein the channel is defined by a NrxNt channel matrix H in a MIMO system having a receiver with Nt transmitter antennas and Nr receiver antennas, x is the Ntx1 transmitted signal vector, the received symbol is represented by a Nrx1 received symbol vector y=Hx+n, n represents a Nrx1 noise vector, such that the detector applies a linear filter W to the received symbols y to compute estimated symbols {circumflex over (x)} as:

{circumflex over (x)}=Wy=WHx+Wn,

where W=(H^HH+σ

_n²I)<sup>−

1</sup>H^H with ZF criterion and σ

_n² represents the noise variance,

such that for ZF estimation;

WH=(H^HH)<sup>−

1</sup>H^HH=I.

20. The receiver of claim 13 wherein the soft bit metric of a first received data stream is divided by a first diagonal element of the noise variance matrix, and the soft metric of a second received data stream is divided by a second diagonal element of the noise variance matrix.

21. The receiver of claim 13 wherein the receiver comprises a wireless MIMO receiver.

22. The receiver of claim 21 wherein the receiver receives multiple data streams signals from a transmitter with multiple antennas.

23. The receiver of claim 13 wherein:

the detector detects the data symbols in the received signal using direct matrix inversion type linear MIMO detection, and

the metric calculator computes the bit metrics by determining distances between the detected data symbols and constellation points.

24. A wireless communication receiver, comprising:

a detector configured to detect symbols by determining distances between received signal points and constellation points, and to determine channel noise variance in a received data signal by applying a linear MIMO detection;

a metric calculator configured to select diagonal elements σ

_v² of a noise variance matrix E[vv^H] and calculate soft bit metrics for the received data signal via a channel by selecting the diagonal elements σ

_v² of the noise variance matrix E[vv^H] as a function of the distances and σ

_v², where the diagonal elements σ

_v² is defined as;

σ

_v²=E[diag{vv^H}]

=σ

_n²diag{WW^H},

W is a linear filter, W^H is a pseudo-inverse of W, σ

_n² represents the noise variance and v is a new noise term and v^H is a pseudo-inverse of v.

25. The wireless communication receiver of claim 24, wherein the soft bit metrics are calculated by dividing the distances by the noise variance.

26. The wireless communication receiver of claim 24, further comprising a decoder that performs LDPC decoding of the received data signal using the soft bit metrics.

27. The wireless receiver of claim 24, wherein the channel is defined by a NrxNt channel matrix H in a MIMO system having a receiver with Nt transmitter antennas and Nr receiver antennas, x is the Ntx1 transmitted signal vector, the received symbol is represented by a Nrx1 received symbol vector y=Hx+n, n represents a Nrx1 noise vector, such that detector applies the linear filter W to the received symbols y to compute estimated symbols {circumflex over (x)} as:

{circumflex over (x)}=Wy=WHx+Wn,

where W=(H^HH+σ

_n²I)<sup>−

1</sup>H^H with minimum mean squared error (MMSE) criterion and H^H is a pseudo-inverse of H.

28. The wireless receiver of claim 24, wherein the channel is defined by a NrxNt channel matrix H in a MIMO system having a receiver with Nt transmitter antennas and Nr receiver antennas, x is the Ntx1 transmitted signal vector, the received symbol is represented by a Nrx1 received symbol vector y=Hx+n, n represents a Nrx1 noise vector,

such that the detector applies a linear filter W to the received symbols y to compute estimated symbols {circumflex over (x)} as:

{circumflex over (x)}=Wy=WHx+Wn,

where W=(H^HH)<sup>−

1</sup>H^H with Zero Forcing (ZF) criterion and H^H is a pseudo-inverse of H.

29. A method of soft bit metric calculation for received data signals via a channel in a telecommunications receiver utilizing LDPC decoding, comprising:

detecting data symbols in the received signals by applying a linear MIMO detection, wherein the linear MIMO detection comprises determining distances between the detected data symbols and constellation points;

determining a noise variance of the channel;

calculating the soft bit metrics as a function of the distances between the detected data symbols and the constellation points, and calculating a channel noise variance matrix E[vv^H], wherein the distances are divided by the noise variance, said calculating further includes selecting the diagonal elements σ

_v² of the noise variance matrix E[vv^H] as a function of the distances and σ

_v², where the diagonal elements σ

_v² is defined as;

σ

_v²=E[diag{vv^H}]

=σ

_n²diag{WW^H},

where W is a linear filter, W^H is a pseudo-inverse of W, σ

_n² represents the noise variance and v is a new noise term and v^H is a pseudo-inverse of v,

wherein the channel is defined by a NrxNt channel matrix H in a MIMO system having a receiver with Nt transmitter antennas and Nr receiver antennas, x is the Ntx1 transmitted signal vector, the received symbol is represented by a Nrx1 received symbol vector y=Hx+n, n represents a Nrx1 noise vector, the method further comprising applying the linear filter W to the received symbols y to compute estimated symbols {circumflex over (x)} as;

{circumflex over (x)}=Wy=WHx+Wn,

where W=(H^HH+σ

_n²I)<sup>−

1</sup>H^H with MMSE criterion, where H^H is a pseudo-inverse of H.

30. A telecommunications receiver, comprising:

a detector that detects data symbols in received signals by determining distances between received signal points and constellation points, and determines noise variance of a channel;

a metric calculator that calculates soft bit metrics for the received signals via the channel, as a function of the distances and the noise variance, wherein the distances are divided by the noise variance, said calculator further selects the diagonal elements σ

_v² of a noise variance matrix E[vv^H] as a function of the distances and σ

_v², where the diagonal elements σ

_v² is defined as;

σ

_v²=E[diag{vv^H}]

=σ

_n²diag{WW^H},

where W is a linear filter, W^H is a pseudo-inverse of W, σ

_n² represents the noise variance and v is a new noise term and v^H is a pseudo-inverse of v; and

a decoder that performs LDPC decoding of the received signals using the soft bit metrics, wherein the channel is defined by a NrxNt channel matrix H in a MIMO system having a receiver with Nt transmitter antennas and Nr receiver antennas, x is the Ntx1 transmitted signal vector, the received symbol is represented by a Nrx1 received symbol vector y=Hx+n, n represents a Nrx1 noise vector, such that the detector applies the linear filter W to the received symbols y to compute estimated symbols {circumflex over (x)} as;

{circumflex over (x)}=Wy=WHx+Wn,

where W=(H^HH+σ

_n²I)<sup>−

1</sup>H^H with MMSE criterion and H^H is a pseudo-inverse of H.

31. A method of soft bit metric calculation for received signals via a channel in a telecommunications receiver utilizing LDPC decoding, comprising:

detecting data symbols in the received signals by applying a linear MIMO detection, wherein the linear MIMO detection comprises determining distances between the detected data symbols and constellation points;

determining a noise variance of the channel;

calculating the soft bit metrics as a function of the distances between the detected data symbols and the constellation points, and calculating a channel noise variance matrix E[vv^H], wherein the distances are divided by the noise variance, wherein calculating the soft bit metrics further includes;

selecting diagonal elements σ

_v² of the channel noise variance matrix E[vv^H] and

calculating the soft bit metrics as a function of the distances and the diagonal elements σ

_v², where the diagonal elements σ

_v² is defined as;

σ

_v²=E[diag{vv^H}]

=σ

_n²diag{WW^H},

W is a linear filter, W^H is a pseudo-inverse of W, σ

_n² represents the noise variance and v is a new noise term and v^H is a pseudo-inverse of v.

32. A telecommunications receiver, comprising:

a detector that detects data symbols in signals received via a channel by determining distances between received signal points and constellation points, and determines noise variance of the channel;

a metric calculator that calculates soft bit metrics for the received signals, as a function of the distances and the noise variance, wherein the distances are divided by the noise variance, wherein the metric calculator calculates the soft bit metrics by;

selecting the diagonal elements σ

_v² of a channel noise variance matrix E[vv^H],

calculating the soft bit metrics as a function of the distances and the diagonal elements σ

_v², where the diagonal elements σ

_v² is defined as;

σ

_v²=E[diag{vv^H}]

=σ

_n²diag{WW^H},

W is a linear filter, W^H is a pseudo-inverse of W, σ

_n² represents the noise variance and v is a new noise term and v^H is a pseudo-inverse of v; and

a decoder that performs LDPC decoding of the received signals using the soft bit metrics.

33. A wireless communication receiver, comprising:

a detector configured for detecting symbols by determining distances between received signal points and constellation points, and for determining channel noise variance in a received data signal by applying a linear MIMO detection;

a metric calculator configured for selecting diagonal elements σ

_v² of a noise variance matrix E[vv^H] and for calculating soft bit metrics for the received data signal via a channel by selecting the diagonal elements σ

_v² of the noise variance matrix E[vv^H], and calculating the soft bit metrics as a function of the distances and the diagonal elements σ

_v², where the diagonal elements σ

_v² is defined as;

σ

_v²=E[diag{vv^H}]

=σ

_n²diag{WW^H},

W is a linear filter, W^H is a pseudo-inverse of W, σ

_n² represents the noise variance and v is a new noise term and v^H is a pseudo-inverse of v.