Method and device for symbol detection

1. A method for symbol detection, comprising:

assigning a received symbol to at least one particular candidate symbol of a set of candidate symbols of a finite candidate symbol alphabet based on a set of metrics between the received symbol and the at least one particular candidate symbol, the set of metrics comprising contributions with respect to channel-based information and contributions with respect to a priori information;

partitioning the set of metrics into a first sub-set of metrics comprising contributions with respect to the in-phase components of the candidate symbols and a second sub-set comprising contributions with respect to the quadrature components of the candidate symbols;

determining a first hypothesis based on the first sub-set of metrics;

determining a second hypothesis based on the second sub-set of metrics;

determining at least one first counter-hypothesis based on the first sub-set of metrics;

determining at least one second counter-hypothesis based on the second sub-set of metrics; and

computing a result based on the first and second hypothesis and based on the at least one first and the at least one second counter-hypotheses.

2. The method of claim 1 , comprising:

separating and combining the metrics contributions with respect to the channel-based information and the contributions with respect to a priori information.

3. The method of claim 1 , comprising:

using metrics according to one of a Manhattan norm, a Euclidean norm and an infinity norm, or powers thereof; and

combining the in-phase contributions and the quadrature contributions of such metrics.

4. The method of claim 1 , comprising:

receiving the a priori information from a channel decoder or any other source prior information.

5. The method of claim 1 , comprising:

receiving the symbol over a multiple-input multiple output communication system.

6. The method of claim 5 , comprising:

representing the particular symbol by a node of a search tree or a path through multiple nodes of a search tree whose layers represent streams of a multiple-input multiple output communication system.

7. The method of claim 6 , wherein the multiple-input multiple output communication system comprises one of the following schemes:

single-user MIMO,

multi-user MIMO,

network-assisted interference cancellation,

blind interference cancellation, and

Alamouti schemes.

8. The method of claim 1 , comprising:

using one of a QAM, a QPSK and a BPSK constellation, or vectors thereof, for forming a finite candidate symbol alphabet.

9. The method of claim 1 , comprising:

determining the at least one particular candidate symbol from the set of candidate symbols such that its metric is minimum with respect to an arbitrary set of candidate symbols.

10. The method of claim 1 , wherein the a priori information comprises a priori logarithmic likelihood ratios.

11. The method of claim 1 , comprising:

assigning logarithmic likelihood ratios to the received symbol based on the metrics between the received symbol and the at least one particular candidate symbol.

12. A device for symbol detection, comprising:

a first circuit configured to determine a set of metrics between a received symbol and candidate symbols of a set of candidate symbols of a finite candidate symbol alphabet, the set of metrics comprising channel-based contributions and a priori based contributions with respect to in-phase and quadrature components of the candidate symbols;

a second circuit configured to partition the set of metrics into a first sub-set comprising contributions with respect to the in-phase components of the candidate symbols and a second sub-set comprising contributions with respect to the quadrature components of the candidate symbols;

a third circuit configured to determine a first hypothesis based on the first sub-set of metrics;

a fourth circuit configured to determine a second hypothesis based on the second sub-set of metrics;

a fifth circuit configured to determine at least one first counter-hypothesis based on the first sub-set of metrics;

a sixth circuit configured to determine at least one second counter-hypothesis based on the second sub-set of metrics; and

a seventh circuit configured to compute a result based on the first and second hypothesis and based on the at least one first and the at least one second counter-hypotheses.

13. The device of claim 12 , wherein the first circuit is configured to determine the first and second sub-sets such that both their sizes are smaller or equal than a square root of a symbol number of a finite candidate symbol alphabet.

14. A device, comprising:

a soft-input soft-output MIMO detector configured to compute soft-output information based on a metric between a channel observation and a candidate symbol of a finite symbol alphabet, the metric comprising contributions with respect to channel-based information and contributions with respect to a priori information;

a soft-input soft-output channel decoder configured to provide the a priori information to the MIMO detector based on the soft-output information; and

a circuit configured to:

partition the metric into a first set of metrics comprising contributions with respect to the in-phase components of the candidate symbols and a second set comprising contributions with respect to the quadrature components of the candidate symbols;

determine a first hypothesis based on the first set of metrics;

determine a second hypothesis based on the second set of metrics;

determine at least one first counter-hypothesis based on the first set of metrics;

determine at least one second counter-hypothesis based on the second set of metrics; and

compute a result based on the first and second hypothesis and based on the at least one first and the at least one second counter-hypotheses.

15. The device of claim 14 , wherein the MIMO detector and the channel decoder are configured to iteratively exchange the detector soft-output information and decoder soft-output information.

16. The device of claim 14 , wherein the MIMO detector is configured to compute the soft-output information based on a tree search technique.

17. The device of claim 14 , wherein the MIMO detector comprises a sphere decoder configured to compute the soft-output information.

18. The device of claim 14 , wherein the MIMO detector is configured to compute the soft-output information based on a minimum mean-squared error algorithm or a Markov Chain Monte Carlo algorithm.