Method and apparatus for channel predicting

1. A channel predicting method in a communication network comprising a time-varying wireless channel, the method comprising:

determining a prediction strategy;

selecting a certain number of channel samples from candidate channel samples in accordance with the prediction strategy, the channel samples each corresponding to one symbol in a time domain;

determining prediction weights in accordance with the prediction strategy and statistical properties of the time-varying wireless channel;

weighting the selected channel samples with the determined prediction weights to obtain a predicting channel sample as a prediction result.

2. The method of claim 1 , wherein the candidate channel samples comprise previously predicted channel samples.

3. The method of claim 1 , wherein the prediction strategy includes a triplet of three parameters (M, I, P), in which M denotes the number of the selected channel samples, I denotes a distance between two contiguously selected channel samples in the candidate channel samples, and P denotes a distance between the predicting channel sample and the selected channel sample that is closest to the predicting channel sample.

4. The method of claim 3 , wherein I and P are fixed, and M is incremented symbol by symbol.

5. The method of claim 1 , wherein the statistical properties of the time-varying wireless channel include spaced-frequency spaced-time correlations.

6. The method of claim 5 , wherein the spaced-frequency spaced-time correlations are based on a delay power spectrum and a spaced-time correlation function of the time-varying wireless channel.

7. The method of claim 1 , wherein the determining prediction weights comprises:

calculating prediction weights corresponding to all possible prediction strategies based on the statistical properties in advance;

selecting a prediction weight corresponding to the determined prediction strategy.

8. The method of claim 1 , wherein the determining prediction weights comprises calculating the prediction weights based on the statistical properties in accordance with the determined prediction strategy.

9. The method of claim 1 , wherein the determining a prediction strategy comprises training some previous channel samples of the candidate channel samples in terms of respective prediction errors to determine the prediction strategy.

10. The method of claim 9 , wherein a Viterbi Algorithm is used in the training.

11. The method of claim 1 , wherein the channel predicting method is performed symbol by symbol.

12. The method of claim 11 , wherein the prediction strategy is different on each symbol.

13. An apparatus for channel predicting in a communication network comprising a time-varying wireless channel, the apparatus comprising:

a prediction strategy determination circuit configured to determine a prediction strategy;

a first selection circuit configured to select a certain number of channel samples from candidate channel samples in accordance with the prediction strategy, each channel sample corresponding to one symbol in a time domain;

a prediction weights determination circuit configured to determine prediction weights in accordance with the prediction strategy and statistical properties of the time-varying wireless channel;

a weighting circuit configured to weight the selected channel samples with the determined prediction weights to obtain a predicting channel sample as a prediction result.

14. The apparatus of claim 13 , wherein the candidate channel samples comprise previously predicted channel samples.

15. The apparatus of claim 13 , wherein the prediction strategy includes a triplet of three parameters (M, I, P), in which M denotes the number of the selected channel samples, I denotes a distance between two contiguously selected channel samples in the candidate channel samples, and P denotes a distance between the predicting channel sample and the selected channel sample that is closest to the predicting channel sample.

16. The apparatus of claim 15 , wherein I and P are fixed, and M is incremented symbol by symbol.

17. The apparatus of claim 13 , wherein the statistical properties of the time-varying wireless channel include spaced-frequency spaced-time correlations.

18. The apparatus of claim 17 , wherein the spaced-frequency spaced-time correlations are based on a delay power spectrum and a spaced-time correlation function of the time-varying wireless channel.

19. The apparatus of claim 13 , wherein the prediction weights determination circuit comprises:

a first calculation circuit configured to calculate prediction weights corresponding to all possible prediction strategies based on the statistical properties in advance;

a second selection circuit configured to select a prediction weight corresponding to the determined prediction strategy.

20. The apparatus of claim 13 , wherein the prediction weights determination circuit comprises a second calculation circuit configured to calculate the prediction weights based on the statistical properties in accordance with the determined prediction strategy.

21. The apparatus of claim 13 , wherein the prediction strategy determination circuit comprises a training circuit configured to train some previous channel samples of the candidate channel samples in terms of respective prediction errors to determine the prediction strategy.

22. The apparatus of claim 21 , wherein a Viterbi Algorithm is used in the training.

23. The apparatus of claim 13 , wherein the apparatus performs channel predicting symbol by symbol.

24. The apparatus of claim 23 , wherein the prediction strategy is different on each symbol.