Frequency estimation based on constellation error in QAM modems

1. A method for frequency estimation in a QAM modem comprising the steps of:

receiving I symbol and Q symbol components of a QAM symbol from a carrier recovery block;

mapping said received QAM symbol as defined by its I symbol and Q symbol symbol components to a nearest QAM plant point using a slicer;

determining an instant error power of said received QAM point by comparing said I symbol and Q symbol components of said received QAM symbol point and I plant and Q plant components of said nearest QAM plant point;

averaging said instant error power over a plurality of incoming QAM symbols in time domain;

excluding a DC component from said averaged error power;

translating said averaged error power without said DC component from said time domain into a frequency domain in order to determine the frequency spectrum of said averaged error power;

selecting the frequency spectrum component of said averaged error power with the maximum amplitude by using a peak detection;

and

determining said frequency offset as said maximum frequency spectrum component of said averaged error power.

2. The method of claim 1 , wherein said step of averaging said instant error power over said plurality of incoming QAM symbols in time domain further includes the step of:

using a digital infinite impulse response (IIR) filter.

3. The method of claim 2 , wherein said step of using said digital infinite impulse response (IIR) filter in order to average said instant error power over said plurality of incoming QAM symbols in time domain further includes the step of:

using a digital exponential averaging filter.

4. The method of claim 1 , wherein said step of averaging said instant error power over said plurality of incoming QAM symbols in time domain further includes the step of:

using a digital low pass filter (LPF).

5. The method of claim 1 , wherein said step of averaging said instant error power over said plurality of incoming QAM symbols in time domain further includes the step of:

using a digital finite impulse response filter (FIR).

6. The method of claim 5 , wherein said step of using said digital finite impulse response filter (FIR) in order to average said instant error power over said plurality of incoming QAM symbols in time domain further includes the step of:

using a moving averaging filter of length N, wherein N is a number of received symbols selected for averaging purposes, N is an integer.

7. The method of claim 6 further including the step of:

selecting a sequence α N of N symbols, wherein said moving averaging filter is configured to use said sequence α N of N symbols for averaging purposes.

8. The method of claim 7 , wherein said step of selecting said sequence α N of N symbols for averaging purposes further includes the step of:

selecting a sequence α N — Last of last received N symbols for averaging purposes.

9. The method of claim 7 , wherein said step of selecting said sequence α N of N symbols for averaging purposes further includes the step of:

selecting a decimated sequence α N,k,R Decimated of received N symbols for averaging purposes, wherein said decimated sequence α N,k,R Decimated includes a sequence of N symbols selected in the following order: {“k”-th incoming symbol, “k+R”-th incoming symbol, “k+2R”-th incoming symbol, . . . , “k+(N−1)* R”-th incoming symbol}; wherein “k” is an integer less than N; and wherein a decimation rate “R” is an integer less than N.

10. The method of claim 1 , wherein said step of translating said averaged error power without said DC component from said time domain into said frequency domain in order to determine the frequency spectrum of said averaged error power further includes the step of:

using a Fast Fourier Transform (FFT).

11. The method of claim 1 , wherein said step of translating said averaged error power without said DC component from said time domain into said frequency domain in order to determine the frequency spectrum of said averaged error power further includes the step of:

using a Discrete Fourier Transform (DFT).

12. The method of claim 1 , wherein said step of translating said averaged error power without said DC component from said time domain into said frequency domain in order to determine the frequency spectrum of said averaged error power further includes the step of:

using a wavelet transform.

13. The method of claim 1 , wherein said step of averaging said instant error power over said plurality of incoming QAM symbols in time domain further includes the step of:

measuring signal-to-noise ratio (SNR) by using said averaged error power and by using an averaged predetermined QAM constellation power.

14. An apparatus for frequency estimation in a QAM modem comprising:

a means for receiving I symbol and Q symbol components of a QAM symbol from a carrier recovery block;

a means for mapping said received QAM symbol as defined by its I symbol and Q symbol symbol components to a nearest QAM plant point using a slicer;

a means for determining an instant error power of said received QAM point;

a means for averaging said instant error power over a plurality of incoming QAM symbols in time domain;

a means for excluding a DC component from said averaged error power;

a means for translating said averaged error power without said DC component from said time domain into a frequency domain in order to determine the frequency spectrum of said averaged error power;

a means for selecting the frequency spectrum component of said averaged error power with the maximum amplitude;

and

a means for determining said frequency offset as said maximum frequency spectrum component of said averaged error power.

15. The apparatus of claim 14 , wherein said means for mapping said received QAM symbol as defined by its I symbol and Q symbol symbol components to a nearest QAM plant point further includes:

a slicer.

16. The apparatus of claim 14 , wherein said means for determining an instant error power of said received QAM point further includes:

a means for comparing said I symbol and Q symbol components of said received QAM symbol point and I plant and Q plant components of said nearest QAM plant point.

17. The apparatus of claim 14 , wherein said means for averaging said instant error power over said plurality of incoming QAM symbols in time domain further includes:

a digital infinite impulse response (IIR) filter.

18. The apparatus of claim 14 , wherein said means for averaging said instant error power over said plurality of incoming QAM symbols in time domain further includes:

a digital exponential averaging filter.

19. The apparatus of claim 14 , wherein said means for averaging said instant error power over said plurality of incoming QAM symbols in time domain further includes:

a digital low pass filter (LPF).

20. The apparatus of claim 14 , wherein said means for averaging said instant error power over said plurality of incoming QAM symbols in time domain further includes:

a digital finite impulse response filter (FIR).

21. The apparatus of claim 14 , wherein said means for averaging said instant error power over said plurality of incoming QAM symbols in time domain further includes:

a moving averaging filter of length N, wherein N is a number of received symbols selected for averaging purposes, N is an integer.

22. The apparatus of claim 14 , wherein said means for averaging said instant error power over said plurality of incoming QAM symbols in time domain further includes:

a decimation moving averaging filter of length N, said decimation averaging filter utilizes a decimated sequence α N,k,R Decimated of received N symbols selected for averaging purposes, wherein said decimated sequence α N,k,R Decimated includes a sequence of N symbols selected in the following order: {“k”-th incoming symbol, “k+R”-th incoming symbol, “k+2R”-th incoming symbol, . . . , “k+(N−1)*R”-th incoming symbol}; wherein “k” is an integer less than N; and wherein a decimation rate “R” is an integer less than N; N is an integer.

23. The apparatus of claim 14 , wherein said means for translating said averaged error power without said DC component from said time domain into said frequency domain in order to determine the frequency spectrum of said averaged error power further includes:

a Fast Fourier Transform (FFT) algorithm.

24. The apparatus of claim 14 , wherein said means for translating said averaged error power without said DC component from said time domain into said frequency domain in order to determine the frequency spectrum of said averaged error power further includes:

a Discrete Fourier Transform (DFT) algorithm.

25. The apparatus of claim 14 , wherein said means for translating said averaged error power without said DC component from said time domain into said frequency domain in order to determine the frequency spectrum of said averaged error power further includes:

a wavelet transform algorithm.

26. The apparatus of claim 14 , wherein said means for selecting the frequency spectrum component of said averaged error power with the maximum amplitude further includes:

a peak detector.

27. The apparatus of claim 14 , wherein said means for selecting the frequency spectrum component of said averaged error power with the maximum amplitude further includes:

a maximum likelihood peak detector.

28. The apparatus of claim 14 , wherein said means for determining said frequency offset further includes:

an algorithm configured to determine said frequency offset by selecting a residual maximum frequency component.