Method and apparatus to track gain variation in orthogonal frequency division multiplexing (OFDM) systems
Gain variations during a packet can lead to significant performance degradation in communications systems that use high order quadrature amplitude modulation (QAM). A method and the associated apparatus track such variations in an OFDM system and completely eliminate any performance degradation. Gain estimation and compensation is employed with the use of pilot subcarriers in the payload of an OFDM data packet. Estimated pilot magnitude ratios are averaged, throughout the processing life of a packet, to yield accurate gain estimations. A gain compensation factor is used to adjust data carriers. An exclusion method is also employed to eliminate pilot carriers which contribute to noise.
1. An orthogonal frequency division multiplexing (OFDM) system comprising:
an OFDM transmitter operable to transmit a signal comprising a data subcarrier k within a symbol m of a packet, the symbol m being one of a plurality of symbols within the packet and the data subcarrier k being associated with a pilot indexed i of a plurality of pilots within the symbol m; and
an OFDM receiver operable to determine a pilot magnitude ratio estimate for the pilot i of the symbol m by normalizing the pilot i by an absolute value of a lead pilot within the symbol m; to determine an estimate pilot magnitude for the pilot i by multiplying the pilot magnitude ratio estimate for the pilot i of the symbol m by a sum of an absolute value of a plurality of pilots in the symbol m, divided by a sum of pilot magnitude ratio estimates for the plurality of pilots in the symbol m; and to determine a gain compensation factor to be applied to the subcarrier k associated with the pilot i of the symbol m using the estimate pilot magnitude for the pilot i of the symbol m.
2. The system of claim 1 , wherein, to determine the pilot magnitude ratio estimate for the pilot i of the symbol m, the OFDM receiver is operable to average one or more previously calculated pilot magnitude ratio estimates for similarly indexed pilots in previous symbols and the normalized magnitude of the pilot i.
3. The system of claim 1 , wherein, to determine the gain compensation factor to be applied to the subcarrier k associated with the pilot i of the symbol m, the OFDM receiver is operable to divide an absolute value of a pilot indexed i for the symbol m at a time zero by the estimate pilot magnitude for the pilot i of the symbol m.
4. The system according to claim 1 , wherein the OFDM receiver is further operable to determine the plurality of pilots in the symbol m to use to determine the estimate pilot magnitude for the pilot i by excluding pilots whose absolute magnitude has been substantially reduced by gain variations.
5. The system according to claim 1 , wherein the OFDM receiver is further operable to determine the plurality of pilots in the symbol m to use to determine the estimate pilot magnitude for the pilot i by ordering pilots according to their absolute magnitude values in the training symbol and including pilots based on descending order of their strength.
6. The system according to claim 1 , wherein the OFDM receiver is further operable to determine the plurality of pilots in the symbol m to use to determine the estimate pilot magnitude for the pilot i by including a pilot indexed j if the pilot indexed j−1 is included and if the absolute value of the pilot indexed j is greater than a sum of the absolute value of all pilots indexed 0 to j−1 multiplied by X−1, X being a square root of j divided by a squared sum of the absolute value of all pilots indexed 0 to j−1.
7. The system according to claim 1 , wherein the OFDM receiver is further operable to determine the plurality of pilots in the symbol m to use to determine the estimate pilot magnitude for the pilot i by multiplying pilots by a threshold t that is less than one and including pilots whose absolute magnitude is greater than the lead pilot, upon multiplication by the threshold t.
8. An orthogonal frequency division multiplexing (OFDM) receiver comprising:
a Fast Fourier Transform (FFT) element operable to produce a frequency representation of an OFDM signal, the OFDM signal comprising a data subcarrier k within a symbol m of a packet, the symbol m being one of a plurality of symbols within the packet and the data subcarrier k being associated with a pilot indexed i of a plurality of pilots within the symbol m;
a gain estimation and compensation element operable to determine a pilot magnitude ratio estimate for the pilot i of the symbol m by normalizing the pilot i by an absolute value of a lead pilot within the symbol m; to determine an estimate pilot magnitude for the pilot i by multiplying the pilot magnitude ratio estimate for the pilot i of the symbol m by a sum of an absolute value of a plurality of pilots in the symbol m, divided by a sum of pilot magnitude ratio estimates for the plurality of pilots in the symbol m; and to determine a gain compensation factor to be applied to the subcarrier k associated with the pilot i of the symbol m using the estimate pilot magnitude for the pilot i of the symbol m.
9. The OFDM receiver of claim 8 , wherein, to determine the pilot magnitude ratio estimate for the pilot i of the symbol m, the gain estimation and compensation element is operable to average one or more previously calculated pilot magnitude ratio estimates for similarly indexed pilots in previous symbols and the normalized magnitude of the pilot i.
10. The OFDM receiver of claim 8 , wherein, to determine the gain compensation factor to be applied to the subcarrier k associated with the pilot i of the symbol m, the gain estimation and compensation element is operable to divide an absolute value of a pilot indexed i for the symbol m at a time zero by the estimate pilot magnitude for the pilot i of the symbol m.
11. The OFDM receiver according to claim 8 , wherein the gain estimation and compensation element is further operable to determine the plurality of pilots in the symbol m to use to determine the estimate pilot magnitude for the pilot i by excluding pilots whose absolute magnitude has been substantially reduced by gain variations.
12. The OFDM receiver according to claim 8 , wherein the gain estimation and compensation element is further operable to determine the plurality of pilots in the symbol m to use to determine the estimate pilot magnitude for the pilot i by ordering pilots according to their absolute magnitude values in the training symbol and including pilots based on descending order of their strength.
13. The OFDM receiver according to claim 8 , wherein the gain estimation and compensation element is further operable to determine the plurality of pilots in the symbol m to use to determine the estimate pilot magnitude for the pilot i by including a pilot indexed j if the pilot indexed j−1 is included and if the absolute value of the pilot indexed j is greater than a sum of the absolute value of all pilots indexed 0 to j−1 multiplied by X−1, X being a square root of j divided by a squared sum of the absolute value of all pilots indexed 0 to j−1.
14. The OFDM receiver according to claim 8 , wherein the gain estimation and compensation element is further operable to determine the plurality of pilots in the symbol m to use to determine the estimate pilot magnitude for the pilot i by multiplying pilots by a threshold t that is less than one and including pilots whose absolute magnitude is greater than the lead pilot, upon multiplication by the threshold t.
15. A method for determining a mid-packet gain compensation to apply to a data subcarrier k within a symbol m of a packet in an orthogonal frequency division multiplexing (OFDM) system, the symbol m being one of a plurality of symbols within the packet and the data subcarrier k being associated with a pilot indexed i of a plurality of pilots within the symbol m, the method comprising:
determining a pilot magnitude ratio estimate for the pilot i of the symbol m by averaging one or more previously calculated pilot magnitude ratio estimates for similarly indexed pilots in previous symbols and a normalized magnitude of the pilot i;
determining an estimate pilot magnitude for the pilot i by multiplying the pilot magnitude ratio estimate for the pilot i of the symbol m by a sum of an absolute value of a plurality of pilots in the symbol m, divided by a sum of pilot magnitude ratio estimates for the plurality of pilots in the symbol m; and
determining a gain compensation factor to be applied to the subcarrier k associated with the pilot i of the symbol m using the estimate pilot magnitude for the pilot i of the symbol m.
16. The method according to claim 15 further comprising determining the plurality of pilots in the symbol m to use in the determining of the estimate pilot magnitude for the pilot i by excluding pilots whose absolute magnitude has been substantially reduced by gain variations.
17. The method according to claim 15 further comprising determining the plurality of pilots in the symbol m to use in the determining of the estimate pilot magnitude for the pilot i by ordering pilots according to their absolute magnitude values in the training symbol and including pilots based on descending order of their strength.
18. The method according to claim 15 further comprising determining the plurality of pilots in the symbol m to use in the determining of the estimate pilot magnitude for the pilot i by including a pilot indexed j if the pilot indexed j−1 is included and if the absolute value of the pilot indexed j is greater than a sum of the absolute value of all pilots indexed 0 to j−1 multiplied by X−1, X being a square root of j divided by a squared sum of the absolute value of all pilots indexed 0 to j−1.
19. The method according to claim 15 further comprising determining the plurality of pilots in the symbol m to use in the determining of the estimate pilot magnitude for the pilot i by multiplying pilots by a threshold t that is less than one and including pilots whose absolute magnitude is greater than the lead pilot, upon multiplication by the threshold t.
20. The method of claim 15 , wherein the determining the gain compensation factor to be applied to the subcarrier k associated with the pilot i of the symbol m comprises: dividing an absolute value of a pilot indexed i for the symbol m at a time zero by the estimate pilot magnitude for the pilot i of the symbol m.