Low-latency channel equalization using a secondary channel
An equalization method has been developed for low latency, low bandwidth wireless communication channel environments. With this method, an exact copy, nearly exact copy, or some facsimile of a message (or associated information), which was transmitted via a low latency, low bandwidth wireless communication channel, is also sent via a backend communication channel such as a fiber optic network. Equalization is generally performed by comparing the originally received message to the copy sent via the backend channel. The original message can incorporate an added channel delay to compensate for the time delay between the primary wireless channel and the backend channel.
1. A method, comprising:
receiving a signal over a primary communication channel, wherein the primary communication channel includes a high frequency radio channel;
receiving at least a representative copy of the signal over a backend communication channel, wherein the backend communication channel has a greater latency and bandwidth than the primary communication channel;
determining a timing performance for the primary communication channel, wherein the timing performance includes a minimum time delay;
determining the minimum time delay at least based on financial trading timing data for a financial market;
determining a maximum delay time advantage limit, wherein the maximum delay time advantage limit is where being faster minimally impacts a rate of return in the financial market;
equalizing the primary communication channel with a first equalizer by comparing the signal with the representative copy, wherein the equalizing includes selecting a modulating and coding (MODCOD) scheme where an equalizer processing time for the equalizer is within the minimum time delay and the maximum delay time advantage limit;
establishing a best case performance of the first equalizer by delaying the signal by a delay during the equalizing;
converging a second equalizer with the first equalizer, wherein the converging occurs in response to the second equalizer at least satisfying the timing performance;
converging tap values of the first equalizer and the second equalizer;
combining the tap values of the first equalizer and the second equalizer to create a metric, wherein the metric includes an estimated inter-symbol interference (ISI) and signal to noise ratio (SNR);
selecting a modulation and coding method based on the metric; and
wherein the modulating and coding scheme is selected based on the estimated ISI and SNR.
2. The method of claim 1 , wherein the second equalizer is identical to the first equalizer.
3. The method of claim 2 , wherein the converging occurs in response to the signal not being additionally delayed.
4. The method of claim 1 , wherein the primary communication channel includes a low bandwidth, low latency communication link.
5. The method of claim 1 , wherein the backend communication channel includes a fiber optic network.
6. The method of claim 1 , wherein the representative copy includes an exact copy of the signal.
7. The method of claim 1 , wherein the representative copy includes characteristic information of the signal.
8. The method of claim 1 , further comprising:
determining the minimum time delay at least based on a nature of the primary channel.
9. The method of claim 8 , further comprising:
changing to a higher modulation format when the nature of the primary channel is benign.
10. The method of claim 8 , further comprising:
determining the minimum time delay based on jitter of a financial market.
11. The method of claim 10 , wherein the financial market includes a high frequency trading system.
12. The method of claim 1 , further comprising:
indexing the modulation and coding method based on the metric.
13. The method of claim 1 , further comprising:
reducing a time domain for the first equalizer by evaluating an effect of removing feed forward delay stages.
14. A method, comprising:
receiving a signal over a primary communication channel, wherein the primary communication channel includes a high frequency radio channel;
receiving at least a representative copy of the signal over a backend communication channel, wherein the backend communication channel has a greater latency and bandwidth than the the primary communication channel;
determining a timing performance for the primary communication channel, wherein the timing performance includes a minimum time delay;
determining the minimum time delay at least based on financial trading timing data for a financial market;
reducing a time domain for an equalizer by evaluating an effect of removing feed forward taps, wherein the reducing includes normalizing all taps in the equalizer where a sum of squares of tap weights for all of the taps is equal to one, wherein the reducing includes creating a metric by subtracting squares of magnitudes of the feed forward taps being removed from the sum of squares; and
selecting a modulating and coding (MODCOD) scheme based on the metric.
15. The method of claim 14 , wherein the backend communication channel extends through an ocean.
16. The method of claim 14 , wherein the determining the minimum time delay based on jitter of the financial market.
17. The method of claim 14 , wherein the financial market includes a high frequency trading system.
18. A method, comprising:
receiving a signal over a primary communication channel, wherein the primary communication channel includes a high frequency radio channel;
receiving at least a representative copy of the signal over a backend communication channel;
determining a timing performance for the primary communication channel, wherein the timing performance includes a minimum time delay;
determining the minimum time delay based on jitter of a financial market;
determining a maximum delay time advantage limit, wherein the maximum delay time advantage limit is where being faster minimally impacts a rate of return in the financial market;
equalizing the primary communication channel with a first equalizer by comparing the signal with the representative copy, wherein the equalizing includes selecting a modulating and coding (MODCOD) scheme where an equalizer processing time for the equalizer is within the minimum time delay and the maximum delay time advantage limit;
establishing a best case performance of the first equalizer by delaying the signal by a delay during the equalizing;
converging a second equalizer with the first equalizer, wherein the converging occurs in response to the second equalizer at least satisfying the timing performance;
converging tap values of the first equalizer and the second equalizer;
combining the tap values of the first equalizer and the second equalizer to create a metric, wherein the metric includes an estimated inter-symbol interference (ISI) and signal to noise ratio (SNR);
selecting a modulation and coding method based on the metric; and
wherein the modulating and coding scheme is selected based on the estimated ISI and SNR.
19. The method of claim 18 , wherein the backend communication channel includes a fiber optic network.
20. The method of claim 18 , further comprising:
determining the minimum time delay at least based on a nature of the primary channel.
21. The method of claim 20 , further comprising:
changing to a higher modulation format when the nature of the primary channel is benign.
22. A method, comprising:
receiving a signal over a primary communication channel, wherein the primary communication channel includes a high frequency radio channel;
receiving at least a representative copy of the signal over a backend communication channel;
determining a timing performance for the primary communication channel, wherein the timing performance includes a minimum time delay;
determining the minimum time delay based on jitter of a financial market;
reducing a time domain for an equalizer by evaluating an effect of removing feed forward taps, wherein the reducing includes normalizing all taps in the equalizer where a sum of squares of tap weights for all of the taps is equal to one, wherein the reducing includes creating a metric by subtracting squares of magnitudes of the feed forward taps being removed from the sum of squares; and
selecting a modulating and coding (MODCOD) scheme based on the metric.
23. The method of claim 22 , wherein the backend communication channel includes a fiber optic network.