Apparatus for free-space optical communications at high data rates

1. A method, comprising:

encoding, by a computing device comprising at least one processor, one or more data frames with a data-link layer forward error correction (FEC) code to produce a plurality of encoded data frames, wherein parameters of the data-link layer FEC code are based on a failure probability of a data-link layer FEC decoder at a receiver (RX) and wherein the parameters and a size of the one or more data frames are selected to meet a performance level;

transmitting, by the computing device, the plurality of encoded data frames from a transmitter (TX) to the RX at least partially through air or space using a plurality of optical beams, wherein the plurality of encoded data frames is distributed using a round-robin distribution over the plurality of optical beams;

receiving, by the computing device, a plurality of encoded data frames at the RX;

identifying, by the computing device, at least one corrupted received encoded data frame and one or more non-corrupted received encoded data frames at the RX; and

reconstructing, by the computing device, the at least one corrupted received encoded data frame using the data-link layer FEC decoder, wherein the data-link layer FEC decoder operates over the non-corrupted received encoded data frames.

2. The method of claim 1 , further comprising:

encoding the plurality of encoded data frames into codewords using a physical layer FEC encoder at the TX; and

decoding a received plurality of codewords into the plurality of encoded data frames received at the RX using a physical layer FEC decoder.

3. The method of claim 2 , further comprising:

interleaving symbols of the codewords at the TX;

modulating interleaved codeword symbols at the TX;

demodulating received codeword symbols at the RX; and

deinterleaving received demodulated codeword symbols at the RX.

4. The method of claim 1 wherein the data-link layer FEC code is an erasure code.

5. The method of claim 4 wherein the erasure code is a fountain code.

6. The method of claim 5 wherein the fountain code is a raptor code.

7. The method of claim 1 , wherein the plurality of encoded data frames are transmitted using intensity modulation techniques.

8. The method of claim 1 wherein the plurality of encoded data frames are transmitted using coherent modulation techniques.

9. The method of claim 1 , further comprising:

declaring that reconstructing the at least one corrupted data frame was unsuccessful; and

requesting retransmission of the corrupted data frame.

10. The method of claim 1 , further comprising a distributor configured to direct individual encoded data frames to individual channels at the TX.

11. The method of claim 1 , further comprising a selector configured to receive encoded data frames from individual channels and provide them to the data-link layer FEC decoder at the RX.

12. The method of claim 1 wherein the plurality of optical beams have distinct spectra.

13. The method of claim 1 wherein the plurality of optical beams are not coincident at the TX.

14. A system, comprising:

a transmitter (TX) having a data-link layer forward error correction (FEC) encoder configured to encode one or more data frames to produce a plurality of encoded data frames using a data-link layer FEC code, wherein parameters of the data-link layer FEC code are based on a failure probability of a data-link layer FEC decoder at a receiver (RX) and wherein the parameters and a size of the one or more data frames are selected to meet a performance level;

a plurality of apertures configured to emit a plurality of optical beams at different wavelengths to transmit the plurality of encoded data frames at least partially through air, wherein the plurality of encoded data frames is distributed using a round-robin distribution over the plurality of optical beams; and

the RX configured to:

receive the plurality of encoded data frames, wherein each different subset of the plurality of encoded data frames is received from each of the plurality of optical beams;

identify at least one corrupted received encoded data frame and one or more non-corrupted received encoded data frames; and

reconstruct the at least one corrupted received encoded data frame using a data-link layer FEC decoder, wherein the data-link layer FEC decoder operates over the non-corrupted received encoded data frames.

15. The system of claim 14 , wherein the data-link layer FEC code is an erasure code.

16. The system of claim 14 , wherein the TX is operable to transmit the plurality of encoded data frames using coherent modulation techniques.

17. The system of claim 14 , wherein the TX is operable to transmit the plurality of encoded data frames using intensity modulation techniques.

18. A method, comprising:

encoding one or more data frames with a data-link layer forward error correction (FEC) code to produce a plurality of encoded data frames, wherein parameters of the data-link layer FEC code are based on a failure probability of a data-link layer FEC decoder;

transmitting the plurality of encoded data frames at least partially through air or space using a plurality of optical beams, wherein the plurality of encoded data frames is distributed using a round-robin distribution over the plurality of optical beams; and

receiving a request for retransmission of a corrupted encoded data frame from the plurality of encoded data frames, wherein the request for retransmission is responsive to:

an identification of the corrupted encoded data frame and one or more non-corrupted encoded data frames from the plurality of encoded data frames; and

an unsuccessful reconstruction of the corrupted encoded data frame using the data-link layer FEC decoder, wherein the data-link layer FEC decoder operates over the non-corrupted encoded data frames.

19. The method of claim 18 , wherein the data-link layer FEC code is an erasure code.

20. A method, comprising:

receiving a plurality of encoded data frames, wherein the plurality of encoded data frames were produced by encoding one or more data frames with a data-link layer forward error correction (FEC) code, and wherein the plurality of encoded data frames are received via one or more apertures for receiving a plurality of optical beams having designated wavelengths, wherein the plurality of encoded data frames distributed using a round-robin distribution over the plurality of optical beams is received from each of the plurality of optical beams, wherein parameters of the data-link layer FEC code are based on a failure probability of a data-link layer FEC decoder and wherein the parameters and a size of the one or more data frames are selected to meet a performance level;

identifying at least one corrupted received encoded data frame and one or more non-corrupted received encoded data frames; and

reconstructing the at least one corrupted received encoded data frame using the data-link layer FEC decoder, wherein the data-link layer FEC decoder operates over the non-corrupted received encoded data frames.