Loss tolerant transmission control protocol

• US 8,537,675 B2
• Filed: 01/04/2011
• Issued: 09/17/2013
• Est. Priority Date: 03/30/2005
• Status: Expired due to Fees

• Current Assignee: AT&T Intellectual Property I LP (AT&T, Inc.), Rensselaer Polytechnic Institute
• Original Assignee: AT&T Intellectual Property I LP (AT&T, Inc.)
• Inventors: Kalyanaraman, Shivkumar, Subramanian, Vijaynarayanan, Tickoo, Omesh, Ramakrishnan, Kadangode K
• Primary Examiner(s): MILLS, DONALD L
• Application Number:
  US12/984,479
• Publication Number:
  US 20110099437A1
• Time in Patent Office: 987 Days
• Field of Search: None
• US Class Current: 370/231
• CPC Class Codes: H04L 1/0007 by modifying the frame length
• CPC Class Codes: H04L 1/0009 by adapting the channel coding H04L1/1812 takes precedence
• CPC Class Codes: H04L 1/1635 Cumulative acknowledgement, i.e. the acknowledgement message applying to all previous messages
• CPC Class Codes: H04L 47/10 Flow control; Congestion control
• CPC Class Codes: H04L 47/135 by jamming the transmission media
• CPC Class Codes: H04L 47/193 at the transport layer, e.g. TCP related
• CPC Class Codes: H04L 47/27 Evaluation or update of window size, e.g. using information derived from acknowledged [ACK] packets
• CPC Class Codes: H04L 47/365 Dynamic adaptation of the packet size

Abstract

A particular device includes a transmitter. The transmitter is adapted to estimate a packet erasure rate for packets of a data window to be transmitted to a receiver. The transmitter is adapted to determine a number of proactive forward error correction (FEC) packets for the data window based on the estimated packet erasure rate. The transmitter is adapted to determine a packet size for the packets in the data window based on a window size of the data window and the determined number of proactive FEC packets. The transmitter is also adapted to transmit the data window to the receiver. The packets in the transmitted data window have a size corresponding to the determined packet size and include the determined number of proactive FEC packets.

Claims

1. A device comprising:

a transmitter configured to;

estimate a packet erasure rate for packets of a data window to be transmitted to a receiver;

determine a number of proactive forward error correction packets for the data window based on the estimated packet erasure rate;

determine a packet size for the packets in the data window based on a window size of the data window and the determined number of proactive forward error correction packets; and

transmit the data window to the receiver, wherein the packets in the transmitted data window have a size corresponding to the determined packet size and include the determined number of proactive forward error correction packets.

2. The device of claim 1, further comprising a second receiver to receive packets from a second transmitter that is external to the device.

3. The device of claim 1, wherein an exponential weighted moving average is used to estimate the packet erasure rate for the packets.

4. The device of claim 1, wherein the determined packet size is decreased after receipt of an explicit congestion notification.

5. The device of claim 1, wherein the transmitter is further configured to:

receive a duplicate acknowledgement;

determine a hole size based on the duplicate acknowledgement; and

send reactive forward error correction packets to the receiver, wherein a number of the reactive forward error correction packets sent to the receiver is based on the hole size and the estimated packet erasure rate.

6. The device of claim 5, wherein the reactive forward error correction packets are sent as a second data window, wherein a window size of the second data window is the same as the window size of the data window.

7. The device of claim 6, wherein zero-segments are used to complete the second data window when the reactive forward error correction packets do not fill the second data window.

8. A method comprising:

setting a size of a block;

determining a number of packets for the block based on a window granularity;

calculating a number of proactive forward error correction packets for the block based on a loss estimate value;

selecting a data window for the block, wherein the number of packets for the block includes the number of proactive forward error correction packets and a number of data packets in the data window;

forming the block that includes proactive forward error correction packets and the data window;

sending the block to a receiver;

receiving at least one acknowledgement from the receiver, wherein the at least one acknowledgement includes an updated loss estimate value;

recalculating the number of proactive forward error correction packets based on the updated loss estimate value and the window granularity;

forming a second block wherein a number of packets for the second block is based on the window granularity, and wherein the number of packets of the second block includes the recalculated number of proactive forward error correction packets and a number of data packets in a second data window; and

sending the second block to the receiver.

9. The method of claim 8, further comprising adjusting the window granularity.

10. The method of claim 8, further comprising increasing the window granularity in response to detection of an explicit congestion notification.

11. The method of claim 10, wherein the number of packets in the block is increased to increase the window granularity.

12. The method of claim 8, further comprising searching a state table to obtain information regarding a state of the data window.

13. The method of claim 12, further comprising providing a number of reactive forward error correction packets based on the state of the data window.

14. The method of claim 8, further comprising operating in a data mode when the loss estimate value is zero and operating in a proactive forward error correction mode when the loss estimate value is non-zero.

15. The method of claim 8, further comprising receiving packets from a second transmitter.

16. A method comprising:

receiving an incoming packet at a receiver from a transmitter, wherein the incoming packet is one of a data packet and a forward error correction packet;

identifying a block associated with the incoming packet at the receiver;

after the identified block is decoded, discarding the incoming packet;

if the identified block is not decoded;

when the incoming packet is a new packet, allocating decoding structures at the receiver;

generating a loss estimate value and a hole size for the block;

sending an acknowledgement of the incoming packet from the receiver to the transmitter, wherein the acknowledgement includes the loss estimate value and the hole size; and

when the incoming packet raises a number of received data packets for the block and received forward error correction packets for the block to a number sufficient to decode the block, decoding the block at the receiver.

17. The method of claim 16, wherein the loss estimate is an exponential weighted moving average.

18. The method of claim 16, wherein the hole size is based on original transmissions and ignores any received retransmitted packets.

19. The method of claim 16, wherein the hole size is based on original transmissions and received retransmitted packets.

20. The method of claim 16, wherein the acknowledgement further includes a hole size for a previous block.