METHODS FOR ACHIEVING TARGET LOSS RATIO

• US 20120201147A1
• Filed: 02/04/2011
• Published: 08/09/2012
• Est. Priority Date: 02/04/2011
• Status: Active Grant

• Current Assignee: Adaptiv Networks Inc.
• Original Assignee: Ipeak Networks Incorporated (Adaptiv Networks Inc.)
• Inventors: Williams, Matthew Robert, Vemulapali, Mohan Krishna
• Granted Patent: US 8,437,370 B2
• Time in Patent Office: Days
• US Class Current: 370/252
• CPC Class Codes: H04L 43/0829 Packet loss
• CPC Class Codes: H04L 43/16 Threshold monitoring
• CPC Class Codes: H04L 47/32 by discarding or delaying data units, e.g. packets or frames
• CPC Class Codes: H04L 47/34 ensuring sequence integrity, e.g. using sequence numbers

Abstract

A method of forwarding data transmissions from a first network to a third network via a second network comprises receiving packets of a first type from the first network, segmenting each packet of a first type into packets of a second type, transmitting the packets of a second type to the second network, and producing and transmitting at least one encoded duplicate of each of the packets of a second type to the second network. The encoded duplicate packet allows a packet of the first type to be recreated in the event that not all the packets of the second type are received at the second network. In the event that a sufficient number of the packets of a second type and the encoded duplicate packets are not received to recreate the packet of a first type, the method determines a loss ratio that represents the number of packets not recreated relative to the number of packets transmitted during a selected time interval, dynamically adjusts the overhead required to produce the encoded duplicate packets to attempt to minimize the loss ratio, detects a significant increase in the determined loss ratio over a plurality of the time intervals, successively reduces the overhead required to produce the encoded duplicate packets in response to the detection of a significant increase in the determined loss ratio, and determines whether the reduced overhead requirement stabilizes the determined loss ratio and, if the answer is affirmative, terminates the successive reduction of the overhead requirement.

Claims

1. A method of forwarding data transmissions from a first network to a third network via a second network, the method comprising

receiving packets of a first type from said first network,

segmenting each packet of a first type into packets of a second type,

transmitting said packets of a second type to said second network,

producing and transmitting at least one encoded duplicate of each of said packets of a second type to said second network, wherein said encoded duplicate packet allows a packet of said first type to be recreated in the event not all said packets of said second type are received at said second network,

in the event a sufficient number of said packets of a second type and said encoded duplicate packets are not received to recreate said packet of a first type,

determining a loss ratio that represents the number of packets not recreated relative to the number of packets transmitted during a selected time interval,

dynamically adjusting the overhead required to produce said encoded duplicate packets to attempt to minimize said loss ratio,

detecting a significant increase in the determined loss ratio over a plurality of said time intervals,

successively reducing the overhead required to produce said encoded duplicate packets in response to the detection of a significant increase in said determined loss ratio, and

determining whether the reduced overhead requirement stabilizes said determined loss ratio and, if the answer is affirmative, terminating the successive reduction of said overhead requirement.

2. The method of claim 1 in which the overhead required to produce said encoded duplicate packets is reduced by adjusting an encoding level in the production of said encoded duplicate packets in response to the detection of a significant increase in said determined loss ratio over said plurality of time intervals.

3. The method of claim 2 in which said encoding level is decreased sufficiently to stop increases in said loss ratio.

4. The method of claim 1 in which said significant increase in the determined loss ratio over a plurality of said time intervals is detected by determining the change in said loss ratio over said time intervals, and comparing said change with a preselected threshold.

5. The method of claim 1 in which said significant increase in the determined loss ratio over a plurality of said time intervals is detected by determining the change in an average of said loss ratio over said time intervals, and comparing said change with a preselected threshold.

6. A system for forwarding data transmissions from a first network to a third network via a second network, the system comprising

a first interface module for sending packets to and receiving packets from said first network;

a second interface module for sending packets to and receiving packets from said second network;

a segmenting module for segmenting each packet of a first type into packets of a second type, and producing and transmitting at least one encoded duplicate of each of said packets of a second type to said second network, wherein said encoded duplicate packet allows a packet of said first type to be recreated in the event not all said packets of said second type are received at said second network, and

a reassembly module for determining whether a sufficient number of said packets of a second type and said encoded duplicate packets are received to recreate said packet of a first type, and in the event that said sufficient number of said packets of a second type and said encoded duplicate packets are notreceived to recreate said packet of a first type,

determining a loss ratio that represents the number of packets not recreated relative to the number of packets transmitted during a selected time interval,

dynamically adjusting the overhead required to produce said encoded duplicate packets to attempt to minimize said loss ratio,

detecting a significant increase in the determined loss ratio over a plurality of said time intervals,

successively reducing the overhead required to produce said encoded duplicate packets in response to the detection of a significant increase in said determined loss ratio, and

determining whether the reduced overhead requirement stabilizes said determined loss ratio and, if the answer is affirmative, terminating the successive reduction of said overhead requirement.

7. The system of claim 6 in which the overhead required to produce said encoded duplicate packets is reduced by adjusting an encoding level in the production of said encoded duplicate packets in response to the detection of a significant increase in said determined loss ratio over said plurality of time intervals.

8. The system of claim 7 in which said encoding level is decreased sufficiently to stop increases in said loss ratio.

9. The system of claim 6 in which said significant increase in the determined loss ratio over a plurality of said time intervals is detected by determining the change in said loss ratio over said time intervals, and comparing said change with a preselected threshold.

10. The system of claim 6 in which said significant increase in the determined loss ratio over a plurality of said time intervals is detected by determining the change in an average of said loss ratio over said time intervals, and comparing said change with a preselected threshold.

11. A computer program product, comprising one or more non-transitory tangible media having a computer readable program logic embodied therein, the computer readable program logic configured to be executed to implement a method of forwarding data transmissions from a first network to a third network via a second network, the method comprising

receiving packets of a first type from said first network,

segmenting each packet of a first type into packets of a second type,

transmitting said packets of a second type to said second network,

producing and transmitting at least one encoded duplicate of each of said packets of a second type to said second network, wherein said encoded duplicate packet allows a packet of said first type to be recreated in the event not all said packets of said second type are received at said second network,

in the event a sufficient number of said packets of a second type and said encoded duplicate packets are not received to recreate said packet of a first type,

determining a loss ratio that represents the number of packets not recreated relative to the number of packets transmitted during a selected time interval,

dynamically adjusting the overhead required to produce said encoded duplicate packets to attempt to minimize said loss ratio,

detecting a significant increase in the determined loss ratio over a plurality of said time intervals,

successively reducing the overhead required to produce said encoded duplicate packets in response to the detection of a significant increase in said determined loss ratio, and

determining whether the reduced overhead requirement stabilizes said determined loss ratio and, if the answer is affirmative, terminating the successive reduction of said overhead requirement.

12. The computer program product of claim 11 in which the overhead required to produce said encoded duplicate packets is reduced by adjusting an encoding level in the production of said encoded duplicate packets in response to the detection of a significant increase in said determined loss ratio over said plurality of time intervals.

13. The computer program product of claim 12 in which said encoding level is decreased sufficiently to stop increases in said loss ratio.

14. The computer program product of claim 11 in which said significant increase in the determined loss ratio over a plurality of said time intervals is detected by determining the change in said loss ratio over said time intervals, and comparing said change with a preselected threshold.

15. The computer program product of claim 11 in which said significant increase in the determined loss ratio over a plurality of said time intervals is detected by determining the change in an average of said loss ratio over said time intervals, and comparing said change with a preselected threshold.