Universal rate control mechanism with parameter adaptation for real-time communication applications

• US 9,088,510 B2
• Filed: 12/18/2012
• Issued: 07/21/2015
• Est. Priority Date: 12/17/2010
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Inventors: Li, Jin, Mehrotra, Sanjeev, Xie, Di
• Primary Examiner(s): Nguyen, Minh-Trang
• Application Number:
  US13/718,114
• Publication Number:
  US 20130128735A1
• Time in Patent Office: 945 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: H04L 1/0002 by adapting the transmission rate
• CPC Class Codes: H04L 1/0015 characterised by the adaptation strategy
• CPC Class Codes: H04L 1/0034 where the transmitter decides based on inferences, e.g. use of implicit signalling
• CPC Class Codes: H04L 43/0882 Utilisation of link capacity
• CPC Class Codes: H04L 47/12 Avoiding congestion; Recovering from congestion
• CPC Class Codes: H04L 47/193 at the transport layer, e.g. TCP related
• CPC Class Codes: H04L 47/20 Traffic policing
• CPC Class Codes: H04L 47/22 Traffic shaping
• CPC Class Codes: H04L 47/2408 for supporting different services, e.g. a differentiated services [DiffServ] type of service
• CPC Class Codes: H04L 47/2416 Real-time traffic
• CPC Class Codes: H04L 47/2441 relying on flow classification, e.g. using integrated services [IntServ]
• CPC Class Codes: H04L 47/25 with rate being modified by the source upon detecting a change of network conditions
• CPC Class Codes: H04L 47/263 Rate modification at the source after receiving feedback
• CPC Class Codes: H04L 47/34 ensuring sequence integrity, e.g. using sequence numbers

Abstract

A “Universal Rate Control Mechanism with Parameter Adaptation” (URCMPA) improves real-time communication (RTC) sessions in terms of delay, loss, throughput, and PSNR. The URCMPA automatically learns network characteristics including bottleneck link capacity, inherent queuing delay, inherent packet loss rates, etc., during RTC sessions. The URCMPA uses this information to dynamically adapt rate control parameters in a utility maximization (UM) framework. The URCMPA operates reliable RTC sessions across a wide range and combination of networks near full throughput rates while maintaining low operating congestion levels (e.g., low queuing delay and low packet loss). Examples of networks applicable for use with the URCMPA include, but are not limited to, combinations of mobile broadband (e.g., 3G, 4G, etc.), WiMAX, Wi-Fi hotspots, etc., and physical networks based on cable, fiber, ADSL, etc. The URCMPA can also dynamically adapt operating congestion levels relative to competing TCP flows to maintain fair use of network resources.

Claims

1. A computer-implemented process for dynamically adapting rate control parameters of a communications session, comprising using a computer to perform process actions for:

instantiating a communications session using a Utility Maximization (UM) framework;

wherein the UM framework includes a first rate control parameter for adjusting an operating congestion level of the communications session;

while the communication session is ongoing, repeatedly performing process actions for;

based on a current sending rate of the communications session, inferring current congestion levels of the communications session from one or more congestion signals derived from any of a current packet'"'"'s congestion signals and a first sliding window of the congestion signals of the communications session,

based on the current sending rate of the communications session, inferring longer-term congestion levels of the communications session from a second sliding window of the congestion signals of the communications session;

dynamically adapting the first rate control parameter as a function of the longer-term inferred congestion levels and a current long-term average sending rate, and

dynamically adjusting the current sending rate as a function of the dynamically adapted first rate control parameter and the inferred current congestion level.

2. The computer-implemented process of claim 1 wherein the UM framework includes a second rate control parameter for controlling a rate of convergence towards a steady-state sending rate of the communications session relative to the current sending rate of the communications session.

3. The computer-implemented process of claim 2 wherein dynamically adjusting the current sending rate as a function of the dynamically adapted first rate control parameter further includes dynamically adjusting the current sending rate as a function the second rate control parameter.

4. The computer-implemented process of claim 2 wherein the second rate control parameter is dynamically adapted while the communication session is ongoing as a function of the inferred current congestion levels.

5. The computer-implemented process of claim 1 wherein the congestion signals include one or more of estimated queuing delay, and estimated packet loss rates.

6. The computer-implemented process of claim 5 wherein the congestion signals further include Explicit Congestion Notification (ECN) marking signals.

7. The computer-implemented process of claim 1 wherein the dynamic update of the first parameter is also a function of an estimated maximum rate capacity of a bottleneck link and minimum congestion signals levels seen when an aggregate rate into the bottleneck link is below the capacity.

8. A system for adapting rate control parameters of a Utility Maximization (UM) based communications session, comprising using a computing device for:

initiating a UM-based communication session using an initial estimate of a bottleneck link rate capacity and an initial estimate of a minimum congestion level when an aggregate rate into the bottleneck link is below the bottleneck link rate capacity;

wherein the UM-based communication session includes a first rate control parameter for adjusting an operating congestion level of the UM-based communications session;

evaluating one or more congestion signals obtained from one or more sequential packets on an ongoing basis throughout the UM-based communication session to infer a current congestion level for those packets;

evaluating one or more congestion signals obtained within sequential instances of a second sliding window throughout the UM-based communication session to infer a long-term value for the congestion level;

for each inferred current congestion level, adapting the first rate control parameter based on the inferred long-term congestion level obtained using the second sliding window and a long-term average sending rate; and

using each adaptation of the first rate control parameter to perform a corresponding adjustment to the current sending rate of the UM-based communication session.

9. The system of claim 8 wherein the one or more sequential packets are repeatedly obtained on an ongoing basis throughout the UM-based communication session using any of per-packet selection and selection of multiple sequential packets via sequential instances of a first sliding window.

10. The system of claim 8 further comprising a second rate control parameter for controlling a rate of convergence of a current sending rate of the UM-based communication session.

11. The system of claim 10 wherein each corresponding adjustment to the current sending rate is performed as a joint function of the first rate control parameter and the second rate control parameter.

12. The system of claim 10 wherein, for each sliding window, the second rate control parameter is adapted based on the inferred current congestion level of each sliding window.

13. The system of claim 8 wherein the congestion signals include one or more of estimated queuing delay, estimated packet loss rates, and Explicit Congestion Notification (ECN) marking signals.

14. A computer-readable storage media having computer executable instructions stored therein for adapting rate control parameters of a Utility Maximization (UM) based communications session, said instructions causing a computing device to execute a method comprising:

instantiating a communications session using a Utility Maximization (UM) framework;

wherein the UM framework includes a first rate control parameter for adjusting an operating congestion level of the communications session;

while the communication session is ongoing, repeatedly performing process actions for;

based on a current sending rate of the communications session, inferring current congestion levels of the communications session from one or more congestion signals derived from any of a current packet'"'"'s congestion signals and a first sliding window of the congestion signals of the communications session,

based on a current sending rate of the communications session, inferring longer-term congestion levels of the communications session from a second sliding window of the congestion signals of the communications session;

dynamically adapting the first rate control parameter as a function of the inferred congestion levels, and a long-term average sending rate; and

,

dynamically adjusting the current sending rate as a function of the dynamically adapted first rate control parameter and the inferred current congestion level.

15. The computer-readable storage media of claim 14 wherein the UM framework includes a second rate control parameter for controlling a rate of convergence towards a maximum sending rate of the communications session relative to the current sending rate of the communications session.

16. The computer-readable medium of claim 15 wherein dynamically adjusting the current sending rate as a function of the dynamically adapted first rate control parameter further includes dynamically adjusting the current sending rate as a function the second rate control parameter.

17. The computer-readable medium of claim 15 wherein the second rate control parameter is dynamically adapted while the communication session is ongoing as a function of the inferred current congestion levels.

18. The computer-readable storage media of claim 14 wherein the congestion signals include one or more of estimated queuing delay and estimated packet loss rates.

19. The computer-readable storage media of claim 18 wherein the congestion signals further include Explicit Congestion Notification (ECN) marking signals.

20. The computer-readable storage media of claim 14 wherein the dynamic update of the first parameter is also a function of an estimated maximum rate capacity of the bottleneck link and minimum congestion signals levels seen when an aggregate rate into the bottleneck link is below the capacity.