FLOW-RATE ADAPTATION FOR A CONNECTION OF TIME-VARYING CAPACITY

• US 20100023634A1
• Filed: 07/28/2008
• Published: 01/28/2010
• Est. Priority Date: 07/28/2008
• Status: Active Grant

• Current Assignee: Masimo Corporation
• Original Assignee: Vantrix Corporation
• Inventors: Cote, Sebastien, Lefebvre, Yves, LABONTE, Francis Roger
• Granted Patent: US 8,001,260 B2
• Time in Patent Office: Days
• US Class Current: 709/231
• CPC Class Codes: H04L 47/10 Flow control; Congestion control
• CPC Class Codes: H04L 47/38 by adapting coding or compression rate
• CPC Class Codes: H04L 65/613 for the control of the source by the destination control signals issued by the client directed to the server or network components specially adapted for selective content distribution H04N21/637
• CPC Class Codes: H04L 65/70 Media network packetisation
• CPC Class Codes: H04L 65/752 adapting media to network capabilities
• CPC Class Codes: H04L 65/80 Responding to QoS

Abstract

A system and methods for adapting streaming data for transmission over a connection of time-varying capacity are disclosed. A streaming server individually adapts transmission rates of signals directed to subtending clients according to measurements characterizing connections from the streaming server to the clients. The measurements may relate to characteristics such as transfer delay, data-loss fraction, and occupancy level of a buffer at a client'"'"'s receiver. A flow controller associated with the streaming server derives metrics from measurements taken over selected time windows to determine a permissible transmission rate from the server to each active client. Metrics related to a specific characteristic may include a mean value over a moving window as well as short and long term tendencies of respective measurements. An adaptable encoder at the streaming server encodes signals to meet permissible transmission rates.

Claims

1. A method of data streaming from an encoder to a decoder through a time-varying connection, the method comprising steps of:

acquiring measurements characterizing said connection;

estimating transmittance variation of said connection according to said measurements;

determining an adjustment of a current encoding rate of said encoder compatible with the transmittance variation to attain a favorable encoding rate; and

instructing said encoder to encode a signal according to said favorable encoding rate.

2. The method of claim 1 wherein said measurements comprise at least one of:

transfer delay between said encoder and said decoder;

data-loss fraction; and

occupancy of a receiving buffer at said decoder.

3. The method of claim 1 wherein said measurements are acquired over a time window of a predefined duration.

4. The method of claim 1 wherein said favorable encoding rate is initialized as a nominal encoding rate.

5. The method of claim 3 wherein said favorable encoding rate is determined by multiplying said nominal encoding rate by a first encoding coefficient, said first encoding coefficient determined according to a first metric derived from said measurements.

6. The method of claim 5 wherein said first encoding coefficient is determined according to a predefined function of said first metric.

7. The method of claim 6 wherein said first metric is a mean value of said measurements over said time window.

8. The method of claim 3 wherein said favorable encoding rate is determined by multiplying said nominal encoding rate by a second encoding coefficient, said second encoding coefficient determined according to a second metric derived from measurement trend during said time window.

9. The method of claim 8 wherein said measurement trend is a slope of a regression line relating said measurements to time during said time window.

10. The method of claim 8 wherein said measurement trend is based on a slope of a regression line relating said measurements to time during said time window and a gradient of measurements during a short period within said time window, said short period selected to be substantially smaller than said time interval.

11. The method of claim 1 wherein said measurements are acquired by sending real-time-transport-protocol (RTP) packets from said encoder to said decoder along said time-varying connection and sending real-time-transport-control-protocol (RTCP) packets from said decoder to said encoder through a control channel.

21. A computer readable medium, comprising instructions stored thereon, when executed by a processor, to perform the steps of the method of claim 1.

12. A method of determining an adaptable encoding rate of a signal transmitted from a data-streaming server to a client device through a time-varying connection, the method comprising steps of:

determining a current encoding rate for said connection;

acquiring transfer-delay measurements over a time window between a first time instant and a second time instant;

acquiring a data-loss-ratio measurement over said time window;

determining a slope of a regression line relating said transfer-delay measurements to respective time instants within said time window;

determining a gradient of selected transfer-delay measurements immediately preceding said second time instant;

determining a first tentative encoding rate according to said slope and said gradient;

determining a second tentative encoding rate according to said data-loss-ratio measurement; and

selecting the lesser of said first tentative encoding rate and said second tentative encoding rate as a preferred encoding rate.

13. The method of claim 12 further comprising steps of:

acquiring a measurement of occupancy of a buffer associated with said client device; and

where said preferred encoding rate exceeds a nominal encoding rate and said occupancy exceeds a predefined buffer-occupancy threshold, reducing said preferred encoding rate to equal said nominal encoding rate.

14. The method of claim 12 wherein said first tentative encoding rate is determined by multiplying said current encoding rate by a first encoding coefficient E* determined according a predefined function:

E*=Γ

(α

,β

), where α

is said slope of said regression line and β

is said gradient.

15. The method of claim 12 further comprising steps of:

comparing said data-loss-ratio measurement, denoted θ

, with a lower bound θ

_min and an upper bound θ

_max of a predefined data-loss-ratio acceptance interval;

where θ

>

θ

_max, determining a second encoding coefficient E** as E**=(1−

θ

),

where θ

<

θ

_min, determining said second encoding coefficient E** as E**=χ

>

1,

where χ

is a design parameter; and

multiplying said current encoding rate by said second encoding coefficient E** to determine said second tentative encoding rate.

16. A system of data streaming comprising:

a streaming server comprising;

at least one processor;

a computer-readable storage medium;

a signal source;

an adaptable encoder for encoding signals produced by said signal source;

a source reporter for formulating downstream control packets directed to a plurality of clients; and

a flow controller for receiving control packets from said plurality of clients and determining individual encoding coefficients for said plurality of clients, where each said encoding coefficient determines a respective encoding rate of said adaptable encoder;

and

a sink reporter associated with each client for formulating upstream control packets directed to said flow controller;

wherein said flow controller processes said upstream control packets to determine a current state of a connection from said streaming server and said each client.

17. The system of claim 16 wherein each downstream control packet sent from the source reporter to a respective client contains a respective packet identifier.

18. The system of claim 17 wherein an upstream control packet sent by a sink reporter of a specific client in response to a specific downstream control packet contains a packet identifier of said specific downstream control packet.

19. The system of claim 17 further comprising a data buffer associated with said each client wherein said upstream control packet contains an indication of occupancy of said data buffer.

20. The system of claim 19 wherein said current state of said connection from said streaming server and said each client is characterized by transfer delay from said streaming server to said each client and a proportion of lost downstream control packets and wherein said flow controller determines said individual encoding coefficients based on at least one of said transfer delay, said proportion of lost downstream control packets, and said occupancy of said data buffer.