Power saving system, method and apparatus for a wireless device

1. An article comprising a non-transitory machine-readable storage medium containing instructions that when executed enable a system to:

determine statistics on packet network traffic;

associate a quality of service metric and a traffic mode for a packet based on the statistics;

predict a packet-free time based on the traffic mode and the quality of service metric; and

send a control directive to enter a low power state based on the predicted packet-free time, the quality of service metric and the traffic mode.

2. The article of claim 1 wherein said instructions that when executed enable a system to determine statistics comprise instructions that when executed enable a system to collect and sample the packet network traffic over a period of time.

3. The article of claim 1 wherein said instructions that when executed enable a system to:

set an initial predicted packet-free time based on whether the traffic mode is sensitive to the quality of service metric; and

update the packet-free time based on each received packet.

4. The article of claim 1 wherein said instructions that when executed enable a system to determine a quality of service metric and a traffic mode for a packet based on the statistics comprise instructions that when executed enable a system to classify the packet network traffic into one or more of an unknown mode, a voice over internet protocol (VoIP) mode, a constant bit rate (CBR) mode, a high-throughput traffic mode or a web-traffic mode or a mixed mode.

5. The article of claim 1 wherein said instructions that when executed enable a system to:

linearly update the predicted packet-free time with each new packet when the traffic mode is sensitive to the quality of service metric; and

exponentially update the predicted packet-free time with each new packet when the traffic mode is not sensitive to the quality of service metric.

6. The article of claim 1 wherein said instructions that when executed enable a system to manage power of different sub-modules in-between packet burst to reduce power consumption.

7. The article of claim 1 , further comprising instructions that when executed enable a system to send one or more of the statistics, the traffic mode and the predicted packet-free time to a network platform.

8. A communications device comprising:

a network interface card comprising:

a sampling and statistics module operative to determine statistics on packet network traffic;

a traffic mode module operative to detect a traffic mode based on the statistics;

a packet-free time prediction module operative to predict a packet-free time based on a quality of service metric associated with the traffic mode; and

a power management mode operative to generate control directives to enter a low power state based on the predicted packet-free time, the quality of service metric and the traffic mode.

9. The device of claim 8 , further comprising:

a platform in communication with a network interface card, wherein the one or more modules of the network interface card communicates with the platform.

10. The device of claim 8 wherein the power management mode of the network interface card adaptively controls power-down of device sub-modules.

11. The device of claim 8 wherein the network interface card dynamically changes a sleep pattern of the communications device each time a new traffic mode is detected.

12. The device of claim 8 , further comprising: a digital display.

13. A method comprising:

detecting, by a processor, a traffic mode based on statistics of packet network traffic;

predicting a packet-free time based on a quality of service metric associated with the traffic mode; and

entering a low power state based on the predicted packet-free time, the quality of service metric and the traffic mode.

14. The method of claim 13 , further comprising determining statistics via collecting and sampling the packet network traffic over a period of time.

15. The method of claim 13 wherein predicting a packet-free time based on the quality of service metric associated with the traffic mode comprises:

setting an initial predicted packet-free time based on whether the traffic mode is sensitive to the quality of service metric; and

updating the packet-free time based on each received packet.

16. The method of claim 13 , further comprising determining statistics based on one or more of the following: packet arrival time, packet counts, packet size in bytes, packet direction and packet recipient.

17. The method of claim 13 , further comprising:

determining one or more of an average number of packets per burst for each transmission and/or reception path, a standard deviation of packets per burst for each transmission and/or reception path, an average of a inter-burst/inter-packet gap for each transmission and reception path, a standard deviation of an inter-burst/inter-packet gap for each transmission and/or reception path, an average packet size for each transmission and/or reception path, a standard deviation of a packet size for each transmission and/or reception path, a number of uplink packets, or a number of downlink packets.

18. The method of claim 13 wherein detecting a traffic mode based on statistics of packet network traffic comprises classifying the packet network traffic into one or more of an unknown mode, a voice over internet protocol (VoIP) mode, a constant bit rate (CBR) mode, a high-throughput traffic mode, a web-traffic mode or a mixed mode.

19. The method of claim 13 wherein predicting a packet- free time based on a quality of service metric associated with the traffic mode comprises:

linearly updating the predicted packet-free time with each new packet when the traffic mode is sensitive to the quality of service metric; and

exponentially updating the predicted packet-free time with each new packet when the traffic mode is not sensitive to the quality of service metric.

20. The method of claim 13 wherein entering a low power state based on the predicted packet-free time, the quality of service metric and the traffic mode comprises determining one or more of: a user preference, available power states, exit latency, break even time for each power state and a history of previous low power states.

21. The method of claim 13 wherein entering a low power state based on the predicted packet-free time, the quality of service metric and the traffic mode comprises saving power by power-managing different sub-modules in-between packet bursts.

22. The method of claim 13 , further comprising:

sending one or more of the statistics, the traffic mode and the predicted packet-free time to a network platform.