Adaptive timers for polling in a mobile wireless device

• US 8,724,612 B2
• Filed: 01/04/2011
• Issued: 05/13/2014
• Est. Priority Date: 01/04/2011
• Status: Active Grant

• Current Assignee: Apple Inc.
• Original Assignee: Apple Inc.
• Inventors: Al-Khudairi, Abdul-Munem
• Primary Examiner(s): Zhao, Wei
• Application Number:
  US12/984,495
• Publication Number:
  US 20120170569A1
• Time in Patent Office: 1,225 Days
• Field of Search: 370/346, 370/310, 709/226
• US Class Current: 370/346
• CPC Class Codes: H04L 69/28 Timers or timing mechanisms used in protocols

Abstract

A method for adapting data retrieval polling in a mobile wireless device is disclosed. The mobile wireless device is configured in an adaptive pull mode and polls a remote server when a data retrieval polling timer expires. Data retrieval information is stored in the mobile wireless device and the data retrieval polling timer is updated to a value based on the stored data retrieval information. The data retrieval information includes time intervals between successive polls or time delay intervals between data arrival at and retrieval from the remote server. Updates are based on weighted moving averages of the data retrieval information.

Claims

1. A method for adapting data retrieval polling in a mobile wireless device, the method comprising:

polling a remote server by the mobile wireless device when a data retrieval polling timer expires;

storing data retrieval information in the mobile wireless device, wherein the data retrieval information comprises a plurality of distinct historical time intervals, and wherein each distinct historical time interval is based on one or more of;

(a) a time when the mobile wireless device retrieves data from the remote server, and

(b) a time when the remote server receives data; and

updating the data retrieval polling timer to a value based on a weighted average of N most recent distinct historical time intervals in the stored data retrieval information, N being an integer greater than one, and wherein updating the data retrieval polling timer comprises;

increasing the data retrieval polling timer value when a most recent data retrieval poll retrieved no data and a current data retrieval polling timer value is less than a maximum polling interval value; and

decreasing the data retrieval polling timer value when the most recent data retrieval poll retrieved data and the current data retrieval polling timer value is greater than a minimum polling interval value.

2. The method as recited in claim 1, wherein each distinct historical time interval is an interval between a data retrieval time when a data retrieval poll returns data and a data storage time when the data was received at the remote server.

3. The method as recited in claim 1, wherein updating the data retrieval polling timer further comprises:

minimizing a moving average of the N most recent distinct historical time interval values; and

minimizing a moving average of the number of data retrieval polls that return no data.

4. The method as recited in claim 1, wherein the weighted average uses averaging weights based on a current time of day and a current day of the week.

5. The method as recited in claim 1, wherein updating the data retrieval polling timer is based on a user selected rate adaptation preference.

6. The method as recited in claim 1, wherein updating the data retrieval polling timer is based on a user selected activity state.

20. The method as recited in claim 1, wherein each distinct historical time interval is an interval between successive data retrieval polls that returned data.

21. The method as recited in claim 1, wherein at least one of the plurality of distinct historical time intervals is an interval based on time stamp information provided by the remote server.

22. The method as recited in claim 1, wherein each distinct historical time interval is an interval between successive data that arrives at the remote server.

23. The method as recited in claim 1, wherein each distinct historical time interval in the weighted average of N most recent distinct historical time intervals is weighted equally.

7. A mobile wireless device configured to adapt data retrieval polling, the mobile wireless device comprising:

a transceiver configured for transmitting and receiving wireless signals from a wireless network; and

an application processor configured for;

enabling an adaptive pull mode for the mobile wireless device;

polling a remote server by the mobile wireless device when a data retrieval polling timer expires;

storing data retrieval information, wherein the data retrieval information comprises a plurality of inter-arrival time intervals, and wherein each inter-arrival time interval is based on one or more of;

(a) a time when data arrives at the remote server and (b) a time when a data retrieval poll returns data; and

updating the data retrieval polling timer to a value based on at least a portion of the plurality of inter-arrival time intervals in the stored data retrieval information, wherein updating the data retrieval polling timer comprises;

increasing the data retrieval polling timer value when a current data retrieval polling timer value is less than a moving average of historical inter-arrival time intervals and a current data retrieval polling timer value is less than a maximum polling interval value; and

decreasing the data retrieval polling timer value when the current data retrieval polling timer value is greater than the moving average of historical inter-arrival time intervals and the current data retrieval polling timer value is greater than a minimum polling interval value, and

wherein the moving average of historical inter-arrival time intervals is based on a weighted average of N most recent inter-arrival time intervals in the stored data retrieval information, N being an integer greater than one.

8. The mobile wireless device as recited in claim 7, wherein each inter-arrival time interval is an interval between successive data retrieval polls that returned data.

9. The mobile wireless device as recited in claim 7, wherein each inter-arrival time interval is an interval between a data retrieval time when a data retrieval poll returns data and a data storage time when the data was received at the remote server.

10. The mobile wireless device as recited in claim 7, wherein the moving average of historical inter-arrival time intervals weights the most recently determined inter-arrival time interval more than the previously determined inter-arrival time intervals.

11. The mobile wireless device as recited in claim 7, wherein the moving average of historical inter-arrival time intervals uses weights based on one or more of:

a current time of day, a current day of the week and a user selected activity state.

19. The mobile wireless device as recited in claim 7, wherein each inter-arrival time interval is an interval between successive data that arrives at the remote server.

24. The mobile wireless device as recited in claim 7, wherein each inter-arrival time interval in the weighted average of N most recent inter-arrival time intervals is weighted equally.

12. A non-transitory computer readable medium for storing non-transitory computer program code executed by a processor for adapting data retrieval timing from a remote server by a mobile wireless device, the non-transitory computer readable medium comprising:

non-transitory computer program code configured for polling the remote server by the mobile wireless device when a data retrieval polling timer expires;

non-transitory computer program code configured for storing data retrieval information in the mobile wireless device, wherein the data retrieval information comprises a plurality of distinct historical time intervals, and wherein each distinct historical time interval is based on one or more of;

(a) a time when data is retrieved by the mobile wireless device and (b) a time when data arrives at the remote server;

non-transitory computer program code configured for updating the data retrieval polling timer to a value based on a weighted average of N most recent distinct historical time intervals in the stored data retrieval information, N being an integer greater than one;

non-transitory computer program code configured for increasing the data retrieval polling timer value when a most recent data retrieval poll retrieved no data and a current data retrieval polling timer value is less than a maximum polling interval value; and

non-transitory computer program code configured for decreasing the data retrieval polling timer value when the most recent data retrieval poll retrieved data and the current data retrieval polling timer value is greater than a minimum polling interval value.

13. The non-transitory computer readable medium as recited in claim 12, wherein each distinct historical time interval is an interval between successive data retrieval polls that returned data.

14. The non-transitory computer readable medium as recited in claim 12, wherein each distinct historical time interval is an interval between a data retrieval time when a data retrieval poll returns data and a data storage time when the data was received at the remote server.

15. The non-transitory computer readable medium as recited in claim 12, wherein each distinct historical time interval is an interval between successive data received for a user of the mobile wireless device at the remote server.

16. The non-transitory computer readable medium as recited in claim 12, further comprising:

non-transitory computer program code configured for updating the data retrieval polling timer by minimizing a moving average of the N most recent distinct historical time interval values.

17. The non-transitory computer readable medium as recited in claim 12, further comprising:

non-transitory computer program code configured for updating the data retrieval polling timer by minimizing a moving average of the number of data retrieval polls that return no data.

18. The non-transitory computer readable medium as recited in claim 12, further comprising:

non-transitory computer program code configured for increasing the data retrieval polling timer value when the current data retrieval polling timer value is less than the weighted average of N most recent distinct historical time intervals and the current data retrieval polling timer value is less than the maximum polling interval value; and

non-transitory computer program code configured for decreasing the data retrieval polling timer value when the current data retrieval polling timer value is greater than the weighted average of N most recent historical time intervals and the current data retrieval polling timer value is greater than the minimum polling interval value.

25. The mobile wireless device as recited in claim 12, wherein each distinct historical time interval in the weighted average of N most recent distinct historical time intervals is weighted equally.