Neural network-based extension of global position timing
View Patent ↗A wireless communication system ( 20 ) includes a base station controller ( 22 ) that receives timing information from a data set ( 26 ) that is generated by a neural network ( 28 ). The data set ( 26 ) allows for generating timing information based upon previous time information received from a GPS ( 24 ) and in one example, is capable of covering a time interval of up to two weeks during which effective communication with the GPS may be interrupted. In one example, the data set is continuously updated so that the base station controller ( 24 ) continuously has up to two weeks of future time information available.
1. A method of maintaining time information for a wireless communications base station, comprising:
using time information provided to a neural network for generating a data set that provides future time information; and
selectively using time information from the data set for conducting a communication involving the base station,
wherein the data set comprises a plurality of coefficients for generating future time information based upon a start time.
2. The method of claim 1 , wherein the data set is useful for a first time interval and including generating another data set for a second, later time interval.
3. The method of claim 2 , including repeatedly generating another data set for subsequent time intervals.
4. The method of claim 1 , including
gathering time information from an external source;
inputting the gathered time information to the neural network; and
generating the data set based upon the inputted time information.
5. The method of claim 4 , wherein the gathered time information extends over a selected period and including
comparing time information from the data set for a period corresponding to the selected period with the gathered time information; and
changing at least one characteristic of the neural network when the data set time information does not correspond to the gathered time information within a selected range.
6. The method of claim 5 , including changing the characteristic of the neural network by changing at least one of a number of layers in the neural network, a number of neurons in the neural network or a complexity factor of the neural network.
7. The method of claim 5 , including repeatedly performing the steps of comparing and changing until the data set time information corresponds to the gathered time information within the selected range.
8. The method of claim 1 , including
receiving time information from an external source;
determining when the external source time information is not available; and
using the data set for time information when the external source time information is not available.
9. The method of claim 8 , including using an initialization time value and the data set to generate time information until the external source time information becomes available.
10. The method of claim 8 , wherein the external source time information comprises global position system time information.
11. The method of claim 1 , including providing at least more than 24 hours of future time information using the data set.
12. The method of claim 11 , including providing at least two weeks of future time information using the data set.
13. The method of claim 1 , wherein the data set corresponds to global position system time information.
14. A wireless communication device, comprising:
a neural network that generates a data set for providing future time information; and
a base station controller that determines time information from a global position system (GPS) source of time information and uses the data set for obtaining time information if the GPS source is unavailable to the base station controller,
wherein the data set comprises a plurality of coefficients for generating future time information based upon a start time.
15. The device of claim 14 , wherein the data set is useful for a first time interval and the neural network generates another data set for a second, later time interval.
16. The device of claim 15 , wherein the neural network repeatedly generates another data set for subsequent time intervals.
17. The device of claim 14 , wherein the neural network receives an input of gathered time information and generates the data set based upon the inputted time information.
18. The device of claim 14 , wherein the data set provides at least more than 24 hours of future time information.
19. The device of claim 18 , wherein the data set provides at least two weeks of future time information.
20. The device of claim 14 , wherein the data set corresponds to GPS time information.