Event driven grant allocation

• US 8,780,823 B1
• Filed: 10/08/2010
• Issued: 07/15/2014
• Est. Priority Date: 10/08/2009
• Status: Active Grant

• Current Assignee: ViaSat Incorporated
• Original Assignee: ViaSat Incorporated
• Inventors: Sebastian, William B.
• Primary Examiner(s): BEHARRY, NOEL R
• Application Number:
  US12/901,250
• Time in Patent Office: 1,376 Days
• Field of Search: 370/329
• US Class Current: 370/329
• CPC Class Codes: H04B 7/18578 Satellite systems for providing broadband data service to individual earth stations for selecting H04W; provisions for broadband connection, H04Q11/0478
• CPC Class Codes: H04W 4/18 Information format or content conversion, e.g. adaptation by the network of the transmitted or received information for the purpose of wireless delivery to users or terminals

Abstract

Systems and methods are described for event-driven predictive scheduling of a return link of a communications system. Embodiments detect an event (e.g., a packet) on a forward link that may indicate future usage of the return link. The event is then associated with an event model that characterizes the event, and the event model is further associated with a predictive model. Scheduling of the return link is implemented and/or adjusted according to the predictive model. For example, return link allocations are adjusted by one or more amounts (e.g., one or more grants are allocated or bandwidth is adjusted by a certain amount) and at one or more designated timing offsets according to the predictive model. According to various embodiments, the predictive model is generated by tracking feedback from the return link and/or by scanning packets on the forward link.

Claims

1. A method for event-driven predictive scheduling of a return link in a communication system comprising:

detecting, at a provider side of the communications system, an event predictor corresponding to a download packet associated with an HTTP object destined for a user terminal over a forward link of the communications system, the event predictor indicating a future event to occur in association with receipt of the download packet at the user terminal;

selecting a prediction model according to the detected event predictor from a plurality of prediction models maintained at the provider side of the communications system, wherein each prediction model of the plurality of prediction models estimates a size or timing of a future communications event associated with the HTTP object;

generating a resource allocation according to the selected prediction model at the provider side of the communications system, the resource allocation indicating a magnitude by which to adjust resources of the return link of the communications system associated with the user terminal, and the resource allocation indicating a timing offset at which to adjust the resources by the magnitude, the timing offset corresponding to an amount of time after occurrence of the event;

scheduling the resources of the return link of the communications system associated with the user terminal at least according to the resource allocation; and

communicating data corresponding to the data packet and data corresponding to the resource allocation to the user terminal over the forward link of the communications system;

wherein selecting the prediction model according to the detected event predictor comprises;

scanning the download packet to extract metadata, the metadata indicating at least one HTTP object likely to be requested subsequent to receiving the download packet at the user terminal; and

selecting the prediction model at least according to the metadata.

2. The method of claim 1, further comprising:

receiving feedback data at the provider side of the communications system over the return link from the user terminal, the feedback data relating to a magnitude and a timing of at least one request issued by the user terminal subsequent to and associated with the event; and

updating at least one of the plurality of prediction models according to the feedback data.

3. The method of claim 1, wherein generating the resource allocation according to the selected prediction model comprises:

generating a grant request, the magnitude being a size of the grant request to allocate on the return link of the communications network.

4. The method of claim 1, wherein generating the resource allocation according to the selected prediction model comprises:

determining a bandwidth adjustment, the magnitude indicated by the allocation request being a magnitude by which to adjust bandwidth on the return link of the communications network for the user terminal.

5. The method of claim 1, wherein:

generating the resource allocation comprises generating an allocation request and communicating the allocation request indicating the magnitude and the timing offset associated with the resource allocation to a return-link scheduler; and

scheduling the resources of the return link of the communications system associated with the user terminal at least according to the resource allocation comprises using the return-link scheduler to adjust allocation of the resources of the return link by the magnitude at a time corresponding to the timing offset.

6. The method of claim 1, further comprising:

receiving at least one allocation request at the provider side of the communications system over the return link from the user terminal; and

scheduling the resources of the return link of the communications system associated with the user terminal further according to the at least one allocation request.

7. The method of claim 1, wherein scanning the download packet to extract the metadata comprises:

scanning the download packet to extract header metadata from a header portion of the download packet; and

scanning the download packet to generate request predictor metadata corresponding to a payload portion of the download packet indicating at least one associated attachment.

8. The method of claim 1, wherein communicating data corresponding to the data packet and data corresponding to the resource allocation to the user terminal over the forward link of the communications system comprises:

generating a combined packet comprising data corresponding to the download packet and data corresponding to the allocation request; and

communicating the combined packet to the user terminal over the forward link of the communications system.

9. The method of claim 1, wherein the forward link and the return link are portions of a satellite link between a base station at the provider side of the communications system and the user terminal.

10. The method of claim 1 wherein detecting, at the provider side of the communications system, the event predictor corresponding to the download packet comprises:

parsing the download packet to identify objects; and

estimating at least one object of the objects not able to be prefetched;

wherein the download packet comprises HTTP header information.

11. The method of claim 10 wherein the prediction model estimates a number, a size, and a timing of requests to be received at the provider side of a communications system in response to the estimating at least one object of the objects not able to be prefetched.

12. The method of claim 11 wherein selection of the prediction model is determined by a specificity of the prediction model and an amount of sample data;

wherein the amount of sample data comprises a number of previous requests for the HTTP object recorded in the prediction model; and

wherein the specificity of the prediction model is based on at least one request recorded in the prediction model from a same session as a session of the download packet.

13. The method of claim 12 wherein the plurality of prediction models comprises at least one model specific to a server as identified by an internet protocol (IP) address of the server and at least one model specific to the HTTP object.

14. A base station system disposed in a satellite communications network and in communication with a user terminal over a satellite link having a forward link and a return link, the system comprising:

an optimizer subsystem, configured to detect an event predictor corresponding to a download packet associated with an HTTP object destined for the user terminal over the forward link, the event predictor indicating a future event to occur in association with receipt of the download packet at the user terminal;

a predictive grant generator subsystem, communicatively coupled with the optimizer subsystem, and configured to select a prediction model according to the detected event predictor, and generate a resource allocation according to the selected prediction model, wherein the prediction model estimates a size or timing of a future communications event, the resource allocation indicating a magnitude by which to adjust resources of the return link and a timing offset at which to adjust the resources by the magnitude, the timing offset corresponding to an amount of time after occurrence of the event; and

a satellite modem termination subsystem, communicatively coupled with the predictive grant generator subsystem and the optimizer subsystem, and configured to;

schedule the resources of the return link at least according to the resource allocation; and

communicate data corresponding to the download packet and data corresponding to the resource allocation to the user terminal over the forward link;

wherein;

the optimizer subsystem is configured to detect the event predictor corresponding to the download packet by scanning the download packet to extract metadata, the metadata indicating at least one HTTP object likely to be requested subsequent to receiving the download packet at the user terminal; and

the predictive grant generator subsystem is configured to select the prediction model at least according to the metadata.

15. The system of claim 14, wherein:

the satellite modem termination subsystem is further configured to receive feedback data relating to a magnitude and a timing of at least one request issued by the user terminal subsequent to and associated with the event and to communicate the feedback data to the predictive grant generator subsystem; and

the predictive grant generator subsystem is further configured to update at least one of the plurality of prediction models according to the feedback data.

16. The system of claim 14, wherein the satellite modem termination subsystem is configured to schedule the resources of the return link at least according to the resource allocation by generating at least one grant request sized according to the magnitude of the resource allocation and timed according to the timing offset of the resource allocation.

17. The method of claim 14, further comprising:

a packetizer, communicatively coupled with the predictive grant generator subsystem, and configured to generate a combined packet comprising optimized data corresponding to the download packet and data corresponding to the allocation request; and

the satellite modem termination subsystem is configured to communicate the data corresponding to the data packet and the data corresponding to the resource allocation by communicating the combined packet to the user terminal over the forward link.

18. A satellite communications system comprising:

a base station in communication with a user terminal over a satellite link having a forward link and a return link, and configured to;

detect an event predictor corresponding to a download packet associated with an HTTP object destined for the user terminal over the forward link, the event predictor indicating a future event to occur in association with receipt of the download packet at the user terminal;

select a prediction model according to the detected event predictor, wherein the prediction model estimates a size or timing of a future communications event;

generate a resource allocation according to the selected prediction model, the resource allocation indicating a magnitude by which to adjust resources of the return link and a timing offset at which to adjust the resources by the magnitude, the timing offset corresponding to an amount of time after occurrence of the event;

schedule the resources of the return link at least according to the resource allocation; and

communicate data corresponding to the download packet and data corresponding to the resource allocation to the user terminal over the forward link;

wherein the base station is further configured to select the prediction model by;

scanning the download packet to extract metadata indicating at least one HTTP object likely to be requested subsequent to receiving the download packet at the user terminal; and

selecting the prediction model at least according to the metadata.

19. The system of claim 18, the base station further configured to:

receive feedback data over the return link from the user terminal, the feedback data relating to a magnitude and a timing of at least one request issued by the user terminal subsequent to and associated with the event; and

update at least one of the plurality of prediction models according to the feedback data.

20. The system of claim 18, further comprising:

the user terminal, communicatively coupled with the base station over the satellite link, and configured to;

receive the data corresponding to the download packet and the data corresponding to the resource allocation from the base station over the forward link;

receive an upload packet from an application in communication with the user terminal at least partially in response to receiving the data corresponding to the download packet; and

upload the upload packet to the base station via the return link according to the resource allocation.