Predictive provisioning of media resources

• US 20040010582A1
• Filed: 06/28/2002
• Published: 01/15/2004
• Est. Priority Date: 06/28/2002
• Status: Active Grant

• Current Assignee: Intel Corporation
• Original Assignee: Intel Corporation
• Inventors: Oliver, Neal C.
• Granted Patent: US 7,451,207 B2
• Time in Patent Office: Days
• US Class Current: 709/224
• CPC Class Codes: H04M 3/493 Interactive information services, e.g. directory enquiries ; Arrangements therefor, e.g. interactive voice response [IVR] systems or voice portals
• CPC Class Codes: H04M 7/00 Arrangements for interconnection between switching centres
• CPC Class Codes: H04M 7/1285 Details of finding and selecting a gateway for a particular call

Abstract

An apparatus and method are provided for predictive provisioning of functional packages based on offered traffic and a predictive model of the offered traffic. According to one embodiment of the invention, a managed agent provides predictions regarding an anticipated need for functional package provisioning based on traffic offered to the managed agent and a predictive model of offered traffic. A provisioning agent receives the predictions and in response to the predictions instructs the managed agent to provision a new functional package. According to another embodiment of the invention, the provisioning agent instructs a second managed agent to provision the media resource needed by the original managed agent. The traffic is then routed from the original man aged agent to the second managed agent.

Claims

1. A system comprising:

a first managed agent including a first set of one or more functional packages, the first managed agent to process offered traffic and to provide predictions regarding an anticipated need for functional package provisioning based on the offered traffic and a predictive model of offered traffic; and

a provisioning agent to receive the predictions from the first managed agent and in response to the predictions to instruct the first managed agent to provision a new functional package.

2. The system of claim 1, wherein the first managed agent is further to adapt the predictive model based on the offered traffic.

3. The system of claim 1, wherein the functional packages comprise one or more of Automatic Speech Recognition (ASR) resources, Text-to-Speech (TTS) resources, Echo Canceller (EC) resources, and Conference Bridge (CB) resources.

4. The system of claim 3, wherein the predictive model is parameterized in terms of at least one of resource type, attributes, grammars, and vocabularies.

5. The system of claim 1, further comprising a second managed agent, the second managed agent including a second set of one or more functional packages.

6. The system of claim 1, further comprising a pipe to route offered traffic from the first managed agent to the second managed agent, and wherein the second managed agent provisions the new functional package.

7. The system of claim 1, wherein the first managed agent comprises a media gateway.

8. The system of claim 1, wherein the provisioning agent comprises a media gateway controller.

9. The system of claim 8, wherein the media gateway controller comprises a softswitch.

10. The system of claim 1, wherein the predictive model comprises a queuing model to model call arrival and holding time, and a Hidden Markov Model (HMM) to model resource allocation, deallocation and operation.

11. A system comprising:

a predictive model of flow patterns of resource usage;

a first managed agent comprising a first set of one or more media resources, the first managed agent to provide predictions regarding an anticipated need for media resource provisioning based on offered traffic and a predictive model of offered traffic, and the first managed agent to adapt the predictive model in response to usage of the system; and

a provisioning agent to receive the predictions from the first managed agent and in response to the predictions to instruct the first managed agent to provision a new media resource.

12. The system of claim 11, wherein the media resources comprise one or more of Automatic Speech Recognition (ASR) resources, Text-to-Speech (TTS) resources, Echo Cancellation (EC) resources, and Conference Bridge (CB) resources.

13. The system of claim 11, wherein the first managed agent comprises a media gateway and the provisioning agent comprises a media gateway controller.

14. An Interactive Voice Response (IVR) system comprising:

a predictive model of flow patterns of user interactions with the IVR system;

a media gateway communicatively coupled to a Public Switched Telephone Network (PSTN), the media gateway comprising a first set of one or more media resources, the media gateway to adapt the predictive model in response to the offered traffic and to provide predictions regarding an anticipated need for media resource provisioning based on the predictive model and the offered traffic; and

a media gateway controller communicatively coupled to the media gateway, the media gateway controller to receive the predictions from the media gateway and in response to the predictions to instruct the media gateway to provision a new media resource.

15. The IVR system of claim 14, wherein the predictive model is parameterized in terms of at least one of resource type, acoustic model identification, attributes, grammars, and vocabularies.

16. A media gateway controller comprising:

a provisioning agent to receive predictions regarding an anticipated need for one or more media resources from a media gateway and in response to the predictions to instruct the media gateway to provision the one or more media resources

17. The media gateway controller of claim 16, wherein the media resources comprise one or more of Automatic Speech Recognition (ASR) resources, Text-to-Speech (TTS) resources, Echo Canceller (EC) resources, and Conference Bridge (CB) resources.

18. The media gateway controller of claim 16, further comprising one or more records of the provisioning state of media gateways under the control of the media gateway controller.

19. A media gateway comprising:

a queuing model of traffic offered to the media gateway;

one or more media resources; and

a resource allocation model to model resource allocation, deallocation, and operation;

wherein the media gateway provides predictions to a media gateway controller regarding an anticipated need for one or more media resources, the predictions based on the queuing model and the resource allocation model.

20. The media gateway controller of claim 19, wherein the media resources comprise one or more of Automatic Speech Recognition (ASR) resources, Text-to-Speech (TTS) resources, Echo Canceller (EC) resources, and Conference Bridge (CB) resources.

21. The media gateway controller of claim 20, wherein the resource allocation model is parameterized in terms of at least one of resource type, attributes, grammars, and vocabularies.

22. A method comprising:

observing offered traffic to a media gateway; and

proactively provisioning the media gateway in response to the offered traffic and a predictive model of offered traffic.

23. The method of claim 22, further comprising modifying the predictive model based on the offered traffic.

24. The method of claim 22, wherein the proactively provisioning comprises proactively provisioning the media gateway with one or more Automatic Speech Recognition (ASR) resources.

25. A method comprising:

receiving offered traffic at a first managed agent; and

providing a prediction of a required media resource based on the offered traffic and a predictive model of offered traffic.

26. The method of claim 25, wherein the first managed agent comprises a media gateway and the media resources comprise one or more of Automatic Speech Recognition (ASR) resources, Text-to-Speech (TTS) resources, Echo Canceller (EC) resources, and Conference Bridge (CB) resources.

27. The method of claim 25, further comprising:

provisioning the functional package at a second managed agent; and

routing traffic from the first managed agent to the second managed agent.

28. A method comprising:

receiving offered traffic at a first media gateway;

providing a prediction of a required media resource to a media gateway controller based on the offered traffic and a predictive model of offered traffic;

adapting the predictive model based on the offered traffic; and

instructing the first media gateway to provision the media resource.

29. The method of claim 28, further comprising:

instructing a second media gateway to provision the new media resource; and

routing traffic from the first media gateway to the second media gateway.

30. The method of claim 28, wherein the predictive model is parameterized in terms of at least one of ASR implementation, attributes, grammars, and vocabularies.

31. An apparatus comprising:

a means for receiving offered traffic at a first media gateway;

a means for providing a prediction of a required media resource to a media gateway controller based on the offered traffic and a predictive model of offered traffic.

32. The apparatus of claim 31, further comprising a means for adapting the predictive model based on the offered traffic.

33. The apparatus of claim 31, further comprising:

a means for provisioning the media resource at a second media gateway; and

a means for routing traffic from the first media gateway to the second media gateway.

34. A machine-readable medium having stored thereon data representing sequences of instructions that when executed cause a machine to:

receive offered traffic; and

provide a prediction of a required media resource based on the offered traffic and a predictive model of offered traffic.

35. The machine-readable medium of claim 34, further comprising instructions to adapt the predictive model based on the offered traffic.

36. The machine-readable medium of claim 34, wherein the predictive model comprises a queuing model to model call arrival and holding time, and a Hidden Markov Model (HMM) to model resource allocation, deallocation, and operation.

37. A method comprising the steps of:

a step for receiving offered traffic at a media gateway;

a step for providing a prediction of a required media resource to a media gateway controller based on the offered traffic and a predictive model of offered traffic; and

a step for instructing the media gateway to provision the media resource.

38. The method of claim 37, further comprising a step for adapting the predictive model based on the offered traffic.

39. The method of claim 38, wherein the predictive model is parameterized in terms of at least one of resource type, attributes, grammars, and vocabularies.