Apparatus, and associated method, for selecting radio communication system parameters utilizing learning controllers

• US 20030091004A1
• Filed: 11/13/2001
• Published: 05/15/2003
• Est. Priority Date: 11/13/2001
• Status: Active Grant

• Current Assignee: Nokia Corporation
• Original Assignee: Nokia Corporation
• Inventors: Tang, Clive
• Granted Patent: US 7,221,654 B2
• Time in Patent Office: Days
• US Class Current: 370/252
• CPC Class Codes: H04L 1/0003 by switching between different modulation schemes
• CPC Class Codes: H04L 1/0021 in which the algorithm uses adaptive thresholds

Abstract

Apparatus, and an associated method, for selecting operational parameters pursuant to which a radio communication system is operable. The operational parameters form, for instance, adaptive switching thresholds used in an adaptive modulation scheme. Separate learning controllers are configured to operate independently and cooperatively to select the adaptive switching thresholds, or other operational parameters. Iterative operation of the learning controllers causes the values to converge to optimal values.

Claims

1. In a communication station having a receive portion operable to receive data communicated thereto upon a radio channel susceptible to distortion, an improvement of apparatus for facilitating selection of a value of a first parameter and a value of at least a second parameter by which to operate upon the receive data once received at the communication station, said apparatus comprising:

a first learning controller coupled to the receive portion to be provided with a performance measure representative of operation of the receive portion upon a set of successively-received sequences of the receive data received at the communication station, said first learning controller for iteratively selecting the value of the first parameter responsive to the performance measure associated with each of the successively-received sequences of the set, each successive performance measure identifying a first parameter value; and

at least a second learning controller also coupled to the receive portion to be provided with the performance measure representative of operation of the receive portion upon the set of the successively-received sequences of the receive data received at the communication station, said second learning controller for iteratively selecting the value of the second parameter responsive to the performance measure associated with each of the successively-received sequences of the set, each successive performance measure identifying a second parameter value.

2. The apparatus of claim 1 wherein the performance measure responsive to which said first learning controller and said at least second learning controller iteratively select the value of the first parameter and the value of at least the second parameter comprise a data throughput indicia of the receive data received at the received portion of the communication station.

3. The apparatus of claim 1 wherein said at least second learning controller comprises said second learning controller and at least a third learning controller, said third learning controller for iteratively selecting the value of a third parameter value responsive to the performance measure associated with each of the successively-received sequences of the set, each successive performance measure identifying a third parameter value.

4. The apparatus of claim 1 wherein the receive data is formatted into successive frames of receive data and wherein each sequence of the receive data operated upon by the receive portion and of which the performance measure representative of operation thereon by the receive portion is provided to said first learning controller and said at least second learning controller, respectively comprises a frame of the receive data.

5. The apparatus of claim 1 wherein the performance measure provided to said first learning controller and to said at least second learning controller comprises an error rate value indicative of errors contained in individual ones of the successively-received sequences of the set of the successively-received sequences of the receive data.

6. The apparatus of claim 1 wherein said first learning controller further comprises a first memory device for storing a first set of responsive-actions, said first learning controller further for iteratively selecting a first selected responsive-action of the first set of responsive-actions responsive to the performance measure provided thereto and associated with each of the successively-received sequences of the set.

7. The apparatus of claim 6 wherein said first learning controller comprises a first algorithm executable at a processing device, the first algorithm defining a linear reward/penalty (LRP) scheme.

8. The apparatus of claim 7 wherein the first algorithm further defines a linear reward/ε

penalty (LRε

P) scheme.

9. The apparatus of claim 8 wherein the first algorithm further defines a linear reward inaction (LRI) scheme.

10. The apparatus of claim 6 wherein the communication station is operable in a communication system in which the receive data communicated to the communication station is modulated pursuant to an adaptive modulation scheme, the receive data selectably modulated pursuant to a first modulation format, a second modulation format, and at least a third modulation format, the first modulation format used during first channel distortion conditions, the second modulation format used during second channel distortion conditions, and the third modulation format is used during third channel distortion conditions, the first selected responsive-action selected by said first learning controller defining a first-selected boundary value defining a first boundary, defined in terms of the performance measure, separating the first channel distortion conditions and the second channel distortion conditions.

11. The apparatus of claim 6 wherein said second learning controller further comprises a second memory device for storing a second set of responsive-actions, said second learning controller further for iteratively selecting a second selected responsive-action of the second set of responsive-actions responsive to the performance measure provided thereto and associated with each of the successively-received sequences of the set.

12. The apparatus of claim 11 wherein said second learning controller comprises a second algorithm executable at a processing device, the second algorithm defining a linear reward/penalty (LRP) scheme.

13. The apparatus of claim 11 wherein the said second learning controller comprises a second algorithm executable at a processing device, the second algorithm defining a linear reward/ε

penalty (LREP) scheme.

14. The apparatus of claim 11 wherein said second learning controller comprises a second algorithm executable at a processing device, the second algorithm defining a linear reward inaction (LRI) scheme.

15. The apparatus of claim 11 wherein the communication station is operable in a radio communication system in which the receive data communicated to the radio communication station is modulated pursuant to an adaptive modulation scheme, the receive data selectably modulated pursuant to a first modulation format, a second modulation format, and at least a third modulation format, the first modulation format used during first channel distortion conditions, the second modulation format used during second channel distortion conditions, and the third modulation format is used during third channel distortion conditions, the first selected responsive-action selected by said first learning controller defining a first-selected boundary value defining a first-selected boundary value defining a first boundary, defined in terms of the performance measure, separating the first channel distortion conditions and the second channel distortion conditions, the second selected responsive-action selected by said second learning controller defining a second-selected boundary value defining a second boundary, defined in terms of the performance measure, separating the second channel distortion conditions and the third channel distortion conditions.

16. The apparatus of claim 11 wherein said first learning controller and said second learning controller operate independently and cooperatively to select the first selected responsive-action and to select the second selected responsive-action, respectively.

17. In a method for communicating at a communication station having a receive portion operable to receive data communicated thereto upon a channel susceptible to distortion, an improvement of a method 86 for facilitating selection of a value of a first parameter and a value of at least a second parameter by which to operate upon the receive data once received at the communication station, said method comprising:

iteratively selecting 88 a value of a first parameter responsive to a performance measure associated with each successively-received sequence of a set of successively received sequences received at the receive portion, each successive performance measure identifying a first parameter value; and

iteratively selecting 92 a value of a second parameter responsive to the performance measure associated with each successively-received sequence of the set, each successive performance measure identifying a second parameter value .

18. The method of claim 17 wherein said operation of iteratively selecting the value of the first parameter and said operation of iteratively selecting the value of the second parameter are performed independently and cooperatively.

19. The method of claim 17 further comprising the operation, subsequent to said operation of iteratively selecting the value of the first parameter, of iteratively selecting a first-selected responsive action responsive to the performance measure associated with each of the successively-received sequences of the set.

20. The method of claim 17 further comprising the operation, subsequent to said operation of iteratively selecting the value of the second parameter, of iteratively selecting a second-selected responsive action responsive to the performance measure associated with each of the successively-received sequences of the set.