Method and system for service rate allocation, traffic learning process, and QoS provisioning measurement of traffic flows

1. A method for determining allocation of resources in a communication system comprising at least two entities, the resources being allocated for transmitting, receiving, or forwarding information, the information being transmitted as flows, wherein the system allocates resources based upon at least one parameter comprising, class of service, delay requirement, arrival rate requirement and service rate requirement of a current flow request which parameters are stored, and wherein, during a call or connection between the at least two entities, the system monitors at least one of the parameters for updating the service rate allocated to meet a Quality of Service QoS requirement of the current flow request in accordance with a result of the monitoring, the updated service rate being allocated to a new flow request of a same type, and wherein

a learning process is provided for adjusting a required service rate of applications and allocating a computed service rate for new flow requests, wherein, based upon a current allocation, the system learns which resources should have been allocated to be able to meet a promised QoS requirement of the current flow request.

2. A method according to claim 1 , wherein the system learns the traffic pattern and adjusts a current airlink performance to allocate a service rate to flows.

3. A method according to claim 1 , wherein a service rate computation algorithm is applied to all classes of traffic in a system.

4. A method according to claim 1 , wherein information gained by monitoring the parameters is used to refine the allocation of future flows but is not used to modify the allocation of flows already admitted.

5. A method according to claim 1 , comprising the step of monitoring how the QoS requirement is met and based on thereon allocating resources to future flows of different types.

6. A method according to claim 5 , wherein a service rate adaptation algorithm is used to predict the flow requirements.

7. A method according to claim 1 , wherein the system uses an algorithm which adapts to each flow request and airlink performance by constantly monitoring parameters associated with each traffic request class.

8. A method according to claim 1 , applied to packet data applications including real-time and/or non-real-time applications in a cellular system, for QoS for bursty applications over a cellular link.

9. A method according to claim 1 , comprising the step of measuring QoS actually experienced by a given data application flow, the measuring being based on an evaluation of a number or proportion of packets for which the QoS was met or not met in past time periods.

10. A method according to claim 9 , wherein measured values are directly compared against a QoS performance objective.

11. A method according to claim 10 , wherein the QoS performance objective is that the QoS performance must be met at least a specified percentage of the time.

12. A method according to claim 11 , wherein a value measured by an entity of the system provides a direct estimate of a percentage of time the QoS is met, the measured value being directly compared with the specified percentage, and resources are increased or decreased to track the specified percentage.

13. A method according to claim 1 , wherein queue lengths are monitored and compared to a threshold to determine if a provided QoS is within an acceptable limit.

14. A method according to claim 13 , wherein delay and an amount or percentage of packets which are below the threshold are is measured.

15. A system for determining allocation of resources in a communication system comprising at least two entities, the system comprising an allocation system which the allocates resources for transmitting, receiving, or forwarding information, the information being transmitted as flows, wherein the system allocates resources based upon at least one parameter comprising class of service, delay requirement, arrival rate requirement and service rate requirement of a current flow request which parameters are stored, comprising a monitoring system, which during a call or connection between the at least two entities, monitors at least one of the parameters for updating service rate allocated to meet a Quality of Service QoS requirement of the current flow request in accordance with a result of the monitoring system, the allocation system allocating the updated service rate to a new flow request of the same type, and wherein

a learning system is provided for adjusting a required service rate of applications and allocating a computed service rate for new flow requests, wherein, based upon the current allocation, the learning system learns how resources of the system should have allocated to be able to meet a promised QoS requirement of the flow request.

16. A system according to claim 15 , wherein the system learns the traffic pattern and adjusts a current airlink performance to allocate a service rate to flows.

17. A system according to claim 15 , wherein the allocation system includes a service rate computation algorithm to be applied to all classes of traffic in a system.

18. A system according to claim 15 , wherein the monitoring system refines resource allocation for future flows but does not modify allocation of flows already admitted.

19. A system according to claim 15 , wherein the monitoring system monitors how the QoS requirements are met and based thereon allocates resources to future flows of different types by the allocation system.

20. A system according to claim 19 , wherein the allocation system includes a service rate adaptation algorithm for predicting the flow requirements.

21. A system according to claim 15 , wherein the system includes an algorithm which adapts to each flow request and airlink performance by constantly monitoring parameters associated with each traffic request class.

22. A system according to claim 15 , applied to at least one packet data applications including real-time and non-real-time applications in a cellular system, for bursty applications over a cellular link.

23. A system according to claim 15 , wherein the QoS of the system is experienced by a given data application flow and measuring of QoS is performed based on an evaluation of a number or proportion of packets for which the QoS was met or not met in past time periods.

24. A system according to claim 23 , comprising a comparison device which compares measured values against a QoS performance objective.

25. A system according to claim 24 , wherein the QoS performance objective is that QoS must be met at least a specified percentage of the time.

26. A system according to claim 25 , wherein a value measured by an entity of the system provides a direct estimate of a percentage of time the QoS is met, the measured value being directly compared with the specified percentage, and resources are increased or decreased to track the specified percentage.

27. A system according to claim 15 , wherein queue lengths are monitored and compared to a threshold to determine if a provided Qos is within an acceptable limit.

28. A system according to claim 2 , wherein delay and an amount or percentage of packets which are below the threshold are measured.