Auto-IP traffic optimization in mobile telecommunications systems

• US 7,466,652 B2
• Filed: 08/13/2004
• Issued: 12/16/2008
• Est. Priority Date: 08/14/2003
• Status: Expired due to Fees

• Current Assignee: Telcordia Technologies Incorporated (Telefonaktiebolaget LM Ericsson)
• Original Assignee: Telcordia Technologies Incorporated (Telefonaktiebolaget LM Ericsson)
• Inventors: Siegell, Bruce, Lau, Richard
• Primary Examiner(s): Pham; Chi H
• Assistant Examiner(s): Chou; Albert T
• Application Number:
  US10/918,256
• Publication Number:
  US 20050041584A1
• Time in Patent Office: 1,586 Days
• Field of Search: 370/235, 370/230.1, 370/328, 709/224, 455/445, 455/456.2
• US Class Current: 370/230.1
• CPC Class Codes: H04L 41/0866 Checking the configuration
• CPC Class Codes: H04L 41/0893 Assignment of logical groups to network elements
• CPC Class Codes: H04L 41/0894 Policy-based network configuration management
• CPC Class Codes: H04L 41/5003 Managing SLA; Interaction between SLA and QoS
• CPC Class Codes: H04L 47/10 Flow control; Congestion control
• CPC Class Codes: H04L 47/193 at the transport layer, e.g. TCP related
• CPC Class Codes: H04L 47/20 Traffic policing
• CPC Class Codes: H04L 47/22 Traffic shaping
• CPC Class Codes: H04L 47/2416 Real-time traffic
• CPC Class Codes: H04L 47/27 Evaluation or update of window size, e.g. using information derived from acknowledged [ACK] packets
• CPC Class Codes: H04W 24/02 Arrangements for optimising operational condition
• CPC Class Codes: H04W 28/02 Traffic management, e.g. flow control or congestion control
• CPC Class Codes: H04W 8/18 Processing of user or subscriber data, e.g. subscribed services, user preferences or user profiles; Transfer of user or subscriber data

Abstract

A system and method for the implementation of fine-grained quality of service in a mobile telecommunications environment uses an Auto-IP Policy Decision Point (PDP) to determine what traffic optimizations actions should be taken in reaction to various network conditions. The Auto-IP PDP uses as inputs information from a G_b interface probe to determine the user identification of an IP data flow, information from the radio access network (RAN) network management system (NMS) regarding network congestion. The cell traffic information is passed onto a traffic analysis and processing engine, the Auto-IP PDP which maps the real-time traffic information into an n-dimensional traffic model. An automated policy decision guiding algorithm is executed on the n-dimensional traffic model and selects policy based on the traffic and cell congestion conditions, without human intervention. Additionally, a consistency check is performed to ensure that new policies are consistent with existing policies. The policy decision outcome is forwarded to the Auto-IP Traffic Optimizer that acts on the network at a point in the network (G_i) that is different from where traffic problem occurs in the RAN. The Auto-IP Traffic Optimizer implements the decisions of the Auto IP-PDP by performing traffic shaping, TCP window clamping or other traffic optimization procedures on a cell-by-cell basis.

Claims

1. In a mobile telecommunications network in which a plurality of mobile stations communicate through a radio access network (RAN) through a network having at least one Serving GPRS Support Node (SGSN) and at least one Gateway GPRS Support Node (GGSN) through the Internet to a plurality of servers, a system for the automatic optimization of traffic comprising:

one or more Gb probes for extracting information from an interface between the RAN and the SGSN including the identity of each user associated with a specific data flow;

a network management system (NMS) for providing congestion and traffic information from the RAN;

one or more IP probes for providing information regarding quantities and types of IP traffic and delays between the mobile stations and the interface;

a customer database for providing information regarding the type and quality of service for each user of a mobile station;

a policy decision point for automatic generation of policy actions to optimize traffic for each data flow on a cell-by-cell basis, said policy decision point using a finite state machine working in conjunction with an n-dimensional traffic model and information from the Gb probes, the NMS and the customer database to determine said policy actions; and

,

a traffic optimizer for implementing the policy actions determined by the policy decision point.

2. The system of claim 1 wherein said n-dimensional traffic model is partitioned into a set of non-overlapping sub-regions corresponding to different states in the finite state machine and predefined policies allow transition between the different states in an automated fashion and each slate transition accompanies a set of Auto-IP actions executed in the traffic optimizer.

3. The system of claim 1 wherein the policy decision point further comprises a means for checking the consistency of new policies with existing policies.

4. The system of claim 1 wherein the traffic optimizer separates traffic arriving from the Internet into components identified by the identity of the cell in the RAN through which the traffic will pass.

5. The system of claim 4 wherein the separated traffic is optimized on a cell-by-cell basis using optimization techniques selected from the group consisting of:

TCP window clamping, admission control mechanisms, compression, TCP ACK spacing, local TCP acknowledgements, IP Traffic shaping, and local TCP retransmissions.

6. The system of claim 1 wherein the traffic optimizer is located at the G_i interface.

7. A method for automatic optimization of traffic flow between users of mobile stations and servers connected through a radio access network (RAN), a packet-based core network and the Internet comprising the steps of:

extracting information from the interface between the RAN and the packet-based network including the identity of each user associated with a specific traffic flow;

collecting information from a network management system (NMS) regarding congestion and traffic information from the RAN;

collecting information from one or more IP probes regarding quantities and types of IP traffic and delays between mobile stations and a Gi interface;

retrieving data from a customer database regarding the type and quality of service for each user of a mobile station;

inputting the extracted RAN information, collected NMS information, and retrieved customer data into a policy decision point;

determining at the policy decision point policy actions to optimize traffic for each data flow on a cell-by-cell basis, said determining based on a finite state machine and an n-dimensional traffic model; and

,

implementing the actions on a traffic optimizer on a cell-by-cell basis.

8. The method of claim 7 wherein the step of determining the actions further comprises:

identifying total quantity of a resource to be managed and a set of traffic classes sharing the resource;

converting the set of traffic classes into an n-dimensional space where n is the number of traffic classes, and the axes represent amounts of traffic of each class;

partitioning the n-dimensional space into regions representing conditions of the network for which traffic constraints are met or not met, wherein each state corresponds to a subset of traffic regions of the n-dimensional space;

for each state not meeting traffic constraints, identifying a set of actions for moving the network into another state meeting the traffic constraints;

creating a state transition diagram with nodes representing each of the traffic regions and with edges between the nodes representing adjacent regions; and

labeling each edge with the type of traffic that causes the transition, and for the edges that move the network from an unacceptable state to an acceptable state, labeling the edge with the name of the action that causes the transition.

9. The method of claim 7 wherein the step of determining the actions to be taken further comprises checking the consistency of new policies with existing policies.

10. The method of claim 7 further comprising the step of separating traffic arriving from the Internet at the traffic optimizer into components identified by the identity of the cell in the RAN through which the traffic will pass.

11. The method of claim 10 wherein the separated traffic is optimized on a cell-by-cell basis using optimization techniques selected from the group consisting of:

TCP window clamping, admission control mechanisms, compression, TCP ACK spacing, local TCP acknowledgements, IP Traffic shaping, and local TCP retransmissions.