Back to results
US 2010/0149971 A1Published application
System and method for multi-services packet network traffic engineering
AssigneeAT&T Intellectual Property I, L.P.
InventorDimas Noriega
Life of this patent
Estimating the term…
Abstract
Systems and methods are described that provide network traffic engineering that obviate network over-provisioning by providing QoS to each traffic class. Embodiments dimension switching router LTE schedulers to ensure that each traffic class receives an appropriate QoS in terms of delay, jitter, Packet Loss Ratio and throughput.
Claims42
A Traffic Engineering (TE) method for a Long Term Evolution (LTE) multimedia IP network that obviates network over-provisioning by providing Quality of Service (QoS) to traffic classes comprising:
- acquiring Key Performance Indicators (KPI) and predefined Service Level Specifications (SLSs);
- partitioning S1/X2 interfaces into a plurality of Ethernet Virtual Connections (EVCs), with each EVC supporting one traffic class;
- dividing the total capacity of each S1/X2 interface among the classes it supports;
- constructing a per-class traffic table for each LTE switch-router interface;
- comparing each LTE switch-router traffic table with a short term traffic forecast table for that LTE switch-router;
- if the comparison is a good fit, acquiring network S1/X2 link current bandwidth profiles for each EVC and if not a good fit, using smoothing filters to remove transitory peaks in an LTE switch-router's per-class traffic table;
- for each S1/X2 link EVC, calculating a Committed Information Rate (CIR) bandwidth and a Peak Information Rate (PIR) bandwidth for each traffic class, class aggregate and link aggregate basis;
- and
- if the S1/X2 link EVC per class CIR and PIR bandwidths satisfy equipment capacity constrains, installing the S1/X2 link EVC per class CIR and PIR bandwidths in their respective LTE switch-routers via an Element Management System (EMS) as part of the network configuration.
6 more elements
A Traffic Engineering (TE) system for a Long Term Evolution (LTE) multimedia IP network that obviates network over-provisioning by providing Quality of Service (QoS) to traffic classes comprising:
- an LTE Connection Broker (LCB) configured for:
- acquiring Key Performance Indicators (KPI) and predefined Service Level Specifications (SLSs);
- partitioning S1/X2 interfaces into a plurality of Ethernet Virtual Connections (EVCs), with each EVC supporting one traffic class;
- dividing the total capacity of each S1/X2 interface among the classes it supports;
- constructing a per-class traffic table for each LTE switch-router interface;
- comparing each LTE switch-router traffic table with a short term traffic forecast table for that LTE switch-router;
- if the comparison is a good fit, acquiring network S1/X2 link current bandwidth profiles for each EVC and if not a good fit, using smoothing filters to remove transitory peaks in an LTE switch-router's per-class traffic table;
- for each S1/X2 link EVC, calculating a Committed Information Rate (CIR) bandwidth and a Peak Information Rate (PIR) bandwidth for each traffic class, class aggregate and link aggregate basis;
- and
- if the S1/X2 link EVC per class CIR and PIR bandwidths satisfy equipment capacity constrains, installing the S1/X2 link EVC per class CIR and PIR bandwidths in their respective LTE switch-routers via an Element Management System (EMS) as part of the network configuration.
7 more elements
A Traffic Engineering (TE) method for a Long Term Evolution (LTE) multimedia IP network that calculates a bandwidth profile parameter for defined IP multimedia network traffic classes comprising:
- defining a set of Ethernet Virtual Connections (EVCs) for the network having elastic bandwidths;
- receiving network traffic measurements and network state information such as per class call blocking;
- for each network LTE switch-router:
- acquiring network Service Level Specifications (SLSs);
- acquiring a switch-router bandwidth profile;
- setting EVC bandwidths for each traffic class;
- reading defined network traffic class A and class B traffic;
- calculating a class A and a class B bandwidth;
- if the class A and class B calculated bandwidths comply with the SLS, setting Committed Information Rates (CIRs) and Peak Information Rates (PIRs) for the class A and B traffic, and if the class A and class B bandwidths do not comply with the SLS, recalculating the class A and B bandwidths;
- reading defined network traffic class C and class D traffic;
- calculating a class C and a class D bandwidth;
- and
- if the class C and class D bandwidths comply with the switching router SLS, setting Committed Information Rates (CIRs) and Peak Information Rates (PIRs) for the class C and D traffic, and if the class C and class D bandwidths do not comply with the switching router SLS, recalculating the class C and D bandwidths.
10 more elements
A Traffic Engineering (TE) system for a Long Term Evolution (LTE) multimedia IP network that calculates a bandwidth profile parameter for defined IP multimedia network traffic classes comprising:
- an LTE Connection Broker (LCB) configured for:
- defining a set of Ethernet Virtual Connections (EVCs) for the network having elastic bandwidths;
- receiving network traffic measurements and network state information such as per class call blocking;
- for each network LTE switch-router:
- acquiring network Service Level Specifications (SLSs);
- acquiring a switch-router bandwidth profile;
- setting EVC bandwidths for each traffic class;
- reading defined network traffic class A and class B traffic;
- calculating a class A and a class B bandwidth;
- if the class A and class B calculated bandwidths comply with the SLS, setting Committed Information Rates (CIRs) and Peak Information Rates (PIRs) for the class A and B traffic, and if the class A and class B bandwidths do not comply with the SLS, recalculating the class A and B bandwidths;
- reading defined network traffic class C and class D traffic;
- calculating a class C and a class D bandwidth;
- and
- if the class C and class D bandwidths comply with the switching router SLS, setting Committed Information Rates (CIRs) and Peak Information Rates (PIRs) for the class C and D traffic, and if the class C and class D bandwidths do not comply with the switching router SLS, recalculating the class C and D bandwidths.
11 more elements
Dependencies are inferred from the claim text.