9/8/2014 1 A Practical Traffic Management for - PowerPoint Presentation

9/8/2014 1 A Practical Traffic Management for Integrated LTE-WiFi Networks Speaker : Rajesh Mahindra NEC Labs America Hari Viswanathan , Karthik Sundaresan, and Mustafa Arslan

9/8/2014 2 Key TrendsData traffic exploding on cellular networksRise in video streaming, social networking Revenue per byte is decreasing Mobile operators embracing WiFi as a key technology to enhance LTE experience Cheap to deploy – unlicensedEasy (fast) to deploy – unplanned Critical to manage flows across APs-Basestations to maximize QoE and resource utilization

Operator-based WiFi deploymentsAbsence of network-wide traffic managementDevices always connect to WiFi when available (static policy)Past focus has been authentication methods over WiFi 9/8/2014 3

Today: Devices always connect to WiFi 9/8/2014 4

Operator-based WiFi deploymentsAbsence of network-wide traffic managementDevices always connect to WiFi when available (static policy)Past focus has been authentication methods over WiFi 9/8/2014 5 Absence of tight data-plane integration 3GPP based deployments have high CAPEX Requires backhauling WiFi traffic through mobile core Increased investment in infrastructure

9/8/2014 6 Today: Resistance to Tight Integration of LTE and WiFi LTE Core-Network ePDG 3GPP standard WiFi Gateway PDN-gateway MME Serving-gateway INTERNET Increased backhaul costs

Operator-based WiFi deploymentsAbsence of network-wide traffic managementDevices always connect to WiFi when available (static policy)Past focus has been authentication methods over WiFi 9/8/2014 7 Absence of tight data-plane integration 3GPP based deployments have high CAPEX Requires backhauling WiFi traffic through mobile core Increased investment in infrastructure Inability to perform dynamic network selection Result Diminishes the potential effectiveness of WiFi Degrades the user Quality of Experience (QoE)

OpportunityState of the Art: Client-side solutions Qualcomm’s CnE, Interdigital SAMStatic policies (application level) enforced locally on each clientQoE requirements provided by the application on the clientClient-side decision making -> inefficient use of network resourcesOperator agnostic mobile service (MOTA), in Mobicom 2011Requires frequent network state information from each base stationIncompatible with standards -> difficult to deployIndividual decisions by client -> sub-optimal 9/8/2014 8 Inability for Mobile Operators to perform effective network-wide traffic management!

9/8/2014 9 Our Idea: A Traffic Management Solution LTE Core-Network WiFi Gateway Traffic Manager Maps user flows to appropriate network(LTE/ WiFi ) Centralized management -> E fficient use of network resources Reduces backhaul costs -> Facilitates dynamic traffic mgmt Operates for each LTE cell -> Scalable Standards agnostic -> Easily Deployable PDN-gateway MME Serving-gateway Network Interface Assignment Switching Service

9/8/2014 10 ComponentsNetwork Interface Assignment Algorithm (NIA)Goal: Dynamically maps user traffic flows to appropriate LTE basestation or WiFi AP Interface switching service (ISS) Goal: Switch current user flows from WiFi AP to LTE or vice versa based on decisions from NIA

Component 1: NIA9/8/201411

Consider an LTE cell and multiple WiFi APs in its coverage area Assign basestation/ AP to each flowMaximize sum of users flows’ QoEQoE captured using “utility”Weighted PF provides differential QoEPricing function supports 2 modelsBased on data usage Based on price/byteProblem Formulation 9/8/2014 12 Weight Throughput Network Pricing

Throughput ModelsLTE basestation performs weighted PFWiFi AP performs throughput based fairnessAlgorithm does not depend on specific schedulerWiFi APs may perform weighted PF9/9/201413

Problem depiction9/9/201414 3Mbps 5 Mbps 1Mbps 2Mbps 4Mbps 8Mbps

Problem depiction9/9/201415 4Mbps 6Mbps 2Mbps 3Mbps 2Mbps 5Mbps

Problem depiction9/9/201416 5Mbps 7Mbps 3Mbps 3Mbps 3Mbps 7Mbps

Network Interface Assignment (NIA)Problem is NP-HardIncluding the simplest topology of an LTE cell and a WiFi APNIA is a two-step greedy heuristicConsiders each AP-basestation in isolationFixes assignment for AP that maximized incremental utilityIterate till all hotspots are coveredComplexity is O(K2S2), where K = # clients, S = # APs 9/8/201417

NIA ExampleTrigger - arrival/departure of clients or periodicStep 1: In each WiFi hotspot, partition clients into two sets, LTE and WiFi, so that sum of utilities is maximized 9/8/201418

9/8/201419NIA ExampleStep 2: Finalize interface assignment for clients in the WiFi hotspot with the highest incremental utility

9/8/201420NIA Example – IterateRepeat 1&2 with the new initial condition until all hotspots are covered Done!

9/8/201421Component 2: Interface Switching Service

9/8/2014 22 Mid-session network switching capability facilitates dynamic traffic mgmt Leverage HTTP characteristics HTTP traffic ( esp video and browsing) dominates (>90% of internet) Session content(s) are downloaded using multiple HTTP requestsVideo streaming use HTTP-PD (Progressive Download) or DASH (Dynamic Adaptive Streaming over HTTP): A HTTP-GET request/chunkBrowsing : A HTTP-GET request/object DASH Server HTTP TCP Multi-resolution video VIDEO VIDEO VIDEO Clients Design Considerations HTTP GET

Interface Switching Service (ISS)9/9/201423 SwitchInterfaceLTEInterface to NIAHTTP based Video streaming/Browsing Control TrafficWiFi LTE HTTP-GET ISS Controller Application / Browser HTTP Proxy Control Logic Mobile Device Internet Other types of traffic can leverage existing 3GPP standards for seamless interface switching Switch to WiFi

Prototype9/9/201424Linux Laptop(Client)NEC LTE BasestationWiFi Gateway DlinkWiFi APOpenEPCLTE Core HTTP requests Chrome Browser Shrpx HTTP Proxy ISS Control ATOM NIA Algorithm Squid HTTP Proxy Squid HTTP Proxy ISS Control

9/8/201425Experiment 1: Large-scale evaluation Topology: 1 LTE basestation and 3 WiFi APsResult: ATOM performs better than client-side solutions

9/8/201426Experiment 2: Benchmarking the ISS Measured the time taken for flows to switch using ISS:HTTP based video streaming flowsHulu (uses HTTP-DASH) v/s Youtube(uses HTTP-PD)Insight: Switching time improves with DASH streamingDASH flows use smaller chunk sizes to ensure adaptive-ness to changing network conditions

9/8/2014 27 Operators have to look towards exploiting multiple access technologies to increase capacity WiFi offers the cheapest alternate to cellular Our Contributions : a traffic management solution that assigns user flows to LTE basestation/ WiFi APs Low complexity, scalable algorithm for flow assignment Network-based solution more effective than client-side solutions HTTP based switching provides dynamic flow assignment at lower costs Summary

9/8/2014 28