Bernhard Koessler Bernhard.koessler@jdsu...– Intra and Inter-Labs Connections 3GPP Legacy Access...

Bernhard Koessler

[email protected]

Don’t Gamble with Your LTE Trials

and Deployment

© 2010 JDSU. All rights reserved. JDSU CONFIDENTIAL & PROPRIETARY INFORMATION2

A Global Company

Employees: 4,600

Locations: 80+ Sales and R&D sites globally

Global Presence: 164 countries

Annual Revenue: $1.3B

Index Membership: S&P 500


JDSU E2E LTE Portfolio

Challenges

Solution

Benefits

• RF Coverage & Quality

• Handovers

• Application Performance

• Throughput & Latency

• Run a successful NEM

Trial

• Interoperability

• E2E performance & latency

• Correlating radio & network

• Turn up & acceptance

• Fix Service Issues

• Reduce Operational

Costs

• Troubleshooting & testing

• Increased

Complexity

• Scalability

• QoS Management

• VoIP

• RF Network coverage and

application measurements

- LTE & legacy technology

• Protocol Analysis &

Network Diagnostics

for 3G & LTE

technologies

• Complete Ethernet

lifecycle

deployment & operation

• Session Trace,

QoSA, Investigator,

DMT

•One tool: UE & Receiver

Measurements

•All Technologies:1

platform

• Integration with SART

•Full End-to-End Analysis

•Reduce Storage Costs

•Lowest Cost Per User

•Integration with Drive

Test

• Troubleshoot: physical

layer up to the

application

• Continuous Threshold

Monitoring

• Lower Resolution Time

• KPI network monitoring

• Real time reactive

investigations

JDSU offers the only solution to test LTE and existing networks with visibility of both radio & network links

• Services fail even when

network looks normal

• E2E service monitoring

• Test as a customer

would

Drive Test Active Test Network Protocol Test Ethernet Backhaul Assurance

• Scheduled & on-

demand active testing

• 24 x 7 monitoring

• Subscriber perspective

• E2E user perspective

• Alarms on test failures

• Continuous, wide

regional coverage


Industry Recognition of JDSU in LTE

Infonetics Research, Mobile and LTE Continuous Research Service,

Interface, Timing and Interoperability Are Key LTE Testing Issues, July 15, 2010

“Nonetheless, leading testing vendors such as… JDSU…are up to the challenge and are already

racing to stay on top of potential interoperability and pending interworking issues. They all know

that collecting data and developing sophisticated end-user models will be essential for LTE to be a

success story.”

Agilent Technologies, LTE and the Evolution to 4G Wireless: Design and

Measurement Challenges, June 2009

Four JSDU engineers contributed chapters to this book that has been

described as “Very impressive and much much more readable than the

standard documents!”*

*Quote from Martin Sauter at

http://mobilesociety.typepad.com/mobile_life/2009/05/book-review-lte-and-the-

evolution-to-4g-wireless.html

MultiService Forum, LTE Interoperability Event, March 15 – 26, 2010

“We were encouraged not only by the results of the interoperability

trials, but by the fact that a neutral test vendor could deliver such clear,

accurate and comprehensive results across all relevant standards”.

- David Hutton, VP, MSF

Informa Telecoms & Media, LTE World Summit,

Amsterdam, May 18, 2010

JDSU solution of Drive Test for LTE and Signaling

Analyzer Real Time (SART) recognized for LTE

test excellence.


Leverage from the Industry

Many industry organization has performed much

testing and (public) reporting already:

– LSTI

– MSF

– NGMN

Leverage this to enable rapid LTE Trials and

Deployments


MSF LTE IOT Event Scenarios

Basic Interoperability

– Attach/Detach /Tracking Area Updates

– IMS Session Registration/Establishment/Termination

– MME Pooling and Serving Gateway Selection

Roaming

– Home-Routed, Local Breakout (Home and Visited Network P-CSCF)

– Intra and Inter-Labs Connections

3GPP Legacy Access

– S4 and Gn/Gp SGSN Access and Interworking

Handover and Relocation

– Intra-LTE S1 and X2-based Handovers with MME and S-GW Relocations

– Inter-RATs with Legacy 3GPP Accesses

Non-3GPP (eHRPD) Access and Interworking

– eHRPD Access and Interworking

Robustness

– Serving and PDN Gateway over S5/S8 and S4/S11 interfaces

– Simulated Attacks to Locate Known and More Importantly Unknown Vulnerabilities

6


LSTI Activities & Members


LTE Data Rates (LSTI)


Predicted End User Data Rates (LSTI)


Control-Plane Latency: Idle to Active time (LSTI)


SPEED!

LTE/SAE High level requirements Reduced cost per bit

More lower cost services with better user experience

Flexible use of new and existing frequency bands

Simplified lower cost network with open interfaces

Reduced terminal complexity and reasonable power consumption

Spectral Efficiency3-4x HSDPA (downlink)2-3x HSUPA (uplink)

LatencyIdle active < 100 msSmall packets < 5 ms

Downlink peak data rates

(64QAM)

Antenna

configSISO

2x2

MIMO

4x4

MIMO

Peak data

rate Mbps100 172.8 326.4

Uplink peak data rates

(Single antenna)

Modulation QPSK16

QAM

64

QAM

Peak data

rate Mbps50 57.6 86.4

MHz

1.4

3

5

10

15

20

Optimized: 0–15 kph

High Performance: 15-120 kph

Functional: 120–350 kph

Under consideration: 350+ kph

Mobility

Multiple Input Multiple Output

MIMO

Mobile MUST Support!!

LTE – What is it?


Downlink peak data rates (64QAM)

Antenna config SISO 2x2 MIMO 4x4 MIMO

Peak data rate Mbps 100 172.8 326.4

Uplink peak data rates (Single antenna)

Modulation QPSK 16 QAM 64 QAM

Peak data rate Mbps 50 57.6 86.4

Challenges - Adaptive Modulation and Coding (AMC)

eNB

UE1

UE2

UE3

UE4

UE5

64 QAM

16 QAM

QPSK

When will Who have What coverage Where?

64 QAM requires 10dB Carrier to Interference ratio (C/I)

How do you achieve that much isolation without adding more sites?


Efficiency is limited by inter-cell interferenceThe geometry factor

Low SNR High SNR

The geometry factor is the ratio between the serving cell power and interfering cells + noise

At the boundary of any two cells the geometry factor will not exceed -3dB

At the boundary of any three cells the geometry factor will not exceed -4.8 dB

No one has yet invented a spatially aware electromagnetic wave that stops at the desired cell boundary!


Geometry factor distribution in urban cells

Geometry factor in dB

Cum

ula

tive d

istr

ibution

0 %

100 % This plot shows the variation in geometry factor across a typical outdoor urban cell

Very high spectral efficiency is only seen when the geometry factor is above 15 dB, which is an environment that 90% of the user population will not experience

In-building penetration loss will degrade performance further

This puts a finite and very lowlimit on indoor performance when using outdoor transmission systems

90% of users 10% of users

-30 300-20 -10 10 20

Most new high data rate/MIMO performance targets require geometry

factors experienced by <10% of the user population

LTE at the Movies

Moray Rumney and Per Kangru

18th May 2009


W1314A NGR LTE Measurements Summary

• LTE Channel Analyzer:• P-SS RSSI, S-SS RSSI

• LTE Physical Layer Cell-ID

• LTE Reference Signal Analyzer:• Reference Signal RSRP, Reference Signal RSRQ

• Reference Signal CINR

– LTE Measurements run in all bands supported on the W1314A Receiver.

– Supports FDD + TDD Mode

– Supports 1.4 MHz, 3 MHz, 5 MHz, 10 MHz, 15 MHz, 20 MHz

– LTE Measurements run together with other technologies (UMTS, GSM, CDMA,

EVDO, WiMAX)

– Preparation for LTE - Spectrum Monitoring/Clearance prior to trial deployments

– R0 and R1 Measurements (MIMO Coverage Assessment)

450MHz to 3.8GHz frequency range (not contiguous)


Receiver Measurements in E6474A


Blue are R0 signals, Yellow are R1 signals.

Receiver measures reference signals from each transmission port.

– RSRP

– RSRQ

– RS-CINR

Transmit Diversity Measurement with W1314A RX

Reference Signal Mapping for Antenna Diversity


E6474A-X Interference Spectrum Event

Scans the entire Spectrum Trace for detecting and characterizing interferers • Triggers when crossing a user defined threshold

– For user-defined period of seconds

– “Do not repeat” option to avoid continuous triggering

• Reports on the triggering peak

– Center frequency

– Power

– 3dB BW around peak

• Displays in the Map, Spectrograph and other Views


EPC Fundamentals – QoS Model

Today’s 3G Services use best effort QoS Classes parameters for all PS services

• An LTE user can have up to 24 bearers, each with its own QoS Class

• In the EPC, the QoS parameters are a function of guaranteed & non-guaranteed bit rates defined by 9 QCI labels.

How does QoS over backhaul match QoS over RAN?

• The radio access promises to flexibly shape capacity… what will this do to the backhaul?

QoS Class

Identifier

L2 Packet

Delay

Budget

L2

Packet

Loss

Rate

Example Services

QCI=1 (GBR) 100 ms 10-2 Conversational Voice

QCI=2 (GBR) 150 ms 10-3 Conversational Video (Live Streaming)

QCI=3 (GBR) 50 ms 10-3 Real Time Gaming

QCI=4 (GBR) 300 ms 10-6 Non-Conversational Video (Buffered Streaming)

QCI=5 (non-GBR) 100 ms 10-6 IMS Signalling

QCI=6 (non-GBR) 300 ms 10-6 Video (Buffered Streaming)

TCP-based (e.g., www, e-mail, chat, ftp, p2p, video, etc.)

QCI=7 (non-GBR) 100 ms 10-3 Voice, Video (Live Streaming) Interactive Gaming

QCI=8 (non-GBR) 300 ms 10-6Video (Buffered Streaming)

TCP-based (e.g., www, e-mail, chat, ftp, etc.)QCI=9 (non-GBR) 300 ms 10-6


PDN -GWSGWeNBUE

EPC Fundamentals – QoS Model

RB-ID S1-TEID

UL-TFT RB-ID

S1-TEID S5/S8-TEID

S5/S8-TEI DL-TFT

UL Service Data Flows DL Service Data Flows

Application / Service Layer

GBR – Service BlockingReal Time

Non GBR – Service Dropping Non Real-Time

TFT = Traffic Flow Template

Radio Bearer S1 Bearer S5 Bearer


21

• 2x2 MIMO

• HARQ; Round Robin

• Best effort user: fixed SNR (23 dB aprox), UDP transfer

• GBR user: GBR = 10 Mbps, variable SNR, HD video streamingWith decreasing SNR, GBR user maintains desired rate by taking

resources from best effort user, until SNR too low to support GBR.

eNB Scheduler and QoS – An LSTI Illustration


JDSU LTE with Video Streaming Testing


Challenges – Managing Voice Quality

Voice is still the most used application

• Primary revenue generator in communication networks

• Voice Quality is a major factor determining customer satisfaction and operator selection

It is important to know where and why the voice quality may be degraded.

Packet Loss

Voice Quality deteriorates with high

loss rates

Good

0 - 0.5 %

Acceptable

0.5 - 1.5 %

Poor

> 1.5 %

Jitter

A variation in packet transit delay due to network conditions

Good

0 - 20 ms

Acceptable

20 - 50 ms

Poor

> 50 ms

Delay

The one-way delay objective for VoIP

connectivity is 100 ms

Good

0 - 150 ms

Acceptable

150 - 400 ms

Poor

> 400 ms

R-Factor MOS

(ITU-T G.107): A value derived from delay,

jitter, and packet loss.

Good

80-90

Acceptable

65-80

Poor

< 65


Challenges – How will Voice be Delivered??

Two Leading Proposals for CS Services Delivery over EPS Access in R8

• CS Fallback

• Voice over LTE Generic Access (VoLGA)

CS Fallback

• Currently selected by 3GPP as a Part of Release 8

• Connected to E-UTRAN uses GERAN or UTRAN to Establish CS & SMS Services

• Co-Exist with IMS-based Services in Same Operator’s Network

• CS Service over IMS takes precedence over CS Fallback

VoLGA – Voice over LTE via Generic Access

• Connected and Uses EPS to Establish One or More CS Services, via Legacy CN Infrastructure

• Does Not Require E-UTRAN Coverage Overlap with GERAN or UTRAN Coverage

• Co-Exist with IMS-based Services and CS Fallback in Same Operator’s Network

• Simultaneous Use of VoLGA & IMS Services Possible, while Not So with CS Fallback

LTE One Voice – The Future of Voice for LTE ?

• Selected for LTE Release 9

• IMS Based Implementation


JDSU LTE with VOIP Testing


UTRAN PS Domain

LTE UTRAN

E2E LTE QoS/QoE Monitoring (VoIP, IPTV, Data)

UMTS Core PS Doain

GPRS Core

SGSN GGSN

Radio Network Subsystem

Iub

UuRNC Iu-PS

NodeB

S1-U

MMEeNB

eNB

Xu

MME

S1-MME

PDN

Gateway

X2

S5

SGi

HSS

S6

S3 S4

PCR

F

S7

PDN

Rx+

SAE-GWS10

S11

MRF CSCF AS

IMS

S1-MME

Gn

Agilent J6900A Triple

Play Analyzer

Passive monitoring

Gi

Agilent E6474A NiXt

Mobile drive test

Active test


Challenges: Diversity of Transport Options

IP

Ethernet

PHY

L2

L3

copperfibre micro-wave

mc-ml ppp

PDHSDH/SONET DSL

P2PPON

ATM

(IMA)

P2PP2MP

WiMAX

DWDM

MPLS

MEDIA

CWDMTECHNIQUE

Details of protocols used for

encapsulation are not shown

in this diagram


Full 7 – Layer

Analysis:

• Decodes

• Protocol statistics

• Node Discovery and Stats

• Connection statistics by

protocol and user

• Network Vitals

• Protocol Vitals

• TCP Connection Analyzer

• Expert Analysis and

Commentators

• Active tests – Ping, ARP,

Playback capture buffer

Supports over 500 protocols

RTP/RTCP analysis with Telephony Network Analyzer

Report center – Log measurements/stats and automatically

create reports

Stream traffic to Disk at up to 5 Gbps

The Network Analyzer Application Solution


Multi-Vendor Interoperability is Crucial!!

eNB

Vendor A

S1 S1

X2

EPC

Vendor B

eNB

Vendor A

Uu

IOT S1

Validate eNodeB

from one vendor

and EPC from

different vendor

S1 S1

X2

EPC

eNB

Vendor AeNB

Vendor B

Uu

IOT X2

Validate eNodeB

from different

vendors


Whole Network Visibility with End2End Call Trace

MSF LTE IOT Event Test Summary: Scenario 1a

– Attach Procedure: E2E Multi-Interface Call Trace

30


Performance Benchmarking and Validation SART KPIs Viewer


Evolved 3GPP Network Architecture

Evolved Packet Core (EPC)

Serving NetworkGSM/GPRS/EDGE Radio Access Network (GERAN)

UMTS Terrestrial Radio Access Network (UTRAN)

Core & Services Network

CS Core (Voice)

SGSN

MGWMSC

VLR HLRSMS

PCRF

NodeB

RNC

BTS

BSC

PCU

Evolved UTRAN (E-UTRAN)

UE

MS

Uu

Abis

Iub

RNC

Iur

UEX2

Uu eNB

eNB

PS Core (Data)

GSM-A

CS

PS

GPRS-Gb

PS

CS

Iu-CS

Iu-PS

S11

SAE-GW

MME

S5

S4S3

S1-UEvolved Node B

Radio Network Subsystem

Base Station Subsystem

S6a

Gx

SGW

SS7/SIGTRAN

RX

SGi

WCDMA HSPA/HSPA+

GSMEDGEEvolved EDGE

Um

S1-C

S12

GnInternet

Operator’s IP Services

(E.g. IMS, PSS…)

P-GW

GGSN

Supported Interfaces

HSS CSCF MRF


Challenges – Different Interfaces = Different Info

Serving Gateway

Mobility Anchoring

MME

NAS Security

Idle State Mobility Handling

SAE Bearer Control

eNB

Inter Cell RRM

RB Control

Connection Mobility Control

Radio Admission Control

eNB Measurement

Configuration & Provisioning

Dynamic Resource

Allocation (Scheduler)

RRC

PDCP

RLC

MAC

PHY S1

PDN Gateway

UE IP Address allocation

Packet Filtering

How do you do RF Performance Engineering & Optimization, when all these functions are

closed within the eNB?

E-UTRAN Evolved Packet CoreInternet


The Air Interface Problem

– How do you see both sides of the eNodeB?• UE and core messaging is very difficult and time-consuming to

analyze manually. This is automated in the JDSU solution.

– How do you identify Control Plane vs. User Plane issues?• Correlated messaging can show messaging delays that can

affect setup, teardown, call progress, and handover – all service affecting.

– How do you see performance affecting conditions ? RAN vs. Network?

• Correlation of Radio Layer with Core metrics and Events can easily show differentials in reported service level quality across radio and core network interfaces. e.g. Throughput, UE BLER


JDSU Methodology for LTE Testing

EPCE-UTRAN eNodeBMME

SGW

PDN

GW

IMS

CSCF

HSS

AS

MGCF

eNodeB

TraceReceiver

KPI’s

Device

KPI’s

& Trace

• Faster and more comprehensive troubleshooting

– Detailed end-to-end picture of the true network issues/performance

– Optimize both the uplink and downlink

– Isolate node Bs with RF problems

• Optimize high speed services

– Detect messages lost over the air interface

– Determine the overall RF performance

Signaling

KPI’s

& Trace

JDSU

Drive TestImpossible to optimize LTE, HSDPA & HSUPA without visibility to both radio & network links

Signaling

KPI’s

& Trace

Signaling

KPI’s

& Trace

JDSU

Signaling Analyzer


Mobile Phone Trace Data

• L1, MAC, RLC, RRC

•MIMO Performance

Signaling Analyzer

E2E Call Trace, KPIs,

TCP/IP Performance

JDSU NiXt Drive Test Integration

RF Receiver Measurements

• RSRP, RSRQ, RS CINR

•Spectrum, CW/channel power

Application Test

• HTTP, FTP, VoIP & Video

streaming MOS

S1-U

MMEeNB

eNB MME

S1-MME

PDN

Gateway

X2

S5

SGi

SAE-GWS10

S11

S1-MME

HSS

S6

PCRF

MRF CSC

F

AS

IMS

Triple Play Analyzer

VoIP MOS, Video MOS,

QoE Analysis


Drivetest integration: Combined Uu & S1 log

S1 network interface trace

correalted to Uu RRC

message from drive test log











Drivetest integration: 2 LTE UE’s downloading in //


Drive Test Uu Data as seen in JDSU SART

During Handover EPCE-UTRAN eNodeB

MME

SGW

PDN

GW

IMS

CSCF

HSS

AS

MGCF


Thank You!Questions?

Bernhard Koessler Bernhard.koessler@jdsu...– Intra and Inter-Labs Connections 3GPP Legacy Access...

Documents

Transcript of Bernhard Koessler Bernhard.koessler@jdsu...– Intra and Inter-Labs Connections 3GPP Legacy Access...