LTE TDD What to Test and Why

© 2012 LitePoint Corp.

© 2 0 1 2 L i t e P o i n t , A T e r a d yn e C o m p a n y. Al l r i g h t s r e s e r ve d .
Agenda
• LTE Overview
• LTE Measurements
• Testing LTE TDD – Where to Begin?
• Building a LTE TDD Verification Plan
• Optimizing a LTE TDD Verification Plan

© 2012 LitePoint Corp. •32

Why LTE? – Something for Everyone
For the user..
Higher Performance (Data Rate)
 Instantaneous downlink peak data rate:
150 Mbit/s in a 20MHz downlink spectrum (5 bit/s/Hz)
 Instantaneous uplink peak data rate:
75 Mbit/
Mbit/s iin a 20MH
20MHz uplink
li k spectrum
t (2
(2.5
5 bit/
bit/s/Hz)
/H )

For the service provider…

Cell capacity – more users per cell
 up to 200 active users per cell (5 MHz) (i.e., 200 active data clients)

1st all-data network: packet-switched

 Simplifies network architecture – no difference between voice & data

© 2012 LitePoint Corp. •33

Key Features of LTE
Multiple access scheme
• Downlink (DL): OFDMA enables maximum spectrum utilization by the base station
• Uplink (UL): SC-FDMA
SC FDMA relaxes the linearity requirements for the handset

Multiple Uplink Transmission Modes

• FDD: balanced DL / UL data traffic by using different channels
• TDD: enables asymmetric
y DL / UL capacity,
p y, sharing
g a single
g channel

Adaptive Modulation and Coding

LTE dynamically changes modulation based on channel conditions to optimize its capacity
• DL modulations: QPSK, 16QAM, and 64QAM
• UL modulations: QPSK and 16QAM

Channel Bandwidth Scalability

Scalable channel bandwidth allows efficient operation in differently-sized allocated spectrum bands

Multiple Antenna Technology – MiMo

Multiple Antenna techniques enables higher data rate, improve network reliability and data capacity

© 2012 LitePoint Corp. •34

OFDM meets Cellular…
• LTE is the first cellular standard to use OFDMA modulation
- Combining time and frequency multiplexing, enabling multiple users to
operate in a single time slot

OFDMA

•OFDM Modulation

© 2012 LitePoint Corp. •35

OFDMA Highlights
• LTE uses OFDMA for the downlink
- Uses a large number of narrow sub-carriers for
multi-carrier transmission
- “Resource blocks” and “elements”
•Each resource block and element is defined in
“frequency” and “time” (1 block = 180 kHz; 0.5 ms)
- Dynamically assigns these resource blocks to
LTE users, thus improving spectrum utilization
- Subcarrier spacing – 15 kHz compared to
312 .5 kHz for WLAN

•The basic LTE downlink physical resource

can be seen as a time-frequency grid:

© 2012 LitePoint Corp. •36

SC-FDMA
• The LTE uplink transmission scheme based on a pre-coded version of OFDMA
known as SC-FDMA (Single Carrier Frequency Division Multiple Access).

• SC-FDMA operates with a lower Peak to Average Power Ratio (PAPR) than OFDM
- High PAPR requires expensive and inefficient power amplifiers

• SC-FDMA reduces the linearity

requirement for power amplifier

• Two LTE UL transmission modes:

- FDD
- TDD

© 2012 LitePoint Corp. •37

LTE-FDD & LTE-TDD
FDD
• downlink and uplink traffic is transmitted simultaneously at separate carrier frequencies
• is the preferred mode by most cellular systems,
systems wherever paired spectrum is available –
easy transition from existing 3G networks

TDD
• transmission in uplink and downlink is at the same carrier frequency
• is a good option where spectrum (carriers) availability is lower
• is necessary when pair spectrum is not available

FDD

fDL
TDD

fUL fDL/UL

time time

© 2012 LitePoint Corp. •38

LTE-FDD vs. LTE-TDD
• The two versions of LTE are actually quite similar
• The only differences are in the physical layer, enabling support of both
TDD and FDD with a single chipset
- All major LTE chipset vendors have released chipsets that support both FDD
and TDD

Parameter LTE-FDD LTE-TDD

Paired spectrum Requires spectrum pairs – TX No spectrum pair required – TX and
and RX on different frequencies RX on the same frequency
UL / DL asymmetry Data capacity determined by Possible to dynamically change
spectrum allocations UL / DL to meet capacity demand
Guard interval impact on Increasing guard interval (due to Increasing guard interval (due to
data capacity distance from base station) does distance from base station) reduces
not impact data capacity data capacity

© 2012 LitePoint Corp. •39

•LTE Growth – China and LTE-TDD will play a key
role

•2011 •2016

•It
It is
i expected
t d that
th t by
b 2016,
2016 Chi
China M
Mobile
bil will
ill representt over 15 percentt off the
th total
t t l LTE
market, with its TDD LTE deployment.

© 2012 LitePoint Corp. •Source: Signals and Systems Telecom 4/2012 •40
New Challenges in LTE – More Bands

Band Frequency Channel Mode

Range Bandwidths
1.4, 3, 5, 10,
33 to 41 <2.69
69 G
GHz TDD
15 20 MH
15, MHz

•More bands means more test time

© 2012 LitePoint Corp. •41
New Challenges in LTE – More Configurations
• LTE has many configurations to test – more test time
- Per channel…

Modulation RB Config PWR Levels

QPSK 50,0 4
QPSK 12,0 4
QPSK 12,38 2
QPSK 1,0 1
QPSK 1,24 1
QPSK 1 49
1,49 1
16 QAM 50,0 2
16 QAM 12,38 2
16 QAM 12 0
12,0 2
16 QAM 50,0 1
64 QAM 50,0 1

•LTE threatens to reduce test throughput…

Higher
© 2012 LitePoint Corp. cost test? •42
New Challenges in LTE – More Bandwidth
• Spectrum Emission Mask (SEM): 6.6.2.1
• Adjacent Channel Leakage Ratio (ACLR): 6.6.2.3

•SE
SE M
Mask
k

•20 MHz
•Limit: -25 dB Channel

•25 MHz •25 MHz

•SEM = 70 MHz Total Bandwidth

© 2012 LitePoint Corp. •43

Testing LTE: Key Requirements
RF Frequency Range
• The test equipment must support the frequency bands 698 MHz - 2690 MHz
• The test equipment must support handsets with an increasing number of antennas

VSA / VSG Bandwidth

• The test equipment
q p must have at least 20 MHz VSA/VSG bandwidth
- LTE requires support for six channel bandwidths (from 1.4 to 20 MHz)
- With LTE-Advanced, this requirement will become 100 MHz
- >70 MHz required for single-shot LTE ACLR & Spectrum Emission Mask testing

MiMo Technology
• Support for accurate MiMo testing is necessary in both R&D and MFG
• I particular,
In ti l it iis essential
ti l tto h
have multiple
lti l VSA / VSG ports
t for
f DL / UL MIMO signal
i l

Transmission Schemes
• Supportt two
S t transmission
t i i schemes
h f downlink
for d li k and
d uplink
li k (OFDMA,
(OFDMA SC
SC-FDMA)
FDMA)
• Support two transmission modes (FDD and TDD)
© 2012 LitePoint Corp. •44
Testing LTE TDD: Where to Begin?
• LTE complexity introduces more than
10x configurations to test
- Testing every scenario is not practical

• In production, we are looking to validate

manufacturing quality

• Goal is to exercise the mobile as much

as possible while minimizing test time

© 2012 LitePoint Corp. •45

Testing LTE TDD: Where to Begin?
• What to test in mobile manufacturing verification:
- Physical layer RF measurements
- TX power
- TX modulation quality
- TX frequency
- TX / RX timing
- RX sensitivity (min / max)

• What NOT to test in mobile

manufacturing verification
- Software
- Digital Design
- Redundant (overlapping) tests or configurations

© 2012 LitePoint Corp. •46

LTE UE Transmitter Tests
Measurement Why is this Important?
LTE network performance is highly dependent on accurate
TX Power
power control
p
Primary TX quality measurement – detects distortions that will
Error Vector Magnitude
ultimately degrade accurate transmission of data
Frequency Error Critically important to avoid communication interference
Ensures that transmission does not interfere with neighboring
ACLR
channels
Occupied Bandwidth Confirms that signal is contained within channel allocation
Ensures that signal in adjacent channels rolls off to minimize
Spectrum Emissions Mask
interference
Carrier Leakage An indication of mismatch in the I/Q modulator
Verifies UE timing accuracy – particularly important for LTE
Transmit Time Mask
TDD since the UL/DL are on the same frequency
In-Band Emissions for Ensures that a UE’s assigned RBs (within a channel) do not
non-allocated RBs interfere with the unassigned RBs in the channel
•3GPP Measurements

© 2012 LitePoint Corp. •47

LTE UE Receiver Tests

• TX measurements give direct access to the signal via the UE antenna

• Unlike TX measurements
measurements, RX signal quality issues remain buried until the
signal is fully decoded
Measurement Notes
Fundamental test of a receiver’s ability to decode the inbound
RX Bit Error Rate (BER)
signal. Typically performed at both min & max RX input power
Receive signal strength is a parameter often measured as part
of calibration.
calibration Since the initial TX power level is calculated per
RX Sensitivity (RSSI)
the measured RSSI, accuracy of this measurement directly
impacts UE power transmission
•3GPP Measurements

© 2012 LitePoint Corp. •48

LTE Test Plan Development

• Several different approaches to develop a test plan:

- Use the 3GPP standard’s recommendations

- Use
U ththe IC manufacturer’s
f t ’ recommendations
d ti

- Use historical data from similar devices

- Apply some reasonable logic to look for

likely failure modes and apply 3GPP spec conditions

© 2012 LitePoint Corp. •49

Building a LTE TDD TX Verification Test Plan
Per-Band
Per Band / Per-Channel
Per Channel
A “reasonable” LTE test plan covered in 21 configurations
Showing config 1-11

•Varies in RB Offset for •Varies TX Power

RB = 1, QPSK channel for RB = 12 QPSK channel

Test Configuration
Parameters 1 2 3 4 5 6 7 8 9 10 11
TX Power +23 +23 +23 +23 +3.2 -30 -40 +23 +3.2 -30 -40
Modulation QPSK QPSK QPSK QPSK QPSK QPSK QPSK QPSK QPSK QPSK QPSK
RB 1 1 1 12 12 12 12 12 12 12 12
RB Offset 0 24 49 0 0 0 0 38 38 38 38
DL Power -57 -57 -57 -57 -57 -57 -57 -57 -57 -57 -57
Measurements 1 2 3 4 5 6 7 8 9 10 11
Power
EVM
EVM Flatness
•RB Offset
Frequency Accuracy
Carrier Feedthrough Extremes
TX Time Mask
Occupied Bandwidth
ACLR
SEM
In-Band Emissions for
Non-Allocated RBs

© 2012 LitePoint Corp. •50

Building a LTE TDD TX Verification Test Plan
Per-Band
Per Band / Per-Channel
Per Channel
A “reasonable” LTE test plan covered in 21 configurations
Showing config 12-21
•Min / Max •Min / Max •Min / Max
Power •Tests Absolute Power for 16 Power for 16
•for QPSK Power Setting QAM QAM
RB = 50 •RB = 12 •RB = 50

Test Configuration
Parameters 12 13 14 15 16 17 18 19 20 21
TX Power +23 -40 +6.4 -5.6 +23 -40 +23 -40 +23 -40
Modulation QPSK QPSK QPSK QPSK 16QAM 16QAM 16QAM 16QAM 16QAM 16QAM
RB 50 50 50 50 12 12 12 12 50 50
RB Offset 0 0 0 0 0 0 38 38 0 0
DL Power -57 -57 -57 -57 -57 -57 -57 -57 -57 -57
Measurements 12 13 14 15 16 17 18 19 20 21
Power
EVM
EVM Flatness •RB Offset
Frequency Accuracy Extremes
Carrier Feedthrough
TX Time Mask
Occupied Bandwidth
ACLR
SEM
In-Band Emissions for
Non Allocated RBs
Non-Allocated

© 2012 LitePoint Corp. •51

 Optimizing the TX Test Plan
•Configurations 1, 3, 12, & 20 test the extremes of modulation and RB allocations / offsets
•Configuration 4 is a “typical” use case

Test Configuration
Parameters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
TX Power +23 +23 +23 +23 +3.2 -30 -40 +23 +3.2 -30 -40 +23 -40 +6.4 -5.6 +23 -40 +23 -40 +23 -40
Modulation QPSK QPSK QPSK QPSK QPSK QPSK QPSK QPSK QPSK QPSK QPSK QPSK QPSK QPSK QPSK 16QAM 16QAM 16QAM 16QAM 16QAM 16QAM
RB 1 1 1 12 12 12 12 12 12 12 12 50 50 50 50 12 12 12 12 50 50
RB Offset 0 24 49 0 0 0 0 38 38 38 38 0 0 0 0 0 0 38 38 0 0
DL Power -57 -57 -57 -57 -57 -57 -57 -57 -57 -57 -57 -57 -57 -57 -57 -57 -57 -57 -57 -57 -57
Measurements 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
Power
EVM
EVM Flatness
Frequency Accuracy
Carrier Feedthrough
TX Time Mask
Occupied Bandwidth
ACLR
SEM
In-Band Emissions for
Non-Allocated RBs

Can be covered
No Need for Mid- Covered by Already Tested Band Covered by Covered by Config 20 &
by any absolute
Channel Offset Config 21 Edges in Configs 1 & 3 Config 21 21, do not need mid-RB
power setting

•Configurations we
definitely want to keep

© 2012 LitePoint Corp. •52

Condensed Test Plan
• Reduced to 7 TX configurations
• Added RX tests
• Increases number of measurements while reducing test time
Test Configuration
Parameters T1 T2 T3 T4 T5 T6 T7
TX Power +23 +23 +23 +23 +3.2 +23 -40
Modulation QPSK QPSK QPSK QPSK QPSK 16QAM 16QAM
RB (UL / DL) 1 1 12 50 12 12 50
RB Offset 0 49 24 0 24 0 0
DL Power -57 -57 -57 -94 -57 -25 -60
Measurements TX1 TX2 TX3 TX4 TX5 TX6 TX7
Power MPR MPR MPR
EVM
EVM Flatness
Frequency Accuracy
Carrier Feedthrough
TX Time Mask
Occupied Bandwidth
ACLR
SEM
In-Band Emissions for
N All t d RB
Non-Allocated RBs
Measurements RX1 RX2 RX3
RX BER
© 2012 LitePoint Corp.RX Level •53
LTE TDD Manufacturing Test

• LTE increases test complexity 5 to 10X

- More measurements, more antennas, wider bandwidth, higher
g p
performance
- IQxstream’s unique architecture makes LTE test simple and fast

• Test plan development for LTE needs to focus on exercising the mobile
device with the minimum test time

• A test plan can be created to maximize

the coverage of the device by using the
test equipment in an efficient manner
- Number of configurations take more test time than number of tests
- Scale test plan to multi-DUT through turnkey non-signaling solutions
- No sacrifice in product quality with shorter per-DUT test times

© 2012 LitePoint Corp. •54