5G Antenna Talk TW

Simulation-Enabled 5G
Antenna Design
Martin Leung
CST COMPUTER SIMULATION TECHNOLOGY | www.cst.com

Agenda
1. 5G Motivation and Challenges
2. Antenna Array Simulation, MIMO
3. Mobile Device Antenna Array Example

Element -> Array Design
Device level performance
System level performance

5G Motivation and Challenges
Consumer device bandwidth requirements increasing at
rapid rates
Historical (4, 3G) bands below 3 GHz increasingly

crowded
Push for 28+ GHz frequency for 5G

subject to relatively high atmospheric attenuation
- smaller wavelength more susceptible to multipath loss

5G Motivation and Challenges
Efficiency is critical, Gain of 12 dB for mobile device and 25 dB targeted
Gain cannot come at the expense of coverage beam steered arrays ideal solution
Diversity Gain can also be leveraged - MIMO
High frequency and array topology pose simulation challenges (memory, system
complexity)

Antenna Arrays Categories
Large Arrays Small Arrays

Small Arrays
All elements need to be simulated.

Distributed Computing
Multiple ports, frequency points, and parameter values
can be simulated simultaneously.
Port 1
CST STUDIO frontend Port 2

with 22 port model
Port 3

Port 22

Array Postprocessing
Use Combine Results and Correlation Coefficient result
templates to achieve a desired beam shape.
Purely postprocessing, no further 3D simulations
needed!
How can I achieve

a pattern like this?
All amplitudes = 1 Amplitudes = ????

Array Postprocessing
Initial beam shape
Template Based
Postprocessing as a
Simulation type
Parameters for
amplitude weights
Desired beam shape Achieved beam shape

Array Feeding Networks
CST DESIGN STUDIO
ideal parts
analytical blocks
other 3D models
CST MICROWAVE STUDIO

(nearfield and farfield)
results updated to take
into account the
attached circuitry!

Large Arrays
Can be approximated as an
infinite array which
requires the simulation of
only a single element.
To include edge effects

and other physical
realities, the entire array
needs to be modeled.

Finite Array Analysis
As the array becomes larger, the center element pattern
starts to resemble the infinite array results.

CST ARRAY WIZARD

CST Array Wizard Sim Proj
Grid Angle 60

Drag&Drop your Unit Array Element

Finite Array Sim. Proj.
Set Elements Active/Passive/Empty

MIMO Antennas

Communication Networks
Communication antennas need to cope with a complex environment
All base-station antennas are Multi-path signal transmission

placed near the horizontal plane may lead to destructive signal
Mean Effective Gain (MEG) overlay, resulting in local deep
dips (called Rayleigh-Fading)
Diversity/MIMO
Tilted reflecting planes change

signal polarization
Cross Polarization Rate (XPR)

Diversity / MIMO Antennas
Simple Maximal Signal Diversity Gain
Multi-path signal transmission may
lead to destructive signal overlay
resulting in local deep dips
(called Rayleigh-Fading)
Antenna 1
Antenna 2
Multiple antennas (antenna diversity)
may overcome this problem
best of (diversity gain)

Diversity / MIMO Antennas
Select from:
Diversity Gain
Envelope Correlation Coefficient
Multiplexing Efficiency
Mean Effective Gain
Load farfield of second antenna
Envelope Correlation from Farfield: Envelope Corr. from S-Parameters:

Mobile Antenna Array
Element -> Array Design

Device level performance
System level performance

Antenna Magus

Mobile Antenna Element Design
Magus Antenna Synthesis generates initial square

patch design for 28 GHz operating frequency

Mobile Antenna Element Performance
Magus also provides

performance estimation for
initial design qualification
Initial design is clearly

approximate Microwave
Studio can be used to optimize
and modify

Mobile Antenna Element Performance
3D Simulation Results show similar performance issues in initial design

Optimization

Results
Retuned for 28 GHz, at the expense of

impedance match quality

Mobile Antenna Array Synthesis
Synthesize planar array for 8 dB gain from

Array Factor @ 28 GHz

Mobile Antenna Array Synthesis

Finite Array Generation

Finite Array Performance

Best Achievable Pattern
Envelope Theta = 0 deg
Phi = 0 deg
Theta = 15 deg
Phi = 30 deg

Achievable Array
Patterns Theta=10
Theta=45 Theta=85

Installed Performance

System Assembly Modelling
System Assembly Modelling provides

convenient layout assembly to bring in
phone + other antenna models
Several copies of array independently

placed and parameterizable

Layout Modelling
After wiring system, switch to layout view
and align antenna arrays

Layout Modelling
Lower hemispherical array needs

to be flipped to aim downward
Rotate on axis for diversity

Simulation Task
In this case, all models will

be solved together in 3D by
MWS T solver

Simulation Performance
3.5 Minutes simulation time per excitation on dual CPU workstation with nVIDIA
Kepler 20 GPU (10% GPU capacity)
3.48 GB of system Memory utilized for initial Matrix Calculation

Installed Performance - Coupling
Both arrays operating (boresight), pattern intact

In band coupling to WIFI antenna, GPS coupling at 5G operating freq
Designing Clean Mobile Devices CST Webinar available at
https://www.cst.com/webinar14-10-23~

Installed Performance - Envelope
Relatively, low coupling to GPS at 28 GHz, but radiation angles towards GPS antenna
are affected
Best Achievable Pattern envelope clearly shows performance degraded

System Level Bio Models

Posable CTIA Hand Models
CST Installation Help Files include both static and posable CTIA Hand (+spacer)
models

System Assembly Modelling
Add posed hand model to System

model and align to spacer in Layout
mode

Simulation Performance
15 Minutes simulation time per excitation

on dual CPU workstation with nVIDIA Kepler
20 GPU (~35% GPU capacity)
8.5 GB of system Memory utilized for initial

Matrix Calculation

Effects of Hand Model
Top hemisphere array is largely unaffected
Bottom array highly dependent on finger proximity

Summary
5G Antenna Device design will require high efficiency devices at
frequencies approaching mm wave
CST Array Simulation Workflows provide smooth design process for

large and small finite arrays
Antenna Magus provides Antenna and Array synthesis for rapid

design exploration
System level simulation increasingly important for antenna

performance
Thank you for your attention

5G Antenna Talk TW

Uploaded by

Copyright:

Available Formats

5G Antenna Talk TW

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

5G Antenna Talk TW

Uploaded by

Copyright:

Available Formats

Simulation-Enabled 5G

CST COMPUTER SIMULATION TECHNOLOGY | www.cst.com

2. Antenna Array Simulation, MIMO

3. Mobile Device Antenna Array Example

CST COMPUTER SIMULATION TECHNOLOGY | www.cst.com

Historical (4, 3G) bands below 3 GHz increasingly

Push for 28+ GHz frequency for 5G

CST COMPUTER SIMULATION TECHNOLOGY | www.cst.com

Diversity Gain can also be leveraged - MIMO

CST COMPUTER SIMULATION TECHNOLOGY | www.cst.com

CST COMPUTER SIMULATION TECHNOLOGY | www.cst.com

CST COMPUTER SIMULATION TECHNOLOGY | www.cst.com

CST STUDIO frontend Port 2

CST COMPUTER SIMULATION TECHNOLOGY | www.cst.com

How can I achieve

All amplitudes = 1 Amplitudes = ????

Desired beam shape Achieved beam shape

CST COMPUTER SIMULATION TECHNOLOGY | www.cst.com

CST MICROWAVE STUDIO

CST COMPUTER SIMULATION TECHNOLOGY | www.cst.com

To include edge effects

CST COMPUTER SIMULATION TECHNOLOGY | www.cst.com

CST COMPUTER SIMULATION TECHNOLOGY | www.cst.com

CST COMPUTER SIMULATION TECHNOLOGY | www.cst.com

CST COMPUTER SIMULATION TECHNOLOGY | www.cst.com

CST COMPUTER SIMULATION TECHNOLOGY | www.cst.com

Set Elements Active/Passive/Empty

CST COMPUTER SIMULATION TECHNOLOGY | www.cst.com

CST COMPUTER SIMULATION TECHNOLOGY | www.cst.com

All base-station antennas are Multi-path signal transmission

Tilted reflecting planes change

CST COMPUTER SIMULATION TECHNOLOGY | www.cst.com

CST COMPUTER SIMULATION TECHNOLOGY | www.cst.com

Load farfield of second antenna

Envelope Correlation from Farfield: Envelope Corr. from S-Parameters:

CST COMPUTER SIMULATION TECHNOLOGY | www.cst.com

Element -> Array Design

CST COMPUTER SIMULATION TECHNOLOGY | www.cst.com

CST COMPUTER SIMULATION TECHNOLOGY | www.cst.com

Magus Antenna Synthesis generates initial square

CST COMPUTER SIMULATION TECHNOLOGY | www.cst.com

Magus also provides

Initial design is clearly

CST COMPUTER SIMULATION TECHNOLOGY | www.cst.com

3D Simulation Results show similar performance issues in initial design

CST COMPUTER SIMULATION TECHNOLOGY | www.cst.com

CST COMPUTER SIMULATION TECHNOLOGY | www.cst.com

Retuned for 28 GHz, at the expense of

CST COMPUTER SIMULATION TECHNOLOGY | www.cst.com

Synthesize planar array for 8 dB gain from

CST COMPUTER SIMULATION TECHNOLOGY | www.cst.com

CST COMPUTER SIMULATION TECHNOLOGY | www.cst.com

CST COMPUTER SIMULATION TECHNOLOGY | www.cst.com

CST COMPUTER SIMULATION TECHNOLOGY | www.cst.com

CST COMPUTER SIMULATION TECHNOLOGY | www.cst.com

CST COMPUTER SIMULATION TECHNOLOGY | www.cst.com

CST COMPUTER SIMULATION TECHNOLOGY | www.cst.com

System Assembly Modelling provides

Several copies of array independently