Research & Development AMI
Research& DevelopmentAMI
AMI R&D
Research& DevelopmentAMI
• A unique ultra low-cost, broadband, phased array, steerable antenna technology to facilitate:• Low cost consumer SATCOM terminals
• Automotive 5G mm wave cellular communication for the connected car
• mm wave based C-V2X.
• Low cost, Hi performance LIDAR & RADAR systems
• Can be the basis for 5G mm wave cellular antenna for mobile phones
• Protected by issued or pending patents
• Initial Development for under internal R&D followed by U.S. Army started 2010
• Over $5M invested to-date but lacked funds to complete development
• Basic POC completed; with funding, engineering prototype Q2 2020
The DTWA Opportunity
AMI R&D
Research& DevelopmentAMI
• Controllable Effective Dielectric Constant - Dielectric Traveling Wave Array controls the propagation constant of the feed lines to effect beam steering
• Large Instantaneous Bandwidth (IBW) - able to have simultaneous Tx & Rx from same aperture
• Low loss – Dielectric filled waveguide feed network
Unconventional Design
AMI R&D
Research& DevelopmentAMI
• COST - Designed to shatter the cost barrier that previously limited electronically-steerable arrays from addressing the consumer market for LEO SATCOM Terminals
• FREQUENCY SCALABLE – Applicable to large frequency ranges. Design is Frequency Scalable – L-band to optical
• VERY BROADBAND – over 3:1 instantaneous bandwidth demonstrated. Obviates need for frequency selectability.
• POWER EFFICIENT – Passive Phased Array. No active elements.
• SUPERIOR NOISE FIGURE – Very Low Noise. Dielectric Feed Structure minimizes losses which improves G/T.
Design Advantages
AMI R&D
Research& DevelopmentAMI
• Initial development started with dielectric waveguide with a variable airgap to change the effective dielectric, thus changing the propagation in the waveguide• Initially implemented in a leaky wave structure with scattering
elements• Evolved to a coupled array fed by a TE11 & a TEM waveguide to
control axial ration• Latest evolution is a Solid State implementation of the variable
dielectric waveguide• There are two (2) versions of this approach:
• Elements: Replace airgap waveguide with Solid State Waveguide• Element-less: Apply our Coherent Concentrator for Solar Energy collection
architecture eliminating elements with the application of a tapered dielectric layer
How We Got Here
AMI R&D
Research& DevelopmentAMI
• 2010 – Started development of Dielectric Traveling Wave Array (DTWA)
technology to address high cost of Phased Arrays
• 2012-4 – Parallel effort for Solar Energy using same theory
• 2012 – Contract with CERDEC S&T to develop a proof of concept at Ku-Band –
Successful [scanned beam at fixed frequency]
• 2013-16 – AMI R&D focused on addressing lessons learned from Ku proof of
concept and design of a Ka-Band version of the DTWA
• 2016-17 – Phase I SBIR for US Army to develop an X-Band array for Blue Force
Tracker to PDR – Successful [completed Phase I, there was no Phase II –
sponsor left Army]
• 2017-18 – Continuing internal development. Developed 3:1 IBW architecture.
Development History
Research& DevelopmentAMI
How DTWA works and what differentiates it from other phased arrays
AMI R&D
Research& DevelopmentAMIHow DTWA Works
X1
X2
X3
X4
X5
X6
X
X1
X2
X3
X4
X5
X6
X
Traditional “Corporate Feed”:• Each path is set to have all energy arrive at
point X at same time for a given angle ofarrival
• To change angle of arrival all the pathlengths need to be changed
VS.
DTWA:• Each path is set to have all energy arrive at
point X at same time for a given angle ofarrival
• To change angle of arrival the speed of thewave in the waveguide is changed
AMI R&D
Research& DevelopmentAMI
• Network is reciprocal, no T/R switching required
• No active devices between array elements and input/output port:
SWaP-C reduction
• Dynamic range limited only by dielectric breakdown voltage
• Phase shifting method doesn’t introduce loss
Why a Passive Array?
DTWA is a passive ESA (PESA)
AMI R&D
Research& DevelopmentAMI
• Control amplitude and phase of each element in array
• Combine element outputs coherently to generate directivity
• Provide transmit and receive functions
• Implement these without dissipative loss or distortion
Review Fundamental Array Requirements
Implementation of ESAs at microwave frequencies has historically been challenging
AMI R&D
Research& DevelopmentAMI
• Phase/gain tracking of many PA/LNA’s is very costly: significant calibration cost
• Unfiltered LNA inputs are exposed to raw RF environment
• Filtering LNA inputs degrades phase/gain tracking
• Dynamic PA phase error during transmit cannot be compensated
• PA output filtering to reduce cosite interference reduces efficiency
• Active component losses must be compensated with amplification
• AM/PM conversion distorts beamforming
Problems With Active Arrays
Passive approach avoids active array issues
AMI R&D
Research& DevelopmentAMI
• Passive RF combining network consists of directional couplers and waveguide
• Phase shifting performed by a solid state waveguide changing the effective dielectric to control the propagation velocity
Dielectric Traveling Wave Line Array Architecture
Passive approach avoids active array issues
Variable dielectric
waveguide
Patch Array
Input/output
AMI R&D
Research& DevelopmentAMIProof of Concept: Variable Effective Dielectric Waveguide
Passive approach avoids active array issues
Air gaps 1mil
Air gaps 10 mil
Propagation velocity increases as we increase gaps…..
E-Field Mag Normal to Waveguide Ground Plane
AMI R&D
Research& DevelopmentAMI
• Passive RF combining network consists of directional couplers and waveguide
• Phase shifting performed by a solid-state waveguide changing the effective dielectric to control the propagation velocity
Bandwidth Enhancement / Beam Squint
Passive approach avoids active array issues
cos 𝜃 =𝛽'((𝑓)𝛽,
−𝑚𝜆𝑑
Waveguide Beamformer Analytical Model
Waveguide dispersioncausing very small beam squint over channel bandwidth
Element spacing causing beam squint
Far field radiation angle
Progressive Delay allows for Instantaneous bandwidth of 3:1
Research& DevelopmentAMI
Foundational Early Work
AMI R&D
Research& DevelopmentAMIDTWA Major Components
Passive approach avoids active array issues
3. Unit Cell -Radiating Element
2. Waveguide Coupler
1. Solid State Variable Effective Dielectric Waveguide
AMI R&D
Research& DevelopmentAMIUnit Cell
Passive approach avoids active array issues
Coupler-fed L-probe circular patch• Two (2) patch probes fed in
quadrature, spaced at λ/2 along independent quadrature feed waveguides feed a circular patch
RHCP
LHCP
Gain, Axial Ratio Elevation Pattern for Three (3) Frequencies
AMI R&D
Research& DevelopmentAMIAxial Ratio Over FoR
+90°
-90°
Ʃ
Ʃ
RHCP
LHCP
• Traveling wave-fed CP element from single waveguide in quadrature difficult because waveguide propagation constant changing through FoR• Only possible to achieve near 0 dB
axial ratio over a narrow FoR (typically broadside)
• Feeding a single radiator with two (2) waveguides in quadrature mitigates this issue
• Quadrature feeds allow for simultaneous R/LHCP – frequency reuse (doubles channel bandwidth)
λg / 4
Ku Brassboard CP Feed
Ka Brassboard CP Feed
AMI R&D
Research& DevelopmentAMIEarly 1000 MHz Instantaneous Bandwidth Design
Normalized Rx Patterns (GHz)
At 1000 MHz Inst. Bandwidth, remaining beam squint causes 0.6 dB frequency distortion
Research& DevelopmentAMI
EvolutionarySolid state architecture
AMI R&D
Research& DevelopmentAMIDTWA Waveguide Overview
• Goal: Variable Dielectric Waveguide• Change effective Dielectric Constant (Er) in order to swing the beam across
elevation
• Two (2) Approaches to accomplish Goal of varying Er1. Air Gap Actuator Approach (baseline – proven)
Vary Air Gap between two (2) dielectric slab waveguides by means of actuation changes effective Er
2. Solid State Varactor (New Approach)Varying the control voltage changes the capacitance of the path from the
upper waveguide thru the Higher Dielectric slab to the lower waveguide, resulting in an alternating impedance of the waveguide as a function of the distance along the waveguide where the propagation constant of the structure is β as controlled by voltage
𝛽 = 𝛽1234562789
AMI R&D
Research& DevelopmentAMIDTWA Waveguide - Air Gap Actuator Approach
Full Waveguide View (Baseline)
3D Isometric View
Side View
Upper WG
Lower WG
Varying Air Gap via Actuators changes the effective Er
AMI R&D
Research& DevelopmentAMIDTWA Waveguide - Air Gap Actuator Approach
Epsilon vs. Air Gap
AMI R&D
Research& DevelopmentAMIDTWA Waveguide - Air Gap Actuator Approach
Epsilon vs. Air Gap (Zoomed In)
AMI R&D
Research& DevelopmentAMIDTWA Waveguide - Air Gap Actuator Approach
WG Loss vs. Frequency
AMI R&D
Research& DevelopmentAMIDTWA Waveguide - Solid State Varactor
Full Waveguide View (New Approach)
3D Isometric View
Side View
Upper WG
Lower WG
Varying Periodic Dielectric Layer w/ Alternating ErLow & ErHigh Slabs
where a varactor layer resided atop the ErHigh Slabs
AMI R&D
Research& DevelopmentAMIDTWA Waveguide - Solid State Varactor
Zoomed In Waveguide View
Upper WG
Varying Periodic Dielectric Layer w/ Alternating ErLow & ErHigh Slabs
Lower WG
ErLow
ErHigh
VaractorLayer
AMI R&D
Research& DevelopmentAMIDTWA Waveguide - Solid State Varactor
Epsilon vs. Varactor Space (Capacitance)
AMI R&D
Research& DevelopmentAMIDTWA Waveguide - Solid State Varactor Approach
WG Loss vs. Frequency
AMI R&D
Research& DevelopmentAMIInfinite Element Configuration
• Dielectric antenna and dielectric feed lower noise temperature• Solid state waveguide enables
low-loss dynamic beam tapering• Tapering can be adjusted on-the-fly
as required to maintain ITU compliance
• V/H/LH/RH POL• No elements – continuous array• Fabricated using proven
microelectronic fabrication methods
AMI R&D
Research& DevelopmentAMITime Domain Multi-Beam
• With a >3:1 IBW, DTWA is able to form multiple beams by utilizing time domain beamforming• An example is for a Low Earth Orbiting (LEO) satellite
system operating in the Ku band. The bandwidth of a signal in such a system might be 12.5 MHz. A DTWA would be able to generate 50 simultaneous beams to validly service 40 or possibly even 50 signals of interest.
AMI R&D
Research& DevelopmentAMIPerformance Predictions:
Bandwidth/Beamwidth/Polarization
• H pol Rx is efficient at the same aspect angle• Infers that all polarizations are efficiently collected at that
angle
30.00 35.00 40.00 45.00 50.00 55.00 60.00 65.00Theta [deg]
-20.00
-15.00
-10.00
-5.00
0.00
5.00
10.00
dB(R
ealiz
edGa
inPhi)
10b_Hpol_profile9_very longH Pol Light Far Field Pattern ANSOFT
AMI R&D
Research& DevelopmentAMIManufacture
• The waveguide structure utilizes ceramic substrates such as BaTi3, TiO2 or other low loss dielectric• Varactors and control lines (ITO) are deposited on the
substrate• All processes are commonly used in semiconductor
industry• Engaged with UIUC to fabricate prototype structures• Transition to commercial foundry with support of partner