17TH CALIFORNIAN-DANISH PHOTONICS WORKSHOP 2016
MARCH 25. 2016
mmW-SPRAWL photonic integrated circuits for microwave photonics
Martijn Heck
27TH CALIFORNIAN-DANISH PHOTONICS WORKSHOP 2016
37TH CALIFORNIAN-DANISH PHOTONICS WORKSHOP 2016
The case for photonic integration
• system performance:speed, sensitivity and stability
• low cost for high volume• decreased size, weight and
power consumption (SWaP)
• component performance• inherent performance trade-
off
• competing technologies• discrete optics• electronics• MEMS, ...
Pros Cons
PIC
tec
hn
olo
gy
pla
yin
gfi
eld
• exponential bandwidth growth• telecommunications• exascale datacenters
• connectivity everywhere• 4G/5G wireless
• accurate metrology, sensing, radar, lidar, ...
47TH CALIFORNIAN-DANISH PHOTONICS WORKSHOP 2016
There is a clear case for telecom and datacom
• increased performance:speed, sensitivity and stability
• increased functionality• low cost for high volume• decreased size, weight and
power consumption (SWaP)
• inherent performance trade-off
• high development cost• no standardization
• competing technologies• discrete optics• electronics• MEMS, ...
• exponential bandwidth growth• telecommunications• exascale datacenters
• connectivity everywhere• 4G/5G wireless• ”internet-of-things”
• ubiquitous sensors, networks
Pros Cons
PIC
tec
hn
olo
gy
pla
yin
gfi
eld
• energy efficiency• reduced cost
• technical feasibility:”good enough”
57TH CALIFORNIAN-DANISH PHOTONICS WORKSHOP 2016
But for most other applications it’s a struggle
• increased performance:speed, sensitivity and stability
• increased functionality• low cost for high volume• decreased size, weight and
power consumption (SWaP)
• inherent performance trade-off
• high development cost• no standardization
• competing technologies• discrete optics• electronics• MEMS, ...
• exponential bandwidth growth• telecommunications• exascale datacenters
• connectivity everywhere• 4G/5G wireless• ”internet-of-things”
• ubiquitous sensors, networks
Pros Cons
PIC
tec
hn
olo
gy
pla
yin
gfi
eld
• performance too low• competing technologies
are (still) better
67TH CALIFORNIAN-DANISH PHOTONICS WORKSHOP 2016
Some state of the art microwave technology
1 100 10k 1M 100M 10G
MLLDs
µOEODRO
lab/rack
analog-to-digital converters
microwave filters
10-GHz oscillators
chip
scal
e
77TH CALIFORNIAN-DANISH PHOTONICS WORKSHOP 2016
Ultra low loss waveguide technology required
1 100 10k 1M 100M 10G
MLLDs
µOEODRO
lab/rack
resonator Q values > 100M
oscillator timing jitter << 10 fs(10 kHz – 10 MHz)
waveguide losses < 0.1 dB/m
87TH CALIFORNIAN-DANISH PHOTONICS WORKSHOP 2016
Ultra-low loss waveguidesbased on silicon nitride
Bauters et al., OSA Opt. Exp., 19 (2011).
World record low loss:(0.045 ± 0.05) dB/m
97TH CALIFORNIAN-DANISH PHOTONICS WORKSHOP 2016
Ultra-low loss waveguides required for high-end (microwave photonic) applications
• increased performance:speed, sensitivity and stability
• increased functionality• low cost for high volume• decreased size, weight and
power consumption (SWaP)
• inherent performance trade-off
• high development cost• no standardization
• competing technologies• discrete optics• electronics• MEMS, ...
• exponential bandwidth growth• telecommunications• exascale datacenters
• connectivity everywhere• 4G/5G wireless• ”internet-of-things”
• ubiquitous sensors, networks
Pros Cons
PIC
tec
hn
olo
gy
pla
yin
gfi
eld
1. ultra-low loss waveguide(ULLW) technology;
2. integration ULLW with III/V or silicon photonics
3. opportunities for high-performance applications
107TH CALIFORNIAN-DANISH PHOTONICS WORKSHOP 2016
Millimeter-wave silicon photonics for future energy-efficient 5G networks
We use silicon photonics for 60 GHz – 90 GHz generation and modulationto address the increasing bandwidth demand for wireless communications. We will use thistechnology to realize energy-efficient wireless transceivers.
117TH CALIFORNIAN-DANISH PHOTONICS WORKSHOP 2016
mmW-SPRAWL concept
Silicon photonic chip
RF
DFB laser
MOD
100 Gbps
filter
60 – 90 GHz
(QAM)
low-frequencyRF oscillator
< 20 GHz bandwidthsilicon modulator
127TH CALIFORNIAN-DANISH PHOTONICS WORKSHOP 2016
Energy efficiency analysis: trade-off photonic and electronic power consumption
Power efficiency
Literature dataRC fitted model
Literature dataFitted model
MOD
higher drive voltage wider comb
higher drive frequency higher RF loss
higher drive frequency lower oscillator efficiency
137TH CALIFORNIAN-DANISH PHOTONICS WORKSHOP 2016
A comprehensive simulation tool foroverall system efficiency
Mohammadhosseini and Heck, to be published
Frequency quadruplingoptimum for 30 GHz
generation
147TH CALIFORNIAN-DANISH PHOTONICS WORKSHOP 2016
mmW-SPRAWL will optimize the energy efficiency of FiWi 5G systems
• increased performance:speed, sensitivity and stability
• increased functionality• low cost for high volume• decreased size, weight and
power consumption (SWaP)
• inherent performance trade-off
• high development cost• no standardization
• competing technologies• discrete optics• electronics• MEMS, ...
• exponential bandwidth growth• telecommunications• exascale datacenters
• connectivity everywhere• 4G/5G wireless• ”internet-of-things”
• ubiquitous sensors, networks
Pros Cons
PIC
tec
hn
olo
gy
pla
yin
gfi
eld
• energy efficiency• reduced cost (?)
• technical feasibility:”good enough”
157TH CALIFORNIAN-DANISH PHOTONICS WORKSHOP 2016
dr. Martijn J.R. HeckAssociate ProfessorDepartment of EngineeringAarhus University
Our research is supported by: