The SKA -Challenges, Opportunities, and Industry
Involvement
Phil CrosbySKA Program Development Office
The SKA – “will be the largest scientific instrument on the planet”
• A ‘next generation’ global radio astronomy facility• To be built in either Southern Africa, or Australia•Will operate between around 70 MHz to 25 GHz• 50 times the sensitivity and 8000 times the survey speed of current instruments• a collecting area of around 1 million square meters over a vast unpopulated area • a combination of 3000-5000 dishes and wide FoV antennas• will employ beam forming technology on a scale not previously explored• Needs data transport system and computing power beyond that available today• Will address fundamental questions about the universe
SKA Key Science Questions• When & how were the first stars and galaxies
formed?
• What is the large scale structure of the universe? ‘Dark Energy’ ‘Dark Matter”
• What is the origin and evolution of cosmic magnetic fields?
• Was Einstein right? Can we detect gravitational waves?
• Planet formation and the ‘Cradle of Life’Will we find ET?
• EXPLORATION OF THE UNKNOWN
System DesignSystem Design Phase 2 construction Phase 2 construction ––mid + low freq mid + low freq
Preliminary SKA specs
External Engineering Review of design
Early ScienceEarly Science SKA mid+lowSKA mid+low
SKASKA--mid+lowmid+lowCompleteComplete
Phase 1 complete
Concept DesConcept Des’’nn
SiteSelect
Establish SKA
Organisation
Phase 1 implementation Phase 1 implementation ––low & mid freqlow & mid freq
06 | 08 | 10 | 12 | 14 | 16 | 18 | 20 |
Reference Design selected
SKA SKA PathfindersPathfindersCompleteComplete
Phase 3 Phase 3 --ConstructionConstructionSystem design SKASystem design SKA--hihiConcept & Tech Development for Concept & Tech Development for
Phase 3 Phase 3 -- SKASKA--hihi
• Target construction cost:for Phases 1+2
Civil worksAntennas & RF systemsSignal transmissionSignal processingSoftware development & computing hardwareDesign, integration, testing, and managementContingency
€ 1.5 billion (2007)
• Expected operating costs:
Salaries (400-500 staff)PowerMaterials & services including dark fibre leaseRenewal of instrumentation and computing(science centres additional)
€ 150 million /year
The SKA timeline & estimated project costs
Funding AgenciesNASANSF
United States of AmericaCIT / JPLCornell U / NAICHarvard U / Smithsonian Centre for AstrophysicsMIT / Haystack ObservatoryNRAONaval Research LabSETIU Cal – BerkleyU IllinoisU New MexicoU WinsconsinVirginia Tech
United KingdomU Manchester (JBO)Oxford UCambridge UU GlasgowLeeds UCardiff U
Funding AgenciesEuropean UnionNational Governments via Research CouncilsRadioNet
SwedenOnsala Space ObservatoryChalmers U of Tech
ItalyNational Institute
for Astrophy
sics
FranceParis ObservatoryU d’OrleansCentre National de la Recherche ScientifiqueOMMIC
NetherlandsAstronJIVEKapteyn Astro Inst Leiden UU Gronegan
PortugalIST-Centra
SpainFundacion Alcala – Nat. Astronomy ObservatoryU Valencia
PolandTorun Centre for Astronomy
GermanyMax-Planck Institute for Radio Astronomy
RussiaPushchino Astro Space Centre
AustraliaCSIRO-ATNFICRARSwinburne UU NSWU MelbourneU SydneyU AdelaideU TasmaniaU WA
Funding AgenciesARCDIISRCSIROWA Gov
New ZealandAuckland U of Tech
CanadaNRCLaval UMcGill UQueens UU CalgaryU MontrealU TorontoYork UACURA CASCA
Funding AgenciesNRCNSERC
South AfricaNRFU KwaZulu-Natal
Funding AgenciesDept Science & TechDept Trade & IndustryNRF
The SKA Global Network
Over 100 Organisations8 SKA Consortia
SKA Program Development Office• Based at University of Manchester, UK• Group of domain specialists• Mission – to deliver a costed design by 2012
Indian ConsortiumRaman Research InstituteTATA Inst for Radio AstronomyNat. Centre for Radio Astronomy Funding Agencies
Dept of Atomic EnergyDept of Science & Tech
ChinaNAO, Chinese Academy of Sciences Funding Agencies
Chinese Academy of SciencesMoSTDept Science & TechTsing hua U
Korea Korean VLBI Network
JapanKagoshima UNAOJ
SKA Movie – 3 minutes
The Challenges of Radio Astronomy
• The signals are extremely weak.– We need huge antennas to capture them– They need to be in special places
• Astronomers ‘compete’ with noise– From radio, TV, phones, machines, etc– From the equipment itself, amplifiers etc
• Signals are buried in the noise– Need smart techniques to ‘resolve’– This means huge computing power
• Large amounts of data to handle– Pushing boundaries in capacity, speed and
storage
Array Telescopes
VLA (USA)
Australia Telescope
E-MERLIN
But better instruments needed to answer the key science questions
ELTALMA
JWST
IXO
SKALOFAR
StationBeamform’g
Analog links …
DSPDSP
DSP
..
Digital links..
AA-lo
1st Stage DSP
O-EDSP
O-EDSP
.....
AA-hi
..
StationBeamform’g
StationBeamform’g
Mass Storage
AA Station
TimeTimeStandardStandard
Central Processing Facility - CPF
User interfacevia Internet
...
To ~250 AA Stations
...
DSP & Control
...
DSP & Control
To ~2400 Dishes
...
Dishes
Control Proc.
Correlator –
AA
& D
ish
PowerPowerSuppliesSupplies
2.7.4 Phase transfer
2.5.2 AA Tiles
2.6.1 LNA
2.7.2 Intra-station data links
16 Tb/s
80 Gb/s
2.7.3 Station-core data links
2.7.5 Monitor & control2.9.2 Computing hardware
2.6.2 Int. receivers2.8.1 Station DSP 2.8.2 Correlator
2.8.4 Non-imaging processors
Control Processors& User interface
Post Processor
2.9.4 Data management
2.9.3 Software engineering 2.9.5 Calibration2.9.6 Science post processing
2.1.x SKA system design2.9.1 SKA computing & software spec
2.9.4 Data management
2.11.1 AA infrastructure
Green SKA
2.4.x Dish design
2.5.1 Wide-band single pixel feeds
2.5.2 Phased array feeds
2.6.1 LNA
2.6.3 Cryo systems
2.8.1 Station DSP
SKA System Diagram on a Page
Diagram by Andrew Faulkner
SKA as e-science
AntennaArray
DSP(“correlator”)
Post-processingHPC
(“imaging”)Tier 0(SKA)
Tier 1(National)
Tier 2(Regional)
Tier 3(Institute)
Europe USA Australia South Africa
Pb/s
0.1 – 1 Tb/s
Gb/s
…………
Tier 4(Researcher)
E-science:Global collaboration in key areas of science, and the next generation of infrastructure that will enable it.More data, more computation, faster networks, more collaboration, exploration of data and models – in silico discovery, floods of public data, GRID computing, ..
• Each pair of antennas is called a baseline• The more different baselines there are, the more detailed the astronomical
image. • Short baselines - antennas are close to each other - provide coarse structure.• Long baselines provide the fine detail, the longer, the finer the detail.
Interferometry using arrays
Correlator
Data Processor
3 km
SKA is an “aperture synthesis”telescope
-A large aperture telescope is ‘synthesized’by sampling the wave-front in the aperture plane
Station
Central ProcessingFacility
Comms links
Max. Distance for Dense AAs
Dishes spread along spiral
Possible Site Schematic
Max. Distance for Central Power Dist’n
Dense AA
Sparse AA
Dishes
Concise Picture of Technology Options
• Numbers of dishes (2000-3000) depends on whether Phased Array Feeds and/or Aperture Arrays are used in the SKA.• Each technology is characterized by a frequency range and field of view.
Aperture Array Technology Production Thinking
Courtesy ASTRON, OPAR
Electronic Sensor
ASTRON Prototype – The Netherlands
Phased Array Feed Prototype
One of three prototypes under development.
Multi-pixel phased array feeds
from Dave Deboer
Dishes+Single Pixel Feeds
Prototype 15 m composite dish
USA Allen Telescope Array 42x6m hydroformed dishes
South Africa
Canada prototype 10 m composite dish
rms error: 0.25 mm
Final Mould Alignment
Composite Dish Manufacturing (Canada)
Removing from MouldMounted on Drive
rms error: 0.25 mm
Metal Dish Manufacturing
12m antenna stretch-formed panels
Patriot Systems
Novel Sheet Metal Structure
Wide field-of-view
Allen Telescope Array
Parkes
Base model SKA
Correlators – Ultimately Built in Industry
Electrical power – not solved yet
•Solar potential is high on both sites.•24-hour coverage is needed.
•requires storage or alternative night-time power source.
•Cost likely to be an issue if not subsidized.•30-50MW required for full SKA•Role for small scale (~100 kW) systems if they exist.
• Systems must withstand occasional flooding.• Priority power not needed for SKA but power-outage notice
might be needed.• Safety grounding issues in desert areas.• Lightning protection required• Equipment subject to unusually high temperatures and large
diurnal swings.• Power for redundant communication needed for emergency
shut-down• Staggered antenna slewing is standard practice for arrays.• Some equipment will require local UPS systems that may
need remote control.
Other Power Issues
The Pre-Cursor Projects
South Africa + 7 countries
Australia
Spectrum Measurements: 80 MHz – 1.6 GHz
SydneyPop. 4 million
NarrabriPop. 6,000
Boolardy
100 MHz 200 MHz
Industry Opportunities
From P. Hall
Summary of Opportunities in the SKA Signal Path
The SKA – “will be the largest scientific collaboration on the planet”
•To meet the SKA timelines, a very high level of industry involvement will be needed, especially R & D, and economical mass production and deployment .• Benefits to industry include opportunities to:-Grow and hone the creative energies of the best professionals-Perfect leading-edge techniques and products in a very demanding application,- Generate and share information in a benign and commercially non-threatening environment-Raise company profile/visibility by association with an innovative, high profile, international mega-science project- Gain early involvement and favourable positioning in a € 1.5 billion (2007) project spanning a wide range of engineering and computing disciplines.
Site works and Infrastructure
• Site studies, and infrastructure engineering
• Site works for design & construction of antennas, support buildings (offices, equipment, accommodation, etc) , cable roll-out, and repeaters
• Electrical supply to chosen site, in order of 50 MW (with a proportion of ‘green’ energy (TBD))
• High-speed (Tb/s) digital fibre optic links for distance regimes extending from 100 m to 3000 km
Industry Opportunities and the SKA
SKA Project Support, Tools, Operations & Maintenance
• Outreach and public education
• Project management, site supervision (works management), and Systems
engineering support
• Radio-frequency interference mitigation using coherent and incoherent
techniques
• High dynamic range (>60 dB) image formation using sparsely-sampled
Fourier plane data
•SKA scheduling, operations and maintenance models
High volume production & deployment
• Low-cost manufacturing of small to medium diameter dishes
• Advanced mechatronic systems for feed positioning and antenna control
• Decade bandwidth feed antennas for dish flux concentrators
• Broadband, active, phased arrays for aperture and focal plane applications
• Low noise wideband RF amplifiers for both cryogenic and uncooled
applications
• Low-noise, highly integrated, receivers for both cryogenic and uncooled
applications
• Low-cost, high-speed (Gs/s) analog to digital converters
Low-medium volume production & deployment
•High-speed digital signal processing engines (correlator) at 24 peta-flops/sec
• Ultra-fast supercomputing at 200 peta-ops/sec
• High speed data transmission at 160 Gb/sec
•Software engineering for robust, intelligent, array control and data processing
• Master oscillator time standards, and distribution
What kinds of firms should consider participating in the SKA?
• Information & Communication Technology (ICT) - hardware, software, digital fibre systems, data management, high-speed / high-volume data processing, control systems, modelling and simulation systems, networked enabled system deployment & management, integrated circuit design, fabrication, and test, telecoms systems.
• Engineering Construction & Maintenance (ECM) – building construction, electrical and mechanical services, R & D services, environmental services, fibre optic, power, civil engineering, land access consultants, remote infrastructure operations and maintenance, site management & planning, surveying services.
• Advanced Aerospace & Radar Technology and Equipment – antenna design and manufacture, image processing, radio astronomy, receiver feed systems, wideband phased arrays, RF devices, RFI mitigation.
• Advanced Materials and Manufacturing - advanced materials, composites, sheet metal fabrication.
• Systems Integration & Maintenance – project design, execution, interface management, risk management, scheduling, operations and maintenance of complex distributed systems.
• Transport, Training, and Other Goods & Services – regional support, recruitment and training, transport and logistics, community consultation and studies, regulatory monitoring.
Informatio1High-spee2ICT / elec3ICT data m4ICT distrib5ICT softw6ICT softw7ICT datab8Integrated9Telecomm10Engineerin11Building c12Electrical13Engineerin14Environme15Fibre optic16Infrastruct17Land acce18Remote in19Site mana20Surveying21Advanced22Antenna d23Image pro24Radio astr25Receiver f26RF device27RFI mitiga28Computer29High perfo30High-spee31Networkin32Advanced33Advanced34Advanced34Engineerin36Systems I37Integrated38Schedulin39Systems i39Transport40Other goo42Regional d43Transport44
Intellectual Property
• SKA has developed a Statement of Intent on IP.– Signed by organizations participating in the SKA.– Establishes ground rules on protection, licensing, and donation of foreground and
background IP for the SKA project.
• An IP strategy will be developed, with registered ownership– Negotiations in the spirit of scientific cooperation.– Signed by legal entities, when that becomes possible.
• SKA must have access to IP developed in the national & regional projects:
– Some of this will be generated in industry.– Where possible, IP licensed early for the SKA, so that it can be used in open
bidding, rather than giving advantage to a particular supplier.
To Summarise…• The SKA is an ‘icon’
project – a BIG leap.• Answers to key science
questions• Global collaboration
institutes & industry• Challenges yet to be
solved• Potential for industry
‘spin-off’ products & IP
Thank You
Phil CrosbySKA Program Development Office