Документ взят из кэша поисковой машины. Адрес оригинального документа : http://www.atnf.csiro.au/people/cjackson/xNTD_Antennas_Sept_05/NTD_xNTD_sheet_Aug05_1.pdf Дата изменения: Thu Sep 1 05:50:47 2005 Дата индексирования: Sun Dec 23 18:12:22 2007 Кодировка: Поисковые слова: вторая космическая скорость

CSIRO technology demonstrator projects and the Square Kilometre Array telescope
The SKA, or Square Kilometre Array telescope, is an international `next-generation' radio telescope. CSIRO is working on a technology demonstrator for the SKA: this sheet gives an overview of the R&D activities under way for that project.

CSIRO's project for an SKA demonstrator
CSIRO is wor king on the phased development of a technology demonstrator for the SKA. There are two stages, both funded via a mixture of grants from the Commonwealth and Western Australian Governments and strategic CSIRO investment. The fir st stage is the prototype development of a two-antenna system, scheduled for completion in mid-2007. The second phase , presently known as the `extended new technology demonstrator' or xNTD, will extend and refine the concept. It will be a radio telescope with the following specification: · total collecting area of 3 500 m2, from 20 dishes, each ~15 m diameter, forming 190 baselines · system temperature (Tsys) of 50 K · frequency range 0.8 ­ 1.8 GHz · instantaneous bandwidth 256 MHz · at least 30 independent beams, each beam being 1 sq degree, giving a total field of view of more than 30 sq deg at 1.4 GHz · maximum baseline ~1000 m · full cross-correlation of all antennas. This demonstrator will be located in a radio-quiet zone at Mileura, Western Australia. The target completion date is mid-2009.

Low-noise , highly integrated receivers for both cr yogenic and uncooled applications The focal-plane arrays will need high-performance low-noise amplifier s (LNAs). There are ver y stringent requirements on the noise temperature (< 40 K), linearity and cost of these. Commercial, off-the-shelf (COTS) components of this type are not readily available . CSIRO is developing RF CMOS `receiver-on-a-chip' prototypes in collaboration with Macquarie Univer sity, plus solutions based on discrete components using RF CMOS and SiGe technologies. High-speed digital signal processing An immense amount of the digital signal processing will take place in the beam-former s and the correlator, both of which have to operate at ultra-fast processing rates. Each antenna will need a beam-former ; the target is to form at least 30 independent, steerable beams. With two polarizations, each feed contributes to at least 10 beams, and at a bandwidth of ~250 MHz the beam-former will have to operate at 10 Tera operations per second. The correlator will make ~32 Tera operations per second. (The challenge for the SKA is to scale this up by a factor of 125 000.) High-speed (Tbps) digital fibre links Each xNTD antenna will have 200 receiver s and generate data at up to 1 Tbps. The xNTD will need low-cost means of transpor ting this data, using both digital transmission and new forms of analogue transmission on fibre optic cables, over both shor t and long distances: · shor t, intra-station data transpor t, up to 100 m. For these relatively shor t links, we need ver y-high-bandwidth efficiency; and · fibre-optic links for spans of 100 m to 3 000 km. Here we are aiming for commercial compatibility ( i.e. conforming to International Telecommunication Union standards).

One possible design for the Square Kilometre Array, in which the individual elements that collect the radio waves are 15-m diameter dishes, grouped into patches 200 metres across. The whole telescope consists of many patches, some separated by thousands of kilometres. Image: Chris Fluke, Centre for Astrophysics and Supercomputing, Swinburne University of Technology

What are the technology challenges?
The xNTD is a full-system `mini SKA station' prototype, so the challenges are wide-ranging. Below are the key areas of R&D. Wide bandwidth feed antennas for dish concentrators, and broadband active phased arrays for aper ture and focal plane applications We are developing a low-cost, 10 x 10 element focal-plane array (FPA) operating over the frequency range 0.8­1.8 GHz, to produce a highpolarisation-purity receiver. Our initial wor k is based on the Vivaldi notch array but other options are also being investigated.

continued over page

CSIRO
August 2005


Low-cost high-speed (Gbps) analogue-to-digital conver tors The xNTD will require uncooled systems for the receiver s, which will have an overall system temperature of < 50 K. The LNAs required have been described ear lier ; the technology required for digitisation also needs development. The bandwidth requirements for the digitisation is at the current forefront of technological development, and is being realised commercially. Ultra-fast supercomputing The xNTD will produce large amounts of raw data. The capacity to store data will be limited, meaning that data must be largely processed in real time. Ideally, a buffer of the past ~20 minutes obser ving will be kept to allow reprocessing of the data for time-critical events only `discovered' after the fact. Low-cost manufacturing of small-to-medium dishes The dishes plus drive systems must be manufactured at an affordable cost. We are investigating if the best solution is to develop an extremely `cheap' low-spec antenna and compensate for deformations and pointing error s with ver y smar t electronics. An entire antenna-plus-receiver system of this kind could be applicable to many areas other than radio astronomy. High-dynamic-range imaging from sparsely-sampled Fourier data Image processing may require dedicated hardware accelerator s. We are investigating progressive FPGA-based processing to reduce data rates. Radio-frequency interference mitigation We are developing RFI and spectral-line-ripple cancellation techniques based on FPA multi-

beaming, to ensure efficient real-time signal processing. System integration The xNTD and the ultimate SKA will involve complex system integration--an area in which Australia can claim wor ld-leading exper tise . A critical aspect is to ensure that the components (dish, FPA, receiver s and data transmission system) are compatible in terms of mechanical specification, self-RFI, power supply and dissipation, local-oscillator distribution and local oscillator stability. Deser t (remote) station power A number of power options will be considered. These include: commercially-emerging technologies such as natural gas micro-turbines, the new generation of photo-voltaic solar cells, wind turbines, and geothermal power ; conventional diesel generator s; and (for the SKA) choosing sites that can be ser ved by existing gas pipelines.

Australian industry participation
Advantages for industr y par ticipants include: · the oppor tunity to engage in a highly creative project to address cutting-edge science · the ability to perfect leading-edge techniques and products for a demanding application, with technologically sophisticated user s, · the ability to generate and share information with R&D par tner s in a benign and commercially non-threatening environment · high visibility within an international, collaborative project · potential for ear ly involvement in (and skilling up for) a one-billion Euro project spanning a wide range of engineering and ICT disciplines.

Contact the Australian SKA Planning Office (ASPO)
Dr Carole Jackson Phone +61-2-9372-4407 Email Carole.Jackson@csiro.au Web www.atnf.csiro.au/projects/ska

© CSIRO. Disclaimer : CSIRO shall not be liable for technical or editorial omissions contained herein. The information is provided in the best of faith but is subject to change without notice.