The Square Kilometre Array

 
Russ Taylor (Univ of Calgary) and Sean Dougherty (NRC), on behalf of the Canadian SKA consortium.
 

International Developments and the Canadian Program

The international SKA program is moving toward a final design phase leading up to a proposed construction start for phase I SKA around 2012-14. The overall timeline for the SKA project is shown in Figure 1, extending to completion of the SKA by 2020, operating at frequencies from a few 100 MHz to 10 GHz. Funded programs in Europe, US, Australia and South Africa underpin the international program activity for the next five-years. Including:

1) The EU funded SKA Preparatory Phase program (PrepSKA)

2) The US SKA Technology Development Program (TDP)

3) The design and construction of SKA pathfinder instruments in Australia and South Africa. These are 1% SKA scale functional telescopes that will demonstrate SKA technology and carry out early SKA science post 2012.

Over the period from 2008 to 2012, about $300M will be spent around the world on design and prototyping. PrepSKA is the umbrella program that plays the role of coordinating the international activity toward the goal of a costed design for the phase-I SKA by 2012. Phase-I SKA will be a build-out of an SKA pathfinder on the selected site. The international design effort will be coordinated by the Central Coordination and Development Team (CDIT) located at the SKA Project Development office hosted by the University of Manchester. Peter Dewdney has joined the SPDO as the SKA international SKA project engineer and will be leading the CDIT.

Figure 1. The International SKA project timeline. The Canadian SKA program is our plan for participation in the international activity during the 2008 – 2012 timeframe. During this period the SKA costed system design will be completed, and the SKA pathfinder telescopes will be constructed, leading to beginning of early science with the pathfinders.

The international consortium developing the SKA involves over 50 institutes in 17 countries. As we enter the final design phase, the governance and oversight of the program has been restructured through a new set of international agreements. These agreements establish the following bodies

1) The international SKA Science and Engineering Steering Committee (SSEC). The SSEC provides scientific and engineering oversight to the project from the international consortium. Canada has two members on the SSEC (Sean Dougherty and Russ Taylor).

2) The PrepSKA Board that will direct the PrepSKA program. The PrepSKA Board has 20 members with 2 members from Canada (Dougherty and Taylor).

3) An SKA Program Development Office (SPDO) funded by the PrepSKA program funds from the European Commission and by a common fund of defined contributions from the international partners. The SPDO hosts the Central Development Team charged with coordinating the international engineering design activity toward a final SKA design.

4) Region Program Development Offices (RPDOs) that are the liaison between the SPDO and regional development efforts.

5) An SKA Forum made up of funding agency representatives and the executive of the SSEC. This will evolve the international governance and funding model for the construction phase of the SKA. Greg Fahlman represents Canada on the SKA Forum.

The Canadian SKA Consortium Board (www.ska.ca), a body established in 2005 via an MOU between NRC-HIA, ACURA and industry partners and with representatives of CASCA, directs the Canadian SKA program on behalf of the Canadian community. Canada brings very significant technology leadership to the international design effort. The remaining sections of this update highlight the major Canadian SKA R&D efforts.

The Canadian program plan integrates this research and development effort into PrepSKA. We have engaged in a partnership with Australia to deploy technology on the Australia SKA Pathfinder (ASKAP – www.atnf.csiro.au/projects/askap), and to develop science programs with ASKAP that will demonstrate the scientific capability of the SKA technology and use ASKAP to make a scientific step toward addressing SKA key science goals. A science case document has been developed by a joint Canadian-Australia team and has been submitted for publication as a special issue of Experimental Astronomy. A brief synopsis has been published in the Publications of the Astronomical Society of Australia (Johnston, et al. 2007, PASA, 24, 174).

SKA Research and Development in Canada

 
CART
 
The Mk2 program of the Composite Applications for Radio Astronomy (CART) project at the NRC-HIA is well underway. 
The Mk2 program is the second 10-m prototype composite reflector and is intended to build upon the success of the first 
prototype. The Mk2 program targets improvements in the structural design, manufacturability, cure characteristics and 
surface accuracy through design and process improvements. The Mk2 design departs from the first prototype in that the 
reflector surface and structural rim are infused as one piece on mould and then a central hub and eight radial beams 
are bonded on (Fig. 1). The surface and rim are a symmetric structure so shrinkage during cure will be symmetric and 
a more accurate parabolic shape is maintained. The hub and beams will be bonded to the surface while it is still on 
the mould adding stiffness to maintain the desired shape. Fabrication of the beams and hub as separate components also 
improves the manufacturability as lay-up is simplified and production of components can occur in parallel.
 
The surface and rim were successfully 
infused at HIA-DRAO in the first week of June (Fig. 2). The beams are in production at Profile Composites in 
Victoria and the hub will be produced at DRAO later in June. The Mk2 is expected to be lifted onto the MV-1 mount in mid-July for metrology and holography. 
 

Figure 2. Rear-view of the Mk 2 10-m CART reflector showing the radial beams and hub that are bonded on the rear surface of the reflector.

Figure 3. Infusion of the Mk2 reflector on June 4, 2008. At this stage of the process, the part is fully infused. This set up is far simpler than that used in the Mk1.

 
PHAD
 
The Phased-Array Feed Demonstrator (PHAD) has been relocated to the DRAO near-field range for testing. In 2007, passive array measurements (no receiver electronics) were completed, but these latest tests have all 180 elements feeding receivers that downconvert signals to baseband. These signals are then transmitted out of the anechoic chamber to a signal processor for sampling and recording. Initial tests will be to measure radiation patterns again, this time terminated with real receivers. Following digitizing, combination of outputs for beam forming will commence. Once confidence in the performance of the system has been established, PHAD will be mounted on the 10-m CART dish for on-the-sky testing using astronomical sources.
 

Figure 4. PHAD array complete with receivers and downconvert system under test in the DRAO anechoic chamber.

Room temperature Low-Noise Amplifiers

The group in the Electrical Engineering department at the University of Calgary (Leo Belostotski and Jim Haslett) have established a world-wide reputation with the excellent performance of their 90nm room temperature LNA designs, with noise temperatures better than 14K between 0.8 to 1.5 GHz. Designs for room-temperature LNAs in 65nm CMOS to operate between 700 and 1400MHz have been completed (Fig. 4).  The fabricated integrated circuits have arrived and will be tested over the next few months.  Simulated noise temperatures of these LNAs are less than 10K. However, due to uncertainties in CMOS MOSFET noise models and to confirm whether the LNAs perform as simulated, careful measurements are still required to determine actual achievable noise temperatures.

Figure 5. 65-nm CMOS room temperature LNA design.

Multidimensional Signal Processing.

Over the past six months, colleagues at NRC-HIA DRAO have established a number of collaborative research projects with Dr. Len Bruton, from the Multidimensional Signal Processing Group at the University of Calgary, and Dr. Pan Agathoklis from the Digital Signal Processing Group at the University of Victoria.  With postdoctoral fellow, Dr. Arjuna Madanayake, and graduate students Thushara Guaratne and Najith Liyanage, they are investigating applications of high-performance IIR and FIR space-time multidimensional filters for enhancing radio astronomy signals.  So far, they have started three projects.  The first involves the design, synthesis and poly-phase FPGA implementation of 3D FIR cone filters for selectively filtering both focal-plane array and aperture phased array space-time signals.  The second project is an investigation of 3D IIR cone-type filters having hexagonal cross-sections and their implementation using FPGA circuits. The third project is a theoretical study of the 3D spectral characteristics of noise coupling, signal distortion and other interfering signals on focal plane arrays.  Novel techniques for attenuating such signals are under investigation.  Abstracts for all three research projects can be found at

http://www-mddsp.enel.ucalgary.ca/People/bruton/research.htm
and further details can be obtained from Dr. Len Bruton at bruton@ucalgary.ca

FPA simulations

Tony Willis, Bruce Veidt and Andrew Gray of NRC-HIA have been simulating astronomical observations made with focal plane arrays (FPA). Using the commercial GRASP antenna design package, they created a simulated array of 90-dipole elements in each of the orthogonal X and Y directions, mounted at the focal point of a 10-m dish and operating at a frequency of 1500 MHz. The spacing between the dipole elements is half a wavelength. This is not meant as a realistic design for an FPA, but as a good test-bed for examining the behaviour of an FPA from the viewpoint of reducing data obtained with an aperture synthesis radio telescope.

One of the goals of the SKA is high dynamic range imaging. Generally this is accomplished with a system whose instrumental response is constant as a function of time and position. One way of attempting to do this with our simulated FPA is to adjust the weights so that a phased-up beam has a constant total power response. The resulting total intensity (Stokes I) beams phased up on boresight and at an offset of 2 x FWHM are shown in Figure 5. However, while the phased up total intensity responses appear similar, the same cannot be said of the corresponding Q instrumental polarization responses (Figure 6). There are significant changes in the instrumental Q response as we move away from boresight. Further work is required to understand how to correct for instrumental polarization, but it is clear that for wide field imaging observations a significant computational effort will be required.

Figure 6a. Simulated Stokes I beams on boresight (left) and 2xFWHM offset from boresight (right).

Figure 6b . Simulated Stokes Q beams on boresight (left) and 2xFWHM offset (right).