Research Projects

Approved Research Projects for SKA SA Masters and Doctoral Bursary Students for 2017

When reviewing the approved projects for 2017, please ensure that you click on “View Project Proposal” just below the Project overview in order for you to see who the prospective Supervisor of the selected project is.

Masters Research Projects: Astronomy

C-BASS southern telescope characterisation and analysis

Overview

The C-Band All Sky Survey (C-BASS) is an experiment that will produce high signalto-noise maps of Galactic synchrotron emission by measuring the entire sky in temperature and polarisation at 5 GHz. Synchrotron emission is a primary source of foreground confusion for cosmic microwave background (CMB) experiments that are aiming to characterise the “B-mode” polarisation imparted by inflationary gravitational waves. A host of experiments are actively working to make a definitive detection of this signal, and precise removal of foregrounds will be one of the greatest analysis challenges. High-fidelity synchrotron templates from C-BASS will be essential for foreground monitoring and cleaning in these CMB experiments.

C-BASS comprises two telescopes that observe the northern and southern hemispheres. The southern system is located at the South African SKA site and has just begun science operations. This telescope in particular will play a critical role for CMB experiments since the vast majority of CMB observations (including BICEP2) are in the south. The student who takes on this project will help characterise the C-BASS southern system by analysing calibration measurements, as well as develop the data analysis pipeline. In contrast to the northern C-BASS system, the southern C-BASS telescope has a digital back end that measures 128 frequency bins within the 4.5–5.5 GHz bandwidth. This spectral information opens new avenues within the data analysis pipeline; for example, RFI flagging can be fine-tuned to take advantage of the multiple frequency measurements.

View project proposal

Drone-based beam characterisation for HERA

Overview

The redshifted 21-cm emission line of atomic hydrogen provides a unique window on reionization history and will become one of our most powerful cosmological probes in the coming decade. Because 21-cm emission maps a volume (rather than a surface), observations contain a tremendous amount of information and thus hold the potential for vastly improving cosmological parameter constraints. One of the main goals of current and upcoming experiments is to make a definitive detection of the 21-cm power spectrum from the epoch of reionization (EoR). Precise characterisation of this signal will allow us to understand how and when the first stars ignited in our universe.

South Africa already hosts the PAPER telescope, which has published the tightest limits yet on the EoR signal. The PAPER team is one of several collaborations that are joining forces to build HERA, a next-generation 21-cm telescope that will comprise over 300 dishes and will represent a significant leap in sensitivity over existing instruments. A 19-element HERA pathfinder has already been constructed in South Africa, and buildup of the final array is in progress. The full HERA array is expected to make the first significant detection of hydrogen during reionization.

Precise beam characterisation of HERA will be essential for interpreting the science data, and this measurement is made challenging by HERA’s large (14-m diameter), stationary dishes. The student who takes on this project will help develop a drone-based calibration system for mapping HERA beams. The work will involve learning how to use the drone, improving the drone and calibrator system for measurements that are specific to HERA, taking the actual beam measurements, and analysing the resulting beam data.

View project proposal

Exploring cosmic dawn from the sub-Antarctic with SCI-HI

Overview

SCI-HI is an experiment that will study cosmic dawn in the universe using low-frequency (< 150 MHz) observations of redshifted 21-cm emission from neutral hydrogen. The experiment, which is illustrated in Figures 1 and 2, is unique in that it comprises a single, portable antenna that observes the signal averaged over a large fraction of the visible sky. Measuring this global signal as a function of frequency/redshift opens a new window into a part of the universe’s history that is very poorly understood. One of the greatest challenges of this measurement is terrestrial radio frequency interference (RFI), which swamps the cosmological signal even when the nearest RFI sources are hundreds of kilometres away. SCI-HI has been funded by the South African National Antarctic Programme (SANAP) for deployments to Marion Island, which lies 2000 km from the nearest continental land masses and potentially offers an exceptionally clean RFI environment. SCI-HI has previously deployed to Guadalupe Island (200 km off the cost of Mexico), and the residual RFI levels were an order of magnitude larger than the expected cosmological signal. We expect the measurement to improve from Marion Island, which is even more remote than Guadalupe Island. Given the small scale of SCI-HI, the student who takes on this project will be able to contribute to a wide range of work spanning both instrumentation and analysis. The student will have the opportunity to participate in the second deployment of SCI-HI to Marion Island, which is scheduled for April 2017. The work will involve on-site operations, post-deployment lab testing of the hardware, improving the data analysis pipeline, and analysing the science data. View project proposal

HIRAX instrumentation and prototype characterisation

Overview

An exciting frontier of radio astronomy is using the redshifted 21-cm emission of neutral hydrogen to reconstruct a three-dimensional map of large-scale structure in the universe. These maps encode a faint imprint, known as baryon acoustic oscillations (BAOs), that correspond to remnant ripples left behind by sound waves echoing through the plasma of the early universe. Measurements from upcoming experiments will constrain BAOs with exquisite precision, opening new views into structure formation and the universe’s expansion history, and shedding light on the mystery of dark energy.

We are in the initial stages of building a new radio telescope array called the Hydrogen Intensity and Real-time Analysis eXperiment (HIRAX)1. HIRAX will measure BAOs by mapping the entire southern sky over a frequency range of 400–800 MHz, and the experiment will be sited in South Africa. The project is complementary to the Canadian Hydrogen Intensity Mapping Experiment (CHIME), which is about to begin surveying the northern sky. HIRAX has received seed funding, and prototype construction (Figure 1) is in progress.

The student who takes on this project will focus on the assembly and characterisation of an eight-element prototype array sited at HartRAO. This prototype is a critical milestone along the path to constructing the full science array, consisting of 1024 elements. The student will perform calibration measurements on the eight-element array, including beam mapping with a drone-based calibration source. The student will analyse the resulting data and will work to improve the HIRAX design for the following prototype phase.

View project proposal

Observations of diffuse radio emission in galazy clusters

Overview

View project proposal

HIRAX Design and Calibration

Overview

The Hydrogen Intensity and Real-time Analysis eXperiment (HIRAX) is an approximately 1,000 element array with approval from the NRF and funding already in place for 128 elements. The main science goals are intensity mapping of neutral hydrogen between redshifts 0.8 and 2.5, radio transient searches, pulsar search and monitoring, and high-redshift neutral hydrogen absorbers. The base design is 1,204 close-packed 6m dishes we plan to site in the Karoo. This project aims to develop GPU code to form beams for pulsar and fast radio burst searches. The project is suitable for an MSc degree.

The HIRAX correlator will consist of GPU nodes that will form n2 correlation products for the neutral hydrogen studies. Separately, we plan to form tied-array beams filling the primary beam for FRB, pulsar, and hydrogen absorber searches. The beamforming in principle should only be about 10-15% more work compared to correlation, however this needs to be demonstrated on GPUs. The goal of this project is to develop that code and test on early hardware. A prototype array already exists at HartRAO, where the student can begin testing code on real data immediately.

View project proposal

Diffuse Radio Emission in ACTPol/MUSTANG Clusters

Overview

View project proposal

HIRAX: Analysis, foregrounds and cosmological cross-correlations

Overview

View project proposal

HI intensity mapping with MeerKAT and the SKA

Overview

HI intensity mapping promises to be an exquisite way of mapping the large scale structure of the Universe. The technique relies on measuring the integrated intensity of HI (21 cm) emission on each pixel on the sky without actually resolving any galaxies. In particular, MeerKAT should be able to measure this large scale emission, if used in single dish mode (instead of cross-correlating the dishes). With the full array of 64 dishes, it should be able to measure the distribution of hydrogen on very large scales and detect the baryon acoustic oscillations. This in turn will allow probing several cosmological parameters such as the nature of dark energy. This project aims to develop calibration techniques for single dish surveys with MeerKAT. In particular, making use of the new beam-former mode developed for MeerKAT to observe pulsars and use them as exquisite calibrators. This analysis is crucial to settle the survey strategy for HI observations with the SKA and a publication in an international peer reviewed journal is expected. Moreover, if the calibration process is successful, observations with MeerKAT early science can provide the first ever direct detection of the HI intensity mapping signal.

View project proposal

HI stacking with maximum likelihood techniques

Overview

View project proposal

VLA and VLBA analysis of Nova V445 Puppis

Overview

The thermonuclear eruption on the surface of a white dwarf (a nova eruption) occurs following extensive accretion of matter from a secondary, less evolved companion star. The accreted, and subsequently expelled, matter has been largely in the form of hydrogen however, in the case of Nova V445 Puppis, which had an eruption in 2000, the matter appeared to be helium with no signs of hydrogen. Nova V445 Pup was shown to be expanding at very high velocities (~8000 km/s) with a bipolar outflow morphology (Woudt et al. 2009, ApJ, 706, 738).

Considering V445 Puppis is the first, and so far only, nova where no hydrogen has been observed poses interesting questions such as, what happened to the hydrogen and may provide a vital link in understanding the progenitors of type Ia supernovae. The nova was observed at various wavelengths however, radio observations have largely remained unpublished. Radio observations provide essential clues to the nature of the system and provide information on fundamental parameters of the nova eruption such as, the ejected mass, kinetic energy and ejecta geometry.

The aim of the research project will be to understand the nature of the radio observations of nova V445 Pup. The student will reduce and analyse archival Very Large Array (VLA) and Very Long Baseline Array (VLBA) observations. The fundamental parameters will be derived based on nonspherical morphokinematical models (Ribeiro et al. 2014, ApJ, 792, 57). These observations will allow comparison with theoretical expectations.

View project proposal

Developing advanced beam models of MeerKAT antennas using interferometer data

Overview

The image is obtained by combining visibilities from various antennas having varied beams. A nominal primary beam correction is applied to the image, to get the true image. However, the interferometric image beam could be widely different from individual antenna beams. Techniques will be developed to measure the combined interferometer beam using the data used for mapping, and errors will be quantified as compared to individual antenna beams. The project will be using MeerKAT AR1 data.

View project proposal

Analyzing The Impact Of Ghosts On Future Cosmological Radio Surveys

Overview

Ghosts are calibration artefacts that are always present in radio maps (even if only below the noise) that are calibrated with incomplete sky models, as would be the case for any blind survey. While the presence of ghosts and the mechanism by which they form has been conclusively established (see recent work by Grobler, Smirnov et al.), their impact on the science performance of future deep surveys needs to be studied. Ghosts show up in specific configurations determined by the interferometer layout, and In principle affect the statistical distribution of galaxies in an image. The aim of this project is to obtain a scientific measure of how much ghosts affect the power spectrum in the intensity mapping of galaxies and to determine if they are serious enough to put a show-stopper in cosmological measurements of the galaxy distribution. The student would attempt to calibrate radio data with incomplete models and hence obtain images with ghosts and obtain a measurement of the n-point correlation functions to measure cosmological parameters. These measurements would be biased due to the presence of calibration ghosts. The magnitude of the measurements would yield information about how well calibration needs to be done (i.e. how deep the sky models need to be) in order to obtain correlations of galaxies which are not corrupted by the presence of ghosts. Other related observational techniques such as intensity mapping. This is directly relevant to SKA1 and SKA2. The results of this work will inform the calibration strategies for the next generation of deep radio surveys with JVLA, MeerKAT and SKA1.

Supervisors: Prof O. Smirnov (osmirnov@gmail.com), Prof F.B. Abdalla (UCL/Rhodes) (fba@star.ucl.ac.uk)

Developing pipelines for optimal treatment of beam-related direction-dependent effects

Overview

With the high sensitivities of the new generation of SKA pathfinder telescopes (MeerKAT, LOFAR, JVLA), as well as SKA1 itself, direction-dependent effects are becoming the major bottleneck for high-fidelity imaging. A number of tools for the treatment of DDEs have been developed over the past years, such as A-projection (Bhatnagar 2008), differential gains (Smirnov 2011), facet-based imaging (Tasse in prep.), etc. Software implementations of these algorithms are now available, and can be applied to real data. The aim of this project is to develop and validate an imaging pipeline based on the latest calibration and imaging tools, using existing JVLA high-dynamic range data, as well as archival KAT-7 data and commissioning data from MeerKAT (subject to availability). This pipeline should incorporate corrections for beam-related effects using the latest A-projection and/or facet-based imager implementations. With the variety of options and parameters available to these new tools, finding the optimal combinations of parameters (such as faceting directions, DD solution intervals, etc.) forms an algorithmic research aspect of its own. The results of this work will inform the calibration and imaging strategies for the next generation of deep radio surveys with JVLA, MeerKAT and SKA1.

Supervisors: Prof O. Smirnov (osmirnov@gmail.com), Dr. C. Tasse (Obs. Paris) (cyril.tasse@obspm.fr)

Masters Research Projects: Engineering

CoBiDA: Cognitive Architecture for Astronomical Big Data Analysis

Overview

Brain has intrigued researchers since the beginning of scientific endeavors. Firstly, beginning of computers saw the advent of exciting developments which culminated to the development of the new discipline of artificial neural networks (ANN). ANNs have been through several generations of major developments, with the recent phase consisting of spiking neural networks based works. Another parallel field of computational neuroscience has been the bio-inspired cognitive architectures (BICA) a field which got major thrust in development. Cognitive architecture (CA) in general and BICA in particular also has a long history and the efforts have been devoted towards trying to emulate the functioning of brain. CAs like SOAR and ACT-R have been under development for many decades and have been applied in various studies. A third direction in cognitive engineering has been the recent developments in communication and radar which are misleadingly termed cognitive communication and cognitive radar. It may also be mentioned here that this 2012-2013 has seen multi-billion dollar investment done separately in the European Union as well as in the USA for the study and understanding of brain.

In this proposed MSc project the student shall investigate the use of cognitive architectures in tracking and extracting information from the large amount of data generated by MeerKAT. In addition to being successful in a range of engineering applications, BICA has been hinted by some SKA partners to have very high potential in handling astronomical datasets.

View project proposal

CogPro: Cognitive Processor for Astronomical Big Data Analysis

Overview

Brain has intrigued researchers since the beginning of scientific endeavors. Firstly, beginning of computers saw the advent of exciting developments which culminated to the development of the new discipline of artificial neural networks (ANN). ANNs have been through several generations of major developments, with the recent phase consisting of spiking neural networks based works. Another parallel field of computational neuroscience has been the bio-inspired cognitive architectures (BICA) a field which got major thrust in development. Cognitive architecture (CA) in general and BICA in particular also has a long history and the efforts have been devoted towards trying to emulate the functioning of brain. CAs like SOAR and ACT-R have been under development for many decades and have been applied in various studies. A third direction in cognitive engineering has been the recent developments in communication and radar which are misleadingly termed cognitive communication and cognitive radar. It may also be mentioned here that this 2012-2013 has seen multi-billion dollar investment done separately in the European Union as well as in the USA for the study and understanding of brain.

View project proposal

Visualising RF signals

Overview

The aim of this project is to visualize radio sources. Almost like the SKA radio telescope itself, this device will scan an area (now just horizontally), and produce images of the radio sources in the vicinity. This device can be used to identify radio frequency interferers in the vicinity of the SKA telescope. Many other interesting data can also be deducted from this device, such as RF Signal “contour maps”, etc. Useful statistics and metrics can be researched. This project will also act as a small scale “Meerkat”, where the student is introduced to many radio astronomy principles.

View project proposal

Co-use of the SKA mid-frequency aperture array (MFAA)

Overview

The SKA Mid-Frequency Aperture Array (SKA-MFAA) is part of Phase 2 of the SKA instrument, provisionally scheduled for deployment in the mid-2020s. The instrument will use aperture arrays, covering the frequency band from approximately 500-1500 MHz. It will be a very wide field of view instrument, characterized by electronically steerable beam(s). A presentation at the recent MFAA/MIDPREP meeting (Cape Town, April 2016) proposed co-use of such an array, in particular for the detection of space junk using passive radar techniques, and this thesis aims investigate the feasibility of such a proposal, and to consider other possible science and/or engineering applications of an MFAA.

View project proposal

Integrating a DBE with, and commissioning of, an SKA midfrequency aperture array (MFAA) tile

Overview

The SKA Mid-Frequency Aperture Array (SKA-MFAA) is part of Phase 2 of the SKA instrument, provisionally scheduled for deployment in the mid-2020s. The instrument will use aperture arrays, covering the frequency band from approximately 500-1500 MHz. It will be a very wide field of view instrument, characterized by electronically steerable beam(s). A presentation at the recent MFAA/MIDPREP meeting (Cape Town, April 2016) proposed that “tiles” be made available to South African educational institutions. These will probably be based on developments to the EMBRACE tile, which is the basic building block of the system, consisting of 72 Vivaldi antennas (6x6x2 polarisations) (Torchinsky et al, A&A, 2016). The system uses a hybrid beam-forming strategy, with the first stages done in RF using beam-former chips and subsequent stages done using digital beamforming. The system supports two independent RF channels, permitting two independently steerable beams on the sky.

The aim of this project is to commission such a system, after integrating it with a suitable Digital Back End (DBE) processor, such as a ROACH board. (EMBRACE uses custom-made DSP hardware based on LOFAR hardware). Additionally, the project should demonstrate the ability to observe very strong radio astronomy sources (eg the Sun) – the project will be undertaken in Stellenbosch, where the RFI environment is poor. The aim is to develop engineering skills in aperture array systems.

View project proposal

Alternative Characterisation Techniques for RFI Antennas

Overview

Antennas designed for RFI and EMC purposes with large bandwidths can pose challenges when characterisation measurements need to be done using conventional far-field characterisation techniques, especially when the antennas are large. This project will investigate alternative methods of characterisation, including time domain techniques and one-port gain measurements in the vicinity of a highly reflective ground plane orthogonal to the antenna main beam.

View project proposal

UAV-Assisted EM Characterisation of SKA-SA Core Site

Overview

The main objective of this research project is to develop a suitable UAV platform that is able to conduct EM measurements across a reasonable area of the SKA-SA Core Site.

The ultimate goal would be to use this for characterisation of existing attenuation maps used to determine the location of allowable intentional radiators on and around the site.

The UAV would have to make use of a suitable measurements platform with acceptable sensitivity characteristics. It can either store measured information on-board or relay the data via a suitable RF link to a local base station. The UAV platform and flight path would have to be developed in order to optimise payload and flight time. Suitable control software for monitoring of data and UAV location can be developed or alternatively information can be fed into a Google Maps/Earth environment.

The measurement system would have to be calibrated to allow comparison of results with alternative data as well as propagation analysis.

View project proposal

Numerical electromagnetic analysis for the SKA mid-frequency aperture array

Overview

Numerical electromagnetic analysis has become an indispensable tool in modern antenna design. The design of the mid-frequency aperture array involves very large computational electromagnetics analyses. In such cases the computational cost can quickly become prohibitive and it becomes necessary to investigate more efficient numerical methods. For this project, the candidate must firstly study cutting-edge array analysis methods. Secondly, new research work that has recently been done at SU on efficient array analysis methods must be implemented as a computer code and further refined. Thirdly, the work must be applied to SKA mid-frequency aperture array prototypes. This challenging project has excellent scope for publication and can serve as a platform for further, advanced studies.

View project proposal

Non-linear Analysis of Radio Astronomical Data

Overview

Radio Astronomy uses the radio frequency spectrum for detecting weak signals emitted by celestial sources. The spectral properties of the signals emitted by a celestial source are completely determined by the physical processes active there. Increasing the sensitivity of the detection and acquisition system necessarily increases its sensitivity to radio interference sources too.

Non-linear Signal Analysis are far more powerful to unlock the information hidden within signals which are not accessible using ordinary linear Signal Analysis.

Non-linear analysis of such radio astronomical data is achieved using qualitative techniques such as time series and spectrum, bispectrum, phase portrait, Poincare section, polar plot, spatiotemporal analysis, wavelet analysis and distance plots, as well as quantitative analysis techniques including the Lyapunov exponent, Kolmogorov entropy and fractal dimension. Recent research in the world shows that these techniques are useful for astronomical data analysis.

View project proposal

Wavelets Analysis for RFI Mitigation

Overview

Radio Astronomy uses the radio frequency spectrum for detecting weak signals emitted by celestial sources. The spectral properties of the signals emitted by a celestial source are completely determined by the physical processes active there. Increasing the sensitivity of the detection and acquisition system necessarily increases its sensitivity to radio interference sources too. Part of the problem of RFI is a direct result of man-made radio emissions which include communications and radar. Although beamforming and –steering help to reduce some of the RFI these are not effective to all types of RFI. It is also important to realise that no universal RFI mitigation methodology exists for use in Radio Astronomy.

View project proposal

The design and implementation of a FPGA-based Wideband Spectrometer instrument for use at the Hartebeeshoek Radio Astronomy Observatory (HartRAO)

Overview

In 2013 our research group in the Faculty of Engineering and the Built Environment was in contact with Michael Gaylard and Sarah Buchner from the HartRAO observatory north of Johannesburg and identified a potential project that would be beneficial to the observatory as an entry point into the new generation of FGPA based Radio Astronomy instrumentation. Unfortunately a student could not be found for the project at that time and this was followed by the tragic passing of Michael Gaylard.

Since then we have been approached by Dr Gordon MacLeod who is now working at the HartRAO telescope. He had heard about our previous interactions and informed me that they were still looking to collaborate on the development of next generation instrumentation for use at HartRAO. Dr MacLeod has mentioned the need for a number of different instruments at HartRAO and that they need to begin developing some expertise to build their own instruments and support the African VLBI network initiative. We have since met to discuss their needs and have identified an initial instrumentation project that we can undertake to kick off our collaboration.

The goal of this research project is to design and build a new generation wideband spectrometer instrument for the HartRAO observatory. The current generation of hardware in use at HartRAO is reaching the end of its lifespan and is not capable of implementing the most modern forms of signal processing algorithms that are being used for Radio Astronomy today.

The proposed instrument will be built using the ROACH II platform which is a FPGA based platform being developed by the Radio Astronomy Community whose collaboration is collectively called CASPER. The capabilities of the spectrometer will be closely matched to the current research activity taking place at HartRAO.

This project will focus on the following primary objectives:

  • Coming to grips with modern FPGA hardware design paradigms including the existing CASPER toolchain.
  • Exploring the use of the ROACH architecture for signal processing tasks.
  • Introducing the research group to the precepts of Radio Astronomy and its signal processing challenges.
  • Produce and implement a modern spectrometer instrument that can realize the full potential of the antennae at HartRAO and will allow for the future addition of new observation capabilities such as Transient Pulsar detection and Pulsar timing.

While wideband spectrometers are widely understood and a number of off-the-shelf implementations exist for the ROACH II platform. However, we are told that the majority of implementations that are capable of meeting the HartRAO requirements rely on addition processing to be conducted on external computing hardware, usually with the use of GPUs. This increases both the latency and complexity of such a system. This project will seek to explore an optimized spectrometer design that can perform as much signal integration as possible onboard the ROACH II. We will explore the existing FPGA design toolchain to quantify resource usage of various components and based of this study produce an optimized spectrometer that meets the research needs of the HartRAO scientists The research will be focused on the field of Real-time Signal Processing for Radio Astronomy as described as a priority area in the Call for Project Proposals Document.

View project proposal

Development of a MeerKAT Extended System Concept for the 14.5 to 22 GHz Range

Overview

View project proposal

Development of a MeerKAT Multiband Feed Horn for the 14.5 to 22 GHz Range

Overview

View project proposal

Dynamic Optimization Algorithms for Data Transport over Optical Fibre Flex Spectrum

Overview

Flex Spectrum Dense Wavelength Division Multiplexed (DWDM) optical fibre networks are next generation technology for handling extremely high data rates of the kind produced by MeerKAT and SKA. In order for these networks to function optimally, novel dynamic algorithms need to be identified and developed for performing critical tasks such as wavelength assignment, wavelength routing, network restoration and protection. The project identifies and researches dynamic optimization control algorithms in conjunction with a dense wavelength division multiplex (DMDM) flex spectrum network architecture as a solution. Algorithms under consideration include genetic algorithms, evolutionary algorithms, particle swarm optimization, ant colony optimization etc. A further research consideration is as to whether the control plane performs best through a centralized or distributed network implementation. Such algorithms and control plane offer the power and flexibility for real time implementation. The focus of the project is within computer science and will be realized through the development and testing of novel control algorithms for future flex spectrum optical communication networks.

View project proposal

Timing and Reference Signal Distribution over Optical Fibre for Telescope Networks

Overview

Optical fibre technology forms the buried backbone of both MeerKAT and SKA. Both telescopes rely on highly stable clock tones to be distributed over optical fibre to each individual antenna. These clock signals are crucial for driving the digitizers, time-stamping the data, and for monitoring and control functions. Complex interplay between a plethora of effects causes instability in the phase of a lightwave clock tone as it propagates within an optical fibre. These effects include temperature fluctuation, birefringence, polarization instability, component noise and others. The project aims to identify and quantify all such individual effects and investigate the complex interplay between them. A full suite of measurement and characterization techniques (including polarimetric, phase noise, time domain etc.) relevant to both the MeerKAT and SKA clock distribution systems will be used.

View project proposal

A conformal phased array antenna for mm-wave water vapour radiometry

Overview

Accurate estimation of tropospheric water vapour (and appropriate path length correction) is, admittedly, of less value at low observation frequencies (MFAA, MeerKAT L-band and SKA Bands 1-3). It will, however, improve the quality of observations > 5 GHz, as is envisioned for SKA Band 5 and MeerKAT X-band. In that sense, this work will generate data to aid the roll-out of “Wide-band single-pixel receivers”, as well as (perhaps) improve on the quality of “C-BASS data processing and analysis”. In addition, the student will be introduced to the principles and practices of array synthesis, which would serve them well in future MFAA involvement.

View project proposal

A generic digital back-end and data management system for water vapour radiometry

Overview

Accurate estimation of tropospheric water vapour (and appropriate path length correction) is, admittedly, of less value at low observation frequencies (MFAA, MeerKAT L-band and SKA Bands 1-3). It will, however, improve the quality of observations > 5 GHz, as is envisioned for SKA Band 5 and MeerKAT X-band. In that sense, this work will generate data to aid the roll-out of “Wide-band single-pixel receivers”, as well as (perhaps) improve on the quality of “C-BASS data processing and analysis”. In addition, the student will be introduced to the principles and practices of “Real-time Signal Processing instrumentation for Radio Astronomy, specifically using FPGA and GPU platforms”.

View project proposal

X-band cryo-cooled low noise amplifiers integrated in SiGe BiCMOS

Overview

This topic relates to the priority area “Wide-band single-pixel receivers”. The cryogenic LNA is the most critical microelectronic component in the receiver front-end, and any performance improvement in this component will improve the sensitivity of the instrument as a whole. This development targets the future X-band extension of MeerKAT, as well as SKA Band 5.

View project proposal

Doctoral Research Projects: Astronomy

Galactic foreground characterisation with C-BASS

Overview

The C-Band All Sky Survey (C-BASS) is an experiment that will produce high signalto-noise maps of Galactic synchrotron emission by measuring the entire sky in temperature and polarisation at 5 GHz. Synchrotron emission is a primary source of foreground confusion for cosmic microwave background (CMB) experiments that are aiming to characterise the “B-mode” polarisation imparted by inflationary gravitational waves. A host of experiments are actively working to make a definitive detection of this signal, and precise removal of foregrounds will be one of the greatest analysis challenges. High-fidelity synchrotron templates from C-BASS will be essential for foreground monitoring and cleaning in these CMB experiments.

C-BASS comprises two telescopes that observe the northern and southern hemispheres. The southern system is located at the South African SKA site and has just begun science operations. This telescope in particular will play a critical role for CMB experiments since the vast majority of CMB observations (including BICEP2) are in the south.

The student who takes on this project will help develop the data analysis pipeline for the C-BASS southern system and will perform an in-depth study of techniques to separate synchrotron emission from CMB polarisation measurements. UKZN is a collaborating institution on two dedicated B-mode experiments (SPIDER, ABS) and several other CMB experiments (Planck, ACTPOL, SPTPOL), and part of this project will entail writing foreground separation code that is tailored to one of these projects.

View project proposal

Studying the Epoch of Reionization with HERA

Overview

View project proposal

Spatially resolved stacking of line observations of the neutral gas in galaxies

Overview

View project proposal

Finding the Cosmic Dawn - Analysis of SCI-HI Data

Overview

How the first stars formed and lit up the universe is unkown. From Cosmic Microwave Background data, we know that a significant fraction of the universe had reionized by redshift  ~ 8 – 9, which means the first ionizing sources in the universe had to have formed before then. These sources have to leave imprints on the neutral hydrogen 21-cm line, which therefore provides a way to study them. There are two main ways to see their impact – either via intensity fluctuations, or through a change in the mean global brightness of the sky as a function of frequency. This project will search for the signature of the very first stars via the global signal, using data from the SCI-HI experiment.

SCI-HI is a single-antenna experiment sensitive to the mean sky brightness across the range of frequencies (100 MHz) where the global signal is expected to reside. One of the challenges in making a global measurement is radio frequency interference (RFI), which can be extremely bright since the signal sits in or near the FM radio band. Therefore, we deploy SCI-HI to remote locations. Its initial deployment was to Guadalupe Island,
200 km off the coast of Mexico, where RFI from the mainland was still found to be a problem. Since RFI was still a problem there, we applied to the South African National Antarctic Programme to deploy SCI-HI to Marion Island, 2,000 km south of South Africa. The proposal was successful, and so we have been funded to deploy SCI-HI to Marion, with the first voyage taking place in April/May 2016. Marion’s southerly latitude provides an additional benefit. Fluctuations in the Earth’s ionosphere can make it hard to remove the (much larger) galactic signal from the EoR one. The ionosphere is driven by solar ionization, so polar locations should have more stable ionospheres. Nights on Marion look to have particularly good ionospheric conditions, with Maximum Usable Frequencies (the frequency at which radio waves reflect of the ionosphere) regularly as low as 5-6 MHz.

This project will analyze data from SCI-HI looking for the signal from the first stars. The student will work closeley with other members of the SCI-HI collaboration. Initially, the student will use the pipeline developed for the Guadalupe run, but will expand it. Data both from Guadalupe, a test run we carried out at the SKA site in the Karoo, and the initial Marion deployment will be used as test cases. The 2017 Marion deployment will gather significantly more data, so repeat observations can be used to search for and remove more possible systematic errors in the data. The student will process the data from the upcoming Marion run and publish results. The student will also spend time looking at ionospheric effects, models for MilkyWay emission, and how the two couple together. We expect this to lead to either a more sensitive detection or significantly improved limits on the global signal.

View project proposal

Diffuse Radio Emission in ACTPol/MUSTANG Clusters

Overview

View project proposal

First light: studying the epoch of reionisation

Overview

The Epoch of Reionization (EoR) is a crucial stage in the evolution of the Universe, signaling the birth of the first stars and galaxies. Several experiments have been done to probe this signal: LOFAR (Netherlands), MWA(Australia), PAPER (EUA/SA) and two future ones are now being developed: HERA (based in SA, already in construction) and the future SKA1-LOW. This PhD project aims to fully study the epoch of reionisation from different angles, both using simulations and actual data. It will be based on 3 main points: i) development of end to end simulations including signal, foregrounds and instrumental effects; ii) development of estimators and cleaning methods capable of extracting the relevant cosmological signal from the data; iii) use of actual observations using HERA in order to inform the simulations (in particular the HERA dishes primary beam). Part of this work will be done in collaboration with the HERA team and the student will be expected to become part of the HERA core team. Moreover, as data becomes available, it is expected that the student will use the know-how obtained from simulations to mine the data and extract some real cosmological constraints. It is also expected that the simulations of the 21cm signal and statistical techniques developed by the student can also be applied to other projects relevant for SKA-LOW, such as the study of the observational properties of high-z radio galaxies.

View project proposal

Gravitationally lensed HI with MeerKAT and the SKA

Overview

Strong gravitational lensing provides the deepest views of the Universe through its magnification of the solid angle of distant galaxies combined with the conservation of surface brightness. It enables studies of highredshift galaxies only possible with next-generation facilities without the lensing phenomenon. To date, HI has only been detected directly at low redshifts, limited by the sensitivity and frequency range of current radio telescopes. MeerKAT and SKA1-MID will dramatically change this picture, pushing out to redshifts of z~1 for the several thousand hour surveys proposed.

However, what has not been considered in the MeerKAT science case, is the ability to detect gravitationally lensed HI emission in high-redshift galaxies. The instantaneous bandwidth and sensitivity of MeerKAT will yield the potential to produce high-impact early science. In our MNRAS Letter (Deane, Obreschkow & Heywood, 2015), we demonstrate that SKA precursors have the potential to make the highest redshift HI detections to date within a fraction of the total duration of the deep HI surveys, provided the appropriate targeted lensed surveys are designed.

The primary objectives of this PhD project are fourfold:

  1. design and optimise a MeerKAT lensing survey on known lensed systems and galaxy clusters
  2. use current instruments (including MeerKAT early array realeases) to make the highest redshift HI detections to date
  3. use our large-scale HI simulation (SAX-Lens) in combination with machine learning techniques to determine source prioritisation best practices for multi-wavelength followup of the expected >10,000 lens candidates in SKA1-MID surveys
  4. investigate the ability of SKA1-LOW to detect the highest redshift HI galaxies

This ambitious programme is enabled by the significant hardware and software successes within the MeerKAT programme and Rhodes University, providing an opportunity to lead this scientific field into the SKA era.

View project proposal

Machine Learning Techniques for Radio Continuum Survey Studies

Overview

The study of the faint radio universe has recently become a very active field of research not only because of the promise of transformational capabilities of the SKA on this field, but also because of the major steps beging taken and planned with SKA pathfinder and precursors. Upcoming radio surveys with the South African SKA precursor MeerKAT (2017-) and with the SKA1 (2020-) will routinely detect radio emission from Star Forming Galaxies (SFG) and Active Galactic Nuclei (AGN) up to high redshifts over large areas of the sky. Such data will dramatically improve sample sizes allowing us to trace the evolution of these radio source populations over cosmic time. In advance of completion of MeerKAT and the SKA1 one can make the first steps in this investigation using existing deep radio data sets both to do science along the way and to develop and test the approaches and analysis algorithms.

To that end, the deepest existing data from JVLA and GMRT surveys led by Prof Taylor are in hand, covering well-studied fields with deep and homogeneous ancillary data, spanning frequencies from 600 MHz right up to 10 GHz, in both polarization and total intensity. Ongoing and future GMRT, JVLA and MeerKAT survey programs incuding the MIGHTEE survey will soon expand upon current capabilities and pave the way for SKA1 science by producing deep surveys of the radio sky detecting millions of faint radio sources.

A key requirement to fully exploit upcoming radio continuum surveys is advanced algorithms for the automatic classification and characterization of very large numbers of radio sources using multi-wavelength (ultraviolet, optical, near infrared, far-infrared, sub-millimeter) data. This is an area where machine learning techniques have recently led to rapid progress, thanks to large homogeneous multi-wavelength datasets, faster computing facilities and improved algorithms.

View project proposal

Development of new calibration techniques for variable sources using the MeerKAT and KAT-7 observations

Overview

(i) Develop a method to identify variable sources while mapping using the existing multi-epoch data of MeerKAT-AR1 and KAT-7. Example of Extragalactic variable sources are blazars and galactic variable sources are pulsars.

(ii) Study the impact of these variability on calibration solutions and dynamic range of the map. These sources will directly effect direction dependent solving techniques, whereby the source variability will get attributed to instrumental and beam related errors, thereby also spoiling flux density measurements of sources in their vicinity.

(iii) Subsequently, develop new calibration techniques taking into account the variability of such sources and improve dynamic range of map.

View project proposal

Using HI stacking to determine the Cosmic Neutral Gas Density at Intermediate Redshift

Overview

View project proposal

Component Separation and data analysis for 21-cm intensity mapping

Overview

A major challenge for all intensity mapping experiments, and Epoch of Re-ionization (EoR) which is a primary goal of SKA, is the separation of the cosmological signal from the much brighter foreground emission. Foregrounds including diffuse Galactic synchrotron and free-free emission as well as bright extragalactic sources, which are several of orders of magnitude brighter than the HI signal. The problem is analogous to component separation, which is a key feature of cosmic microwave background (CMB) data analysis, such as for the WMAP and Planck space missions. Indeed, many of the algorithms developed within the CMB community are being considered for intensity mapping such as parametric fitting, Principal Component Analysis (PCA), Independent Component Analysis (ICA) and Generalized Morphological Component Analysis (GMCA) among others.

The student will develop and test component separation techniques for intensity mapping experiments. This will require detailed end-to-end simulations of the data, including real non- idealities such as mis-calibration, 1/f noise and beam effects. The first methods will involve using standard techniques (e.g. parametric fitting, PCA) to utilize the spectral smoothness of the foregrounds. The impact of systematic effects will be quantified. We will also develop new algorithms, that will ideally use both the spectral and spatial information in the data. We will also implement and review the other component separation techniques, such as PCA method, ICA method, and GNILC method. A key aspect will be the quantification of the separation on the likelihood, and therefore on cosmological parameters, using monte carlo simulations. This will have wide applications including SKA EoR studies, COmapping and HI intensity mapping.

View project proposal

Sparsity-based methods for multi-frequency, rotation measure synthesis deconvolution

Overview

Recent progress in compressive sensing (CS) methods has yielded a number of deconvolution algorithms based on the sparsity assumption of the sky that seem far better at recovering extended emission than CLEAN and its multiscale variations. For example, the MORESANE (Dabbech et al. 2015, arXiv:1412.5387) algorithm as implemented and released by J. Kenyon (Rhodes PhD student, see http://github.com/ratt-ru/PyMORESANE) has already given us the deepest-ever images of Cyg A in S-band (using JVLA data). Other promising algorithms include PURIFY (Carrillo et al. 2014) and LOFAR-CS (Garsden at al. 2015, arXiv:1406.7242 — one of the key developers of the latter did a postdoc at Rhodes/SKA SA and remains a collaborator of the project). New work by Ferrari et al. (2015, arXiv:1504.06847, same team as MORESANE) has extended this work into the multi-frequency regime (and other teams are working on similar extension). We propose to develop, jointly with the Ferrari et al. team (and possibly other teams, e.g. using Morphological Component Analysis at AIM), a sparsity-based deconvolution method for full-polarization multi-frequency rotation measure (RM) synthesis. Relevance to MeerKAT and the SKA. MeerKAT and SKA1 will enable wide-band radio observations to unprecedented sensitivity. A lot more faint extended and polarized emission will be detectable, but the wide frequency band makes it crucial to incorporate the frequency and polarization axis into the deconvolution process if the full potential of these instruments is to be realized.

Supervisors: Prof O. Smirnov (osmirnov@gmail.com), Dr Julien N. Girard (RATT/Rhodes & AIM/CEA-Saclay) (jgirard@ska.ac.za /julien.girard@cea.fr)

Understanding the Limits of Interferometric Techniques for Epoch of Reionisation Detection

Overview

The Epoch of Reionisation (EoR) is the next frontier for the cosmologist. The faint signal arising from this epoch can be measured by radio interferometers at low frequencies capturing the red shifted signal from the 21cm line. However, an outstanding issue not clearly resolved by any current EoR experiment is that calibration issues could plague this signal and prevent a clear detection from taking place. In particular, it is possible that direction dependent effects (DDEs) will change the nature of this signal and smear any possible detection. We propose to investigate how DDEs affect this cosmological signal, and whether the statistical properties of this signal are maintained after DDEs are calibrated out, and more specifically if the magnitude of the fluctuations is maintained. The student would obtain simulations of the 21cm signal, pass them through a radio interferometry simulator, and obtain visibilities which would be corrupted with a direction dependent signal. This signal would be calibrated with DD-solution algorithms such as SAGECAL or StefCal, and the signal would be remeasured and compared to the original input. The result of this research would place clear constrains on our ability to measure the EoR signal using current and future interferometers, and inform the calibration strategy of future EoR experiments.

Supervisors: Prof O. Smirnov (osmirnov@gmail.com), Prof F.B. Abdalla (UCL/Rhodes) (fba@star.ucl.ac.uk)

Bayesian Inference for Polarimetric Calibration

Overview

Recent work on BIRO (Bayesian Inference for Radio Observations) by Lochner at al. (2015) and Natarajan et al. (in prep.) has demonstrated the promise of Bayesian inference techniques in the domain of radio interferometric calibration. We now propose to apply BIRO to polarization calibration. Accurate polarization calibration in radio interferometry is a difficult problem, as the instrumental polarization leakage terms interact with the time-variable gain terms, and also need to be disentangled from intrinsic source polarization. The matrix-based radio interferometry measurement equation (RIME) provides a complete mathematical model for forward modelling of this process; inverting the model to go from observed visibilities back to polarization properties is a different matter. The Bayesian approach seems ideally suited to this, its particular strength being the ability to map out correlations between instrumental and scientific parameters while imposing prior constraints. Relevance to MeerKAT and the SKA. A number of key MeerKAT and SKA1 surveys will rely on accurate polarimetric calibration to achieve their science goals. The problem is especially severe with the linearly polarized feeds employed by KAT-7, MeerKAT and SKA1. Urgent progress on this problem is required.

Supervisors: Prof O. Smirnov (osmirnov@gmail.com), Prof. R. Taylor (UCT/UWC) (russ@ast.uct.ac.za)

Doctoral Research Projects: Engineering

AlGrABA: Algebraic Graph Theory based Astronomical Big Data Analysis

Overview

Applying algebra to study the properties of graphs is a classic area of research in mathematics. With the advent of big-data type challenges the use of graph is gathering interest amongst the data processing engineers. This follows from the fact that bigdata mostly originates because of a large number of sensing nodes. Some of the recent works have looked into developing signal processing framework based on graphs. Tools, inspired by such work, has shown much promise.

Representing astronomical data as a graph at each instant of data-capture is intuitive. Hence, we hypothesize that by exploring ways to 1) represent and 2) manipulate the astronomical recordings in a graph framework will help not only in presenting an elegant framework to handle and store massive amount of data, it will also help in making the data analytics processes more robust and less prone to noisy and false recognition-results.

In this proposed PhD project the student will work on the hypothesis that using algebraic graph theory based approaches will make astronomical data handling and processing faster and more robust.

View project proposal

Wideband Feed for SKA

Overview

A high fidelity feed with an operating bandwidth of more than an octave remains an unsolved problem in radio telescope design. Several feeds have recently been developed around the world with wide operating bandwidths (Eleven Feed, Quad-Ridged Horns, etc), but with inferior performance to the so-called octave band horns that are used in the MeerKAT system.

Given the wide range of science goals of the SKA, however, some trade-offs between different performance metrics of wide-band feeds must be considered. This is, however, normally a difficult task since much of the information might not be available in detail beforehand.

A solution to this issue is to perform formal multi-objective optimization of the feeds, so that the system engineer has access to the exact trade-off levels encountered for each antenna technology. Traditional methods of performing such optimizations are computationally prohibitively expensive, due to the long simulation times of full wave solvers, and the very large number of evaluations required to properly explore the (often high dimensional) design space. Surrogate based optimization (SBO) is a technique well suited to solve such problems. Here a coarse model is sought, which is very fast to evaluate, but still relatively accurate and based on the physical fine simulation model. A surrogate model is constructed by aligning the coarse and fine models in sub-regions of the design space – normally close to the desired optimum (or Pareto front in multi objective optimization problems).

The goal of this project is thus to develop coarse and surrogate models, for use in multi-objective SBO, for all the antennas to be considered for the wide band single pixel receivers. Once these models are available the full trade off space, or so called Pareto front, for all the performance metrics of the feeds may be calculated. The Pareto fronts will provide quantitative information on the performance limitations of current feed technologies. When the design methods are in place, new technologies can be rapidly designed, in a multi-objective sense, for possible use as a wide band single pixel feed in the SKA. One such technology option is a sinuous antenna currently under development as a Masters project at Stellenbosch University, showing great promise.

View project proposal

Novel methods of fast numerical electromagnetic analysis for the SKA mid-frequency aperture array

Overview

Numerical electromagnetic analysis has become an indispensable tool in modern antenna design. The design of the mid-frequency aperture array involves very large computational electromagnetics analyses. In such cases the computational cost can quickly become prohibitive and it becomes necessary to investigate more efficient numerical methods. For this project, the candidate must firstly study cutting-edge array analysis methods. Secondly, existing methods must be implemented as a computer code. Thirdly, new methods must be developed which are more efficient than the present state-of-the-art, with regards to analysis of the SKA mid-frequency aperture array prototypes. This work will likely focus on macro-basis function methods and iterative solvers. This challenging project has excellent scope for journal publications.

View project proposal

Multi-band LNA for the SKA

Overview

The frequency bandwidth allocated to the SKA spans from 70 MHz to 25 GHz. This bandwidth is divided in three bands that are derived from the SKA Science drivers and which are mapped to a receptor type (e.g. phase arrays for the low-band from 70 MHz to 450 MHz) and to a receiver type. Highly sensitive low noise amplifiers (LNAs) are required for the SKA and, owing to its large bandwidth, LNAs are optimized for each receptor and receiver type. It is then desirable to investigate and to design an integrated multi-band LNA for the SKA. The noise figure in each band is expected to be close to the minimum achievable and the relevant semiconductor process (CMOS or SiGe HBT) is expected to be used. The concept will be verified by simulation and fabrication.

View project proposal

Data Transport over Optical Fibre for SKA using Advanced Modulation Flexible Spectrum Technology

Overview

Flexible Spectrum Dense Wavelength Division Multiplexed (DWDM) optical fibre networks are next generation technology for handling extremely high data rates of the kind produced by MeerKAT and SKA. Wavelength Division Multiplexing refers to the transmission of multiple optical fibre wavelengths within a single optical fibre. Flexible spectrum refers to the fact that the channel wavelength assignment is not statically fixed, but is dynamically optimized in real time to ensure optimum network performance at all times. In order for these networks to function optimally, novel hardware and dynamic algorithms need to be identified and developed to perform critical tasks.Such tasks include wavelength assignment, signal routing, network restoration and network protection. This project focuses on developing, implementing and testing modulation formats, hardware technologies and algorithms for optimizing the operation of such networks.

View project proposal

Improving noise isolation between SoC integrated LNAs and ADCs in bulk CMOS

Overview

This topic relates to the priority area “Mid‑frequency Aperture Array technologies”. Successful low-noise co-integration of LNAs and ADCs in bulk CMOS will make a vital and meaningful contribution to MFAA development.

View project proposal

Optimal direct impedance matching between LNAs and ADCs

Overview

This topic relates to the priority area “Mid‑frequency Aperture Array technologies”. Successful low-noise co-integration of LNAs and ADCs in bulk CMOS will make a vital and meaningful contribution to MFAA development.

View project proposal