Electrical and Computer Engineering

 

Project Number: 2019-5

Project Leader: Deborah Munro

Host Department: Mechanical Engineering

Project Title: Biomedical Sensor Systems for Measuring Bone Fracture Healing--MEMS Sensor Fabrication Focus

Project outline: I am developing a wireless, battery-free implantable sensor for measuring the progress of bone healing. The sensor is a MEMS-fabricated interdigitated capacitive sensor that is encased in a waterproof housing and adhered to an implanted fracture plate or spinal implant. The goal is to get data much sooner about healing so that patients can return to normal activities sooner or clinicians can do interventions sooner if something is wrong. There are two positions available, one to do housing design and testing and the other to fabricate MEMS sensors.

I have established access with UC’s Electrical and Electronics Engineering Nanofabrication Laboratory, and I will begin microfabricating sensors for my application. You would be trained to use all the equipment in the laboratory and conduct testing, design electronic circuits (in collaboration with my partners at the Auckland Bioengineering Institute), and develop a sensor system for implanting in sheep in a pilot and full-scale animal study at Lincoln University.

https://www.canterbury.ac.nz/engineering/schools/mechanical/research/munro/mems-sensors/

Specific Requirements: Suitable for a mechanical, mechatronics, or electrical engineering student with an interest in microelectromechanical systems (MEMS) and working in a cleanroom environment.

 

 

Project Number: 2019-6

Project Leader: Deborah Munro

Host Department: Mechanical Engineering

Project Title: Biomedical Sensor Systems for Measuring Bone Fracture Healing--Material Testing Focus

Project outline: I am developing a wireless, battery-free implantable sensor for measuring the progress of bone healing. The sensor is a MEMS-fabricated interdigitated capacitive sensor that is encased in a waterproof housing and adhered to an implanted fracture plate or spinal implant. The goal is to get data much sooner about healing so that patients can return to normal activities sooner or clinicians can do interventions sooner if something is wrong. There are two positions available, one to do housing design and testing and the other to fabricate MEMS sensors.

https://www.canterbury.ac.nz/engineering/schools/mechanical/research/munro/orthopaedic-medical-devices/

Specific Requirements: This project is suitable for mechanical, mechatronics, and electrical engineering students with an interest in hands-on research and material testing.

 

 

Project Number: 2019-23

Project Leader: Michael Hayes, Bill Heffernan

Host Department: ECE (with EPECentre)

Project Title: Instrumentation for groundwater flow measurement

Project outline: As part of the National Science Challenge, UC's ECE Department is involved in a project to try to measure low flow velocities of groundwater in aquifers.   The technique being investigated is based on the concept put forward by Michael Faraday, in which a voltage is induced across water flowing through an AC magnetic field.  Ideally, the intention is to trial the system developed on the Kaitorete spit, which links Te Waihora (Lake Ellesmere) and the Pacific Ocean.

The magnetic field is created by a large electromagnet driven from some power electronics.   The goal of this project is to design and construct an embedded system for controlling the power electronics and synchronising its operation with the measurement electronics.  A critical aspect is that the controlling circuit has low jitter and thus will require a high-performance microcontroller perhaps in conjunction with an FPGA.  The microcontroller needs to remotely communicate with a remote PC and avoid ground loops.

Specific Requirements: 2nd or 3rd professional year ECE or Mechatronics student - preferably a student considering studying towards a higher degree.  Good knowledge of embedded systems and circuit layout fundamentals (e.g., from ENCE361, ENCE461)

Knowledge of PCB layout software, e.g., Kicad, Altium.

 

 

Project Number: 2019-24

Project Leader: Michael Hayes, Bill Heffernan

Host Department: ECE (with EPECentre)

Project Title: Building and testing a novel power converter to drive an induction coil with a trapezoidal current waveform

Project outline: As part of the National Science Challenge, UC's ECE Department is involved in a project to try to measure low flow velocities of groundwater in aquifers.  The technique being investigated is based on the concept put forward by Michael Faraday, in which a voltage is induced across water flowing through an AC magnetic field.  Ideally, the intention is to trial the system developed on the Kaitorete spit, which links Te Waihora (Lake Ellesmere) and the Pacific Ocean.

To produce such an AC magnetic field, a coil is driven with an AC current waveform.  In some instances a flat-topped waveform (approximating a square wave) is preferable to a sinusoidal one.  However, the coil is essentially an inductance with some series resistance and as all electrical engineering students know, you cannot instantaneously change the current in an inductor.  Instead, we have devised a relatively simple and low cost switching power electronics circuit which provides high di/dt to commutate rapidly from negative to positive current half-cycles and constant current to maintain the steady state current during the half-cycles.

This circuit has gradually evolved as the project has progressed, with the result that it is rather large and cumbersome and, most importantly, not readily portable.   The purpose of this project is to design, build and test a new, compact version of the existing circuit, which can be taken out of the lab and deployed in situ.

Specific Requirements: 2nd or 3rd professional year ECE or Mechatronics student - preferably a student considering studying towards a higher degree.

Good knowledge of power electronics and circuit layout fundamentals (e.g., from ENEL372, ENEL471)

Knowledge of PCB layout software, e.g., KiKad, Altium.

 

 

Project Number: 2019-49

Project Leader: Debbie Munro

Host Department: Mechanical Engineering

Project Title: Sensor for Tracking Number of Sterilisations for Medical Device

Project outline: We are developing a sensor medical device to count how many times a device is sterilised. This is an ongoing research effort between Dr Munro and Enztec. Student will help develop design solution and conduct testing on campus in the autoclave (heat steam) sterilisation facility. This project is suitable for mechanical, electrical, or mechatronics engineering student.

Specific Requirements: Student should have a willingness to learn basic microcontroller, sensor, and data capturing skills, including RFID.

 

 

Project Number: 2019-73

Project Leader: Sharee McNab

Host Department: Electrical and Computer Engineering

Project Title: Solar Generation Optimisation Tool and Review of Platforms for Optimisation of Consumer Photovoltaics and Battery Energy Storage Systems

Project outline: Typically residential solar photovoltaic (PV) installations are installed with a fixed mounting of tilt and orientation to optimise generation. Alternatively PV systems (more commonly for commercial or utility-scale solar) can be installed with a tracking system that changes the tilt to optimise generation along one or more axes, however these systems have much greater installation and maintenance costs.

An alternative middle ground is proposed that has a small number of tilt options on a fixed mounting so that the tilt can be manually adjusted throughout the year. The project seeks to develop a tool to optimise generation (or perhaps revenue) for a location using two or three fixed mounting options to take into account the variation in the sun path throughout the year.

A secondary part of the project would be to review home energy management systems available on the domestic market for assisting consumers optimise their PV and battery energy storage system (BESS) by deciding the best times to charge and discharge their BESS. A survey of the level of sophistication employed should be undertaken, for example whether systems use weather forecasting, spot pricing inputs or the ability to provide other paid grid support services to deliver the most economic solution.

Specific Requirements: Engineering student or alternatively Physics/Chemistry

 

 

 

Project Number: 2019-89

Project Leader: Chris Hann, Robin McNeill

Host Department: Electrical and Computer Engineering

Project Title: Development of S/X-Band subsystems for Awarua Satellite Ground station 2.4m antenna

Project outline: This project will investigate:

1. Design and build of tapered corrugated horn feeds for X-Band and S-Band, and a dichroic sub-reflector for exiting antenna

2. Writing GnuRadio flowcharts for our SDR at Awarua for use in S-Band.

3. Reviewing and fixing current space operations software at the station

4. Develop a scheduler for space operations software

The feed design will be done from Canterbury and testing at Awarua.

Results from this project will be used in both satellite tracking and for tracking of orbital rockets.

Specific Requirements: * Electromagnetics course or equivalent required

* Knowledge of partial differential equations

* Practical experience with wireless communication systems

 

 

 

Project Number: 2019-106

Project Leader: Ciaran Moore

Host Department: Electrical and Computer Engineering

Project Title: Surface plasmon resonance sensor testbed

Project outline: Surface plasmon resonance (SPR) sensors are optical devices that respond to a fluid's refractive index. Potential applications include air quality monitoring, detection of fertiliser run-off in waterways and measuring the concentration of pesticides in beehives at risk of colony collapse. While recent research has optimised the sensors construction to maximise their sensitivity, selectivity (the ability to respond only to a material of interest) remains generally poor. Coating the sensors with thiol-terminated aptamers may be able to address this shortcoming; however, before the effects of the aptamers on the sensors characteristics can be quantified, a stable SPR test rig is required that gives reliable, repeatable measurements.

Therefore, the goal of this project is to develop a portable, parallel SPR sensor housing that allows multiple SPR sensors, or sensors with different aptamer coatings, to be characterised. This test rig will comprise a light source, holders for various SPR sensors, including planar thin film stacks and metalised optical fibers, and a number of optical sensors to measure the intensity of reflected light from the SPR sensors. This project will suit a student with an interest in practical lab work and experimentation.

Specific Requirements: N/A

 

 

 

Project Number: 2019-107

Project Leader: Ciaran Moore

Host Department: Electrical and Computer Engineering

Project Title: Personal fitness sensor platform

Project outline: Wearable devices that record exercise and track fitness are becoming increasingly popular as manufacturers such as Apple, Fitbit and Garmin offer systems with various features and monitoring abilities. At the same time, the potential of biometric information as a form of identity is being developed: fingerprints, long used in forensics, are now used to unlock smartphones; more recently facial scans have been put to similar use. The combination of these two applications introduces novel possibilities of using fitness data as a personal identifier. Unlike fingerprints or facial structure, parameters such as pulse shape and frequency are not externally visible to observers and thus have potential to be a more secure form of identification.

Therefore the goal of this summer project is to develop an extensible personal fitness sensor platform that can be used to gather various biometric statistics about the wearer. The aim is not necessarily to uniquely identify individuals; rather, the feasibility of collecting and processing relevant information is to be investigated.

This project will suit a student with an interest in embedded systems and electronics, including breadboarding, circuit simulation, schematic design and PCB layout. Requirements include previous completion of ENCE361 or equivalent experience with programming microcontrollers and good written communication skills.

Specific Requirements: Prior completion of ENCE361: Embedded Systems, or similar experience programming microcontrollers.

 

 

 

Project Number: 2019-108

Project Leader: Ciaran Moore and Volker Nock

Host Department: Electrical and Computer Engineering

Project Title: Microfluidic Machine Learning Circuits

Project outline: Microfluidic microelectromechanical systems (µF-MEMS) are versatile structures that have great potential to revolutionise disease diagnosis and monitoring, gas sensing and drug delivery. One promising development is the ability to perform computation using droplet flows. These systems offer an additional level of automation and sophistication compared to conventional microfluidic circuits.

Meanwhile convolutional neural networks (CNNs) are seeing widespread growth and adoption for a range of applications, including disease detection, image processing, and the estimation of flow rates and concentrations of mixed solutions. Recent work has shown that CNNs need not be confined exclusively to software, with the implementation of a passive image classifier CNN in a slab of glass.

This project aims to integrate machine learning capabilities directly into µF-MEMS by implementing the key building blocks of CNNs as microfluidic circuits. Specifically, this project will investigate designs for:

Non-linear (diode- or valve-like) circuits, to mimic the non-linear activation layers used in CNNs.

Scattering circuits, to couple outputs from one layer of artificial neurons to multiple inputs in the next layer.

Normalisation circuits, to scale inputs to a desired range of output.

Work will involve planning, fabricating, and testing microfluidic devices, with the ultimate goal of implementing a pre-trained convolutional neural network as a series of microfluidic channels.

Specific Requirements: N/A

 

 

Project Number: 2019-113

Project Leader: Richard Clare & Steve Weddell

Host Department: Electrical & Computer Engineering

Project Title: Wavefront Estimation with Machine Learning

Project outline: A large part of research conducted by the Computational Design and Adaptation group in ECE deals with the design of efficient and accurate sensors to measure distortion when imaging through a homogeneous medium, such as air turbulence. This is part of a field of study called Adaptive Optics (AO). Applications of the sensors and resulting instruments developed for AO include near-negation of the effects of turbulence when used by medium, large, and extra large, ground-based telescopes.

Wavefront phase cannot be viewed or imaged directly. However, light intensity or irradiance, rather than phase, can be observed directly. Wavefront sensors (WFSs) are a family of devices that convert a measured quantity, such as irradiance, into phase. From when it is first generated to just before it enters our atmosphere, light, in the the form of a wave front, is undistorted. The effects of light passing through turbulent media, such as air, changes the optical path. If a star is imaged though a lens, like a pair of binoculars for example, changes to the optical path result in a blurred or displaced image. We design, build and test wave front sensors to measure these optical path length differences, so we can correct for image distortion in real time.

A commonly used wavefront sensor is the Shack-Hartmann WFS. This requires an N x N array of small lenses, referred to as a lenslet array, which is placed in the pupil plane of the optical system. The resulting N2 sub-images effectively divides the incident wavefront into zones, where the wavefront slope of each zone, or sub-array, can be measured. In effect, regions of a telescope pupil are populated with mini-lenses that focus regionalised areas of light onto a CCD, which is then used to measure local displacements. Displacements, in the form of x and y measurements for each array, are taken and then used to form a matrix, which can then be fitted using least-squares to a basis function set.

This project will use existing Shack-Hartmann wavefront sensor simulations to produce Shack-Hartmann images from known wavefronts. However, instead of reconstructing the wavefront from matrix multiplications, as is usually done, we want to use machine learning to estimate wave front phase.

Specific Requirements: This project will be suitable for a Computer Engineer, Mechatronics or Electrical Engineering student who has some experience with machine learning or is taking the Computer Science Machine Learning Course, COSC401.

 

 

 

Project Number: 2019-121

Project Leader: Steve Weddell and Richard Clare

Host Department: Electrical and Computer Engineering

Project Title: Upgrading a Schmidt Camera for Space Situational Awareness

Project outline: Wide field photography of large portions of the night sky is important for engineers and scientists to capture events, such as detailed images of orbiting satellites and detecting large space debris objects in the field of space situational awareness. In addition, very rare galactic events, such as supernova explosions, are constantly being monitored for capture around the globe. Without wide field photometry, these and many other exoatmospheric events would not otherwise be possible to record. This is because precise knowledge of where such events occur is very difficult; spatial errors over such large distances make accurate prediction virtually impossible.

To aid our space situational awareness research, a colleague has recently donated a wide-field Schmidt camera to our group. The correct term used to describe this instrument is a Schmidt camera, however use of the word "camera" is only half the story; the Schmidt design encompassed both a camera and a telescope, where the latter has a very wide field of view. For example, our instrument supports a focal ratio of f/2.0, which equates to a field of view of several degrees of sky. As a result, such an instrument would only be used for photographing large areas of the night sky and could not be used for general, direct observations, e.g., using an eyepiece to provide high magnification of an object.

Our Schmidt "camera" has an aperture of 28cm, a primary mirror, a tube, and fittings, such as camera holder. However, given the age of this instrument, the intended camera used a photographic film, which will need to be replaced with a small CCD or sCMOS image sensor, and suitable interface to to laptop or PC. Other repairs will be required, and not all of these will be electrical or electronically based. For example some CAD design work, manufacturing, possibly with 3D printing, and fitting of new components, will be required.

Our group has a budget to restore this Schmidt instrument and we want to start observing satellites with it by mid 2020. We realise not all required work can be completed over this summer period, however funds may be available next year for a possible extension. We intend to fit-out the instrument with several small cameras, which will be used to acquire atmospheric turbulence data from multiple natural point sources, i.e., stars. In addition, we want to capture high-speed images of a range of satellites for image restoration. Intended use of this instrument will be from our UC campus, which will supplement observations performed on larger telescopes (not Schmidt designs) at the University of Canterbury Mt. John Observatory near Lake Tekapo.

Specific Requirements: A background or interest in astronomy and/or digital imaging methods would be useful. An appreciation of optics would also be helpful. Demonstrated familiarity with various CAD software packages for mechanical and electrical design is needed. Correspondingly, students with very good grades in 1st and 2nd Pro design courses are encouraged to apply for this project.