PhD project in Hardware Security for Approximate Computing

As IBM’s recent 2nm chip is pushing Moore’s law to its limit, conventional computing techniques are struggling to offer high performance computing within power consumption constraints. Approximate computing generates results that are good enough rather than always fully accurate. It was reported in 2018 as one of the top ten technologies that could change the world. Approximate computing offers up to an order of magnitude reduction in power consumption by accepting a margin of error in calculations, which reduces the accuracy of a specific result to within an acceptable threshold. Inspired by the fault tolerant capability of the human brain, approximate computing can accept errors in calculation without affecting the results of certain human perception and recognition related computation, including artificial intelligence (AI), deep learning (DL), machine learning (ML), signal processing and even some cryptographic schemes, in which noisy data, redundant information and inaccurate results are tolerable for the computation. Indeed, leading companies are undertaking research into potential products and services based on approximate computing, e.g., Google’s DL chip, IBM’s RAPID on-chip AI accelerators, etc. The biggest difference between accurate computing and approximate computing is the introduced errors in the result. An accurate computing design is supposed to generate precise results and any error, should it occur, would be unintentional. In contrast, an approximate computing design, by definition, may introduce errors and give many different results. However, a major threat to approximate computing is that the access control mechanisms for manipulating acceptable errors, which are added to designs to improve power consumption performance, provide a new attack vector over equivalent exact designs. This project will explore how to achieve secure and lightweight security designs for approximate computing. The project offers a bespoke research and training programme that aims to develop students into cross-disciplinary thinkers and leaders who will influence the roadmaps of future advanced technologies and their applications. They will have a balanced understanding of ICT (security and data analytics) in the context of their application to advanced technologies and high value designs.

Silicon physical unclonable functions (PUFs), which exploit manufacturing variations of silicon chips, offer a promising mechanism that can be used in many security, protection and digital rights management applications. Such a primitive has a number of desirable properties from a security perspective, such as the ability to provide a low-cost unique identifier for an integrated circuit (IC) or to provide a variability aware circuit that returns a device specific response to an input challenge. This gives it an advantage over current state-of-art alternatives such as secure non-volatile memory (NVM) or trusted platform modules (TPMs). No special manufacturing processes are required to integrate a PUF into a design. This lowers the overall cost of the security for the IC enabling the PUF to be utilised as a hardware root of trust for all security or identity related operations on the device. An approximate dynamic random-access memory (DRAM)-based PUF design was proposed. It is the only PUF design based on approximate memory to enhance the security of approximate computing. Unfortunately, it has been found vulnerabilities. Currently, no comprehensive research has been conducted into the security of approximate computing or into countermeasures that protect such designs. The use of approximation strategies can even provide an advantage to attackers, since common security components in a standard computing system will likely be excluded from approximated components, as existing security algorithms cannot work in an approximated way. Intrinsic security strategies, such as PUFs that use circuitry already present in the system, without modification, and purely by means of software, are promising to address the challenges in approximate computing. Ultimately, how to achieve secure and lightweight security designs for approximate computing is a critical challenge that this proposal seeks to address.

Funding Information

To be eligible for consideration for a Home DfE or EPSRC Studentship (covering tuition fees and maintenance stipend of approx. £19,237 per annum), a candidate must satisfy all the eligibility criteria based on nationality, residency and academic qualifications.

To be classed as a Home student, candidates must meet the following criteria and the associated residency requirements:

• Be a UK National,
or • Have settled status,
or • Have pre-settled status,
or • Have indefinite leave to remain or enter the UK.

Candidates from ROI may also qualify for Home student funding.

Previous PhD study MAY make you ineligible to be considered for funding.

Please note that other terms and conditions also apply.

Please note that any available PhD studentships will be allocated on a competitive basis across a number of projects currently being advertised by the School.

A small number of international awards will be available for allocation across the School. An international award is not guaranteed to be available for this project, and competition across the School for these awards will be highly competitive.

Academic Requirements:

The minimum academic requirement for admission is normally an Upper Second Class Honours degree from a UK or ROI Higher Education provider in a relevant discipline, or an equivalent qualification acceptable to the University.

Entrance requirements

Graduate
The minimum academic requirement for admission to a research degree programme is normally an Upper Second Class Honours degree from a UK or ROI HE provider, or an equivalent qualification acceptable to the University. Further information can be obtained by contacting the School.

International Students

For information on international qualification equivalents, please check the specific information for your country.

English Language Requirements

Evidence of an IELTS* score of 6.0, with not less than 5.5 in any component, or equivalent qualification acceptable to the University is required (*taken within the last 2 years).

International students wishing to apply to Queen’s University Belfast (and for whom English is not their first language), must be able to demonstrate their proficiency in English in order to benefit fully from their course of study or research. Non-EEA nationals must also satisfy UK Visas and Immigration (UKVI) immigration requirements for English language for visa purposes.

For more information on English Language requirements for EEA and non-EEA nationals see: www.qub.ac.uk/EnglishLanguageReqs.