|Funding for:||UK Students, EU Students|
|Funding amount:||£15,009 an industrial top-up of at least £6,000 p.a|
|Placed On:||31st July 2018|
|Closes:||31st July 2019|
Functional verification ensures the correct functioning of complex semiconductors. Companies invest a great deal of money and effort in functional verification for various reasons – to ensure a good user experience, to comply with regulators, to avoid the considerable costs of recalls and litigation and, in safety critical applications, to prevent the loss of life.
The two state-of-the-art approaches to functional verification are constrained-random verification and formal verification. The former is inefficient, with many simulation cycles spent exploring the same state space in much the same way; guiding the tool into the interesting corner cases present in complex systems typically requires considerable input from engineers. On the other hand, formal verification can find corner cases with little manual steering, but complexity limits mean that it can only be applied exhaustively to relatively small blocks.
We aim to use advanced learning to direct the generation of stimuli so that the interesting corner cases on a large complex design can be reached in an automated way. In this project we intend to employ machine learning techniques to help find correlations between observed coverage and stimulus. In particular, we will investigate statistical techniques capable of uncertainty propagation to predict the likelihood of given stimulus to achieve verification goals. This information is expected to help direct stimulus generation in a series of iterative cycles, each refining the information available to the learner. The programme of research includes training the test generator to stay within a set of constraints, so it produces valid stimulus; training the test generator to trigger checkers in the design; and training the test generator to achieve coverage goals. Our objective is to generate otherwise hard to find input sequences and to identify complex DUT configurations. We expect that this significantly improves the level of automation and effectiveness of test generation.
How to apply:
Please make an online application for this project at [http://www.bris.ac.uk/pg-howtoapply]
In the Funding section, please state “I would like to be considered for an iCASE award from the Computer Science Department.” and specify the title of the scholarship.
Interested candidates should apply as soon as possible.
A good 2:1 or first-class degree in Computer Science, Computer Systems Engineering, Informatics, Microelectronic Design or a similar discipline.
Excellent programming skills. A good understanding of computer architecture and machine learning techniques will be required for the project.
A background in at least one of the following areas: machine learning, processor architecture and design, simulation-based testing, model-based design or testing techniques. A competent presenter, writer and communicator, willing and able to work with our industrial collaborator.
The scholarship covers full UK/EU (EU applicants who have been resident in the UK for 3 years prior to 1 September 2019) PhD tuition fees and a tax-free stipend at the current RCUK rate (£15,009 in 2019/20). EU nationals resident in the EU may also apply, but will only qualify for PhD tuition fees.
Informal enquiries please contact Prof Kerstin Eder - Kerstin.Eder@bristol.ac.uk
General enquiries please contact firstname.lastname@example.org
Type / Role: