Overseas Travel Grant: Focussed Challenges in Feature Selective Validation, FSV. (OASIS)

Electromagnetic simulation is used in designing and analyzing systems ranging from integrated circuits to aircraft. This project is particularly interested in the use of electromagnetic simulation to support electromagnetic compatibility, EMC, studies (the way systems interact through the coupling of fields and on transmission lines) and signal integrity and power integrity, SIPI, (affecting the quality of digital signals usually in printed circuit boards or integrated circuit systems). Such simulation has a vast array of purposes, for example, it will look to identify whether one part of the system will cause interference with another part of the system, that electronic signals actually behave in the way that the designers want or to understand what happens if lightning strikes an aircraft in flight. In order to do any of these things, those relying on the results must be confident in the output of the simulations. This might be confidence in the solver but more often confidence in the implementation. Properly written simulation tools will always give the right answer to the question they have been asked; but the important thing is have they been asked the right question. This development of confidence in the tools and implementation involves validation and verification. There are limited methods and algorithms in common use, to provide quantitative support in that validation process, with subjective approaches such as 'eyeballing' graphs still being widespread. One approach that has been most widely used in EMC is the Feature Selective Validation (FSV) technique, developed by the PI and his co-workers. As the quality of software improves, the performance of the computing power increases and the demands on data-rich analysis evolve, it is important that FSV is enhanced to adapt to those changes and anticipate the needs of the community in the coming years. This travel grant provides support to document those needs with industries that themselves challenge the simulation developers for enhancements. This information will be combined with information from complementary companies in the UK to produce a roadmap for developments in FSV, along with generating other possible outputs addressing immediate issues. This roadmap will be used to guide the research of the PI but also provide a starting point for other research and industry groups to build a validation tool suitable for emerging needs in EMC and SIPI electromagnetic simulation.

Impact will be generated by encouraging more researchers to become involved in solving the wider industrial problems that it is anticipated that this project will document. Through this, and particularly through the actions resulting from the roadmap, Industry will have an augmented toolkit to enable it to have more confidence in the electromagnetic solvers being used, and hence potentially resulting in reduced design cycle times, more right-first-time designs and better understanding of EMC related phenomena. Having some important industrial participants in this work will allow them to achieve direct benefit from the improved approaches to electromagnetic validation but will also lead to diffusion to the organisations with whom they work.

From the point of view of the UK research base, this project will help ensure that the UK is seen as the hub for this emerging subject.

The use of FSV is currently embedded in IEEE Std 1597.1. The work of this project will inform the next revision cycle for this standard. Similarly, other standards, such as the upcoming IEEE Std 1848 (Techniques & Measures to Manage Functional Safety and Other Risks With Regard to Electromagnetic Disturbances) will benefit from the subsequently updated FSV and a better understanding of the needs of industry. Also, the upcoming IEEE Std 370 (Electrical Characterization of Printed Circuit Board and Related Interconnects at Frequencies up to 50 GHz) mentions FSV but does not include it as a central data comparison technique. An impact of this project, particularly the visit to Intel is that that position may change and the next revision cycle of Std 370 may include more FSV related material.

An impact that should not be overlooked is that this project catalyses discussions with a range of companies that are likely to identify more unanswered questions that, themselves, can contribute to future research and research applications.