Despite their advantages, devices that use PMUTs are not simple to build. End- product performance is affected by every mechanical component in the solution, as well as the interaction between the signals from each electromechanical structure in an array of hundreds or even thousands of individual PMUTs. Given these considerable challenges,

it isn’t difficult to understand why the legacy computer aided engineering (CAE) software providers are struggling to provide solvers that can virtually prototype advanced devices in full-3D in a reasonable time-frame, and why the major players in ultrasonic fingerprint sensing are spending massive sums of money on dozens of prototyping iterations to refine their products.

THE FUTURE OF ENGINEERING But this doesn’t have to be the case. In OnScale’s vision of the future of engineering, product developers are unshackled from the limitations of their desktop hardware or internal high- performance computing (HPC) clusters. Tey are able to bring to bear unlimited computing power to solve their most difficult problems – from their laptops. OnScale’s offering consists of

advanced fully coupled multi-physics solvers, engineered specifically to run on commodity hardware in a massively parallel fashion for more than two decades. Integrated with a highly efficient cloud backend hosted by Amazon, OnScale allows design spaces that were previously unreachable to be explored with legacy solvers. Tere is perhaps no better example of how this kind of predictive power can be

used to reduce cost, time to market and risk than the ultrasonic fingerprint sensor. Legacy solvers are limited to simulating simple 2D slices of such devices, forcing engineers to slog through time consuming analytical and expensive empirical analyses to see the considerable 3D effects that are not visible through the keyhole view their solvers provide.

IMPRESSIVE ANALYSIS Recently, OnScale has demonstrated the capability to simulate a full-3D ultrasonic fingerprint sensor with thousands of individual PMUTs, including not only the transducer, but also the entire OLED/ glass stackup and even the finger itself. Moreover, this analysis was able to be completed in a matter of hours using a highly efficient message passing interface (MPI) to allow hundreds of cloud cores to solve multiple physical domains (electromechanical, acoustic wave, temperature) simultaneously. Te acoustic energy of every active element was modelled in the time domain such that interactions between them could be captured. As the waves propagated through the OLED and glass stack, ultimately reaching the ridges of the finger, reflections at each interface were captured and their effects on the final received signal at each PMUT element were precisely determined.

A PMUT creates acoustic energy from an electrical impulse

From the full-3D data, OnScale was

able to reconstruct the actual fingerprint image using a digital signal processing (DSP) algorithm. While previously an engineer would

have to wait for physical prototypes and empirical data to begin to test their DSP and modify it each time a revision was made to the part, now they can optimise their algorithms on a virtual prototype before any silicon is produced. Te implications of this capability are immense.

Te number of prototype iterations, which cost hundreds of thousands of dollars, can be significantly reduced. Software and firmware can be written, tested and debugged using virtual data. Most importantly, foundry process and technology platform improvements can be explored well in advance, enabling fundamental process and product strategies to be mapped that had previously been so difficult to envision and execute. Very soon, the idea of first-pass silicon in MEMS devices will become a reality as modelling and simulation capabilities approach the reliability and efficiency currently relegated to realm EDA and similar disciplines. OnScale is leading that charge, with tools designed to empower its users with the power of a supercomputer – on-demand. Te future of engineering has arrived, and it exists at the intersection of powerful, parallelised multi-physics solvers and the vast scalability of big compute in the cloud.

Ryan Diestelhorst is with Onscale.

16 An ultrasonic fingerprint sensor images the ridges of a finger using acoustic waves

Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20  |  Page 21  |  Page 22  |  Page 23  |  Page 24  |  Page 25  |  Page 26  |  Page 27  |  Page 28  |  Page 29  |  Page 30  |  Page 31  |  Page 32  |  Page 33  |  Page 34  |  Page 35  |  Page 36  |  Page 37  |  Page 38  |  Page 39  |  Page 40  |  Page 41  |  Page 42  |  Page 43  |  Page 44  |  Page 45  |  Page 46  |  Page 47  |  Page 48  |  Page 49  |  Page 50  |  Page 51  |  Page 52  |  Page 53  |  Page 54  |  Page 55  |  Page 56  |  Page 57  |  Page 58  |  Page 59  |  Page 60