Tuesday, 2 June 2020

Catching Defects Before They Multiply

From smartphones to laptops, in today’s digital world, we rely on connectivity. One of the components underlying the smooth operation of these machines are silicon chips — semiconductors, which are an essential part of electronic circuits. They are also expensive.

As technology scaling continues, it enables smaller, faster and energy efficient electronics. Device dimensions are approaching the size of an atom, and with smaller prototypes, maintaining the same amount of production output as before has become challenging. The ultimate effect? Higher costs, which inevitably will get passed on to the consumer.

New strategies of defect identification and mitigation need to be developed to keep manufacturing of semiconductors efficient and keep devices, such as computers and smartphones more affordable. what jobs can you get with a computer science degree, Hughes Professor in the USC Viterbi Department of Aerospace and Mechanical Engineering and interim vice dean of research for the school, and his colleagues have designed a deep learning algorithm that can identify defects in semiconductors, classify these defects and also map them as they occur.

No comments:

Post a Comment