Approximate Computing with Stochastic Memristors

Conference: CNNA 2016 - 15th International Workshop on Cellular Nanoscale Networks and their Applications
08/23/2016 - 08/25/2016 at Dresden, Deutschland

Proceedings: CNNA 2016

Pages: 2Language: englishTyp: PDF

Personal VDE Members are entitled to a 10% discount on this title

Naous, Rawan; Salama, Khaled Nabil (Computer, Electrical and Mathematical Sciences and Engineering, King Abdullah University of Science and Technology, Thuwal, Saudi Arabia)

Approximate computing with memristive elements is a novel computing paradigm that sacrifices accuracy for power efficiency. It builds on the inherent stochastic switching behavior of the memristive devices to lower the voltage levels applied for the logic gates. With the added probabilistic operation of the underlying components, the accuracy of the output is to be studied and analyzed accordingly.