In our world today, a number of organizations, individuals, and institutions require computers that can perform advanced signals processing. Currently, digital computation is used in signals processing however digital computation is rigid (because of its Boolean nature and the continuous nature of signals) and consumes large amounts of power. In response to this, DARPA has created Project UPSIDE whose goal is to research probabilistic computing methods (which are inherently analog) that can perform the necessary calculations more efficiently while consuming less power. A secondary goal of the program is to design their computers so that they have self-configuring inputs. With the rise of neurogrid computing, one can see that analog computation will fit nicely alongside DARPA's model and view of the future of computing. At the end of this project I aim to have an open-sourced analog computer that can fit this need.
Enjoy this project?