Self-Scaling Evolution of analog computation circuits with digital accuracy refinement

We introduce SCALER, a two-pronged strategy utilizing digital resources for refining intrinsic evolution of analog computational circuits. A Self-Scaling Genetic Algorithm is proposed to adapt solutions to computationally-tractable ranges in hardware-constrained analog reconfigurable fabrics. Differential Digital Correction is developed utilizing an error metric computed from the evolved analog circuit to reconfigure the digital fabric intrinsically thereby enhancing precision. We demonstrate our methods by evolving square, square-root, cube, and cube-root analog computational circuits on the Cypress PSoC-5LP System-on-Chip. Results indicate that the Self-Scaling Genetic Algorithm improves an error metric on average 7.18-fold, up to 12.92-fold for computational circuits that produce outputs beyond device range. Overall, Differential Digital Correction can reduce computational error by 23.1% compared to the performance of the evolved analog circuit.