Particle Swarm Optimization for Reconfigurable Sensor Electronics - Case Study: 3 Bit Flash ADC

Sensor electronics is ubiquitous in embedded systems, yet its performance is susceptible to static and dynamic deviations. Even costly and time consuming laser trimming still can´t deal with all the occurring deviations. Recently, analog reconfigurable electronics offers a solution to compensate these effects. The state of the art uses genetic algorithm (GA) to find an arbitrary topology to fulfil the given specifications, which can cause hardware with unpredictable behavior. Considering the robustness of the reconfiguration approach, we used the particle swarm optimization (PSO) by Tawdross, P. and Konig, A. (2005) as an alternative to GA for reconfiguration of programmable sensor electronics by Tawdross, P. et al. (2005) on basic block level. In order to evolve a reliable hardware with predictable performance, standard circuit topologies are employed. In this paper, we abstract our design environment from amplifier level to the functional block level. We demonstrate our methodology by a reconfigurable 3-bit flash analog to digital convert (ADC), which can recover successfully from static and dynamic deviations