Towards digital circuit approximation by exploiting fault simulation

Abstract

In the recent years, Approximate Computing (AxC) has emerged as a new paradigm for energy efficient design of Integrated Circuits (ICs). AxC is based on the intuitive observation that, while performing exact computation requires a high amount of resources, allowing a selective approximation or an occasional relaxation of the specifications can provide significant gains in area, performances and energy efficiency. This work focuses on a case study about functional approximation of digital circuits. The functional approximation aims at modifying the circuit structure so that the original function F will be replaced by G whose implementation leads to area/energy reduction at the cost of reduced accuracy (i.e., some errors can be observed at the outputs of G). In this paper, we investigate an approach for the functional approximation exploiting fault simulation. Preliminary results show the potentiality of this approach in terms of area reduction.