Zhang dynamics and Newton-Raphson iteration for time-varying and static reciprocals computation

As a floating-point divider, Newton-Raphson iteration (NRI) is an effective method with a quadratic convergence. In this paper, we propose a class of novel dynamic models called Zhang dynamics (ZD) for online time-varying reciprocals computation. Moreover, we discover the relationship between the resultant ZD models and NRI for the static reciprocals computation. Through illustrative examples, we verify the efficiency of the ZD models and their superior properties to the traditional NRI for online time-varying reciprocals computation.