Development of an IDS hardware unit for real-time learning applications

In artificial neural networks (ANNs) and fuzzy inference systems (FISs), hardware implementation is significantly effective in improving real-time performance by utilizing their parallel processing structures. Thus, numerous hardware solutions for ANNs and FISs have been provided for time-critical control applications. The ink drop spread (IDS) method is a modeling technique that has been proposed as a new paradigm of soft computing. The structure of IDS models is similar to that of ANNs: they comprise distributed intermediate units referred to as IDS units. In this paper, the hardware design of the IDS unit is presented and it is demonstrated that the hardware implementation is effective in enhancing the real-time performance of IDS modeling systems.