Membership function generator based on bit-shift operation for self-organizing relationship (SOR) network Hakaru Tamukoh ⁎ , Keiichi Horio, Takeshi Yamakawa Graduate School of Life Science and System Engineering, Kyushu Institute of Technology, 2-4, Hibikino, Wakamatsu, Kitakyushu 808-0196, Japan Abstract. A2 -r shape of a membership function generator is presented. The proposed membership function generator requires a bit-shifter and a NOR operation only, and thus suits digital hardware implementation well. A fast defuzzification method employing active units is also presented. It reduces a calculation cost for the defuzzification. In this paper, we have successfully applied proposed methods to a self-organizing relationship network. Simulation results show that the proposed method has as good and approximation ability of nonlinear functions as the ordinary method. © 2007 Elsevier B.V. All rights reserved. Keywords: Membership function; Fuzzy inference; Self-organizing relationship network; Digital hardware 1. Introduction A self-organizing relationship (SOR) network has two modes: a learning mode and an execution mode [1]. Thus, a SOR network requires two hardwares corresponding to two modes. The former mode is similar to that of a self-organizing map (SOM), and it has been developed by the authors [2]. Consequently, the remaining task is developing the latter mode, i.e., execution mode. To achieve this, it is important to develop membership function (MF) hardware. Therefore, this paper aims to show how to implement the MF hardware on a digital architecture. 2. Self-organizing relationship network In the learning mode of the SOR network, randomly picked up input–output vector pairs, along with their evaluations, are used as the learning vectors. The detailed learning algorithm International Congress Series 1301 (2007) 180 – 183 www.ics-elsevier.com ⁎ Corresponding author. Tel.: +81 93 695 6123; fax: +81 93 695 6133. E-mail address: tamukoh@brain.kyutech.ac.jp (H. Tamukoh). 0531-5131/ © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.ics.2006.12.034