An Efficient Membership Function Representation for High-Resolution Fuzzy Systems

適合高解析度模糊系統之有效率的歸屬函數

黃世旭(Shih-Hsu Huang) ； 賴建元(Jian-Yuan Lai)

《中原學報》 29卷3期 (2001/09) Pp. 285-291

https://doi.org/10.6358/JCYU.200109.0285

摘要

近幾年，模糊邏輯已在不同應用領域，有著極爲廣泛的運用；例如，網路流量控制、航空飛行控制、影像處理等，都有運用模糊系統實現其應用之例子。在一個模糊系統中，主要是以歸屬函數特徵化其模糊規則，並將歸屬函數儲存於記憶體。因此，若要建構高解析度之模糊系統，系統設計者需要使用極大之記憶體空間。本篇論文，我們提出一個有效率之歸屬函數表示法，以減少記憶體之需求。爲了能與既模糊邏輯處理器溝通，我們提出了一個硬體介面以動態產生歸屬函數。我們基本的想法，是要簡化歸屬函數的表示法，但不影響模糊計算之準確度，亦不影響模糊邏輯處理器進行模糊推論所需的時間。實驗結果顯示，此新的表示法，可減少大量之記憶體空間，但硬體介面僅需數百個邏輯閘。因此，實驗結果證實此新表示法之可行性。

關鍵字

歸屬函數 ； 儲存空間最小化 ； 介面設計 ； 硬體架構 ； 模糊集合 ； 高解析度模糊系統 ； 模糊邏輯處理器

並列摘要

In the past few years, there have been an increasing number of systems based on fuzzy logic in various fields of applications, such as ATM network control, GPS data processing, aircraft flight control, image processing, and so on. In a fuzzy system, the fuzzy rules are characterized by membership functions and stored in the memory. To model a high-resolution fuzzy system, the system designer needs a large memory space to store all the membership functions. In this paper, we will present an efficient membership function format, which may significantly reduce the memory requirement. In order to cope with existing fuzzy inference processors, a dedicated hardware is designed to tackle the interface. The basic idea behind our approach is to simplify the representation of membership function without affecting the precision of fuzzy computation or slowing the speed of fuzzy inference. Experimental data shows that the proposed approach achieves very good results. The new format may save a lot of memory requirement if compared with the typical method that stores all the grades, while the cost of the dedicated hardware interface is only several hundreds of logic gates.

並列關鍵字

Membership Function ； Storage Minimization ； Interface Design ； Hardware Architecture ； Fuzzy Set ； High-Resolution Fuzzy Systems ； and Fuzzy Processors