On the feasibility of the fuzzy system approach to signal processing

• Main Authors: Ming-Huei Peng, 彭明輝
• Other Authors: Chi-Cheng Jou
• Format: Others
• Language: en_US
• Published: 1994
• Online Access: http://ndltd.ncl.edu.tw/handle/14182924107304576600

碩士 === 國立交通大學 === 控制工程系 === 82 === This thesis investigates the feasibility of fuzzy system approaches to signal　processing. Noise cancellation, adaptive equalization, image coding, and image　preprocessing are sample problems used to evaluate the feasibility.　The focus of the present exercise is on justifying the capabilities and effetiveness of fuzzy system approaches, and unifying already existing fuzzy system approaches in the problem of optimality. Experimental results demonstrate that　fuzzy system approaches are more expensive than conventional approaches and they may not outperform conventional approaches in all of our studies. We illustrate that fuzzy system approaches are effective for problems that are highly nonlinear and complex.