A VLSI implementation of a fuzzy-inference engine: toward a expert system on a chip
Information Sciences: an International Journal
A digital fuzzy processor for fuzzy-rule-based systems
Hardware implementation of intelligent systems
Different Fuzzy Parameter Selection Based on Multiple Criteria for Microcontroller
EUC '08 Proceedings of the 2008 IEEE/IFIP International Conference on Embedded and Ubiquitous Computing - Volume 01
Rapid prototyping methodology and environments for fuzzy applications
ICCS'03 Proceedings of the 2003 international conference on Computational science
| Hi-index | 0.01 |
This paper studies the timing and memory usages of fuzzy control applications on embedded systems. Particularly, we are interested in the timing and memory usages of various fuzzy applications implemented on microcontroller. Our study is based on the PIC microcontroller platform. We develop a fuzzy library used to build a concurrent fuzzy system prototype. Then, we compare the memory usage and the timing property of our prototype concurrent fuzzy systems implemented by interrupt-based scheme and Micro C-based with the baseline one. The results show that implementation using interrupt-based has overheads on the program memory about 11 bytes per system and about 160 bytes for data memory. Timing overhead for the interrupt-based is about 72% compared to the baseline. When compared to using Micro C, which has more data overheads, it has an advantage on real-time features and small task switching overhead. However, implementation of the Micro C and fuzzy systems are limited on PIC due to the data memory size.