Fuzzy controller for error control of on — Chip communication

M. Jasmin, T. Vigneswaran · 2017 International Conference on Algorithms, Methodology, Models and Applications in Emerging Technologies (ICAMMAET) · 2017

On - chip communication suffer from bit error especially at highly dense implemented system on chip. Reducing the rate of bit error in On-chip communication is carried out by error detection and correction codes. But it has its own limitation. So reconfigurable multiple code based system is suitable for heterogeneous system design. But selecting correct code for the correct subsystem or for current requirements involve good decision system design. More over such selection is fuzzy in nature because heterogeneous subsystem demands different combination of input demands. Here, in this paper a fuzzy controller is designed to select correct error control scheme for the current subsystem demand. The fuzzy controller will take user application demand in terms of minimum required bit rate error, maximum time to compute error code or decode on the receiver side and the correlation level of the input dataset. Fuzzified such input by input membership function is processed by the designed fuzzy controller with 27 rules and selects the appropriate coding scheme. One of the main elements under reconfigurable coding scheme is the communication of selected coding scheme at the given time to decoding unit to enable the decoding process. An efficient frame format with less overhead is designed to communicate the selected coding scheme to the receiver subsystem in order to decode the data. The designed fuzzy controller is tested by random input data and the functionality is verified.

Read the paper · More papers on PaperTik