Amorphous Slack Methodology for Autonomous Fault Handling in Memory and Image Processing Applications

The proposed amorphous fault handling methodology utilizes run-time redundancy in order to enhance the survivability of the design. Here we are proposing a fault isolation algorithm, this algorithm will be solicited upon fault detection, this fault detection scheme uses health metric of application, this is applicable if and only if signal to noise metric is known and also the application should not possess associated metric for recognition of faulty behavior. This is demonstrated by designing a memory controller, which is being implemented using a HDL designer and modelsim software and also we are designing a sobel edge detector and same is implemented using MATLAB.