Fault Classification and Diagnosis Approach Using FFT-CNN for FPGA-Based CORDIC Processor

Abstract

Within the realm of digital signal processing and communication systems, FPGA-based CORDIC (Coordinate Rotation Digital Computer) processors play pivotal roles, applied in trigonometric calculations and vector operations. However, soft errors have become one of the major threats in high-reliability FPGA-based applications, potentially degrading performance and causing system failures. This paper proposes a fault classification and diagnosis method for FPGA-based CORDIC processors, leveraging Fast Fourier Transform (FFT) and Convolutional Neural Networks (CNNs). The approach involves constructing fault classification datasets, optimizing features extraction through FFT to shorten the time of diagnosis and improve the diagnostic accuracy, and employing CNNs for training and testing of faults diagnosis. Different CNN architectures are tested to explore and construct the optimal fault classifier. Experimental results encompassing simulation and implementation demonstrate the improved accuracy and efficiency in fault classification and diagnosis. The proposed method provides fault prediction with an accuracy of more than 98.6% and holds the potential to enhance the reliability and performance of FPGA-based CORDIC circuit systems, surpassing traditional fault diagnosis methods such as Sum of Square (SoS).

Real-Time Direction Judgment System for Dual-Frequency Laser Interferometer

“…Finally, a set of orthogonal signals with equal amplitudes and containing only the Doppler frequency shift ∆ f is obtained. The arctangent calculation is then performed on this set (Equation ( 9)) to obtain the phase solution 2π∆ f t. In this study, the coordinate rotation digital computer (CORDIC) algorithm was selected to perform the arctangent calculations on the FPGA [20][21][22][23][24]. The CORDIC algorithm has been extensively studied and will not be explained in detail in this article.…”

Transformer Fault Location Classification Using FFT Based 1D-Convolutional Neural Network Model

“…Training datasets are often segmented into smaller portions, necessitating running algorithms on each fragment for predictions. This process can be time-consuming, thus to overcome this FFT optimization, we're streamlining these computations, reducing overall training time [20]. To enhance real-time analysis, we're integrating FFT algorithms to replace the conventional forward Convolution propagation in our models.…”