Research on Integrated Circuit Fault Detection Based on Improved Particle Swarm Optimization (SNA)

In view of the shortcomings and defects of the current theory of transient interference signal identification of integrated circuits, this paper applies machine learning to signal recognition, takes the number of original scan chains and the scan compression ratio as the configuration parameters, selects the best algorithm for different target parameters, quickly predicts the target parameters of the DFT structure in the selectable range, and efficiently deduces the optimal configuration parameters of DFT. The wavelet analysis technology is used to decompose the signal wavelet to extract the signal features, and the dynamic particle swarm optimization is used to search for the transient interference signal according to the features, so as to complete the identification of the transient interference signal of the integrated circuit based on the dynamic particle swarm, which has a good application prospect in the recognition of the transient interference signal of the integrated circuit.