Reducing Complexity of Wavelet Based EMI Source Discovery via Bayesian Hash Coding

Locating the electromagnetic interference (EMI) sources of an electronic system remains to be one of the most critical issue for electromagnetic compatibility (EMC) design. Computational models were developed to help designers to locate existing EMI sources. In this paper we present a Bayesian model to formalize the process, and an optimized hash coding method to reduce the inference complexity. After wavelet transformation, high dimensional sub-signals are mapped into low dimensional hash vector. Similar vectors are then stored in a hash bucket and this can be viewed as a preprocessing step whose prior knowledge of the signals can be adopted into subsequent discovery step through Bayesian equation. Experimental results prove that the convergence of our method is at least an order of magnitude faster than other state-of-the-art models.