Algorithmic Optimization of BDDs and Performance Evaluation for Multi-level Logic Circuits with Area and Power Trade-offs

Binary Decision Diagrams (BDDs) can be graphically manipulated to reduce the number of nodes and hence the area. In this context, ordering of BDDs play a major role. Most of the algorithms for input variable ordering of OBDD focus primarily on area minimization. However, suitable input variable ordering helps in minimizing the power consumption also. In this particular work, we have proposed two algorithms namely, a genetic algorithm based technique and a branch and bound algorithm to find an optimal input variable order. Of course, the node reordering is taken care of by the standard BDD package buddy-2.4. Moreover, we have evaluated the performances of the proposed algorithms by running an exhaustive search program. Experi-mental results show a substantial saving in area and power. We have also compared our techniques with other state-of-art techniques of variable ordering for OBDDs and found to give superior results.

Performance analysis of ASR system in hybrid DNN-HMM framework using a PWL euclidean activation function

Automatic Speech Recognition(ASR)is the process of mapping an acoustic speech signal into a human readable text format.Traditional systems exploit the Acoustic Component of ASR using the Gaussian Mixture Model-Hidden Markov Model(GMM-HMM)approach.Deep NeuralNetwork(DNN)opens up new possibilities to overcome the shortcomings of conventional statistical algorithms.Recent studies modeled the acoustic component of ASR system using DNN in the so called hybrid DNN-HMM approach.In the context of activation functions used to model the non-linearity in DNN,Rectified Linear Units(ReLU)and maxout units are mostly used in ASR systems.This paper concentrates on the acoustic component of a hybrid DNN-HMM system by proposing an efficient activation function for the DNN network.Inspired by previous works,euclidean norm activation function is proposed to model the non-linearity of the DNN network.Such non-linearity is shown to belong to the family of Piecewise Linear(PWL)functions having distinct features.These functions can capture deep hierarchical features of the pattern.The relevance of the proposal is examined in depth both theoretically and experimentally.The performance of the developed ASR system is evaluated in terms of Phone Error Rate(PER)using TIMIT database.Experimental results achieve a relative increase in performance by using the proposed function over conventional activation functions.