Parallel reconfigurable computing and its application to hidden Markov model

Abstract

Parallel processing techniques are increasingly found in reconfigurable computing, especially in digital signal processing (DSP) applications. In this paper, we design a parallel reconfigurable computing (PRC) architecture which consists of multiple dynamically reconfigurable computing units. The hidden Markov model (HMM) algorithm is mapped onto the PRC architecture. First, we construct a directed acyclic graph (DAG) to represent the HMM algorithms. A novel parallel partition approach is then proposed to map the HMM DAG onto the multiple DRC units in a PRC system. This partitioning algorithm is capable of design optimization of parallel processing reconfigurable systems for a given number of processing elements in different HHM states.