Embedded hardware solution for principal component analysis

Over the past few years, the realm of portable and embedded computing has expanded to include a wide variety of applications on handheld devices. However, these devices have stringent area and power requirements. Coupled with increasing pressure to decrease cost and shorten time to market, the design constraints of these devices pose a serious challenge to the designers. We are investigating the utilization of FPGA-based hardware for compute intensive applications in portable and embedded computing. In this work, we introduce an FPGA-based hardware solution for Principal Component Analysis (PCA), a classic technique to reduce the dimensionality of data by transforming the original data set to a new set of variables called principal components representing the key features of the data. Experiments are performed using a benchmark dataset on handwriting analysis, in order to evaluate the feasibility and efficiency of our embedded hardware solution.