Implemantion of some data processing algorithms in embedded systems

Instrumentation systems usually include data collection, processing and presentation units. On any measurement process, these units are generally implemented on separate hardware due to insufficient processing power for data processing algorithms. This makes difficult to implement online measurement systems. Due to rapid improvements on embedded systems, desired high processing power can be found on microprocessor systems. In this perspective, a small library that includes some data processing algorithms used in high level programming languages such as Matlab, is implemented for Arm-Cortex M3 microcontroller. At the beginning feed forward neural networks and nearest neighbor search algorithms are implemented as library. In order to provide easiness in use, Matlab function prototypes are implemented without any changes.