Development and implementation of a feature-based automatic classification algorithm for communication standards in the 868 MHz band

Wireless communication systems used for home automation applications in Europe suffer from the recurring standardization problem. Devices from different manufacturers often use different and incompatible protocol waveforms for communication. In this paper, we present the development and the implementation of a feature-based automatic classification algorithm for communication standards that are commonly used in the European 868 MHz short range device (SRD) frequency band, that can help to overcome this problem. The received signal is initially preprocessed and then fed to a feature extraction unit that calculates statistical parameters from the signal. Those features are then used by a classifier that is based on a simple binary decision tree. The performance of the algorithm is verified via Monte-Carlo simulation and compared to the bit error rates of the considered communication standards as a function of the signal-to-noise ratio. Finally, the algorithm is implemented in a software defined radio (SDR) platform and the measurement results are compared to the simulation-based success rates.