Módulo para aquisição de sinais de eletroencefalograma com transmissão via protocolo sem fio

Abstract

Biopotentials are very useful for detection of patologies and for social inclusion of individuals with special needs. The present project consists in the development of a module of EEG acquisition, with wireless transmission. Electroencephalogram signals present amplitudes in the range of µV and are composed, mostly, by frequencies lower than 100Hz. Thus, a circuit capable of providing great amplification of inputs signals and, at the same time, reject noises like those from the power grid is necessary. For wireless transmission, the ZigBee protocol was chosen, mainly due to its low current of operation. Its implementation was made using the IRIS-XM2110 board, produced by Crossbow. The program responsable for the communication was developed using the nesC programming language, specific for the TinyOS operational system. Several difficulties were faced to learn this language and to develop the circuit. The current module contains a eight channel circuit, capable of acquiring EEG signals, binded to eight internal IRIS’s ADC channels. The system is capable of acquiring with a sample rate around 100Hz, which means loss of some signal’s details, but guarantees visualization of its major components. While one IRIS mote does the acquirement and transmission, another mote configured as basestation receives these data and send them to a PC through serial interface. At last, the user’s interface consists in a program based on MATLAB, capable of reading data from serial bus and plotting these values in time and frequency domains. The experiments made validate the adopted topology, through comparing acquired EEG signals and the ones found in literature in terms of differences between alert and rest states. Besides, signals of electrocardiogram and electrooculogram have been sampled, in order to prove the easy adequation of the module for other biopotentials. This project has also the possibility of continuity, improving the aquired signals and using the module in applications related to rehabilitation.