A System Of Multi-channel Wireless EEG Signal Acquisition

Electroencephalogram (EEG) is the intracranial or cortex potential change formed bycerebral neurons’ electrical activities. Without any stimulation, the potential change recorded onthe cortex or scalp by unipolar or bipolar electrode is named electroencephalogram (EEG). Withthe development of signal processing and computer technologies, EEG signal processing hasbeen developed rapidly1.BCI(BrainComputerInterface)focuses on the study of the establishment of a communicationchannel between human brain and extermal devices through which information exchange takesplace.BCI covers many fields,such as biology,medicine,computer and communication.Clinical data recorded on human scalp can hardly reflect the electrical activity of cerebralneurons accurately. However, in the animal experiments, we can record intracranial EEG signal,which has high spatial resolution but few interference; they can reflect the electrical activitiesdeep in the brain. So, we establish animal model for EEG record and analysis2.The main content of our paper introduces a system of multi-channel wireless EEG signalacquisition and processing. The whole system is divided into three parts,the collection circuit oforiginal analog signal, the A/D conversion circuit and the block of data compression and coding,the module of wireless transmitting and receiving1.The front part of our system is the design of analog signal collection.First of all, thepotential of EEG signal is very small,if we want to analysis the EEG signal in the back-endcircuit,all the amplifiers should have a high magnification enough.Secondly, the EEG signal isvery easily influenced by the external noise around our system.In order to eliminate the electrodevoltage,we use the high-pass filter in the front-end. We filter out almost all the high frequencyinterference by the RF circuit and then through a low pass filter to reduce the noise. Finally, weanalyze the noise of system that we design,such as thermal noise, the input current noise,theinput voltage noise and the noise of instrument amplifier. The A/D data conversion and the compression of the analog signal are the two parts ofdigital signal processing block. The A/D chip is AD7993which is produced by Analog, and itseffective bits is ten, and has characteristic of the high transmission speed, low power, dualchannel continuous approximation. Its highest data transmission rate can reach1MSPS.In thealgorithm,EEG signal is first coded by DPCM that removes the temporal redundancy of thedata,then adaptive Huffman coding is used to compress the data further by removing thestatistical redundancy.The choice of wireless communication protocol plays a very important role in thetransmission of EEG signal.This system is based on Cortex-M3microprocessor STM32W108which integrates IEEE802.15.4wireless communication protocol.The main purpose of IEEE802.15.4is building a net which transmitation speed is two hundred and fifty kilobit per secondand communication distance is ten meters.The main feature of this wireless communicationprotocol is low cost，simple，flexible and general use.In the last part of this paper, we introduce the related simulation and some actual testingresults. And then we draw some conclusions and the outlook based on the overall design of thissystem. Meanwhile this paper introduces and analyzes the related questions and unfinished tasks.