Data Acquisition And OProcessing Of SEMG Signals For Bionic Arm

Data Acquisition And Processing of SEMG Signals for Bionic ArmAt present,the total number of persons with disabilities is large, the number of upper limb disabilities accounted fo a large part of it. Physical disabilities bring a lot of inconvenience to their life, so it is a meaningful thing to design a bionic arm, it can improve the quality of life of persons with disabilities.SEMG is recorded from the surface of skeleton muscle by electrodes, it is the bioelectricity discharged by neuromuscular activities. The artificial limb controlled by EMG can reflect the person's limb movement information, it also can realize multi-freedom prosthetic control, so EMG is more suitable for the bionic arm. Using existing, expensive muscle electricity collection device to acquisition of multi-channel SEMG is widely used at present. This method is only applicable to the analysis of the SEMG, it is not suitable for the bionic arm. We will design an acquisition and processing system of SEMG for bionic arm.The system we designed can complete the function of SEMG amplification, A/D conversion, data storage, data transmission and processing. System design includes hardware design and software programming, hardware design includes analog circuit design and digital circuit design, software design includes MCU programming and MATLAB programming.SEMG signal is a very weak signal, and it is susceptible to noise interference. Analog hardware circuitry not only to magnify the SEMG signal, but also to suppress the interference signal effectively. We use multi-stage amplifier, the first stage amplifier uses the differential input to improve the circuit's common mode rejection ratio(CMRR), the second stage amplifier amplify the weak SEMG signal with the AD620. In order to suppress high frequency noise and power frequency interference, we design a 4-order Chebyshev low pass filter circuit and industrial frequency trapped wave circuit. The filter circuit can be very good to complete the high-frequency interference and 50Hz frequency interference filtering function.The C8051F320 microcontroller is the core of digital circuits, digital section includes the A/D conversion, data storage and data transmission part. First, the SEMG analog signal will be converted into the digital signal. The paper use the ADC module integrated in the C8051F320 to do A/D conversion. We design A/D acquisition program by setting register, using double-ended input mode, the sampling frequency is 1000Hz.The digital signal should be saved for subsequent using. We use multiple sources of SMBus interface to complete consecutive reading and writing for the 24C512.The function of data transfer part of the system is make the PC read and write SEMG data stored in the 24C512. We use serial port of PC to read and write data, serial port of PC uses RS232 level, but MCU system uses TTL level, so we use MAX232 to convert RS232 level to TTL level. In the MCU, it can send data with fixed baud rate by setting he corresponding port register. In the PC, we use callback function for the serial data of the MATLAB to read the data in the MCU through the serial port.In the PC, we use the MATLAB to design digital filter, we designed450Hz low pass filters, 10Hz high-pass filters and industrial frequency filters by the MATLAB. Then we use these digital filters to filter the SEMG signal. Experiments show that the filter works well and it can meet our needs.Finally, the paper carried out motion recognition experiments, we use the SEMG acquisition system which we designed to do dual-channel data acquisition, then extract the SEMG signal feature values to constitute the feature vector for motion recognition. The recognition rate obtained by experiment is more than the rate which got the SEMG signal by using of available SEMG acquisition instrument. It proves the effectiveness of the systems we designed, the system can meet the need of the SEMG signal acquisition for bionic arm.