Development Of Ultra-light Wearable Ear Blood Oxygen Real-time Monitoring System

Blood oxygen saturation is an important physiological parameter of human respiration and circulation,which represents oxygen intake and circulation status.The change of its value has a great reference for response of human health status.At present,the type of oximeter is mainly finger-chip,which measures on the finger and is not suitable for real-time monitoring;in recent years,smart watches with heart rate and blood oxygen monitoring functions have been continuously introduced.However,blood oxygen signals are more sensitive to motion noise,so the actual monitoring effect is not ideal.The combination of blood oxygen monitoring and wearable devices still has great room for improvement.In view of these deficiencies,this paper selects the measurement site in the ear and develops a wearable ear blood oxygen real-time monitoring system based on the reflection-type of blood oxygen probe.The system consists of front-end signal acquisition equipment and monitoring APP.When the blood oxygen value is lower than the normal value,an alarm will be given in time.By studying the low-power performance of hardware and the underlying characteristics of low-power bluetooth protocol stack,a small-volume,ultra-light blood oxygen signal acquisition circuit is developed,and the blood oxygen probe can automatically adjust the working state and is combined with the bluetooth dormancy awakening mechanism.The professional antenna simulation software and vector network analyzer are used to determine the ?-type matching network parameters of the antenna.After optimized.The minimum reflection coefficient is-39dB,and the bandwidth at-10dB is 0.08GHz.The stable wireless chain with the monitoring mobile phone is realized.Road transmission,which ensures the wireless transmission link is stable and reliable.In view of the common noise in pulse wave signal,and considering the data processing ability,algorithm complexity and real-time of the embedded system,this paper proposes the signal pre-processing algorithm based on smoothing and morphological filtering and adaptive pulse wave peak detection algorithm,which have been successfully transplanted into the hardware system.The blood oxygen calculation curve is obtained by method of direct human body calibration.At the same time,in order to ensure the stability of the monitoring system and avoid the phenomenon of data mis-transmission,leakage transmission,full storage in the process of dynamically processing a number of ring storage areas are established in the chip to realize the dynamic cache of data flow.The test results show that the system works stably.Compared with the market finger-type oximeter,the heart rate error is 2 times/minute under static conditions,and the blood oxygen saturation error is within 2%in the range of 85%to 100%.The overall hardware volume is 15 mm*15 mm*3.5 mm,and the weight is 0.89 g.An 80 mAh battery can enable the system working for over 12 hours,which meets the requirement of daily monitoring.