目的针对电子鼻在温湿度波动较大的医疗环境和户外等场景中因传感器漂移导致检测失效的问题,提出小样本补偿模型,解决传统方法依赖大量漂移数据、难以适应长期非线性漂移的瓶颈。方法构建传感器漂移适应中的图神经网络(graph neural net...目的针对电子鼻在温湿度波动较大的医疗环境和户外等场景中因传感器漂移导致检测失效的问题,提出小样本补偿模型,解决传统方法依赖大量漂移数据、难以适应长期非线性漂移的瓶颈。方法构建传感器漂移适应中的图神经网络(graph neural network used in sensors drift adaptation,GNNSD)模型,融合深度残差卷积与图神经网络,采用数据增强与关系推理机制,在公开传感器漂移数据集上开展小样本分类实验。结果GNNSD模型在K=1设置下实现84.12%平均准确率,较最优对比算法FEDA提升9.93%。消融实验表明模型架构具有合理性。结论该模型通过多尺度特征与图结构关系推理的协同机制,当每个类别的参考样本数量只有1个时也可实现较高分类精度,为医疗监测、跨境筛查等生物安全场景提供低样本依赖的漂移补偿解决方案。展开更多
Piezoresistive pressure sensors based on silicon have a large thermal drift because of their high sensitivity to temperature. The study of the effect of the temperature and doping level on characteristics of these sen...Piezoresistive pressure sensors based on silicon have a large thermal drift because of their high sensitivity to temperature. The study of the effect of the temperature and doping level on characteristics of these sensors is essential to define the parameters that cause the output characteristics drift. In this study, we adopted the model of Kanda to determine the effect of the temperature and of doping level on the piezoresistivity of the Silicon monocrystal. This is to represent P(N,T) and for p-type silicon as functions of impurity concentration for different temperatures. This allows us to see the effect of temperature and doping concentration on the output characteristics of the sensor. Finally, we study the geometric influence parameters and doping on these characteristics to optimize the sensor performance. This study allows us to predict the sensor behavior against temperature and to minimize this effect by optimizing the doping concentration.展开更多
The extended gate field effect transistor (EGFET)has many advantages such as the fabrication is easy,low cost, easy to operate etc.The EGFET was applied to biosensor in recent years.In this study,the tin oxide (SnO_2)...The extended gate field effect transistor (EGFET)has many advantages such as the fabrication is easy,low cost, easy to operate etc.The EGFET was applied to biosensor in recent years.In this study,the tin oxide (SnO_2)pH sensitive membrane was deposited on ITO glass,when the surface voltage which pH membrane changes,the gate voltage and current channel of MOSFET will change immediately to detect concentration of the glucose sensor.In this study we have devoted to research about the calibration of the circuit measurement for the glucose sensor,and study the calibration system of the drift and hysteresis.展开更多
文摘目的针对电子鼻在温湿度波动较大的医疗环境和户外等场景中因传感器漂移导致检测失效的问题,提出小样本补偿模型,解决传统方法依赖大量漂移数据、难以适应长期非线性漂移的瓶颈。方法构建传感器漂移适应中的图神经网络(graph neural network used in sensors drift adaptation,GNNSD)模型,融合深度残差卷积与图神经网络,采用数据增强与关系推理机制,在公开传感器漂移数据集上开展小样本分类实验。结果GNNSD模型在K=1设置下实现84.12%平均准确率,较最优对比算法FEDA提升9.93%。消融实验表明模型架构具有合理性。结论该模型通过多尺度特征与图结构关系推理的协同机制,当每个类别的参考样本数量只有1个时也可实现较高分类精度,为医疗监测、跨境筛查等生物安全场景提供低样本依赖的漂移补偿解决方案。
文摘Piezoresistive pressure sensors based on silicon have a large thermal drift because of their high sensitivity to temperature. The study of the effect of the temperature and doping level on characteristics of these sensors is essential to define the parameters that cause the output characteristics drift. In this study, we adopted the model of Kanda to determine the effect of the temperature and of doping level on the piezoresistivity of the Silicon monocrystal. This is to represent P(N,T) and for p-type silicon as functions of impurity concentration for different temperatures. This allows us to see the effect of temperature and doping concentration on the output characteristics of the sensor. Finally, we study the geometric influence parameters and doping on these characteristics to optimize the sensor performance. This study allows us to predict the sensor behavior against temperature and to minimize this effect by optimizing the doping concentration.
文摘The extended gate field effect transistor (EGFET)has many advantages such as the fabrication is easy,low cost, easy to operate etc.The EGFET was applied to biosensor in recent years.In this study,the tin oxide (SnO_2)pH sensitive membrane was deposited on ITO glass,when the surface voltage which pH membrane changes,the gate voltage and current channel of MOSFET will change immediately to detect concentration of the glucose sensor.In this study we have devoted to research about the calibration of the circuit measurement for the glucose sensor,and study the calibration system of the drift and hysteresis.
