Browsing by Author "Mark White, Member"

Filter results by typing the first few letters
Now showing 1 - 2 of 2
  • No Thumbnail Available
    An Adaptive Electronic Interface for Gas Sensors
    (2002-01-22) Cavanaugh, Curtis; H. Troy Nagle, Chair; Edward Grant, Member; Mark White, Member
    This thesis focuses on the development of an adaptive electronic interface for gas sensors that are used in the NC State electronic nose. We present an adaptive electronic interface that allows for the accurate mapping of the sensor's voltage output to sensor resistance profiles. The adaptive interface uses a linearized Wheatstone bridge in a constant current configuration. The balancing of the bridge and the adjustment of the subsequent gain stage is performed using programmable variable resistors. The programmable resistors are controlled by a LabVIEW® program. The same control program also determines and records all the resistor values in the interface circuit. The resistance of each sensor is accurately computed by LabVIEW® using the interface-circuit, resistor values, and the voltage output of the circuit. Compensating for sensor drift can be done in LabVIEW® by adjusting the programmable resistor values so that a zero-voltage output is produced during the reference cycle. By doing this zero adjustment between each 'sniff' of an odorant, the baseline drift can be minimized.A single channel of the adaptive electronic interface has been designed and tested. The interface can be calibrated so that it is 99% accurate when performing sensor resistance measurements. A new conducting polymer sensor chamber has also been designed and tested. The new radial flow sensor chamber was minimizes the dead volume in the chamber and also deliver the odorant to each sensor at the same time. Two operating modes were compared: continuous-flow and sniff-and-hold. Both modes gave good classification performance while testing four different coffee samples. Experimental testing indicates that sensor response is highly correlated with the sample flow rate. Future work to more fully characterize this correlation is recommended.
  • No Thumbnail Available
    Using An Artificial Neural Network to Detect Activations During Ventricular Fibrillation
    (1998-07-02) Young, Melanie Talanda; Susan M. Blanchard, Chair; Mark White, Member; H. Troy Nagle, Member; W. F. McClure, Member
    Ventricular Fibrillation (VF) is a severe cardiac arrhythmia that canresult in sudden death, a leading cause of death in the United States. DuringVF, the electrical activity of the heart becomes disordered, the ventriclescontract erratically, and an insufficient supply of blood is pumped to the body.Identification of the electrical activation sites during VF is important for theunderstanding and improved treatment of the disorder. Unipolar electrogramsof four pigs were recorded following the induction of VF. The data from theVF recordings was preprocessed using a Rule-Based Method (RBM) , a Current Source Density (CSD) method, and a Transmembrane CurrentMethod (TCM) to separate local activations from distant activity. RBM usesthe magnitude of the derivative of the voltage to identify activations. CSD is ascalar quantity that represents the magnitude of the current source or sink.TCM estimates a value proportional to the transmembrane current. Afeedforward artificial neural network (ANN) using backpropagation was trained to identify the local activations in the electrograms of VF based on the RBM and CSD calculations. Another feedforward ANN usingbackpropagation was trained using data preprocessed with not only RBM andCSD, but also TCM. In order to improve the ability of the ANNs to detect localactivations, a new training method, called staged training, was utilized. Instaged training, the ANNs were trained in stages using different sets oftraining examples. Examples were included in a particular training set basedon the minimum magnitude of the voltage derivative. When training was done in stages, both the ANN with RBM and CSD data only and the ANN with RBM,CSD, and TCM data were able to more accurately distinguish activations.Overall, the ANN, which used only RBM and CSD data, produced better results than the ANN that also included TCM data.