A HIGH-VOLTAGE NEURAL STIMULATOR COMBINED WITH A LOW-VOLTAGE RECORDER
HIGH-VOLTAGE NEURAL STIMULATOR COMBINED WITH A
LOW-VOLTAGE RECORDER
Ulrich Bihr, Jens Anders, Joachim Becker and Maurits Ortmanns
Institute of Microelectronics, University of Ulm, Ulm, Germany
Ulrich.Bihr@uni-ulm.de
Abstract: This paper presents a high-voltage (HV) neural stimulator combined with a low-voltage (LV) neural recorder. In many bidirectional neural implementations with a high voltage compliance for the stimulation is it not possible to have a high density due to the high power consumption in the recording part with the same HV supply.
This realization shows a stimulator with a HV supply of 7.5V to enable high voltage compliance together with a neural recorder, which uses a LV supply of 1.65V to minimize the power consumption of the recording. In addition, the recorder split the neural signal into the localfield- potentials and the action potentials. With a separation and an individual amplification of the two sub signals to the maximum swing, it is possible to reduce the dynamic range and thus resolution of the required ADC. This results in a major data reduction and reduces the power consumption of the recording part. The implementation presents a stimulator with a maximum stimulation current of 15mA with
5-bit resolution and 7-bit dynamic range combined with a recorder, which consumes 52W, and provides an input referred noise of 3.8Vrms for the LFPs and the APs.
Keywords: Neural stimulator, neural recording, front-end circuit, separation, switched-capacitor
Introduction
Parallel neural recording combined with neural stimulation enables neuroscientist the possibility to study the complex neural networks in more detail. A combined system has the major advantage that it is possible to stimulate at one electrode and record the neural reaction of the neural network at the same electrode afterwards. With this information, the waveform of the stimulation can be optimized and this gives the possibility to stimulate the
LOW-VOLTAGE RECORDER
Ulrich Bihr, Jens Anders, Joachim Becker and Maurits Ortmanns
Institute of Microelectronics, University of Ulm, Ulm, Germany
Ulrich.Bihr@uni-ulm.de
Abstract: This paper presents a high-voltage (HV) neural stimulator combined with a low-voltage (LV) neural recorder. In many bidirectional neural implementations with a high voltage compliance for the stimulation is it not possible to have a high density due to the high power consumption in the recording part with the same HV supply.
This realization shows a stimulator with a HV supply of 7.5V to enable high voltage compliance together with a neural recorder, which uses a LV supply of 1.65V to minimize the power consumption of the recording. In addition, the recorder split the neural signal into the localfield- potentials and the action potentials. With a separation and an individual amplification of the two sub signals to the maximum swing, it is possible to reduce the dynamic range and thus resolution of the required ADC. This results in a major data reduction and reduces the power consumption of the recording part. The implementation presents a stimulator with a maximum stimulation current of 15mA with
5-bit resolution and 7-bit dynamic range combined with a recorder, which consumes 52W, and provides an input referred noise of 3.8Vrms for the LFPs and the APs.
Keywords: Neural stimulator, neural recording, front-end circuit, separation, switched-capacitor
Introduction
Parallel neural recording combined with neural stimulation enables neuroscientist the possibility to study the complex neural networks in more detail. A combined system has the major advantage that it is possible to stimulate at one electrode and record the neural reaction of the neural network at the same electrode afterwards. With this information, the waveform of the stimulation can be optimized and this gives the possibility to stimulate the
Bibliography: filter structure for neural recording applications,” in Circuits and Systems (ISCAS), 2012 IEEE International Symposium on, pp. 2231 –2234, may 2012. IEEE Journal of, vol. 47, pp. 244 –256, jan. 2012.