Neuromodulation  alters the nerve activity by delivering electrical or pharmaceutical agents directly to a target area. It has long been investigated after the success of the cochlear implant. Targeted neural systems mostly have dense and electro-chemical sensitive environments, which make it harder for selective stimulation. As proven from prior studies, the charge needed to evoke neuronal activity is mostly less than hundreds of micro Ampere(<250μA). Electrolytes surrounding the tissue, and the impedance variation between the tissue and electrode lessen the accuracy and the efficiency. To guarantee high-resolution stimulation and to maintain the cell’s health, it needs to adopt a smaller electrode with high voltage compliance current stimulating system. Also, for the stimulation current pulse, it is reported that monophasic pulses lead to charge buildup at the electrode/electrolyte interface, which is causing damage to the electrode and surrounding tissue. Biphasic pulses maintain charge balance and are used to safely stimulate tissue.

The research here is about building application-specific neural stimulator and conduct experiment through animal model.

In a neural sensor network where there are many low-power micro-systems such as IoT, an address and wireless telemetry  is essential for each independent device. In such a system, it should be possible to give a small size a fairly large bit size. In addition, it is good if the address can be encoded by post-processing with one copy device due to the characteristics of the system. In addition to the existing address assignment methods such as eFuse or PUF ID Generation, we are trying to devise a method that can assign addresses on the circuit. Also, adopting backscattering communication allows our system to have ultra-low power communication.