AN INTEGRATED MULTICHANNEL NEURAL RECORDING SYSTEM WITH SPIKE OUTPUTS By YUAN LI A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA

of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy AN INTEGRATED MULTICHANNEL NEURAL RECORDING SYSTEM WITH SPIKE OUTPUTS By Yuan Li May 2007 Chair: John G. Harris Major Department: Electrical and Computer Engineering The increasing need for implantable devices in biological signal recording systems requires compact, ultra low-power and low-noise electronics. The continuous evolution of integrated circuit technologies partially helps in meeting these requirements by allowing for the realization of more complex functions in a given silicon area. The shrinking power supply helps reducing the power consumption, on the other hand, it also brings new challenges in terms of noise. The purpose of this research is to investigate the feasibility of a multi-channel neural recording system with spike outputs. The multi-channel system imposes four major constraints: large communication bandwidth, simple and compact circuit, low noise and low power. The proposed solution is to design a novel and simple current generator which can be directly used for biphasic spike representation. The current generator can convert the input alternating current voltage (AC) voltage to output AC current, while eliminating the direct current voltage (DC) voltage. By using this current generator, the total system is more compact and less noisy. Integrating the AC current, the output of each channel is a biphasic spike train. These multi-channel spike trains are then transmitted using the address event representation (AER), which can improve the communication efficiency. Preliminary theoretical analysis, simulation results and chip measurements show suitability for neural recording applications.