in vivo Neuronal Imaging (in collaboration with Schnitzer Lab)
We are currently developing fully-integrated and portable fluorescence microscopy devices with higher performance and improved functionality for in vivo neuronal imaging applications.
Folded Multiple Capture: An Architecture for High Dynamic Range, High Speed Image Sensors
We have been interested in high speed, high dynamic range (HDR) image sensor architectures with high signal fidelity and low power consumption. Such requirements are often encountered in infrared (IR) imaging applications in the automotive, industrial, surveillance, tactical, and medical and diagnostic instrumentation arenas. We have developed a new HDR architecture, Folded Multiple Capture (FMC), targeted towards these applications. A prototype implemented in a 0.18um CMOS process achieves >138dB dynamic range at high speeds (1000 frames/sec) with high SNR (>60dB) and low power consumption (25.5nJ per pixel readout). The prototype can be readily extended to a fully integrated imaging system leveraging 3D-IC technology.
Publications
Per-Pixel Background Subtraction for Imaging Applications
We have developed a per-pixel background subtraction scheme based on Pulse Frequency Modulation (PFM). The scheme is targeted for imaging applications that suffer from high background, for example, IR applications which often involve imaging scenes with small target signal variations over a large background offset, and non-invasive in vivo fluorescence imaging applications where the large background is due to tissue autofluorescence. We have achieved significant performance improvements over previous current subtraction schemes.
Publications
Carbon Nanotube Sensors
Carbon nanotube sensors, exhibiting great promise over conventional solid-state sensors for chemical and biological sensing applications, were modeled and their design space analyzed. It is shown that single molecule analyte detection can be readily achieved with 1nm diameter nanotubes up to 30um in length.
Publications
Undergraduate Senior Thesis: CMOS Imager with Polarization Difference Imaging
Primary Advisor: Prof. Jan Van der Spiegel, Department of Electrical and Systems Engineering
Co-Advisor: Prof. Nader Engheta, Departments of Electrical and Systems Engineering and Bioengineering
Polarization-difference imaging (PDI) is a biologically-inspired technique that improves the visibility of objects in scattering media by amplifying the PD signal from targets whose PD magnitude is distinct from that of the background. The project was inspired by the quest to integrate a PDI system with a CMOS camera module. A prototype implemented in a 0.5um CMOS process demonstrated the proof-of-concept and is amenable for future integration with micropolarizers that separate the orthogonal polarization components of incident light. A fully integrated imaging system has applications in nautical, aeronautical and space sensor systems, where imaging in harsh light scattering conditions is required.
A High Voltage CMOS ADSL Line Driver
Advisor: Prof. C. Andre T. Salama, Department of Electrical and Computer Engineering
In the summer of 2001, I was involved in the design and implementation of a high voltage CMOS ADSL line driver for short loop Subscriber Line Interface Circuits (SLICs). The project was motivated by the desire to find an economical conventional CMOS solution to replace existing BiCMOS implementations, and was enabled by the developments in high voltage submicron CMOS technology. The line driver implemented in a conventional 50V 0.8um CMOS process exhibited significantly better performance than previous designs implemented in more complex and expensive BiCMOS processes.
Publications
Copyright © 2006-2008 Kunal K. Ghosh kghosh at stanford dot edu