Welcome to the Integrated Silicon Systems (ISS) Lab. We are performing research in low-power analog, mixed-signal, and RF IC design; biomedical interfaces; and neuromorphic engineering.
Progress in semiconductor technology over the last couple of decades has brought a wide range of applications within the realm of possibility. While transistors have shrunk dramatically, the physical size of electronic systems remains limited by power sources. In order to realize the next generation of micro-scale electronic systems, members of the ISS lab are developing circuit techniques to dramatically reduce the power consumption of sensor-interface, communication, and computation sub-systems. Our research projects can be broadly broken down into two application areas.
Sensing Systems There is a growing interest in developing small wireless sensors to perform measurements where previous instrumentation could not be deployed. Examples include implanted devices to translate neural signals into control signal for prosthetic limbs, and devices to observe the behaviors and environments of small animals or insects in flight. Often, the power source is the biggest obstacle to achieving the size, weight, and lifetime needed for a viable system. To enable the next generation of ultra-miniaturized sensing systems, we are developing low-power circuits such as amplifiers for acquiring neural signals and wireless communication circuits.
Analog Computation By using the physics of transistor operation, analog circuits can compute in a way that is both elegant and much more efficient than their digital counterparts. While analog circuits introduce errors due to noise, non-linearity, etc., some applications are quite tolerant to these effects. Machine learning (ML) is one such application. The naturally parallel nature of many ML algorithms also makes it well-suited to analog computation. We are building analog learning circuits that increase speed and reduce power in order to enable both large-scale and small-scale learning systems.