In-sensor optoelectronic computing using electrostatically doped silicon

Complementary metal-oxide-semiconductor (CMOS) image sensors are a visual outpost of many machines that interact with the world. While they presently separate image capture in front-end silicon photodiode arrays from image processing in digital back-ends, efforts to process images within the photodiode array itself are rapidly emerging, in hopes of minimizing the data transfer between sensing and computing, and the associated overhead in energy and bandwidth. Electrical modulation, or programming, of photocurrents is requisite for such in-sensor computing, which was indeed demonstrated with electrostatically doped, but non-silicon, photodiodes. CMOS image sensors are currently incapable of in-sensor computing, as their chemically doped photodiodes cannot produce electrically tunable photocurrents. Here we report in-sensor computing with an array of electrostatically doped silicon p-i-n photodiodes, which is amenable to seamless integration with the rest of the CMOS image sensor electronics. This silicon-based approach could more rapidly bring in-sensor computing to the real world due to its compatibility with the mainstream CMOS electronics industry. Our wafer-scale production of thousands of silicon photodiodes using standard fabrication emphasizes this compatibility. We then demonstrate in-sensor processing of optical images using a variety of convolutional filters electrically programmed into a 3 × 3 network of these photodiodes.

Piezoelectric MEMS Acoustic Transducer with Electrically-Tunable Resonant Frequency

The paper presents a technique to obtain an electrically-tunable matching between the series and parallel resonant frequencies of a piezoelectric MEMS acoustic transducer to increase the effectiveness of acoustic emission/detection in voltage-mode driving and sensing. The piezoelectric MEMS transducer has been fabricated using the PiezoMUMPs technology, and it operates in a plate flexural mode exploiting a 6 × 6 mm doped silicon diaphragm with an aluminum nitride (AlN) piezoelectric layer deposited on top. The piezoelectric layer can be actuated by means of electrodes placed at the edges of the diaphragm above the AlN film. By applying an adjustable bias voltage Vb between two properly-connected electrodes and the doped silicon, the d31 mode in the AlN film has been exploited to electrically induce a planar static compressive or tensile stress in the diaphragm, depending on the sign of Vb, thus shifting its resonant frequency. The working principle has been first validated through an eigenfrequency analysis with an electrically induced prestress by means of 3D finite element modelling in COMSOL Multiphysics®. The first flexural mode of the unstressed diaphragm results at around 5.1 kHz. Then, the piezoelectric MEMS transducer has been experimentally tested in both receiver and transmitter modes. Experimental results have shown that the resonance can be electrically tuned in the range Vb = ±8 V with estimated tuning sensitivities of 8.7 ± 0.5 Hz/V and 7.8 ± 0.9 Hz/V in transmitter and receiver modes, respectively. A matching of the series and parallel resonant frequencies has been experimentally demonstrated in voltage-mode driving and sensing by applying Vb = 0 in transmission and Vb = −1.9 V in receiving, respectively, thereby obtaining the optimal acoustic emission and detection effectiveness at the same operating frequency.

Measurements of a Silicon Photonic ROADM (Project Report CIAN2-3-Y3)

A multi-university partnership led by UCSD collaborated with Sandia National Labs in an NSF-funded silicon photonics multi-project wafer (MPW) project. This is a report of the ROADM +VOA (reconfigurable optical add drop multiplexer + variable optical attenuator) device made using silicon photonics, including passive and doped silicon waveguides and metalization.