The objective of this research is to create a new class of power-efficient imaging photoreceivers to reach an aggregated data rate of 50 gigabits per second and beyond for broadband wireless optical communication. The approach is to develop a large array of planar tessellated photodetectors with a rate of 10 gigabits per second per pixel. The array includes diversity selection circuits, which are implemented using an optical multi-input/multi-output configuration.
With respect to intellectual merit, this project addresses two well-known challenges in optical wireless communication, channel scintillation and optical beam alignment. The research explores power-efficient InGaAs metal-semiconductor-metal photodetectors that are integrated with silicon integrated circuits for multifunctional operation (decision, amplification, and computing), and the implementation of multi-input/multi-output architectures for signal processing of massive amounts of data. This research is expected to augment the body of knowledge on carrier drift characteristics in the photodetectors at low voltage bias and the physics of complex three-dimensional optoelectronic device structures. This research also explores heterogeneous integration of compound semiconductor device on silicon circuits using three-dimensional wafer bonding for photoreceiver applications.
With respect to broader impacts, the research has the potential to benefit a variety of applications, including wireless networks with reduced security vulnerabilities, networks able to transfer high-resolution digital images in the midst of equipment sensitive to electromagnetic interference in healthcare settings; and optical interconnects in large-scale data centers. A cross-university curriculum is addressed, covering multi-disciplinary topics in information theory, system-level design, analog circuits, device physics, fabrication and integration. Various programs are used to recruit students from underrepresented groups. K-12 outreach efforts are coordinated through the Tufts Center for Engineering Educational Outreach.
The ever increasing demands for seamless, real-time, integrated networks have spurred extensive research interests in wireless technology. Future generation cyber networks will rely on not just one, but complementary wireless access technologies that are secure, fast, inexpensive, and utilize complex information processing systems. Optical wireless technology is becoming an increasingly popular alternative to traditional radio frequency (RF) technology, because optical signals has virtually unlimited, unregulated bandwidth with the potential to enable data transmission in excess of 100Gigabits-per-second. To date, optical wireless systems are primarily implemented using discrete optical transmitters and receivers that are developed for guided fiber optics. These devices are bulky in size, and inadequate in performance to meet the specific needs of high-speed network applications. To reach the full potential of optical wireless technology, the technology requires that imaging receivers achieve a high level of integration, multi-GHz bandwidth per pixel, and power-efficiency. This research project has made several contributions to two important, unresolved challenges in optical wireless imaging receiver design: (1) The realization of detector arrays that possess optimal characteristics in both image receiving, which requires high fill factor, large signal-to-noise (S/N) ratio and wide field-of-view, and data receiving and processing capability at tens of gigabits bandwidth, which is beyond the performance limits of state-of-the-art CMOS and CCD imagers; (2) Silicon-based analog/digital processing circuits to enable a scalable, reconfigurable, and power-efficient architecture. The research will establish a benchmark of an integrated, high performance and low cost platform for FSO deployment. We believe that this work will make far-reaching ideas in optical MIMO theory plausible, directing the field toward a new transformative technology. This collaborative research program conducted by Tufts University in Medford, MA and Rensselaer Polytechnic Institute in Troy, NY is the first study to consider the design of imaging receivers at data rates in excess of 10-Gb/s per pixel with vast applications to interconnect and free-space optical system technology. This work has contributed to broadband wireless communication technology, including mixed-signal integrated circuits for multifunctional operations (decision, amplification, and computing), and the implementation of optical signal processing systems for massive date signal processing. At the device level, the study of the novel photodetector structures and associated asymmetric electric field distributions and nonlinearities has increased our understanding of the basic physics of complicated 3D device structures. We have also developed novel techniques for metal bonding metallurgy from Au (on InGaAs PD arrays) to Cu (on silicon microchips). The research program has also produced a generalized methodology that employs the characterization of the fundamental sensitivity limits of imaging receivers as a function of detector area, field-of-view, array size, bit rate, and limits/capabilities of silicon microelectronic technology to determine the optimal receiver architecture for various applications. The research program has inspired continuous efforts in several applications, including military and space technologies, such as focal plane technology, optical radar detection, and inter-satellite as well as satellite-to-ground communication. The enabling technologies developed through this program will have significant broader impact in a wide variety of arenas, including homeland security efforts to reduce wireless network vulnerabilities; health-care networks that transfer high-content digital imagers and process high-bandwidth applications in the midst equipment sensitive to electromagnetic interference; and ad-hoc wireless networking within computation-intensive devices.
