اشترك بالحزمة الذهبية واحصل على وصول غير محدود شمرا أكاديميا
تسجيل مستخدم جديدImplementing an insect brain computational circuit using III-V nanowire components in a single shared waveguide optical network
88
0
0.0
(
0
)
تأليف
David O. Winge
اسأل ChatGPT حول البحث
ﻻ يوجد ملخص باللغة العربية
Recent developments in photonics include efficient nanoscale optoelectronic components and novel methods for sub-wavelength light manipulation. Here, we explore the potential offered by such devices as a substrate for neuromorphic computing. We propose an artificial neural network in which the weighted connectivity between nodes is achieved by emitting and receiving overlapping light signals inside a shared quasi 2D waveguide. This decreases the circuit footprint by at least an order of magnitude compared to existing optical solutions. The reception, evaluation and emission of the optical signals are performed by a neuron-like node constructed from known, highly efficient III-V nanowire optoelectronics. This minimizes power consumption of the network. To demonstrate the concept, we build a computational model based on an anatomically correct, functioning model of the central-complex navigation circuit of the insect brain. We simulate in detail the optical and electronic parts required to reproduce the connectivity of the central part of this network, using experimentally derived parameters. The results are used as input in the full model and we demonstrate that the functionality is preserved. Our approach points to a general method for drastically reducing the footprint and improving power efficiency of optoelectronic neural networks, leveraging the superior speed and energy efficiency of light as a carrier of information.
قيم البحث
اقرأ أيضاً
Nanowire (NW) crystal growth via the vapour_liquid_solid mechanism is a complex dynamic process involving interactions between many atoms of various thermodynamic states. With increasing speed over the last few decades many works have reported on var
ious aspects of the growth mechanisms, both experimentally and theoretically. We will here propose a general continuum formalism for growth kinetics based on thermodynamic parameters and transition state kinetics. We use the formalism together with key elements of recent research to present a more overall treatment of III_V NW growth, which can serve as a basis to model and understand the dynamical mechanisms in terms of the basic control parameters, temperature and pressures/beam fluxes. Self-catalysed GaAs NW growth on Si substrates by molecular beam epitaxy is used as a model system.
III-V nanowires are useful platforms for studying the electronic and mechanical properties of materials at the nanometer scale. However, the costs associated with commercial nanowire growth reactors are prohibitive for most research groups. We develo
ped hot-wall and cold-wall metal organic vapor phase epitaxy (MOVPE) reactors for the growth of InAs nanowires, which both use the same gas handling system. The hot-wall reactor is based on an inexpensive quartz tube furnace and yields InAs nanowires for a narrow range of operating conditions. Improvement of crystal quality and an increase in growth run to growth run reproducibility are obtained using a homebuilt UHV cold-wall reactor with a base pressure of 2 X 10$^{-9}$ Torr. A load-lock on the UHV reactor prevents the growth chamber from being exposed to atmospheric conditions during sample transfers. Nanowires grown in the cold-wall system have a low defect density, as determined using transmission electron microscopy, and exhibit field effect gating with mobilities approaching 16,000 cm$^2$(V.s).
We report the capability to simulate in a quantum mechanical tight-binding (TB) atomistic fashion NW devices featuring several hundred to millions of atoms and diameter up to 18 nm. Such simulations go far beyond what is typically affordable with tod
ays supercomputers using a traditional real space (RS) TB Hamiltonian technique. We have employed an innovative TB mode space (MS) technique instead and demonstrate large speedup (up to 10,000x) while keeping good accuracy (error smaller than 1 percent) compared to the RS NEGF method. Such technique and capability open new avenues to explore and understand the physics of nanoscale and mesoscopic devices dominated by quantum effects. In particular, our method addresses in an unprecedented way the technological relevant case of band-to-band tunneling (BTBT) in III-V nanowire MOSFETs and broken gap heterojunction tunnel-FETs (TFETs). We demonstrate an accurate match of simulated BTBT currents to experimental measurements in a [111] InAs NW having a 12 nm diameter and a 300 nm long channel. We apply the predictivity of our TB MS simulations and report an in-depth atomistic study of the scaling potential of III-V GAA nanowire heterojunction n and pTFETs quantifying the benefits of this technology for low-power, low-voltage CMOS application. At VDD = 0.3 V and IOFF = 50 pA/um, the on-current (Ion) and energy-delay product (ETP) gain over a Si NW GAA MOSFET are 58x and 56x respectively.
We propose a hierarchical architecture for building logical Majorana zero modes using physical Majorana zero modes at the Y-junctions of a hexagonal network of semiconductor nanowires. Each Y-junction contains three physical Majoranas, which hybridiz
e when placed in close proximity, yielding a single effective Majorana mode near zero energy. The hybridization of effective Majorana modes on neighboring Y-junctions is controlled by applied gate voltages on the links of the honeycomb network. This gives rise to a tunable tight-binding model of effective Majorana modes. We show that selecting the gate voltages that generate a Kekule vortex pattern in the set of hybridization amplitudes yields an emergent logical Majorana zero mode bound to the vortex core. The position of a logical Majorana can be tuned adiabatically, textit{without} moving any of the physical Majoranas or closing any energy gaps, by programming the values of the gate voltages to change as functions of time. A nanowire network supporting multiple such logical Majorana zero modes provides a physical platform for performing adiabatic non-Abelian braiding operations in a fully controllable manner.
Electromagnetic signals in circuits consist of discrete photons, though conventional voltage sources can only generate classical fields with a coherent superposition of many different photon numbers. While these classical signals can control and meas
ure bits in a quantum computer (qubits), single photons can carry quantum information, enabling non-local quantum interactions, an important resource for scalable quantum computing. Here, we demonstrate an on-chip single photon source in a circuit quantum electrodynamics (QED) architecture, with a microwave transmission line cavity that collects the spontaneous emission of a single superconducting qubit with high efficiency. The photon source is triggered by a qubit rotation, as a photon is generated only when the qubit is excited. Tomography of both qubit and fluorescence photon shows that arbitrary qubit states can be mapped onto the photon state, demonstrating an ability to convert a stationary qubit into a flying qubit. Both the average power and voltage of the photon source are characterized to verify performance of the system. This single photon source is an important addition to a rapidly growing toolbox for quantum optics on a chip.
سجل دخول لتتمكن من نشر تعليقات
التعليقات
جاري جلب التعليقات
سجل دخول لتتمكن من متابعة معايير البحث التي قمت باختيارها
