Design of a Temporal Learning Chip for Signal Generation and Classification

The design and analog VLSI implementation of a recurrent neural network with integrated temporal learning is presented. The learning algorithm is forward in time, and is implemented strictly as instantaneous, local weight updates. PSpice simulations of networks with 4 to 6 neurons demonstrate robust learning of trajectory generation and classification tasks. A scalable 2-D VLSI architecture is described and a prototupe 4-neuron recurrent neural network with learning has subsequently been fabricated in MOSIS TinyChip 2 micron technology. Experimental results of the chip validate the learning performance with convergence in the millisecond range. Specific experimental results of learning circular and figure-8 dynamic trajectories are included.     

Polymeric waveguide polarization splitter with a buriedbirefringent polymer

A waveguide polarization splitter is demonstrated based on a low-loss polymer waveguide and a birefringent polyimide. Crosslinkable fluorinated polymers with an excellent stability and a low absorption loss are utilized for the device. The polyimide is buried under one branch of the Y-branch waveguide to enhance the birefringence between the TE and TM modes. By the adiabatic mode evolution, the TE mode is coupled to the branch with the polyimide strip, while the TM mode propagates through the other branch without the polyimide. For the device with a branch angle of 1/400 rad, we obtained a crosstalk less than -20 dB and a fiber-to-fiber insertion loss of 3.8 dB     

Threshold voltage dependence of LOCOS-isolated thin-film SOINMOSFET on buried oxide thickness

Effects of buried oxide thickness on short-channel effect of LOCOS-isolated thin-film SOI n-MOSFETs have been investigated. Devices fabricated on SOI substrate with thin (100 nm) buried oxide have smaller roll-off of threshold voltage than those fabricated on SOI substrate with thick (400 nm) buried oxide. This is caused by a different boron concentration at the silicon film that results from the difference of stress with the buried oxide thickness. In the case of thin buried oxide, higher volumetric expansion of the field oxide causes higher stress at the interface between the silicon film and the surrounding oxide, including field and buried oxide, which prevents boron atoms from diffusing beyond the interface     

Multiaxial and Transparent Strain Sensors Based on Synergetically Reinforced and Orthogonally Cracked Hetero‐Nanocrystal Solids

Wearable strain sensors are widely researched as core components in electronic skin. However, their limited capability of detecting only a single axial strain, and their low sensitivity, stability, opacity, and high production costs hinder their use in advanced applications. Herein, multiaxially highly sensitive, optically transparent, chemically stable, and solution‐processed strain sensors are demonstrated. Transparent indium tin oxide and zinc oxide nanocrystals serve as metallic and insulating components in a metal–insulator matrix and as active materials for strain gauges. Synergetic sensitivity‐ and stability‐reinforcing agents are developed using a transparent SU‐8 polymer to enhance the sensitivity and encapsulate the devices, elevating the gauge factor up to over 3000 by blocking the reconnection of cracks caused by the Poisson effect. Cross‐shaped patterns with an orthogonal crack strategy are developed to detect a complex multiaxial strain, efficiently distinguishing strains applied in various directions with high sensitivity and selectivity. Finally, all‐transparent wearable strain sensors with Ag nanowire electrodes are fabricated using an all‐solution process, which effectively measure not only the human motion or emotion, but also the multiaxial strains occurring during human motion in real time. The strategies can provide a pathway to realize cost‐effective and high‐performance wearable sensors for advanced applications such as bio‐integrated devices.     

Tuning Interfacial Properties by Spontaneously Generated Organic Interlayers in Top‐Contact‐Structured Organic Transistors

Controlling the interfacial properties between the electrode and active layer in organic field‐effect transistors (OFETs) can significantly affect their contact properties, resulting in improvements in device performance. However, it is difficult to apply to top‐contact‐structured OFETs (one of the most useful device structures) because of serious damage to the organic active layer by exposing solvent. Here, a spontaneously controlled approach is explored for optimizing the interface between the top‐contacted source/drain electrode and the polymer active layer to improve the contact resistance (R_C). To achieve this goal, a small amount of interface‐functionalizing species is blended with the p‐type polymer semiconductor and functionalized at the interface region at once through a thermal process. The R_C values dramatically decrease after introduction of the interfacial functionalization to 15.9 kΩ cm, compared to the 113.4 kΩ cm for the pristine case. In addition, the average field‐effect mobilities of the OFET devices increase more than three times, to a maximum value of 0.25 cm² V^?1 s^?1 compared to the pristine case (0.041 cm² V^?1 s^?1), and the threshold voltages also converge to zero. This study overcomes all the shortcomings observed in the existing results related to controlling the interface of top‐contact OFETs by solving the discomfort of the interface optimization process.     

Design of 24 GHz Radar with Subspace‐Based Digital Beam Forming for ACC Stop‐and‐Go System

For an adaptive cruise control (ACC) stop‐and‐go system in automotive applications, three radar sensors are needed because two 24 GHz short range radars are used for object detection in an adjacent lane, and one 77 GHz long‐range radar is used for object detection in the center lane. In this letter, we propose a single sensor‐based 24 GHz radar with a detection capability of up to 150 m and ±30° for an ACC stop‐and‐go system. The developed radar is highly integrated with a high gain patch antenna, four channel receivers with GaAs RF ICs, and back‐end processing board with subspace based digital beam forming algorithm.     

A Single Droplet‐Printed Double‐Side Universal Soft Electronic Platform for Highly Integrated Stretchable Hybrid Electronics

Soft features in electronic devices have provided an opportunity of gleaning a wide spectrum of intimate biosignals. Lack of data processing tools in a soft form, however, proclaims the need of bulky wires or low‐performance near‐field communication externally linked to a “rigid” processor board, thus tarnishing the true meaning of “soft” electronics. Furthermore, although of rising interest in stretchable hybrid electronics, lack of consideration in multilayer, miniaturized design and system‐level data computing limits their practical use. The results presented here form the basis of fully printable, system‐level soft electronics for practical data processing and computing with advanced capabilities of universal circuit design and multilayer device integration into a single platform. Single droplet printing‐based integration of rigid islands and core–shell vertical interconnect access (via) into a common soft matrix with a symmetric arrangement leads to a double‐side universal soft electronic platform that features site‐selective, simultaneous double‐side strain isolation, and vertical interconnection, respectively. Systematic studies of island‐morphology engineering, surface‐strain mapping, and electrical analysis of the platform propose optimized designs. Commensurate with the universal layout, a complete example of double‐side integrated, stretchable 1 MHz binary decoders comprised of 36 logic gates interacting with 9 vias is demonstrated by printing‐based, double‐side electronic functionalization.     

Fine‐Grained FSMD Power Gating Considering Power Overhead

As a fine‐grained power gating method for achieving greater power savings, our approach takes advantage of the finite state machine with a datapath (FSMD) characteristic which shows sequential idleness among subcircuits. In an FSMD‐based power gating, while only an active subcircuit is expected to be turned on, more subcircuits should be activated due to the power overhead. To reduce the number of missed opportunities for power savings, we deactivated some of the turned‐on subcircuits by slowing the FSMD down and predicting its behavior. Our microprocessor experiments showed that the power savings are close to the upper bound.     

Low Complexity Decoder Architecture for Low-Density Parity-Check Codes

In this paper, we propose a low complexity decoder architecture for low-density parity-check (LDPC) codes using a variable quantization scheme as well as an efficient highly-parallel decoding scheme. In the sum-product algorithm for decoding LDPC codes, the finite precision implementations have an important tradeoff between decoding performance and hardware complexity caused by two dominant area-consuming factors: one is the memory for updated messages storage and the other is the look-up table (LUT) for implementation of the nonlinear function Ψ(x). The proposed variable quantization schemes offer a large reduction in the hardware complexities for LUT and memory. Also, an efficient highly-parallel decoder architecture for quasi-cyclic (QC) LDPC codes can be implemented with the reduced hardware complexity by using the partially block overlapped decoding scheme and the minimized power consumption by reducing the total number of memory accesses for updated messages. For (3, 6) QC LDPC codes, our proposed schemes in implementing the highly-parallel decoder architecture offer a great reduction of implementation area by 33% for memory area and approximately by 28% for the check node unit and variable node unit computation units without significant performance degradation. Also, the memory accesses are reduced by 20%.     

Functionally Antagonistic Hybrid Electrode with Hollow Tubular Graphene Mesh and Nitrogen‐Doped Crumpled Graphene for High‐Performance Ionic Soft Actuators

Ionic soft actuators, which exhibit large mechanical deformations under low electrical stimuli, are attracting attention in recent years with the advent of soft and wearable electronics. However, a key challenge for making high‐performance ionic soft actuators with large bending deformation and fast actuation speed is to develop a stretchable and flexible electrode having high electrical conductivity and electrochemical capacitance. Here, a functionally antagonistic hybrid electrode with hollow tubular graphene meshes and nitrogen‐doped crumpled graphene is newly reported for superior ionic soft actuators. Three‐dimensional network of hollow tubular graphene mesh provides high electrical conductivity and mechanically resilient functionality on whole electrode domain. On the contrary, nitrogen‐doped wrinkled graphene supplies ultrahigh capacitance and stretchability, which are indispensably required for improving electrochemical activity in ionic soft actuators. Present results show that the functionally antagonistic hybrid electrode greatly enhances the actuation performances of ionic soft actuators, resulting in much larger bending deformation up to 620%, ten times faster rise time and much lower phase delay in a broad range of input frequencies. This outstanding enhancement mostly attributes to exceptional properties and synergistic effects between hollow tubular graphene mesh and nitrogen‐doped crumpled graphene, which have functionally antagonistic roles in charge transfer and charge injection, respectively.