In this paper, we have investigated the integration process at room temperature and device characteristics of one transistor type nonvolatile memory with organic semiconductor field-effect transistor (OFET) integrated with resistive random access memory (ReRAM). The threshold voltage (V_TH) of pentacene-based OFET with LaB_xN_y gate insulator is controlled by the ReRAM characteristics of LaB_xN_y gate insulator. The bottom-gate type pentacene-based OFET was fabricated on SiO₂/Si(100) substrate. The nitrogen-doped LaB₆ bottom gate electrode was deposited by RF sputtering and patterned. Then, LaB_xN_y gate insulator was deposited by the RF sputtering followed by the pentacene and Au source and drain electrode deposition by the evaporation. The Set/Reset operations of ReRAM were confirmed by the drain voltage sweep of ±2 V. Furthermore, V_TH shift of -0.9 V was observed by the Set operation of ReRAM so that the nonvolatile memory characteristics were realized for the 1 transistor type ReRAM/OFET.

In this work, an 8-bit signed approximate adder (SAA) and a quantization- and bit-pruning-aware training method (QBAT) are proposed to reduce the substantial area and power consumption caused by adder trees for digital Computation-in-Memory (DCiM). QBAT achieves efficient bit pruning and minimizes the accuracy loss caused by the approximation. The SAA reduces area and power consumption by 20% and power-delay product (PDP) by 36.7%. With QBAT, the proposed design achieves 95.5% and 96.4% inference accuracy for Resnet-18 and Resnet-50 models on the CIFAR-10 dataset.

This paper presents topology optimization of microstrip lines using twin deep neural networks (DNNs) for prediction of scattering parameters and its accuracy evaluation. Topology optimization can be accelerated by using a DNN that acts as a surrogate model for time-consuming EM simulations. However, if the prediction accuracy of the DNN for performance prediction is not high enough, the optimization will fail due to misleading caused by prediction errors. To reduce the risk of optimization failure, the present method introduced an additional DNN to evaluate the accuracy of the performance prediction. The proposed method is shown to be effective in avoiding misleading and speeding up the optimization process through numerical and experimental results.

We propose a configuration for Brillouin optical correlation-domain reflectometry (BOCDR) in which the electrical signal processing unit is placed remotely and connected to the optical measurement unit via a long-distance optical fiber. This arrangement eliminates the need for bulky electrical devices at the measurement site. We investigate the effect of changing the connecting fiber length and demonstrate that Brillouin frequency shift distributions can be obtained with separations of up to approximately 20 km under the tested conditions. We also show the importance of compensating for propagation losses when using long-distance fibers.

In a 100VDC/10A resistive circuit, break arcs are generated between electrical contacts. Silver electrical contacts with airflow ejection structure are separated at a constant speed. The duration and motion of the break arcs are investigated for different contact shapes. Three types of contact shapes are used. Contact diameters and airflow hole diameters are changed. Following results are shown. An effective shape of the contacts was found to shorten the arc duration. The arc duration was shortened by decreasing the diameter of the contacts and increasing the diameter of airflow hole for the airflow ejection. The results are discussed in terms of the motion characteristics of the cathode spots on the cathode surface and the shapes of the break arcs.