A fully integrated CMOS pulse transceiver with digital beam-formability for mm-wave active imaging is presented. The on-chip pulse transmitter of the transceiver includes an eight-element antenna array connected to eight pulse transmitters and a built-in relative pulse delay calibration system. The receiver employs a non-coherent detection method by using a FET direct-power detection circuit integrated with an antenna. The receiver dipole-patch antenna derives from the transmitter antenna but is modified with an on-chip DC-bias tail by shorting two arms of the dipole. The bandwidth of the receiver antenna with the DC-bias tail is designed to achieve 50.4-GHz in simulation and to cover the bandwidth of transmitter antennas. The output of the receiver antenna is connected to a resistive self-mixer followed by an on-chip low pass filter and then an amplifier stage. The built-in relative pulse delay calibration system is used to align the pulse delays of each transmitter array elements for the purpose of controlling the beam steering towards imaging objects. Both transmitter and receiver chips are fabricated in a 65-nm CMOS technology process. Measured pulse waveform of the receiver after relatively aligning all Tx's pulses is 0.91 mV (peak-peak) and 3-ns duration with a distance of 25mm between Rx and Tx. Beam steering angles are achieved in measurement by changing the digital delay code of antenna elements. Experimental results show that the proposed on-chip transceiver has an ability of digital transmitted-pulse calibration, controlling of beam-steeting, and pulse detection for active imaging applications.

A TM₀₁₀ cavity power combiner is presented, which achieves direct interface to microstrip lines via magnetic field coupling. A prototype is fabricated and its S-matrix measured. From the S-parameters we calculate that it shows less than 0.85 dB insertion loss over 250 MHz bandwidth at X-band. The return power to the input ports is less than -15 dB over this bandwidth. We verify the insertion loss estimation using S-matrix, by measuring transmission S-parameter of a concatenated 2-port divider-combiner network. Similarly analyzed is the case of performance of power combiner when one of the input fails. We find that we can achieve graceful degradation provided we ensure some particular reflection phase at the degraded port.

This paper presents an extremely low-profile front-end configuration for a base station at quasi-millimeter wave band. It consists of integrated modules of patch antennas and substrate integrated waveguide filters using two printed circuit boards, and transmitter modules using compact GaAs pHEMT three-dimensional monolithic millimeter-wave integrated circuits. The transmitter modules are located around the integrated modules. This is because the proposed front-end configuration can attain extremely low profile, and band-pass filtering performance at quasi-millimeter wave band. As a demonstration of the proposed configuration, 26-GHz-band 4-by-4 elements front-end module is fabricated and tested. The fabricated module has the thickness of about 1 cm, while that offers the attenuation of more than 30 dB with 2 GHz offset from 26 GHz. The proposed configuration can provide base station that can be effective in offering sub-millimeter wave and millimeter-wave bands broadband services for 5G mobile communications systems.

Three-dimensional integrated circuits (3D ICs) that employ through-silicon vias (TSVs) integrating multiple dies vertically have opened up the potential of highly improved circuit designs. However, various types of TSV defects may occur during the assembly process, especially the clustered TSV faults because of the winding level of thinned wafer, the surface roughness and cleanness of silicon dies,inducing TSV yield reduction greatly. To tackle this fault clustering problem, router-based and ring-based TSV redundancy architectures were previously proposed. However, these schemes either require too much area overhead or have limited reparability to tolerant clustered TSV faults. Furthermore, the repairing lengths of these schemes are too long to be ignored, leading to additional delay overhead, which may cause timing violation. In this paper, we propose a region-based TSV redundancy design to achieve relatively high reparability as well as low additional delay overhead. Simulation results show that for a given number of TSVs (8*8) and TSV failure rate (1%), our design achieves 11.27% and 20.79% reduction of delay overhead as compared with router-based design and ring-based scheme, respectively. In addition, the reparability of our proposed scheme is much better than ring-based design by 30.84%, while it is close to that of the router-based scheme. More importantly, the overall TSV yield of our design achieves 99.88%, which is slightly higher than that of both router-based method (99.53%) and ring-based design (99.00%).

In an IoT system, neural networks have the potential to perform advanced information processing in various environments. To clarify this, the robustness of a restricted Boltzmann machine (RBM) used for deep neural networks, such as a deep belief network (DBN), was studied in this paper. Even if memory or logic errors occurred in the circuit operating in the RBM while pre-training the DBN, they did not affect the identification rate of the DBN, showing the robustness of the RBM. In addition, robustness against soft errors was evaluated. The soft errors had almost no influence on the RBM unless they were as large as 10¹² times or more in the 50-nm CMOS process.