Abstract:The Integrated Circuits (ICs) for millimeter-wave transceiver front-ends have attracted great attention in recent year, as the deployment of fifth-generation mobile communication systems (5G) enters millimeter-wave band regime (FR2) and the application of millimeter-wave radar in autonomous vechicle technology.Meanwhile, the rapid development of silicon-based device fabrication technology has greatly improved the cut-off frequency of transistors, providing a foundation for low-cost, high-performance silicon-based millimeter-wave ICs.This article summarizes recent research and development progress and trends of silicon-based millimeter-wave front-end ICs for communications and radar applications.

Abstract:This article aims to propose a center-tapped differential inductance model, which can effectively replace the electromagnetic simulation file for circuit simulation without affecting the circuit simulation results.The article applies the proposed center-tapped differential inductor model to the oscillator circuit, and compares the model results with the test results to verify the effectiveness of the inductor model.The test results show that the model can replace S-parameters to a certain extent, and keep consistent with the test results.

Abstract:Indoor positioning technology is becoming more and more widely deployed in recent years.An important technology trend for mobile communication terminals is that Ultra-WideBand (UWB) transceiver SoCs (System-on Chips) being integrated into the system to support indoor positioning.As a key building block in UWB transceivers, the Low Noise Amplifier (LNA) is of full challenges to design mainly due to the full-band (3.1-10.6 GHz) coverage, wide-band input matching and flat Noise Figure (NF) requirements.Conventional LNA topologies typically suffer from severe compromise among noise figure, gain and input matching, resulting in very limited Figure of Merit (FoM).This paper presents a two-stage LNA for UWB application.The first stage is based on transformer-feedback input matching while the second stage employs multi-functional output driver, achieving flat NF and high gain.Designed in TSMC 65 nm CMOS, simulation results show that the LNA achieves S₁₁<-10 dB, gain >17 dB, NF of 2.71±0.28 dB, and 1 dB compression point of -17.5 dBm while consuming 32.8 mW from a 1.2 V supply, corresponding to FoM-I of 2.32 and FoM-II of 0.41, respectively.

Abstract:Based on 100 nm GaN-on-Si MMIC technology, this paper presents the design of a 20-26 GHz Variable-Gain Low-Noise Amplifier (VGLNA).This LNA adopts a three-stage common source topology and realizes gain control by adjusting the gate bias of the second and third stages.The test results show that the amplifier achieves a >20 dB gain tuning range and a ±1.5 dB gain flatness within the working frequency range, and dissipates power from 126 mW to 413 mW.When biased with a -1.1 V gate voltage and 3.5 V drain voltage, the amplifier achieves a small-signal gain of more than 20 dB in the whole band, a Noise Figure (NF) ranging from 2.95 dB to 3.5 dB, and an average output 1dB compression point(OP1dB) around 14.5 dBm.The chip area is 2 mm².

Abstract:This paper proposes an improved Wilkinson power divider design scheme to solve the low port isolation and large insertion loss in the front-end power divider module of the multi-channel transceiver in wireless communication.Based on the principle of Wilkinson power divider, this design uses the 1/4 signal wavelength transmission line itself to increase the space between the power distribution ports, thereby reduces the insertion loss caused by additional transmission lines.And the HFSS electromagnetic field simulation tool is employed to obtain the wavelength of the signal in the transmission line.According to the proposed scheme, both 1 to 2 and 1 to 4 Wilkinson power dividers are designed.The simulation results show that this design can effectively reduce the insertion loss and improve the port isolation of the power divider.

Abstract:A three-stage differential common source structure is used to design a W-band power amplifier based on 65-nm CMOS technology, and two way current-mode power combining structures are used for power combining to increase output power.In order to achieve single-ended to differential conversion, impedance matching, and DC power supply at the same time, the matching network adopts a transformer structure.The simulation results show that under a supply voltage of 1 V, the power amplifier has a small signal gain of 12.7 to 15.7 dB, a 3-dB bandwidth of 84 to 104 GHz, a saturated output power of 14.6 dBm, and a peak power added efficiency of 9.7%.The power amplifier has good large signal performance, and the core area of the chip is only 0.115 mm².

Abstract:A 2.5-2.7 GHz Power Amplifier (PA) with high efficiency and low harmonic distortion in InGaP/GaAs HBT technology is designed.Multiple LC tanks are introduced into the output matching network of the presented PA to suppress the high-order harmonic distortions and improve efficiency.The simulation results show that, the PA achieves a high gain of 37.6 dB and a saturated output power of 32 dBm with Power Added Efficiency (PAE) of 36% at the saturated situation.The second and the third harmonic are both less than -60 dBc.

Abstract:This paper proposes a Transverse Electric and Magnetic field (TEM) cell that meets the IEC standard to test the electromagnetic compatibility of Integrated Circuits (IC).The TEM cell is featured with less than -12 dB of reflection coefficient and better than -2.5 dB of transmission coefficient in frequency range of 0-3 GHz.The Computer Simulation Technology (CST) software is used to simulate the impedance, S-parameters, field uniformity and the influence of Equipment Under Test (EUT) on the field of the TEM cell, moreover, the TEM cell is fabricated and compared with commercial TEM cell.In addition, an electromagnetic compatibility test board is made according to the IEC standard, which is then used to measure the IC radiation emission with the TEM cell.The analysis of the measurement results indicates some aspects to improve the electromagnetic compatibility of the integrated circuits.

Abstract:A 24-30 GHz monolithic integrated Single-Pole Double-Throw (SPDT) switch based on 100 nm Gallium Nitride (GaN) High Electron Mobility Transistor (HEMT) process is presented in this paper.This switch adopts a quarter-wavelength microstrip line with parallel HEMT devices topology.By using two-stage parallel HEMT design, the SPDT achieves greater isolation while maintaining good Insertion Loss (IL).With a control voltage of 0/-5 V, the measured IL of this SPDT is less than 1.5 dB, the isolation is greater than 28 dB, and the input power 1 dB compression point is greater than 27 dBm within 24-30 GHz 5G millimeter-wave frequency bands.The measurement results well verify the simulations.

Abstract:A double-shunt branch switch structure based on the filter optimization method is introduced to lower the insertion loss.Each branch consists of a series quarter-wavelength impedance converter and a shunt switch device.A distributed Single-Pole Double-Throw (SPDT) switch MMIC treated as a filter issue with this optimization method is implemented in a 0.15 μm GaAs pHEMT technology for the millimeter-wave applications.The chip size of the presented switch is 2.1 mm×1.1 mm.Measurement results show that the SPDT switch exhibits less than 1.3 dB of insertion loss, greater than 23dB of isolation and better than 10 dB of input and output return loss in the frequency range from 23 to 30 GHz.

Abstract:In silicon chip fabrication, each of the metal layers is required to have evenly distributed metal density for chemical mechanical polishing to ensure high yield.Filling dummy metal is a widely-employed solution to such requirement, however, the dummy metal will not only introduce significant parasitic capacitance but also induce eddy current due to electromagnetic coupling with the circuit components.Consequently, the inductive components such as inductors and transformers typically suffer from quality factor degradation after the dummy metals are filled.In this paper, a new approach for filling dummy metals is proposed.The dummy metals around the inductor are filled by metal stripes perpendicular to the current flowing direction, which can reduce the loss due to eddy current and thus alleviate the quality factor degradation.Simulation results show that the proposed method can improve the quality factor by 7.2% for the secondary coil and 17.6% for the primary coil of inductors working at 15 GHz to 70 GHz.

Abstract:A novel ultra-broadband high-gain antipodal Vivaldi antenna is designed with a bandwidth of 3.3-40 GHz frequency.The same radiation structure, which consists of one microstrip line and two exponential slot lines, is printed on both sides of the dielectric plate.In order to effectively constrain the surface current in the radiation direction, a trapezoidal dielectric substrate is added in the radiation direction, which can not only correct the gain peak offset angle of E-plane pattern, but also improve the radiation intensity of antenna.The measured results show that the VSWR (Voltage Standing Wave Ratio) is smaller than 2 at the frequency band of 3.3-40 GHz, the frequency doubling bandwidth is up to 12, and the gain is 1.0-12.6 dB.In addition, the antenna has the characteristics of good symmetry in E-plane pattern, small cross-polarization ratio, simple structure, easy processing and low cost, thus it has high research significance and application prospects in antenna with high bandwidth and high gain.

Abstract:Tourists expect to get travel experiences like low-cost, convenient itinerary, high comfort and so on.Meanwhile, they have different tourist interests such as history and culture, natural landscape, food and shopping, etc.Therefore, a personalized travel route recommendation method based on an improved shuffled frog leaping algorithm is proposed in this paper.A model is established to optimize the personalized travel route recommendation problem, and an improved shuffled frog leaping algorithm is designed based on the characteristics of the model.By adjusting the controllable accuracy, incorporating new selection criteria and handling abnormal solutions in time, the population diversity is increased and the risk of missing the optimal solution is reduced, which enhance the local search ability and searching accuracy of the algorithm.The personalized travel route recommendation for three-day tour in Nanjing is taken as an instance to verify the proposed method.Relevant data are collected, including admission fees, opening hours, and accommodation expenses for well-known tourist attractions in Nanjing.The results show that compared with basic shuffled frog leaping algorithm, the proposed method can recommend touring routes with higher accuracy and better meet the users' individual interests.

Abstract:Meteorological disasters will not only bring huge economic losses, but also cause the outbreak of public opinion and even social panic.The Internet era brings opportunities as well as challenges to the governance of public opinion.In recent years, Long Short-Term Memory network (LSTM) has shown advantages in text sentiment analysis, yet it has problems such as incomplete semantics and low accuracy in feature extracting.Convolutional Neural Network (CNN) has been introduced to make up for this shortcoming, but it still cannot considere the syntactic dependence between words.Here, we combine LSTM and Broad Learning System (BLS) with incremental learning algorithm as its core, and thus propose an LSTM-BLS text sentiment analysis model.Then the 2020 cold wave event with gale process and great temperature drop in central and eastern China is taken as an example to analyze the public sentiment tendency when sudden meteorological disasters occur.The results show that the proposed LSTM-BLS reaches a high accuracy of 97.42%, which is 17.23 and 13.46 percentage points higher than the baseline models of K-means and Support Vector Machine (SVM), respectively, and 7.13 and 4.17 percentage points higher than the existing deep learning models of LSTM and CNN-LSTM, respectively.

Abstract:To deal with the sensor failures commonly occurred in spacecraft attitude control system, we make the fault reconstruction based on the augmented P-type learning observer, and design a fault-tolerant controller for the failure system.First, the state space model of the spacecraft attitude control system is established considering both sensor faults and unknown interference inputs.The fault is really the equivalent of two related terms, which are the fault estimate at the previous moment and the output error at the current moment.The observer is designed using Lyapunov theorem, and the LMI (Linear Matrix Inequality) is utilized to give a systematic design of the observer parameter matrix.Then, the integral sliding mode surface is designed, and the equivalent control law obtained is substituted into the system state space model, and then a fault-tolerant control law including adaptive fault compensation control term and linear output feedback control term is established.Finally, the effectiveness of the designed observer and fault-tolerant controller is verified by a simulation example.

Abstract:Most automatic weather stations sample visibility with optical devices, which are vulnerable to interference from rain, snow and dust, resulting in the inaccuracy or missing of visibility data.To address this and provide complete data for meteorological prediction, this paper proposes a Support Vector Machine-Back Propagation (SVM-BP) neural network based data quality control method for visibility data correction and filling.First, the grey correlation analysis is used to select meteorological elements closely related with visibility.Second, the visibility data of different terrains are estimated by SVM and the BP neural network independently, which are then combined by optimal combination weights.Historical weather visibility data from Anhui Meteorological Bureau are used to verify the proposed method.The results show that compared with the independent SVM or the BP neural network, the combined estimation has smaller mean error and higher overall accuracy.The proposed SVM-BP neural network method provides an effective tool for visibility data filling, thus lays theoretical basis for weather forecasting, weather analysis, meteorological research and public service.

Abstract:The dual-channel bandwidth allocation is studied considering estimation performance.The networked control system is modeled as a Markov jump delay one in order to reduce the influence of time delay on signal transmission.The dual-channel transmission strategy is adopted to improve the data arrival rate and reduce the packet loss.The limited bandwidth is flexibly allocated to two channels to fully exploit the channel capacity.Then a dual-channel quantitative transmission scheme is proposed and a state estimator is designed.The sufficient condition is derived for the time-varying error estimation system to satisfy the H_∞ performance, and the gain matrix is given for the estimator.Finally, the effectiveness of the proposed strategy is illustrated by a numeral example.