Fujian Star-net (China)

A three-dimensional (3D) integrated waveguide filtering power amplifier (PA) is introduced in this paper. An augmented <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${N} +4$</tex-math> </inline-formula> coupling matrix that incorporates the harmonic control (HC) network and the transistor is synthesized to characterize the filtering PA circuits. Due to the introduced HC node, the characteristic polynomials and coupling matrix are resynthesized. By leveraging the coupling matrix, the PA circuits can achieve a standard Chebyshev filtering response at the fundamental frequencies while simultaneously introducing arbitrary transmission zeros (TZs) at harmonic frequencies with precisely controlled phase. This technique greatly expands the flexibility and practicality of waveguide PAs, particularly because harmonic manipulation plays a crucial role in determining efficiency and saturated power. The performances of return loss and gain can be predicted by the coupling matrix. Furthermore, this technique facilitates seamless integration between on-chip transistors and 3D high-Q resonators, as well as HC circuits. Consequently, the need for additional planar matching circuits is eliminated, resulting reduced losses, improved efficiency, and a more compact system architecture. The technique is validated by using the 0.25-<inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\mu $</tex-math> </inline-formula>m GaN HEMTs microwave integrated circuits (MICs) at 3.3–3.5 GHz. The measured drain efficiency is from 62.8% to 67.7% at the saturated power of 38.8–39.1 dBm, and the adjacent channel leakage power ratios (ACLR) are −34.8/−33.5 dBc. The synthesis approach could boost progress of the PAs in high-frequency and high-power scenarios, where 3D waveguide circuit is preferred for its intrinsic low-loss and high-power capacity.

In this letter, a filtering Doherty power amplifier (DPA) monolithic microwave integrated circuit (MMIC) with an on-chip multifunctional power divider (PD) is presented. The PD is implemented at the interstage of the DPA, achieving harmonic control, filtering, and phase compensation functions simultaneously. By integrating the filter into the interstage, separate functional circuits can be eliminated, as well as achieving improved efficiency and minimized chip area. A two-stage DPA MMIC operating at the 2.6–3.8 GHz is realized using 0.25-<inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\mu $ </tex-math></inline-formula>m gallium nitride (GaN) high-electron mobility transistor (HEMT) technology. The fabricated device exhibits small-signal filtering characteristics and a flat gain response, achieving a saturated output power of 33.1–34.1 dBm with 52.8%–55.1% drain efficiency (DE). The DE maintains 40%–45% at 6-dB output back-off (OBO), while linearity measurements with 64-quadrature amplitude modulation (QAM) signals show performance with <2.76% error vector magnitude (EVM) and −32 dBc adjacent channel leakage ratio (ACLR).

This article presents a compact mixer-based true power detector integrated within a Ka-band power amplifier (PA). By performing capacitive voltage sensing and inductive current sensing at the primary coil, the compact power detector eliminates the need for additional area-consuming passive components while still allowing the PA to maintain a single-ended antenna output. The control of magnetic coupling with the secondary coil is achieved by exploiting magnetic flux cancellation rather than increasing the physical distance between coils. The active mixer multiplying the voltage and current uses current bleeding technique that simultaneously suppresses the flicker noise and enhances linearity, leading to a state-of-art dynamic range. Without phase shifters or resistor termination, the overall power detector achieves intrinsic phase compensation by matching the delays between the voltage and current sensing paths. Fabricated in a 40-nm CMOS process, the power detector occupies a core area of <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$3520~\mu $ </tex-math></inline-formula>m2 and covers a detectable power range from −18 to +15 dBm across 25–31 GHz with 12.1-mW power consumption.

This article presents RF and microwave power amplifiers (PAs) that utilize hetero-dielectric (HD) lumped elements in matching circuits. The proposed innovative model achieves a higher <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$Q$</tex-math> </inline-formula>-factor due to its unique electromagnetic (EM) properties compared with conventional on-chip (OC) lumped elements and microstrip circuits. For the first time, formulas for calculating the parameters of the equivalent lumped elements of the model are given. This HD lumped element with a high <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$Q$</tex-math> </inline-formula>-factor reduces the matching circuit loss and further improves the performance of the PA. The design of the PA based on an HD model has been theoretically explored and experimentally validated using 0.25-<inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\mu $</tex-math> </inline-formula>m GaN HEMTs microwave integrated circuits (MICs) at 2.8–3.6 GHz. The drain efficiency (DE) ranges from 53.6% to 63.2% at an output power of 33.5–35.3 dBm, and the adjacent channel leakage power ratios (ACLRs) are −38.1dBc. High <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$Q$</tex-math> </inline-formula>-factor passive elements based on this model are expected to play an important role in realizing OC low-insertion loss filters and high-efficiency PAs.

We propose and demonstrate a microwave photonic distributed radar system for target positioning based on joint polarization and sideband multiplexing to enable four radar transceivers to share a single optical wavelength. The system consists of an indoor unit and an outdoor unit with four transceivers. The indoor unit is responsible for the generation of a linearly frequency-modulated (LFM) optical waveform which is split and sent via optical fibers to the four individual radar transceivers at the outdoor unit. After photodetection at the transceivers, four LFM microwave waveforms are generated and transmitted. When reflected by a target, four echo signals are received which are polarization and sideband multiplexed and transmitted back to the indoor unit. At the indoor unit, the multiplexed optical signal is demultiplexed and undergoes a photonic de-chirp process. The time delays are estimated by measuring the frequencies of the de-chirped signals, which are then used for target positioning. The proposed radar system is validated through a 3-D spatial target positioning experiment. The results show that the system can locate a target with a spatial positioning accuracy within ±3 cm.

This article studies a mobile edge computing (MEC) with one edge node (EN), where multiple unmanned aerial vehicles (UAVs) act as users which have some heavy tasks. As the users generally have limitations in both calculating and power supply, the EN can help calculate the tasks and meanwhile supply the power to the users through energy harvesting. We optimize the system by proposing a joint strategy to unpacking and energy harvesting. Specifically, a deep reinforcement learning (DRL) algorithm is implemented to provide a solution to the unpacking, while several analytical solutions are given to the power allocation of energy harvesting among multiple users. In particular, criterion I is the equivalent power allocation, criterion II is designed through equal data rate, while criterion III is based on the equivalent transmission delay. We finally give some results to verify the joint strategy for the UAV‐aided multiuser MEC system with energy harvesting.

In this paper, a miniaturized and broadband dual-band millimeter-wave filtering antenna is proposed. Four pairs of shorting pins are introduced into the four quadrants of the proposed antenna to realize the miniaturization of the antenna and generate a radiation null in the stopband. A pair of shorting pins are fabricated on both sides of the probe to generate the other two radiation nulls outside upper band. The proposed antenna has dual operation bands of 25&#x2212;30 GHz and 37&#x2212;45.5 GHz for |S11| of less than &#x2212;10 dB. Relying on these superior performances, the proposed antenna has great advantages in 5G phased array applications.

Low drop-out (LDO) voltage regulators are widely used in many portable applications, such as cell phones, laptops, and earplugs. However, the design and simulation of the circuit takes a long time. Hence, the behavioral model for an LDO regulator is greatly needed. In this paper, the hybrid modeling method has been illustrated for generating a behavioral model for an LDO regulator circuit. The hybrid modeling method integrates a table-based modeling method and a circuit simplification method for generating an LDO linear regulator behavioral model. Based on research of the automation topologies and algorithms a behavioral model generation tool for an LDO linear regulator, LDOCad, is also described, including the software architecture and different modules of LDOCad. This tool provides the modeler, or circuit designer, with an approach to quickly and automatically generate a behavioral model for an LDO linear regulator. When the netlist processing algorithm, the table generation algorithm, and the modeled node and topology device extraction algorithm are executed, the hybrid model can be generated through the model topology formulator. The algorithms are generic and robust. The models tested demonstrate an accurate match with the performance of the original circuits and achieve from a 5 to 19 times speed improvement in simulations.

Kawasaki disease is the leading cause of pediatric acquired heart disease. Coronary artery abnormalities are the main complication of Kawasaki disease. Kawasaki disease patients with intravenous immunoglobulin resistance are at a greater risk of developing coronary artery abnormalities. Several scoring models have been established to predict resistance to intravenous immunoglobulin, but clinicians usually do not apply those models in patients because of their poor performance. To find a better model, we retrospectively collected data including 753 observations and 82 variables. A total of 644 observations were included in the analysis, and 124 of the patients observed were intravenous immunoglobulin resistant (19.25%). We considered 7 different linear and nonlinear machine learning algorithms, including logistic regression (L1 and L1 regularized), decision tree, random forest, AdaBoost, gradient boosting machine (GBM), and lightGBM, to predict the class of intravenous immunoglobulin resistance (binary classification). Data from patients who were discharged before Sep 2018 were included in the training set (n = 497), while all the data collected after 9/1/2018 were included in the test set (n = 147). We used the area under the ROC curve, accuracy, sensitivity, and specificity to evaluate the performances of each model. The gradient GBM had the best performance (area under the ROC curve 0.7423, accuracy 0.8844, sensitivity 0.3043, specificity 0.9919). Additionally, the feature importance was evaluated with SHapley Additive exPlanation (SHAP) values, and the clinical utility was assessed with decision curve analysis. We also compared our model with the Kobayashi score, Egami score, Formosa score and Kawamura score. Our machine learning model outperformed all of the aforementioned four scoring models. Our study demonstrates a novel and robust machine learning method to predict intravenous immunoglobulin resistance in Kawasaki disease patients. We believe this approach could be implemented in an electronic health record system as a form of clinical decision support in the near future.

A service-oriented architecture (SOA)-based inter-domain optical virtual private network (OVPN) service architecture is proposed for coordinated scheduling of distributed computing. Different OVPN partition scenarios across two GMPLS domains were experimentally demonstrated in the 3TNet testbed.