Exfo Electro-Optical Engineering (Canada)

This study utilized metal–organic chemical vapor deposition technology to grow β-Ga2O3 epitaxial layers on sapphire substrates and fabricate lateral β-Ga2O3 field-effect transistors (FETs). To enhance the performance of these FETs, a β-Ga2O3/(Al0.33Ga0.67)2O3 structure was incorporated. The resulting FETs demonstrated high on-state current (Ion), low on-state resistance (Ron), and a high on–off ratio of drain current, indicating excellent two-dimensional electron gas (2DEG) characteristic offered by β-Ga2O3/(Al0.33Ga0.67)2O3 structure. By optimization of the thickness and doping concentration of β-Ga2O3, high-performance 2DEG β-Ga2O3 FETs were achieved, which satisfied the stringent requirements for power devices. Notably, the heavily doped β-Ga2O3/(Al0.33Ga0.67)2O3 structure led to a significant improvement in performance, with the saturation current (ID,sat) increasing from 0.16 to 11.4 mA/mm, a remarkable 6925% increase, and the breakdown voltage rising from 331 to 687 V, an increase of 107.25%. The relative performance of the device with and without AlGaO was also simulated. These enhancements were primarily due to the effective reduction of leakage current density and 2DEG formation by the β-Ga2O3/(Al0.33Ga0.67)2O3 layer, which was grown by the importance of precise growth parameters and structural design in advancing the performance of β-Ga2O3 FETs, making them highly promising for high-power applications.

We perform diverse experimental validation of an optical spectrum analysis GOSNR estimation method, showing good agreement (typically within 0.5 dB) relative to the conventional transceiver-based method over metro and long-haul distances and two fiber types.

Channel modeling is considered as a fundamental step in the design, deployment, and optimization of vehicular wireless communication systems. For typical vehicular communication scenarios in urban areas, dense multipath may exist in the wireless channels. The propagation graph (PG) method is an efficient approach to simulate multipath radio propagation. In this paper, we extend the PG method into a Graph Neural Network (GNN) enabled data-driven method for calculating channel transfer function (CTF) and channel impulse response (CIR) in a given space. ChebNet, a classical GNN, is utilized for estimating the scattering coefficients of the edge gains in the PG method. The proposed GNN-enabled method performs better than baseline algorithms, such as multilayer perceptron (MLP), simulated annealing (SA) algorithm, and genetic algorithm (GA) in effectively estimating a large number of scattering coefficients in PG. Mean absolute errors of the proposed method are provided and evaluated in this paper. Additionally, the potential future research directions of the GNN-enabled PG method for channel modeling are discussed.

We are delighted to introduce the third part of the Special Section on Intelligent Autonomous Transportation Systems with 6G, which aims to provide the scientific community with a comprehensive overview of innovative technologies, advanced architectures, and potential challenges for the 6G-supported Intelligent Autonomous Transport Systems. Twenty articles were selected for publication in this issue. All the articles were rigorously evaluated according to the standard reviewing process of the IEEE Transactions on Intelligent Transportation Systems. The evaluation process considered originality, technical quality, presentational quality, and overall contribution. We will introduce these articles and highlight their main contributions in the following.

The evolution of deep learning has promoted the popularization of smart devices. However, due to the insufficient development of computing hardware, the ability to conduct local training on smart devices is greatly restricted, and it is usually necessary to deploy ready-made models. This opacity makes smart devices vulnerable to deep learning backdoor attacks. Some existing countermeasures against backdoor attacks are based on the attacker’s ignorance of defense. Once the attacker knows the defense mechanism, he can easily overturn it. In this paper, we propose a Trojaning attack defense framework based on moving target defense(MTD) strategy. According to the analysis of attack-defense game types and confrontation process, the moving target defense model based on signaling game was constructed. The simulation results show that in most cases, our technology can greatly increase the attack cost of the attacker, thereby ensuring the availability of Deep Neural Networks(DNN) and protecting it from Trojaning attacks.

The papers in this special section examine vehicle connectivity and automation via 5G mobile communication networks. Autonomous vehicles are reshaping the future of transportation through tight integration between automotive and communication echostreams. Undoubtedly, smarter cars employing powerful computational and robotic platforms will require new phases of connectivity with surroundings. Therefore, vehicular communication is evolving from current models of serving passengers with entertaining contents over cellular communications to providing critical information that supports self-driving cars. The new connectivity models support gigabit downloading speeds of reliable communications at ultra-low latencies that provide the necessary information for autonomous car navigation, emergency communication, monitoring of vehicle machinery status, etc. This evolving paradigm still requires definitions considering supporting technologies, deployment scenarios, and performance indicators.

The articles in this special section addresses copper technologies and deployment practices beyond those currently available. Legacy copper infrastructure originally designed to provide voice services (twisted pair network) or television services (coaxial network) has performed vastly beyond specification and continues to deliver increased broadband speeds. Today’s mature technologies like vectored VDSL2 and G.fast for twisted pair and DOCSIS 3.1 for coaxial networks, are able to surpass present-day demand for broadband speeds in excess of several hundreds Mb/s per end user. The ability to off er gigabit speeds has been fueled by advances in digital signal processing and by a steady migration toward fiber rich access networks where only the last “mile” into the homes remains copper based (FTTx). We often get the question why the industry continues to consider copper access, while fiber is perceived as superior in terms of transmission properties.

We propose a tunnel-injection structure, in which WS2 quantum dots (QDs) act as the injector and InGaN/GaN quantum wells (QWs) act as the light emitters. Such a structure with different barrier thicknesses has been characterized using steady-state and time-resolved photoluminescence (PL). A simultaneous enhancement of the PL intensity and PL decay time for the InGaN QW were observed after transfer of charge carriers from the WS2-QD injector to the InGaN-QW emitter. The tunneling time has been extracted from the time-resolved PL, which increases as the barrier thickness is increased. The dependence of the tunneling time on the barrier thickness is in good agreement with the prediction of the semiclassical Wentzel–Kramers–Brillouin model, confirming the mechanism of the tunnel injection between WS2 QDs and InGaN QWs.

We discuss in-service OSNR measurement options independent of coherent receiver metrics. We review a spectral analysis technique for discriminating ASE when fiber nonlinearities contribute “Gaussian-like” noise and evaluate its performance in practical network scenarios.

A novel in-service OSNR monitoring technique is experimentally validated on tightly filtered polarization-multiplexed signals for a range of network operating conditions and reconfigurations. The results demonstrate sufficient accuracy and robustness for measurements in practical networks.