Wiley

Electronics Letters

Published by Wiley and The Institution of Engineering and Technology

Online ISSN: 1350-911X

·

Print ISSN: 0013-5194

Disciplines: General & introductory electrical & electronics engineering

Journal websiteAuthor guidelines

Top read articles

64 reads in the past 30 days

Large language model hypothesis architecture
The large language model for telecom operators' customer service
Fass index generation prompt principle
Reinforcement learning model training
LLM+LangChain technology roadmap

+3

Design of a large language model for improving customer service in telecom operators

May 2024

·

230 Reads

·

2 Citations

Ma Xiaoliang

·

Zhao RuQiang

·

Liu Ying

·

[...]

·

Du Dequan
Download

Aims and scope


Electronics Letters is a premier destination for the best research across all electronic and electrical engineering related fields for over 50 years. We are a fully open access rapid-communication journal which is highly read and cited worldwide.

Recent articles


Superlattice (SL) PFET process flows and schematics. (a) Fabrication flows for Ω‐channel SL PFETs and conventional SL PFETs (Ref. device). (b) Epitaxial Si0.7Ge0.3/Si stacks. (c) α‐Si dummy gate formation. (d) Omega‐shaped channel formation. (e) High‐k/metal‐gate (HKMG) deposition and metallization
Results of wet etching using tetramethylammonium hydroxide (TMAH) on stacked SiGe/Si layers at room temperature. (a) Scanning electron microscope (SEM) image after12.5% TMAH wet etching for 10 min. (b) SEM image after 25% TMAH wet etching for 10 min. (c) Si etching depths at different TMAH treatment times. (d) Etch selectivity of Si and SiGe at different TMAH concentrations
Transmission electron microscope and scanning electron microscoe images of superlattice (SL) fin field‐effect transistors (FinFETs). (a) Structures with different Lg values after gate formation. (b) Ω‐channel SL FinFET structure within the gate. (c) Ref. device structure within the gate. (d) Ω‐channel SL PFET structure cut across the fin direction
Electrical characterization of Ω‐channel superlattice p‐type fin field‐effect transistor versus Ref. device. (a) IDS—VGS curves at Vd = −0.9 V. (b) Variation in SSlin at different values of Lg. (c) Vth characteristics at Lg = 100 nm. (d) Gm characteristics for Ω‐channel PFET and reference device at Lg = 40 nm. (e) Total parasitic resistance (RTOT) data and linear fitting curves for the devices. (f) Ion and Ioff as functions of Lg
Low‐temperature SOI SiGe/Si superlattice FinFET with omega‐shaped channel and self‐allied silicide for 3D sequential IC
  • Article
  • Full-text available

November 2024

Xu‐Lei Qin

·

Guan‐Qiao Sang

·

Lei Cao

·

[...]

·

Hua‐Xiang Yin

In this letter, to improve the performance and reduce leakage currents of bulk low‐temperature multi‐layer SiGe/Si superlattice (SL) fin field‐effect transistors (FinFETs), a p‐type omega‐shaped channel (Ω‐channel) SL FinFET is realized by etching a Si/Si0.7Ge0.3 triple‐layer stacked structure in a replacement metal gate (RMG) module on a silicon‐on‐insulator (SOI) substrate. In addition, a self‐allied Ni0.9Pt0.1 silicide process and a low‐thermal‐budget (≤400°C) integration procedure were performed on the Ω‐channel SL FinFET. Test results demonstrate that the on‐state current (Ion) is increased by 5.5 times (from 78 to 429 µA/µm) and the off‐state current (Ioff) is reduced by 78.8% (from 5.2 × 10⁻³ to 1.1 × 10⁻³ µA/µm) when compared with the corresponding currents of traditional bulk SL FinFETs.


(a) A portion of a composite slab featuring an infinite planar array of randomly oriented helices embedded in concrete as the host medium. (b) The random orientation of the helix axis unit vector in one unit cell, along with the random variables φ and θ. In our example, R = 12 mm, S = 6 mm, helix turns = 1.75, unit cell size = 25 × 25 mm², and d = 30 mm
(a–c) Nine neighbouring unit cells of each infinite periodic planar array embedded in the concrete host medium (the blue cube) corresponding to the x, y, and z‐axis basis‐like functions. (d–f) The basis‐like functions Bx(f), By(f), and Bz(f), which represent the shielding effectiveness of the arrays shown in (a–c), respectively
(a) Comparison between the results of the proposed method and full‐wave simulations of six distinct arrays of randomly oriented helices. (b) Blue (the left ruler): RMS error between the results of the proposed method and the results of the six different full‐wave simulations. Green (the right ruler): Effective electrical conductivity of the composite
(a) A helical steel particle. (b) The planar array of randomly oriented helices embedded in the concrete paste in the mould during manufacturing. (c) The reinforced (with steel additives) and the control (without steel additives) samples after the drying process. (d) The test setup. (e) The iron shelter
A comparison between the results of the proposed method and the experimental measurement. This figure illustrates the difference in shielding effectiveness between the reinforced sample (containing the array of randomly oriented steel helices) and the control sample (without any steel additives) for both the proposed method and the experimental measurement
Stochastic approach to evaluate the shielding effectiveness of composite materials loaded with randomly oriented helices

November 2024

·

3 Reads

This letter proposes a method to calculate the expected value of the shielding effectiveness of a composite material slab, including a planar array of randomly oriented metal helices as chiral particles. Helix‐loaded concrete composite materials are advantageous for electromagnetic interference shielding applications. While planar periodic arrays of chiral particles have been extensively studied, this research discusses a non‐periodic array in which the helical particle axis direction in each unit cell is random. This method estimates the slab's shielding effectiveness by defining it as a random variable expanded into three independent basis‐like functions. Comparisons with full‐wave simulations show acceptable agreement across a wide frequency range, with a total RMS error of 0.9 dB. This study additionally includes an experimental measurement to validate the proposed method. This technique provides the initial step in developing an analytical method to analyse and estimate the transmitted power of general structures using a stochastic approach. Examples include concrete composites containing randomly distributed helical steel particles with significant shielding effectiveness.


Components of the computational imaging system aided by RMS
Schematic of composite resolution. δ=max{δr,δχ}$\delta = \text{max}\lbrace \delta _{\text{r}},\delta _\chi \rbrace$
The calculated composite resolution at different SNR. The coloured part of the figure indicates that the two point source targets were successfully distinguished. (M=100,Zr=1m$M=100,Z_{\text{r}}=1\text{ m}$)
The calculated composite resolution with different number of measurements. The coloured part of the figure indicates that the two point source targets were successfully distinguished. (SNR=20dB,Zr=1m$\text{SNR}=20\text{ dB},Z_{\text{r}}=1\text{ m}$)
The calculated composite resolution at different imaging distances. The coloured part of the figure indicates that the two point source targets were successfully distinguished. (SNR=20dB,M=100$\text{SNR}=20\text{ dB},M=100$)
Statistical resolution analysis of microwave computational imaging

November 2024

·

4 Reads

This letter presents a model based on the Cramér–Rao bound theory for calculating the theoretical lower limit of imaging resolution of microwave computational imaging aided by re‐configurable metasurface. The model takes into account two crucial parameters: the signal‐to‐noise ratio and the number of measurements in the imaging process. By simultaneously considering the signal‐to‐noise ratio and the number of measurements, as well as other parameters, the model allows for a comprehensive assessment of the imaging resolution of the imaging system. In addition, both position estimation and intensity estimation are considered, and a new composite statistical resolution concept is proposed, which is more suitable for microwave computational imaging.


Packet generation architectures: (a) single‐mode label, (b) two‐mode label switching, and (c) two‐mode label multiplexing
Label assignment to payloads (a) without label time shift, (b) with half‐label time shift, and (c) with one label time shift
System configuration (a) single‐mode labels and (b) two‐mode superimposed labels
Optical label generation and processing
Label error ratios (a) two‐mode labels without half‐label time shift and (b) two‐mode labels with half‐label time shift
Two‐mode multiplexed Fourier‐based labels

November 2024

·

4 Reads

A two‐mode optical packet switching architecture using an optical multiport mode‐division multiplexing label generator and processor is proposed. By combining optical labels on two modes, the number of labels is increased exponentially. Beat noise significantly affects the label error rate, when multimode crosstalk is maximum. This impairment can be mitigated by time shifting the optical labels of the two modes by half of the label duration.


The flow chart of the EROA algorithm for solving the sensor placement problem in the WDN
The sensor placement scheme of the EROA algorithm on the Net3 model
Pressure sensor placement in water distribution networks based on enhanced Rafflesia optimization algorithm

November 2024

·

27 Reads

The pressure sensor placement in the water distribution networks (WDNs) is a typical NP‐Hard problem. The goal is to monitor the running state of the whole network area by placing a small number of pressure sensors. This letter proposes an enhanced Rafflesia optimization algorithm (EROA) to solve this problem. The simulation results show that compared with the original ROA algorithm and the traditional intelligent optimization algorithms, the proposed EROA algorithm exhibits superior optimization capabilities and solution efficiency, and can provide a more reasonable sensor placement scheme.


A motion compensation method for the single transmitter and multiple receivers mobile radar system

November 2024

·

3 Reads

The single transmitter and multiple receivers (STMR) mobile radar system is a special distributed radar system, which has stronger detection and survival capabilities. However, the high‐speed movement of the radar platforms easily produce the across‐range cell migration (RCM) of radar echo signal, which cause envelope shift effects within the coherent processing interval. Here, a motion compensation method for the STMR mobile radar system in the sea environment, to solve the problem of RCM is proposed. First, the target points and clutter points with the same Doppler velocity are obtained by the radon Fourier transform method. Second, using the range/Doppler difference between clutter and target under multi‐radar views, the target and clutter points are separated by the spatial range mapping. The simulation results show that the method can effectively improve the coherent accumulation effect of the target in the sea environment with low signal‐to‐clutter ratio (SCR).


Compact circularly polarized metasurface antenna based on characteristic mode analysis

October 2024

·

12 Reads

This article introduces a novel, single‐layer, low‐profile, circularly polarized metasurface antenna. Through characteristic mode analysis, the optimization of a 3××\times3 rectangular metasurface structure was performed. Two orthogonal modes were selected as the operating modes, and improvements to the metasurface structure were made to achieve a 90 phase difference between these modes. The adoption of a coplanar waveguide structure enables the excitation of the metasurface, facilitating circularly polarized radiation performance for the two selected modes. The antenna's dimensions are 40 mm ××\times 40 mm ××\times 3 mm (0.67 λλ\lambda ××\times 0.67 λλ\lambda ××\times 0.05 λλ\lambda), featuring a −10 dB impedance bandwidth of 27.8% (4.49–5.94 GHz). The 3 dB axial ratio bandwidth is 14.2% (5.25–6.05 GHz), with a maximum gain of 7.26 dBi.


Doubly grounded buck‐boost PV grid‐connected inverter without shoot‐through problem

October 2024

·

7 Reads

Transformerless grid‐connected inverters (GCIs) are universally utilized in the PV system. However, they have the shoot‐through issue and common mode leakage current (CMLC). To address the above‐mentioned issues, a transformerless common‐ground GCI without shoot‐through issue is proposed. The proposed topology has one common ground between the PV array and the grid that eliminates the CMLC. The topology is the combination of two buck‐boost converters, so it has the boost capability. Each buck‐boost converter operates in half‐line period, so the proposed inverter is a single‐stage system. Only two switches operate in high frequency in positive half‐line period and only one switch operates in high frequency in negative half‐line period. Therefore, efficiency of the system can be improved. There is no current flowing through the body diode of the MOSFET. Additionally, the presented inverter has no shoot‐through problem. The operating principle is elaborated. At last, simulation and experimental results of a 250 W proposed inverter validate the theoretical analysis. The maximum efficiency of the system is 96.2%.


Pilot spoofing attack detection probability with the proposed scheme, where σ2=0$\sigma ^2=0$ dB and ε=0.05$\epsilon =0.05$
Pilot spoofing attack detection performance with different schemes, where ε=0.05$\epsilon =0.05$
The normalized mean square error of different channel estimation schemes, where K=4$K=4$
The achievable downlink sum secrecy rate with different precoding schemes, where η=0$\eta =0$ dB and K=20$K=20$
Pilot spoofing attack detection and channel estimation for secure massive MIMO

October 2024

·

5 Reads

Pilot spoofing attack (PSA) is an active eavesdropping attack in massive multiple‐input multiple‐output systems, where the eavesdroppers transmit the same pilot sequence as the legitimate user does to the base station to confuse the normal channel estimation during the uplink channel training phase. With the contaminated channel estimations, more information will be leaked to eavesdroppers in the downlink transmission phase. However, it is a challenging issue to detect the PSA attack due to the similarity of the received signals and the variations of the wireless channels. Here, a new PSA detection scheme by using the difference of two different estimators is presented, that is, the least square estimator and the minimum mean square error estimator, without requiring any priori of eavesdroppers. Following that, a new data‐aided channel estimation scheme is proposed to eliminate the PSA effect. Simulation results demonstrate that the proposed PSA detection scheme outperforms the conventional energy detector, and more accurate legitimate channel estimation and higher sum secrecy rates can be obtained with the proposed scheme.


High‐contrast QPSK pattern recognition device consisting of a 4××\times4 MMI coupler

October 2024

·

9 Reads

Pattern recognition of optical label is utilized not only in packet switches but also in the field of data security due to its wide range of applications. In particular, pattern recognition using waveguide devices is highly valued because it eliminates the need for encoder processing. However, existing methods suffer from the problem of degraded contrast ratio because half of the input power are emitted from non‐target waveguides. In this article, a method is proposed to improve the contrast ratio employing a 4××\times4 multi‐mode interference (MMI) coupler with multiple input signals. The theoretical results reveal that the proposed method improves the contrast ratio from 3.0 dB to 9.5 dB and ∞\infty dB for the three‐input and four‐input cases, respectively, for quadrature phase‐shift keying (QPSK)‐modulated signal. Furthermore, numerical simulation was performed through the three dimension finite‐difference time‐domain (3D‐FDTD) method, and the proposed scheme successfully discriminated each QPSK pattern. The power value |E|2E2|E|^2 at all ports closely matches the theoretical value, and the ratios of 9.5 dB and >>>30.0 dB are numerically obtained.


Machine learning‐driven microwave imaging for soil moisture estimation near leaky pipe

October 2024

·

4 Reads

Characterizing soil moisture around drip irrigation pipes is crucial for precise and optimized farming. Machine learning (ML) approaches are particularly suitable for this task as they can reduce uncertainties caused by soil conditions and the drip pipe positions, using features extracted from relevant datasets. This letter addresses local moisture detection in the vicinity of dripping pipes using a portable microwave imaging system. The employed ML approach is fed with two dimensional images generated using back projection as a radar‐based algorithm and the Born approximation as an inverse scattering method, based on spatio‐temporal (collected data at various positions over the soil surface and at different time points.) measurements at various frequencies. The study investigates the performance of K‐nearest neighbour (KNN) and convolutional neural networks (CNN) algorithms for moisture classification based on these imaging techniques. We also explore the potential of KNN and CNN for moisture estimation around the plant roots and in the presence of pebbles. In general, CNN outperforms KNN in moisture content detection from microwave data, especially after applying imaging algorithms. A combination of CNN as the ML approach and the back projection algorithm to provide training data, yielded 20%20%20\% accuracy more than other models for moisture content estimation. Also, the practical results demonstrate that our method can detect soil moisture with an estimation error of less than 10%.


Root mean square error of the predicted ratings with different number of features
Prediction success rate versus number of features
Root mean square error of the predicted ratings with different data deficiency degrees
Prediction success rate versus data deficiency degrees
An anti‐interference decision‐making method for communication waveform based on collaborative filtering

Aiming at the problem of efficient and reliable transmission for communication system in complex electromagnetic environment, an anti‐interference decision‐making method for communication waveform based on collaborative filtering (CF) is proposed here. The required minimum signal‐to‐noise ratio to meet a specific bit error rate is adopted as a key indicator to evaluate the performance of certain waveform–channel combinations. The main challenge that is solved is to predict transmission performance ratings that cannot be obtained in practice. With CF of waveform features and interference features, the missing training data can be accurately predicted. The convergence speed and prediction accuracy of the proposed algorithm are verified by simulation, and the influence of different feature quantities and different data deficiency degrees on the stability and accuracy of the algorithm is discussed.


Configuration and geometry of the proposed bidirectional antenna array. (L = 55, W = 75, H = 0.5, D = 1, P = 128, FL = 28, FW = 1, FH = 0.5, FD = 1, unit: mm.)
Current distribution of the proposed bidirectional antenna array. (a) Simulated current distribution. (b) Diagram of current distribution on radiating units A. (c) Diagram of current distribution on radiating units B
Simulated directivity and patterns with key parameters at 2.4 GHz. (a) Directivity versus different L and P. (b) Directivity and −1 dB directivity bandwidth versus N. (c) Radiation patterns on YOZ plane. (d) Radiation patterns on XOZ plane
Simulated impedance and S11 versus different array scales and bent radiuses at 2.4 GHz. (a) Input impedance versus N. (b) S11 versus N. (c) Bent structure. (d) S11 versus R
Measured results of the proposed bidirectional antenna array. (a) Photograph. (b) Simulated and measured S11 and gain. (c) Radiation patterns on YOZ plane. (d) Radiation patterns on XOZ plane
Normal‐mode planar meander‐line antenna array with bidirectional broadside radiation pattern

October 2024

·

1 Read

In this letter, a normal‐mode planar bidirectional broadside antenna array with the merits of high gain, broadband, low aerodynamic drag, and high assembly robustness is proposed. First, the bidirectional antenna array is implemented based on a novel design strategy. The expected radiation field generated by currents along the same direction is superimposed while the unexpected radiation field is offset due to the out‐of‐phase currents, achieving a normal‐mode high‐gain bidirectional pattern. Then, the bidirectional antenna array is designed with two metallic meander lines, featuring low aerodynamic drag and cost‐efficiency. Moreover, a simple feed structure using an impedance transformer parallel strip line is adopted for a wide bandwidth. Besides, the bidirectional antenna array exhibits a stable impedance when the array scale increases. The stable impedance is also observed when the antenna array is bent to different angles, ensuring ease of installation and high assembly robustness. To validate the design strategy, a prototype with an aperture length of 5.8λ at 2.4 GHz is fabricated. The proposed antenna array is designed to operate from 2.2 to 2.6 GHz, with a maximum gain of 13 dBi around the center of frequency, which can be a promising candidate in long and narrow communication scenarios.


Design of miniaturized and highly selective frequency selective rasorber based on compact spiral resonator

October 2024

·

4 Reads

A miniaturized, high‐selectivity, wideband A‐T‐A frequency selective rasorber (FSR) based on compact spiral resonators (CSR) is proposed here. The lossy layer of the FSR consists of the CSR and a Jerusalem cross, with four resistors soldered onto the surface. The underlying layer is a bandpass frequency selective surface (FSS) with an extremely narrow slit. The dimensions of the FSR unit are only 0.079 λL × 0.079 λL × 0.078 λL. The extremely narrow‐slitted FSS and CSR together provide a narrowband frequency response, enhancing the selectivity of the FSR. Additionally, a prototype of an FSR was fabricated and tested, achieving results highly consistent with simulations.


Image is divided into an m × n grid of Nr subregions. Scale‐invariant feature transform (SIFT) features are extracted from the entire image, with red points representing the selected sparse features
Generalized Hough transform (GHT) voting process for matching pairs between sparse features in a specific subregion and features extracted during initialization. The voting space is organized with △o and △s, and the most‐voted bin indicates the rough pose estimate
Adjacent subregions with the same rough pose are connected to form locally connected regions (LCRs)
Precision versus alignment error threshold curves for different variations and the overall performance across all sequences
Robust planar object tracking using SIFT and GHT with spatial locality

October 2024

·

2 Reads

Planar object tracking (POT) is crucial in vision‐based robotic applications. Despite significant advancements, tracking planar objects under real‐world conditions remains a challenge owing to various factors. The method proposed by the authors leverages SIFT features and generalized Hough transform, which is enhanced by spatial locality, to mitigate background clutter and false matching. The experimental results demonstrate that this method significantly outperforms existing approaches, achieving higher precision across various alignment error thresholds.


The structure of flexible traction power supply system
Multiple converters control strategy
Optimized result of flexible traction power supply system from 8:00 to 9:00. (a) Power of utility grid in TS1. (b) Power of utility grid in TS2. (c) Power and SOC of battery (BAT) in TS1. (d) Power and SOC of ultracapacitor (UC) in TS1. (e) Power and SOC of BAT in TS2. (f) Power and SOC of UC in TS2
Experiment result of flexible traction power supply system. (a) Power flow controller (PFC) current of TS1. (b) Photovoltaic (PV) current and SOC of ultracapacitor (UC) and battery (BAT) in TS1. (c) PFC current of TS2. (d) PV current and SOC of UC and BAT in TS2
Two‐stage economic dispatch of railway FTPSS considering system voltage characteristic

October 2024

·

2 Reads

The flexible traction power supply system (FTPSS), integrating power flow controller (PFC), energy storage system, and photovoltaic (PV) system, is able to cancel the neutral section, realize the collaborative operation of multiple traction substations (MTSs), and enhance the efficient utilization of regenerative braking energy (RBE) and renewable energy in electrified railways. However, co‐dispatching energy among MTSs influences the power flow distribution within the traction network, impacting the traction network voltage (TNV). Maintaining optimal energy dispatch while ensuring compliance with TNV standards is critical. Therefore, a two‐stage energy optimal dispatch and control strategy is proposed for FTPSS here. The first stage develops a synergistic energy optimal dispatch model for MTSs considering TNV constraint and power flow constraints, aimed at maximizing the utilization of RBE and PV while minimizing the electricity cost of FTPSS. In the second stage, a control strategy for multiple converters is designed to ensure the dynamic response of FTPSS. Finally, the effectiveness of the proposed strategy and model is substantiated through field load data and experimental validation, and the electricity cost of the FTPSS is reduced by about 21.4%, meanwhile the TNV is maintained in the range of 22.5–29 kV.


The working overview of the deep learning‐based zero trust network access testbed system
Comparison of CPU usage during a distributed denial of service attack using deep learning‐based zero trust network access (DL‐ZTNA), non‐DL‐ZTNA, and Software Defined Perimeter methods
Throughput comparison during distributed denial of service attack using deep learning‐based zero trust network access, zero trust network access without deep learning, and Software Defined Perimeter methods
Attack detection probability over time of the proposed deep learning‐based zero trust network access (DL‐ZTNA), non‐DL‐ZTNA, and Software Defined Perimeter methods
Safeguarding IoT networks against DDoS attacks using deep learning based zero trust network access

October 2024

·

2 Reads

Here, a deep learning‐based zero trust network access (DL‐ZTNA) system to enhance the security of the Message Queuing Telemetry Transport (MQTT) protocol within Internet of Things (IoT) applications was proposed. Combining multi‐head convolutional neural networks and attention‐based bi‐directional long short‐term memory networks with ZTNA provides real‐time security analysis of device behaviour. This behaviour‐based approach ensures that only authorized devices can access network resources and continuously monitors for potential threats like distributed denial of service (DDoS) attacks. The proposed DL‐ZTNA system revokes device access when a threat is detected and prevents further malicious activities. Evaluation in a testbed environment showed improvements in CPU usage efficiency, throughput, and attack detection probability compared to traditional methods. This highlights the system's effectiveness in securing MQTT‐based IoT networks against DDoS attacks while maintaining high performance, showcasing the potential of integrating deep learning techniques into ZTNA system for addressing security challenges in IoT environments.


Overall framework diagram of the FAFTransformer model, which consists of FCBlocks stacked in a residual way. FCBlocks first uses Fourier transforms to calculate the period of a variable, then reshapes a one‐dimensional variable into a two‐dimensional variable using inception convolution and attention mechanisms to capture time dependencies and associations between variables from k different periods. Then, the period fusion layer is used to fuse the characteristic information learned during different periods. FFT, fast Fourier transform
FCBlock is the main module of the model. The composition and data changes of the FCBlock module are shown. FFT, fast Fourier transform
Different from TimesNet, which uses a 0 value to fill directly, this paper selects the data near the same time of the period within the variable to fill
Period fusion module diagram
Visualize the predicted results of each model; (a) represents the predicted results of ELC data and (b) represents the predicted results of ETTm2 data; the blue lines are the ground truth, and the orange lines are the model prediction
FAFTransformer: Multivariate time series prediction method based on multi‐period feature recombination

October 2024

·

5 Reads

Multivariate time series forecasting is widely used in various fields in real life. Many time series prediction models have been proposed. The current forecasting model lacks the mining of correlation between sequences based on different periods and correlation of periodical features between different periods when dealing with data. In this paper, we propose a multivariate data prediction model FAFTransformer based on the reorganization of multi‐periodic features, which first extracts the multi‐periodic information of the time series using the method of frequency domain analysis. The temporal dependencies within sequences are then captured using convolution based on different periods, and the correlations between sequences are learned by combining the multivariate attention mechanism to obtain the intra‐sequence and inter‐sequence correlations under the same period. Finally, period fusion is proposed to capture the correlation of period characteristics between different periods. The experimental results show that the model achieves the best results on multiple datasets compared to the latest seven predictive models for time‐series data.


W‐band inline waveguide‐to‐coupled microstrip line transition using a compact differential antenna

October 2024

·

12 Reads

Here, a transition technique for converting signals from rectangular waveguide to coupled microstrip lines (CML) is presented, which provides a solution for differential antenna array feeding of automotive radar. The key to the proposed transition structure is to connect the coupling patches together with a short‐bridge structure, thus forming a quasi‐loop structure. Through the short bridge, the radiation efficiency of the antenna is greatly enhanced, and the mode conversion from TE10 of rectangle waveguide to differential mode of coupled microstrip line is realized effectively under the extremely compact size. Finally, the designed transition structure is manufactured and measured. The experimental results show that the back‐to‐back insertion loss is lower than 0.65 dB and the return loss is higher than 16.8 dB in the W‐band (75–110 GHz), which is the best performance and the simplest design among the relevant transition transitions reported so far.


HD‐sEMG electrodes
Data reconstruction and feature extraction
SVIT‐silent speech recognition framework
SVIT‐SSR: A sEMG‐based vision transformer approach for silent speech recognition

October 2024

·

1 Read

Silent speech recognition (SSR) based on surface electromyography (sEMG) is a voice interaction technology proposed for scenarios requiring silent operations. This article abstracts the SSR task based on sEMG into a short‐term image sequence classification task. Time‐frequency domain feature extraction and data reconstruction on the muscle activity segment data is performed. Additionally, the temporal and spatial dimensions to capture the intrinsic correlation representation of muscle activity is analysed. The SVIT‐SSR model is proposed based on the vision transformer (VIT) framework. Finally, experiments to identify 33 types of typical silent speech commands in the SSR dataset are designed. The results demonstrate that the proposed model achieves an accuracy of 96.67 ± 1.15%, outperforming similar algorithms.


(a) Circuit‐schematic of the conventional MB with possible series capacitor locations (1)–(8). (b) Optimal (S11 = 0 at f0) (k, Zk) curve and corresponding FBW as a function of k for the MB. Maximum FBW as a function of k for the C‐loaded MB. (c) C‐loaded MB with series capacitors C1 and C2 at (6) and (3) and coupled lines with electrical length θ. (d) Red: Maximum FBW state for the conventional MB where FBW = 109% for k = 0.83, Zk = 105 Ω. Blue: Maximum FBW S‐parameters for the C‐loaded MB under the condition that k ≤ 0.83 with FBW = 137% for k = 0.83, Zk = 90 Ω, C1 = 0.23 pF, C2 = 1.8 pF, and θ = 40°
(a) Broadside‐coupled EMS with a solid ground plane. (b) EMS with perforated grounded plane (WCut is the width of the perforation). (c) Heatmap of FBW over the (k, Zk) parameter space for θ = 45°, C1 = 0.25 pF, and C2 = 1 pF. Furthermore, the Zk and k of the EMS is overlaid as a function of W/S and B/S for WCut = 0, and as a function of W/S and WCut/S for B/S = 10
Vertically integrated interdigital capacitors in the inkjet two‐material AM process. (a) Top view and side view of a capacitor having 8 vertically integrated inter‐digital fingers, for tCap = 70 µm, S = 50 µm, and G = 100 µm. (b) Capacitance of a 4‐layer interdigital (C1), 8‐layer interdigital (C2), and parallel plate capacitors as a function of occupied area
(a) 3D geometry of the C‐loaded MB. Lconn = 1200 µm, Wconn = 160 µm, WPatt = 4300 µm, WCPW = 200 µm, SCPW = 100 µm, HGND = 187 µm, tTop = 50 µm, tVia = 150 µm, and H = 2B + S + t + tTop = 1117 µm. (b) EM‐simulated S‐parameters, AI, and PI
(a) Optical and X‐ray images of the manufactured C‐loaded Balun A (P3 is 50 Ω terminated) where WDev = 4.3 mm, LDev = 6.0 mm. (b) EM‐simulated and RF‐measured S‐parameters (top) of the two resistively terminated baluns (Balun A: Corresponding to S21A, Balun B: Corresponding to S21B) and EM‐simulated and RF‐measured AI and PI (bottom). Grey area: operational BW
Inkjet‐printed capacitively loaded Marchand baluns with enhanced fractional bandwidths

October 2024

·

11 Reads

A comprehensive design methodology for the realization of ultra‐wideband and highly miniaturized capacitively loaded Marchand baluns (MBs), suitable for integration into balanced antenna systems is presented. A design technique to maximize the fractional bandwidth (FBW) for a given manufacturing process is demonstrated for the first time. Furthermore, a novel MB integration scheme using broadside‐coupled lines and 3D vertically integrated capacitors in a two‐material inkjet printing process is uniquely proposed. The concept has been experimentally validated within the S and C bands. It exhibits a footprint of 0.128 × 0.098 λg², centre frequency of 4.4 GHz, and FBW of 121%. Its power loss, phase imbalance and amplitude imbalance were measured between 0.8 and 2.2 dB, 3 ± 2°, and 0.4 ± 0.4 dB, respectively.


Results of the far‐field radiation pattern reconstructed for a rectangular horn antenna
Flowchart of the proposed far‐field reconstruction method using single‐surface phaseless spherical measurements
Results of the far‐field radiation pattern reconstructed for a rectangular horn antenna
Results of the far‐field radiation pattern reconstructed for a 4‐port loop antenna
Far field reconstruction of antennas via single‐surface phaseless spherical near‐field scans: A novel approach based on dipole equivalence

October 2024

·

8 Reads

A novel approach is proposed in this paper to reconstruct the far‐field radiation pattern from the phaseless electric field of an antenna scanned on a single near‐field sphere. It adopts the dipole equivalence approach to project the near‐field electric field into a spherically distributed array of electric dipoles. A term representing the error from the linearly correlated portion in the least‐square problem associated with the dipole equivalence, namely the linear correlation error, is introduced. It is demonstrated that by iterative search to minimize the linear correlation error using the covariance matrix adaptation evolution strategy, near‐field phase distributions can be found efficiently from the magnitude‐only near‐field, and the far‐field radiation pattern can be computed. Two representative case studies are given here to validate the proposed method. Results demonstrate good agreements between computations and simulations.


Architecture of our device‐to‐device (D2D)‐empowered cellular network model
The simulation results: (a) Radiation patterns of the proposed antenna at 21 GHz. (b) The convergence behaviour of the proposed method. (c) Average transmit power of device‐to‐device (D2D) links. (d) Average sum rate of D2D links. (e) Average sum rate of cellular links. (f) Average outage probability of D2D links.
Joint antenna selection and resource allocation for mm‐wave directional D2D communications using distributed deep reinforcement learning

October 2024

·

21 Reads

In this paper, with the promising assumption of using adaptive directional microstrip antenna on user equipment, the problem of joint antenna selection, spectrum assignment, and transmit power allocation for mm‐wave device‐to‐device communications underlying cellular networks is tackled, with the goal of enhancing system throughput and energy efficiency. To address the complexity of this problem, a method based on multi‐agent distributed deep reinforcement learning in which an autonomous intelligent agent is deployed for each user equipment is proposed. The performance evaluation demonstrates its superiority over existing strategies, resulting in improved system performance, reduced outage probability, and enhanced energy efficiency.


The (a) integral and (b) details of the proposed absorptive frequency‐selective reflector structure. LH = 3.8 mm, LH1 = 1.2 mm, Ll = 3.2 mm, Ll1 = 0.5 mm, wH = 0.6 mm, dH = 2.2 mm, dL = 2.5 mm, r1 = 1.2 mm, r2 = 0.9 mm, w1 = 0.2 mm, d = 0.1 mm, g = 0.3 mm, RH = 125 Ω, and RL = 75 Ω
Reflection coefficients of the proposed absorptive frequency‐selective reflector under the normal incidence
(a) Prototype of the fabricated absorptive frequency‐selective reflector. (b) Prototype of the assemble parts
The comparison of the measured monostatic radar cross‐section reduction with the simulated reflection coefficients
Wideband absorptive frequency‐selective reflector based on split‐ring resonator embedded with chip resistor

October 2024

·

4 Reads

Here, a wideband absorptive frequency‐selective reflector (AFSR) based on a split‐ring resonator (SRR) embedded with chip resistors is proposed. The wide absorption band is achieved by the multimode resonances of two dipoles, and the insertion of SSRs is able to bypass the losses in the passband, thus allowing to open a wide window in addition to achieving broadband absorption performance. The upper structure consists of a dipole in series with two SSRs, while the lower dipole connects one SSR. The upper and lower structures are grouped together and rotated by 90° around the Z‐axis to form the final dual‐polarization 3D‐AFSR. The simulated results demonstrate that reflection coefficients below −10 dB are seen between 6.74–9.31 and 16.91–19.61 GHz under normal incidence, with a total fractional bandwidth of 46.8%. It also has a wide 3 dB reflection band which is obtained from 10.69 to 15.44 GHz, and the entire profile is 0.17λL, where λL is lowest operating frequency.


The architecture of the proposed class‐wised domain decoupling‐guided adversarial feature learning (CDD‐AFL) model
Class‐wised domain decoupling‐guided adversarial feature learning for cross‐age face recognition

October 2024

·

3 Reads

How to extract age‐independent identity features from face images has long been the major challenge in cross‐age face recognition task. In this letter, a class‐wised domain decoupling‐guided adversarial feature learning network to extract age‐independent face features is proposed. In this model, an adversarial feature learning module is designed to decompose identity features and age features. Considering that artificially designed age interference suppression strategies are difficult to fit the complex relationship between identity features and age features, an adversarial training strategy incorporated with an attention mechanism in the adversarial feature learning module is introduced to suppress age interference adaptively. Besides, under the idea of maximizing the distribution difference of identity feature domain and age feature domain, a class‐wised domain decoupling module is further designed to guide the model to extract age‐independent identity features. The proposed model has been validated on the well‐known public datasets CALFW and CACD‐VS, where it achieved remarkable recognition accuracy rates of 94.5% and 99.5%, respectively. This represents improvements of 4.2% and 0.1% over the state‐of‐the‐art single‐task‐based models, clearly showcasing the effectiveness of the proposed class‐wised domain decoupling‐guided adversarial feature learning model.


Journal metrics


0.7 (2023)

Journal Impact Factor™


47%

Acceptance rate


2.7 (2023)

CiteScore™


31 days

Submission to first decision


$2,540 / £1,980 / €2,310

Article processing charges

Editors