REV Journal on Electronics and Communications
Not a member yet
230 research outputs found
Sort by
Implementation of ChaCha20-Poly1305 on Self-Organization Data Framing for Enhancing IoT Communication
Get PDFOur study presents the implementation of a lightweight Authenticated Encryption with Associated Data (AEAD) and a self-organizing data framing in Internet of Things (IoT) communication. The frame structure on the node side offers extra fields for managing data bytes, calculating the Cyclic Redundancy Check (CRC), and detecting the frame start and stop. Between an STM32 and an ESP32, the data framing is applied via the Universal Asynchronous Receiver-Transmitter (UART) protocol. High performance and security are achieved by implementing the ChaCha20-Poly1305 AEAD on the frame structure for nodecloud communication. Through logging, the frame structure is successfully identified on ESP32 and the IoT cloud. ChaCha20-Poly1305 algorithms are fully evaluated in cycle counting on the node-cloud communication using the System for Unified Performance Evaluation Related to Cryptographic Operations and Primitives (SUPERCOP) framework. The data framing gives the MCUs’ communications a foundation for error detection and synchronization failure prevention. Lastly, the IoT system’s node-cloud communications are encrypted and authenticated via the ChaCha20-Poly1305 process
A Deep Learning Model for Splicing Image Detection
Get PDFWith the advancement of digital technology, manipulating images has become relatively easy through many photo editing techniques. One of the techniques is the splicing image method, which crops parts of images and puts them into another image creating a new composite image. The image splicing detection system is soon regarded as an exciting topic for many researchers to solve the problems of forgery images on the Internet, especially in social networks. ResNet-50 and VGG-16 are powerful architectures of convolutional neural networks, but they reveal many weaknesses when operating on low-end computers. The ultimate goal of this research is to create a model for image splicing detection working well in limited memory machines. The study proposes the model, which is the improvement of VGG-16 applying residual network (ResNet). As a result, the proposed model achieves a test accuracy of 92.5% while the ResNet-50 gives an accuracy of 85.6% after 20 epochs of training 9,319 images from the CASIA v2.0 dataset, which are used for forgery classification. The result proves the efficiency of the proposed model for image splicing detection, especially when working on low-end computers
Deep Learning Based Cooperative MIMO Systems for Wireless Body Area Networks
Get PDFWireless Body Area Network (WBAN) is widely applied in various fields, including healthcare, sports, wellness, and assistive technologies, by offering the benefits of convenience, reliability, low latency, privacy, and customization. However, the propagation characteristics of the WBAN channel can impact the reliability of transmission, which is particularly crucial in healthcare systems. To address this issue, this article presents a novel approach using deep learning-based cooperative Multiple-Input Multiple-Output (MIMO) systems that leverage the autoencoder (AE) technique. In our proposed approach, we utilize the AE-based cooperative MIMO systems with two different techniques: Amplify-and-Forward (AE-AF) and Decode-and-Forward (AE-DF). The AE-AF scheme operates without needing training parameters at the relay node, whereas the AE-DF scheme necessitates training parameters at the relay node. Both schemes aim to overcome challenges such as multipath propagation phenomena, thereby enhancing the performance of on-body communication systems. Additionally, we introduce two combinators, Minimum Mean Square Error (MMSE) scheme and Radio Transformation Network (RTN), to effectively mitigate co-channel interference (CCI) in the received signal streams and improve the bit error rate performance of the AE-AF and AE-DF systems. We assess the performance of these systems in scenarios with and without direct links. Simulation results demonstrate significant performance improvements compared to baseline cooperative MIMO systems using MMSE combining, namely AF-MMSE and DF-MMSE systems. Notably, the proposed systems employing RTN combination, including both direct and relay paths, achieve a 7.5 dB gain over the baseline when all nodes are equipped with two transceiver antennas
Prune and Quantize Semantic Segmentation Network for Aerial Objects Recognition
Get PDFSemantic segmentation of aerial and satellite images is crucial for applications in environmental management, urban planning, and traffic safety. While deep learning techniques with convolutional neural networks (CNNs) and attention mechanisms have achieved superior accuracy compared to traditional methods, they often struggle with model complexity and resource constraints. This paper introduces two novel techniques - pruning and quantization - to enhance both the performance and efficiency of semantic segmentation models for remote sensing images (RSIs). Pruning reduces model complexity by eliminating less significant weights, while quantization decreases memory usage by converting weights into a more compact format. We applied these techniques to the DeepLabV3+ model with ResNet18 and ResNet50 backbones and assessed their performance across multiple RSI datasets. Our results show that pruning and quantization effectively balance accuracy and computational efficiency, achieving a mean IoU of 81.24% with a memory footprint of 135.19 MB for pruning, and 81.04% mean IoU with a memory footprint of 33.79 MB for quantization on the ISPRS Vaihingen dataset. These methods offer a viable solution for deploying semantic segmentation models on resource-constrained hardware.
Terahertz Resistor-coupled Arrayed Resonant-tunneling-diode Oscillators with High DC-to-RF Efficiency using Metal-insulator-metal Capacitor
Get PDFResistor-coupled arrayed resonant-tunneling-diode (RTD) oscillators have emerged as promising candidates for high-performance Terahertz sources. Despite various advantages, the DC-to-RF efficiency of such oscillators is still low. The reason is the loss caused by resistors, which form the ends of the slot antennas. This paper presents a method to increase the DC-to-RF efficiency of resistor-coupled arrayed oscillators. Edged resistors at the ends of slot antennas are covered by metal-insulator-metal capacitors, which shunt the currents flowing through those resistors. Thus, the conduction loss caused by edged resistors is reduced. Owing to low conduction loss, high output power is obtained, resulting in a high DC-to-RF efficiency. It is estimated that the proposed oscillator’s output power and DC-to-RF efficiency reach 1.53 mW and 0.44% at 450 GHz, respectively. A high power density of up to 62 mW/mm2 is another advantage of the proposed RTD oscillator. With the improved characteristics, we believe the proposed oscillators could promote various Terahertz applications
Investigate Discriminative AutoEncoder in Few-shot Learning-based Anomaly Detection
Get PDFDiscriminative AutoEncoder (DisAE) plays a crucial role in enhancing the adaptability and gener- alization of few-shot learning methods (DisAEFL) for detecting rare anomalies. DisAE captures meta- knowledge from multiple known tasks, facilitating rapid adaptation in DisAEFL. Key factors like the discriminative parameter (a) and normal proportion (pn) significantly impact DisAEFL performance. However, their influence on the DisAE manifold and DisAEFL’s efficacy in rare cyberattack detection remain understudied in cybersecurity. This study presents an investigative approach to probe DisAE’s influence on DisAEFL’s performance in addressing rare, unseen cyberattacks, aiming to gain insight into the DisAE manifold and outline future research directions. Through intensive analysis, we focus on parameters a and pn, detailing how to examine them to observe DisAE’s effects on DisAEFL. Two main experiments are conducted to investigate their influences. Experimental results on the NSL-KDD dataset reveal a strong correlation between these parameters and both the DisAE manifold and DisAEFL performance. These findings suggest strategies for more efficiently constructing the DisAE manifold to enhance DisAEFL’s adaptability and generalization. Overall, this study contributes to advancing anomaly detection methodologies in cybersecurity by shedding light on the interplay between DisAE, DisAEFL, and crucial parameters
Performance Analysis of Quine-McCluskey Method on CPU
Get PDFThe Quine-McCluskey method is a widely used procedure to minimize Boolean functions. Although the method can be programmed on computers, it takes a long time to return the set of essential prime implicants, thus slowing the analysis and design of digital logic circuits. In this paper, we first propose three ways of data representation for prime implicants in memory, followed by our performance analysis for each representation. We then propose a multithreading scheme to find all prime implicants of a Boolean function. The scheme aims to accelerate step 1 of the method on multicore platforms. After that, we propose an algorithm for step 2 of the Quine-McCluskey method to select the minimal number of essential prime implicants. The evaluation shows that the mask-based representation achieves the highest performance when the input number is small. When the input number is more than or equal to 20, the best data representation is bitarray-based. The bitarray-based representation achieves a 5x higher performance than the ASCII-based representation when the input number equals 24 and the fill factor equals 0.002. The number of essential prime implicants can be reduced up to 45% of the total prime implicants generated in step 1 of the method for a 16-input Boolean function at a fill factor of 0.05
An Ultra-high Quality Factor Terahertz Photonic Crystal Cavity
Get PDFHigh quality factor Terahertz (THz) cavities are highly desired for many THz applications. This paper presents an ultra-high quality factor terahertz planar photonic crystal cavity at 300 GHz range. Two approaches are employed to reduce the losses in the cavity increasing the quality factor of the cavity. Firstly, short embedded photonic crystal waveguides are employed to reduce the in – plane loss. Secondly, a novel way of hole displacement is adopted for four edged holes of the L3 – type photonic crystal cavity to decrease the radiation loss. An ultra – high quality factor of 65000 at a resonant frequency of 315.3 GHz was achieved for the designed cavity. This result could enable promising applications such as THz sensing
