Faculty Profiles - NOGAMI Yasuyuki

Papers - NOGAMI Yasuyuki

Division display 1 - 20 of about 188 ／ All the affair displays >>

Optimizing IoT Intrusion Detection Using Balanced Class Distribution, Feature Selection, and Ensemble Machine Learning Techniques

Muhammad Bisri Musthafa, Samsul Huda, Yuta Kodera, Md Arshad Ali, Shunsuke Araki, Jedidah Mwaura, Yasuyuki Nogami

Sensors 24 ( 13 ) 2024.7

Publishing type：Research paper (scientific journal)

Internet of Things (IoT) devices are leading to advancements in innovation, efficiency, and sustainability across various industries. However, as the number of connected IoT devices increases, the risk of intrusion becomes a major concern in IoT security. To prevent intrusions, it is crucial to implement intrusion detection systems (IDSs) that can detect and prevent such attacks. IDSs are a critical component of cybersecurity infrastructure. They are designed to detect and respond to malicious activities within a network or system. Traditional IDS methods rely on predefined signatures or rules to identify known threats, but these techniques may struggle to detect novel or sophisticated attacks. The implementation of IDSs with machine learning (ML) and deep learning (DL) techniques has been proposed to improve IDSs’ ability to detect attacks. This will enhance overall cybersecurity posture and resilience. However, ML and DL techniques face several issues that may impact the models’ performance and effectiveness, such as overfitting and the effects of unimportant features on finding meaningful patterns. To ensure better performance and reliability of machine learning models in IDSs when dealing with new and unseen threats, the models need to be optimized. This can be done by addressing overfitting and implementing feature selection. In this paper, we propose a scheme to optimize IoT intrusion detection by using class balancing and feature selection for preprocessing. We evaluated the experiment on the UNSW-NB15 dataset and the NSL-KD dataset by implementing two different ensemble models: one using a support vector machine (SVM) with bagging and another using long short-term memory (LSTM) with stacking. The results of the performance and the confusion matrix show that the LSTM stacking with analysis of variance (ANOVA) feature selection model is a superior model for classifying network attacks. It has remarkable accuracies of (Formula presented.) and (Formula presented.) and overfitting values of (Formula presented.) and (Formula presented.) on the two datasets, respectively. The model’s ROC is also shaped with a sharp bend, with AUC values of (Formula presented.) and (Formula presented.) for the UNSW-NB15 dataset and the NSL-KD dataset, respectively.

DOI： 10.3390/s24134293

Scopus

PubMed

researchmap

Optimizing IoT Intrusion Detection Using Balanced Class Distribution, Feature Selection, and Ensemble Machine Learning Techniques.

Muhammad Bisri Musthafa, Samsul Huda, Yuta Kodera, Md. Arshad Ali, Shunsuke Araki, Jedidah Mwaura, Yasuyuki Nogami

Sensors 24 ( 13 ) 4293 - 4293 2024.7

Tolerance Evaluation Against Deep Learning Side-Channel Attack on AES in Automotive Microcontroller With Uncertain Leakage Model

Masaki Himuro, Naoto Kawahara, Yoshiyuki Amanuma, Kengo Iokibe, Yasuyuki Nogami, Yoshitaka Toyota

2024 IEEE Joint International Symposium on Electromagnetic Compatibility, Signal &amp; Power Integrity: EMC Japan / Asia-Pacific International Symposium on Electromagnetic Compatibility (EMC Japan/APEMC Okinawa) 2024.5

A Secure Authentication for Plant Monitoring System Sensor Data Access

Samsul Huda, Yasuyuki Nogami, Md. Biplob Hossain, Yang Jie, Le Hoang Anh, Muhammad Bisri Musthafa, Maya Rahayu, Takuma Akada

2024 IEEE International Conference on Consumer Electronics (ICCE) 2024.1

Performance Analysis of ELiPS-Based CP-ABE with Optimized Decryption Functions

Le Hoang Anh, Yuta Kawada, Samsul Huda, Yuta Kodera, Yasuyuki Nogami

Smart Innovation, Systems and Technologies 404 SIST 345 - 354 2024

Publishing type：Research paper (international conference proceedings)

CP-ABE emerges as an advanced cryptographic method that offers secure and precise access control in Internet of Things and cloud storage applications. Nevertheless, the original CP-ABE algorithm relied on the PBC library with a security level limited to 80 bits and is now outdated, making it vulnerable to attacks. To address this issue, we have introduced ELiPS-based CP-ABE, which boosts the level of security to 128 bits using the ELiPS library. While this enhancement improved the overall performance, the decryption functions remained computationally intensive. To tackle this, we previously optimized the decryption function by minimizing final exponentiations and inversions. In this paper, we evaluate and analyze the impact of these optimizations on decryption efficiency. Moreover, we compare the ELiPS-based CP-ABE with these improvements to the initial version and the original PBC-based CP-ABE. As a result, the combination of both optimization techniques resulted in an average 43.06% overall reduction in decryption time compared to the initial version of the ELiPS-based CP-ABE scheme, while in total execution, it led to a 25.27% improvement. Furthermore, there was an average 53.81% overall reduction in total execution time compared to the original PBC-based CP-ABE method.

DOI： 10.1007/978-981-97-5810-4_30

Scopus

researchmap

Securing Data Transmission in Noisy Channels: A Synergistic Approach using Kyber and BCH Codes

Linh Nguyen, Andri Santoso, Samsul Huda, Yasuyuki Nogami, Tuy Tan Nguyen

11th IEEE International Conference on Consumer Electronics - Taiwan, ICCE-Taiwan 2024 399 - 400 2024

Publishing type：Research paper (international conference proceedings)

The proliferation of the Internet of Things leads to the sharing of more data among diverse stakeholders, ranging from individual users to large enterprises. However, ensuring the security of this data exchange while simultaneously minimizing errors during transmission presents a complex challenge. Due to the involvement of numerous devices and users capable of data sharing, the dynamic and heterogeneous nature of the environment leads to constant changes. In this paper, we integrate the Kyber algorithm for post-quantum security with Bose-Chaudhuri-Hocquenghem (BCH) codes for error correction. We validate our approach by transmitting images from the widely- used CIFAR-10 dataset in encrypted mode over noisy communication channels. The system demonstrates effective image encryption and robust error correction capabilities.

DOI： 10.1109/ICCE-Taiwan62264.2024.10674282

Scopus

researchmap

Optimizing CRYSTALS-Dilithium in Rust: Radix-4 NTT and Assembly-level Comparison with Official C Implementation

Shunri Kudo, Yasuyuki Nogami, Samsul Huda, Yuta Kodera

11th IEEE International Conference on Consumer Electronics - Taiwan, ICCE-Taiwan 2024 395 - 396 2024

Publishing type：Research paper (international conference proceedings)

Post-Quantum Cryptography (PQC) is becoming popular since it offers protection against quantum attacks. CRYSTALS-Dilithium is being considered as a promising NIST PQC standard candidate. It utilizes the Radix-2 Number Theoretic Transform (NTT). Its reliance on lattice-based cryptography and number-theoretic principles positions CRYSTALS-Dilithium as a strong contender for providing robust security in the era of quantum computing. Besides, the Radix-4 NTT divides polynomials into more parts than the Radix-2 NTT, making calculations more efficient by reusing values. In this paper, we improve the computational efficiency of CRYSTALS-Dilithium by implementing the Radix-4 NTT algorithm in both C and Rust programming languages. We verified the effectiveness through experiments in terms of signature and verification speeds. Additionally, we analyzed the assembly code generated by both languages to understand the performance differences.

DOI： 10.1109/ICCE-Taiwan62264.2024.10674226

Scopus

researchmap

A Study to Improve Calculation Efficiency of CKKS Encoding with Parallelization

Taichi Saito, He Yixuan, Samsul Huda, Junting Xiao, Yuta Kodera, Yasuyuki Nogami

11th IEEE International Conference on Consumer Electronics - Taiwan, ICCE-Taiwan 2024 397 - 398 2024

Publishing type：Research paper (international conference proceedings)

CKKS homomorphic encryption is frequently utilized in cloud-based applications due to its ability to handle both real and complex numbers, ensuring the protection of sensitive data. However, the security of CKKS encryption relies on an encoding process to convert input data into a polynomial. While this process involves the Inverse Fast Fourier Transform (IFFT), there are two notable issues with IFFT: first, most multiplications between complex numbers require four multiplications between real numbers, and second, there are delays in processing due to data hazards. In this paper, we propose an effective way to handle these issues by substituting them using parallelization based on the sum of complex conjugates. This approach aims to reduce the multiplication cost, thereby improving processing time compared to IFFT.

DOI： 10.1109/ICCE-Taiwan62264.2024.10674426

Scopus

researchmap

An In-depth Analysis of Kerberos and Blockchain Integration on VANETs' Security and Performance

Maya Rahayu, Md Biplob Hossain, Samsul Huda, Md Arshad Ali, Yuta Kodera, Yasuyuki Nogami

11th IEEE International Conference on Consumer Electronics - Taiwan, ICCE-Taiwan 2024 391 - 392 2024

Publishing type：Research paper (international conference proceedings)

Vehicular Ad-Hoc Networks (VANETs) are essential for modern transportation systems, but their openness exposes them to cyber threats. Authentication in VANETs is challenged by dynamic topology and high mobility, requiring robust mechanisms. Previously, we proposed an authentication system for VANETs using blockchain and Kerberos. Kerberos authenticator messages are stored in a blockchain ledger accessible to the Trusted Authentication Server (TAS) and Roadside Units (RSUs), resulting in minimal signaling overhead and authentication delay. However, the performance of the blockchain itself was not evaluated. In this paper, we investigate the blockchain's performance in a simulated VANET environment with 100 vehicles, 4 RSUs, and 1 TAS. Using Ethereum blockchain and Omnet++ simulation, we assess the blockchain feasibility in VANET authentication scenarios.

DOI： 10.1109/ICCE-Taiwan62264.2024.10674056

Scopus

researchmap

Investigating Vulnerabilities in RSA-like Cryptosystems Through Lattice-based Attacks on Small Exponent Keys

Sou Kumagai, Shota Kanzawa, Samsul Huda, Yuta Kodera, Yasuyuki Nogami

11th IEEE International Conference on Consumer Electronics - Taiwan, ICCE-Taiwan 2024 585 - 586 2024

Publishing type：Research paper (international conference proceedings)

RSA-like cryptosystems, based on the original RSA algorithm, are gaining attention as potential alternatives to traditional techniques. They offer better security, efficiency, flexibility, and compatibility in cryptography. However, recent studies have uncovered vulnerabilities in these variants, particularly to Wiener-type attacks, despite initial expectations of security comparable to RSA. In this paper, we further investigate the security vulnerabilities of a variant of this technique, namely the Murru-Saettone (MS) cryptosystem. Our investigation employs lattice-based attacks to exploit weaknesses in systems with small private keys. It demonstrates that if the private exponent is smaller than N1/4, where N is the product of two distinct balanced primes, the small exponent of the cryptosystem can be recovered.

DOI： 10.1109/ICCE-Taiwan62264.2024.10674633

Scopus

researchmap

A Highly Secure and Accurate System for COVID-19 Diagnosis from Chest X-Ray Images

Tuy Tan Nguyen, Tianyi Chen, Ian Philippi, Quoc Bao Phan, Shunri Kudo, Samsul Huda, Yasuyuki Nogami

Midwest Symposium on Circuits and Systems 980 - 984 2024

Publishing type：Research paper (international conference proceedings)

Global healthcare systems face growing pressure as populations rise. This can lead to longer wait times and an increased risk of treatment delays or misdiagnosis. Artificial intelligence (AI) diagnostic systems are being developed to address these challenges, but concerns exist about their accuracy and data security. This study introduces a robust AI telehealth system that offers a two-pronged approach. It utilizes a cutting-edge image analysis method, vision transformer, to enhance diagnostic accuracy, while also incorporating post-quantum cryptography algorithm, Kyber, to ensure patient privacy. Furthermore, an interactive visualization tool aids in interpreting the diagnostic results, providing valuable insights into the model's decisionmaking process. This translates to faster diagnoses and potentially shorter wait times for patients. Extensive testing with various datasets has demonstrated the system's effectiveness. The optimized model achieves a remarkable 95.79% accuracy rate in diagnosing COVID-19 from chest X-rays, with the entire process completed in under five seconds.

DOI： 10.1109/MWSCAS60917.2024.10658795

Scopus

researchmap

SnapSafe: Enabling Selective Image Privacy Through YOLO and AES-Protected Facial Encryption with QR Code

Andri Santoso, Samsul Huda, Tuy Tan Nguyen, Yuta Kodera, Yasuyuki Nogami

2024 International Technical Conference on Circuits/Systems, Computers, and Communications, ITC-CSCC 2024 2024

Publishing type：Research paper (international conference proceedings)

In the digital age, organizations commonly use photos to share their activities for brand visibility. However, it is crucial to safeguard the facial information of individuals depicted in these photos to prevent potential misuse. In this paper, we propose a single-party security system for enabling selective image privacy with a focus on facial regions, called SnapSafe. It adopts the YOLOv8 deep learning model for face detection, AES encryption for security, and QR technology to store the encrypted face region coordinates. Our evaluation assesses the effectiveness of the system through the usability of its face locking and unlocking processes, alongside an analysis of its running time. The results demonstrate the system's proficiency in executing face protection tasks with high accuracy and minimal time overhead, thus indicating its suitability for real-time applications.

DOI： 10.1109/ITC-CSCC62988.2024.10628222

Scopus

researchmap

Reducing Fruitless Cycles in Pollard's Rho Method with SFM for Efficient ECDLP Attacks on BN Curves

Ryuichi Kato, Takuro Manabe, Shota Kanzawa, Samsul Huda, Yuta Kodera, Takuya Kusaka, Yasuyuki Nogami

2024 International Technical Conference on Circuits/Systems, Computers, and Communications, ITC-CSCC 2024 2024

Publishing type：Research paper (international conference proceedings)

Pollard's rho method is a widely used probabilistic algorithm for solving the Elliptic Curve Discrete Logarithm Problem (ECDLP). To enhance its efficiency, the skew Frobenius mapping (SFM) has been incorporated into the method, reducing the average number of steps required to find a collision. However, the combination of Pollard's rho method with SFM introduces the occurrence of fruitless cycles, which are computational loops that do not contribute to solving the ECDLP and hinder the overall attacking efficiency. This paper presents a comprehensive analysis of fruitless cycles in Pollard's rho method with SFM, focusing on their impact when applied to Barreto-Naehrig (BN) curves. We propose a scheme to mitigate the occurrence of fruitless cycles by optimizing the seed point selection to be adaptive according to situation. To validate the effectiveness of our proposed method, we perform comprehensive experiments and compare the results with the conventional Pollard's rho method with SFM. The results indicate that the proposed approach significantly reduces the occurrence of fruitless cycles, leading to faster and more efficient ECDLP attacks on BN curves.

DOI： 10.1109/ITC-CSCC62988.2024.10628268

Scopus

researchmap

Evaluation of IDS model by improving accuracy and reducing overfitting using stacking LSTM

Muhammad Bisri Musthafa, Samsul Huda, Md Arshad Ali, Yuta Kodera, Yasuyuki Nogami

Digest of Technical Papers - IEEE International Conference on Consumer Electronics 2024

Publishing type：Research paper (international conference proceedings)

Utilizing Machine Learning (ML) techniques to construct models for Intrusion Detection Systems (IDS) has emerged as a leading strategy to enhance security measures. Previously, we applied an ensemble of Support Vector Machines (SVM) for model training, achieving an accuracy detection rate of 88.6% on familiar datasets. However, it performs poorly on unknown datasets due to overfitting problems. This work aims to improve accuracy and tackle overfitting problems by implementing stacked Long Short-Term Memory (LSTM) networks. Furthermore, analysis of variance (ANOVA) with F-test is used to find the important feature. The synthetic minority oversampling technique (SMOTE) algorithm is applied to obtain balanced data. Several experiments were conducted to analyze the performance of the proposed approach, including investigating stacked LSTM with tuning hyperparameters. The proposed method achieved the highest prediction accuracy of 97.23% and an overfitting value of 0.24%.

DOI： 10.1109/ICCE59016.2024.10444231

Scopus

researchmap

A Blockchain-based Approach with zk-SNARKs for Secure Email Applications.

Md. Biplob Hossain, Maya Rahayu, Md. Arshad Ali, Samsul Huda, Yuta Kodera, Yasuyuki Nogami

Int. J. Netw. Comput. 14 ( 2 ) 225 - 247 2024

Evaluation of IDS model by improving accuracy and reducing overfitting using stacking LSTM.

Muhammad Bisri Musthafa, Samsul Huda, Ali Md. Arshad, Yuta Kodera, Yasuyuki Nogami

ICCE 1 - 5 2024

IoT-Enabled Plant Monitoring System with Power Optimization and Secure Authentication

Samsul Huda, Yasuyuki Nogami, Maya Rahayu, Takuma Akada, Md Biplob Hossain, Muhammad Bisri Musthafa, Yang Jie, Le Hoang Anh

Computers, Materials and Continua 81 ( 2 ) 3165 - 3187 2024

Publishing type：Research paper (scientific journal)

Global food security is a pressing issue that affects the stability and well-being of communities worldwide. While existing Internet of Things (IoT) enabled plant monitoring systems have made significant strides in agricultural monitoring, they often face limitations such as high power consumption, restricted mobility, complex deployment requirements, and inadequate security measures for data access. This paper introduces an enhanced IoT application for agricultural monitoring systems that address these critical shortcomings. Our system strategically combines power efficiency, portability, and secure access capabilities, assisting farmers in monitoring and tracking crop environmental conditions. The proposed system includes a remote camera that captures images of surrounding plants and a sensor module that regularly monitors various environmental factors, including temperature, humidity, and soil moisture. We implement power management strategies to minimize energy consumption compared to existing solutions. Unlike conventional systems, our implementation utilizes the Amazon Web Services (AWS) cloud platform for reliable data storage and processing while incorporating comprehensive security measures, including Two-Factor Authentication (2FA) and JSON Web Tokens (JWT), features often overlooked in current agricultural IoT solutions. Users can access this secure monitoring system via a developed Android application, providing convenient mobile access to the gathered plant data. We validate our system’s advantages by implementing it with two potted garlic plants on Okayama University’s rooftop. Our evaluation demonstrates high sensor reliability, with strong correlations between sensor readings and reference data, achieving determination coefficients (R2) of 0.979 for temperature and 0.750 for humidity measurements. The implemented power management strategies extend battery life to 10 days on a single charge, significantly outperforming existing systems that typically require daily recharging. Furthermore, our dual-layer security implementation utilizing 2FA and JWT successfully protects sensitive agricultural data from unauthorized access.

DOI： 10.32604/cmc.2024.058144

Scopus

researchmap

An implementation of ELiPS-based Ciphertext-Policy Attribute-Based Encryption

Le Hoang Anh, Yuta Kawada, Samsul Huda, Md. Arshad Ali, Yuta Kodera, Yasuyuki Nogami

2023 Eleventh International Symposium on Computing and Networking Workshops (CANDARW) 2023.11

A Proposal of IoT Application for Plant Monitoring System with AWS Cloud Service.

Samsul Huda, Yasuyuki Nogami, Takuma Akada, Maya Rahayu, Md. Biplob Hossain, Muhammad Bisri Musthafa, Le Hoang Anh, Yang Jie

SmartNets 1 - 5 2023

Evaluation of machine learning based optimized feature selection approaches and classification methods for Intrusion Detection System.

Muhammad Bisri Musthafa, Ali Md. Arshad, Samsul Huda, Yuta Kodera, Takuya Kusaka, Yasuyuki Nogami

ICCE-Taiwan 285 - 286 2023

　 PREV - NEXT 　

< NOGAMI Yasuyuki >

Papers - NOGAMI Yasuyuki