ICT Express最新文献

In this study, we present the Discriminative Temporal Shift Module (D-TSM), an enhancement of the Temporal Shift Module (TSM) for action recognition. TSM has limitations in capturing intricate temporal dynamics due to its simplistic feature shifting. D-TSM addresses this by introducing a subtraction operation before the shifting. This enables the extraction of discriminative features between adjacent frames, thereby allowing for effective action recognition where subtle motions serve as crucial cues. It preserves TSM’s foundational philosophy, prioritizing minimal computational overhead and no additional parameters. Our experiments demonstrate that D-TSM significantly improves performance of TSM and outperforms other leading 2D CNN-based methods.

Implementation of malware detection using Artificial Intelligence (AI) has emerged as a significant research theme to combat evolving various types of malwares. Researchers implement various detection mechanisms using shallow and deep learning models to counter new malware, and they continue to develop these mechanisms today. However, in the field of malware detection using AI, there are difficulties in collecting data, and it is difficult to compare research content and performance with related studies. Meanwhile, the number of well-organized papers is not sufficient to understand the overall research flow of these related studies. Before starting new research, researchers need to analyze the current state of research in the malware detection field they want to study. Therefore, based on these requirements, we present a summary of the general criteria related to malware detection and a classification table for detection mechanisms. Additionally, we have organized many studies in the field of various types of malware detection so that they can be viewed at a glance. We hope that the provided survey can help new researchers quickly understand the research flow in the field of AI-based malware detection and establish the direction for future research.

Recently, as data demand has increased owing to the rapidly increasing demand for wireless devices and the influence of data traffic, various technologies are being developed to support it. Among them, millimeter-wave (mmWave) frequencies with rich spectra and high data-transmission rates suffer from the problem of large path loss. Accordingly, there is a growing interest in unmanned aerial vehicles (UAVs) and reconfigurable intelligent surfaces (RISs), which can be utilized advantageously to reconstruct wireless communication environments. Therefore, this work considers a large-scale system comprising a number of users and Flying RISs, combining UAVs and RISs to increase algorithm utilization. We propose a deep neural network-based algorithm that places Flying RISs in an appropriate location so that they can support as many users as possible. Simulation results confirmed that the proposed technique could place Flying RISs in an efficient location with higher accuracy and speed in large-scale systems compared to existing techniques.

In CRNs, it is crucial to develop an efficient and reliable spectrum detector that consistently provides accurate information about the channel state. In this work, we investigate a CSS in a fully-distributed environment where all secondary users (SUs) are equipped with directional antennas and make decisions based solely on their local knowledge without information sharing between SUs. First, we establish a stochastic sequential optimization problem, which is an NP-hard, that maximizes the SU’s detection accuracy by the dynamic and optimal control of the energy sensing/detection threshold. It can enable SUs to select an available channel and sector without causing interference to the primary network. To address it in a distributed environment, the problem is transformed into a decentralized partially observed Markov decision process (Dec-POMDP) problem. Second, in order to determine the best control for the Dec-POMDP in a practical environment without any prior knowledge of state–action transition probabilities, we develop a multi-agent deep deterministic policy gradient (MADDPG)-based algorithm, which is referred to as MA-DCSS. This algorithm adopts the centralized training and decentralized execution (CTDE) architecture. Third, we analyzed its computational complexity and showed the proposed approach’s scalability by the polynomial computational complexity, in terms of the number of channels, sectors, and SUs. Lastly, the simulation confirms that the proposed scheme provides enhanced performance in terms of convergence speed, accurate detection, and false alarm probabilities when it is compared to baseline algorithms.

With the rapid technological development, identifying the attackers behind cyber-attacks is getting more sophisticated. To cope with this phenomenon, the current process of cyber-threat attribution includes features like tactics techniques and procedures (TTP), tools, target country/ company and application. They do not include attacker context and motives; thus, they demand more refined traits. Adding behavioral features to this process is essential to better understand the attacker’s context, motivations and goals. This research study accentuates the impact of adding behavioral features with existing technical features in determining the actual actor. The behavioral features are extracted from Threat actor encyclopedia, a dataset published by Thai CERT. This research investigation also analyzes the impact of hybrid features (technical & and behavioral). For this procedure, the best features are chosen by implementing feature selection techniques. For empirical results, we use the threat actor encyclopedia, a data set published by Thai Cert, for extraction of behavioral attributes. With this augmentation, we achieve elevated results of 97%, 98.8%, 97%, and 97.2% in terms of accuracy, precision, recall and F1-measure using machine/deep learning algorithms.

In this study, the secrecy performance of reconfigurable intelligent surfaces (RIS)-aided wireless networks in the existence of multiple interferers towards the destination is investigated. In particular, three critical issues in the design of secure RIS-assisted networks are examined: effects of interferers, operation of multiple eavesdroppers (colluding and non-colluding), and benefit of RISs. To examine their effects, the analytical expressions of secrecy outage probability are derived in a closed form. Additionally, asymptotic analyses at a high signal-to-noise ratio (SNR) regime are provided. Finally, the analytical results are validated through numerical simulations.

While the spreading of the Internet of Things continues beyond expectations, the security of networking technologies used in this context remains an open issue. This paper provides a comprehensive overview of the state of the art on the security of Low Power Wide Area Networks (LPWANs), with a focus on Sigfox, LoRaWAN, and Narrowband Internet of Things. The paper covers five main areas: (1) security requirements and their implementation in these networks, such as authentication, encryption, access control, and key management; (2) categorization of attacks and threat modeling, with the identification of the attack vectors and the presentation of an attack categorization and analysis; (3) a detailed explanation of attacks documented on Sigfox, LoRaWAN, and Narrowband Internet of Things, examining the underlying vulnerabilities exploited, outlining potential consequences, and discussing countermeasures proposed to mitigate these attacks; (4) security enhancements proposed to address vulnerabilities in each network; (5) the integration of LPWANs with 5G and the consequent security challenges. This survey constitutes an important and missing resource for the study and the development of secure Internet of Things solutions based on Low Power Wide Area Networks, raising awareness of potential threats, and guiding future research efforts towards strengthening the security of these networks and of the broader IoT landscape.

Synthetic aperture radar (SAR) is a crucial active imaging technology in remote sensing, offering valuable information for applications like climate monitoring, environmental analysis, and ship surveillance. Ship detection in SAR images remains challenging due to diverse vessel types and environmental interference, especially in inshore areas, despite the proven effectiveness of deep learning-based algorithms. This paper presents an efficient deep learning method named you only look once-shuffle reparameterized blocks with dynamic head (YOLO-SRBD) based on YOLOv8. Additionally, post-processing incorporates the soft non-maximum suppression to enhance precision. Experiments conducted on SAR image datasets demonstrate that the proposed method surpasses the original YOLOv8 both qualitatively and quantitatively, highlighting its feasibility for practical applications. The detection accuracy of the proposed YOLO-SRBD in the high resolution SAR images dataset rose from 89.9% to 91.3%, and the average precision increased from 66.7% to 74.3%, showing significant performance enhancement.

The surgical navigation system enhances surgical safety and accuracy by providing precise guidance. However, traditional pose estimation algorithms lack real-time performance and accuracy. To address this issue, a multi-average Long Short Term Memory (LSTM) prediction network is designed to maintain sensitivity in estimating the position of surgical instruments and track their random motion trends. Additionally, the spatial coordinates of positioning markers are applied back to the imaging plane, reducing the recognition range and improving algorithm running speed. Experimental results show that the average time of estimation is less than 1ms while ensuring the prediction effect.

Beyond 6G services and applications demand high and efficient processing capacity due to the massive connectivity of users equipment (UEs). However, the high computational capability and energy consumption of UEs are limited, which becomes a main challenge to overcome. Multi-access edge computing (MEC) has recently been studied widely as it can potentially assist complex tasks executed at UEs. Furthermore, several techniques have been proposed to optimize task offloading among users. Thus, another challenge in MEC is emerging due to the fact that mobile users do not always have a line-of-sight (LoS) to the base station (BS) due to the blocking object. Therefore, it can affect users data rate and result in incremental energy consumption. This research introduces the concept of reconfigurable intelligence surfaces (RIS) to support multiple-input-single-output (MISO) base stations (BS) in both uplink (UL) and downlink (DL) using BCD algorithms. While previous studies concentrate on enhancing task offloading and neglecting inter-user interference, this study suggests an optimization approach for UL and DL data rates, as well as minimizing task offloading delays. The results indicate that optimizing task placement, phase shift, and precoding can reduce the duration of task offloading.