ICT Express最新文献

Recently, QUIC for the secure and faster connections has standardized but it is unclear that QUIC can cope with website fingerprinting (WF), a technique to infer visited websites from network traffic, since most existing efforts targeted TCP-induced traffic. To this end, we propose a novel QUIC WF technique based on Automated Machine Learning (AutoML). In our approach, we revisit traffic features appeared in literature, but relies on an AutoML framework to achieve best practice without manual intervention. Through experiments, we show that our technique outperforms state-of-the-art WF techniques with an F1-score of 99.79% and a 20-precision of 92.60%.

Federated learning (FL) has emerged as a promising distributed machine learning technique. It has the potential to play a key role in future Internet of Things (IoT) networks by ensuring the security and privacy of user data combined with efficient utilization of communication resources. This paper addresses the challenge of maximizing energy efficiency in FL systems. We employed simultaneous wireless information and power transfer (SWIPT) and multi-carrier non-orthogonal multiple access (MC-NOMA) techniques. Also, we jointly optimized power allocation and central processing unit (CPU) resource allocation to minimize latency-constrained energy consumption. We formulated an optimization problem using a Markov decision process (MDP) and utilized a deep deterministic policy gradient (DDPG) reinforcement learning algorithm to solve our MDP problem. We tested the proposed algorithm through extensive simulations and confirmed it converges in a stable manner and provides enhanced energy efficiency compared to conventional schemes.

Although estimating the azimuth using a geomagnetic sensor is very useful, the estimation error may be very large due to the surrounding geomagnetic disturbance. We proposed a novel method for preprocessing appropriately for geomagnetic and inertial sensor data to be suitable for the proposed Artificial Neural Network model and training method for the model. As a result, the probability of azimuth estimation error within 1 degree is 96.4% with regression estimation. For classification estimation, when the azimuth estimation probability is 90% or more, the probability that the azimuth estimation error is within 1 degree is 100%.

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.

The network of resource constraint devices, also known as the Low power and Lossy Networks (LLNs), constitutes the edge tire of the Internet of Things applications like smart homes, smart cities, and connected vehicles. The IPv6 Routing Protocol over Low power and lossy networks (RPL) ensures efficient routing in the edge tire of the IoT environment. However, RPL has inherent vulnerabilities that allow malicious insider entities to instigate several security attacks in the IoT network. As a result, the IoT networks suffer from resource depletion, performance degradation, and traffic disruption. Recent literature discusses several machine learning algorithms to detect one or more routing attacks. However, IoT infrastructures are expanding, and so are the attack surfaces. Therefore, it is essential to have a solution that can adapt to this change. This paper introduces a comprehensive framework to detect routing attacks within Low Power and Lossy Networks (LLNs). The proposed solution leverages deep learning by combining Restricted Boltzmann Machine (RBM) and Long Short-Term Memory (LSTM). The framework is trained on 11 network parameters to understand and predict normal network behavior. Anomalies, identified as deviations from the forecast trends, serve as indicators of potential routing attacks and thus address vulnerabilities in the RPL.

Digital twin (DT) technologies have been increasingly important and useful for wireless communications. In particular, to support soaring wireless traffic with limited frequency spectrum assigned to legacy cellular systems, efficient operation and management of wireless resources as well as preemptively prediction of future spectrum usage are crucially important. For such purpose, DT networks for current fourth-generation (4G) and fifth-generation (5G) networks are constructed in this paper, by simultaneously utilizing measurement data from user equipments (UEs) and geographical information and characteristics of 4G and 5G base stations (BSs) within a specific observation area. Representative case studies are provided to demonstrate the usefulness of DT enabled cellular network management and prediction. As a real or near real time DT application, the impact and benefit of dual connectivity and dynamic spectrum sharing between 4G and 5G networks are analyzed. As a non-real time DT application, long-term improvement of 5G networks such as densification of BSs, implementation of advanced multiple input and multiple output technologies, and assignment of additional spectrum are analyzed and compared.

In this paper, we design some new LDPC-coded orthogonal modulation (OM) schemes for high data rate transmissions (HDRT) in the navigation satellite systems. We analyze their error-performance utilizing soft-decision bit metrics and compare them with those of L61 and L62 signals in the quasi-zenith satellite system (QZSS) for centimeter-level augmentation services (CLAS). Compare to the L62 signals of QZSS, both schemes have higher data rates (14.6% increase) and essentially the better error performance at high SNR region. At the region where frame error rate (FER) $=10^{-3}$, one of the proposed schemes has better error performance of 1.4 dB in terms of carrier-to-noise ratio $(C/N_0)$.

In this paper, we have investigated the performance of energy efficiency (EE) for Intelligent Transportation Systems (ITS), which recently emerged and advanced to preserve speed as well as safe transportation expansion via a cooperative IRS-relay network. To improve the EE, the relay model has been integrated with an IRS block consisting of a number of passive reflective elements. We analyze the ITS in terms of EE, and achievable rate, with different signal-to-noise ratio (SNR) values under Nakagami-m fading channel conditions that help the system to implement in a practical scenario. From the numerical results it is noticed that the EE for the only relay, IRS, and proposed cooperative relay-IRS-aided network at SNR value of 100 dBm is 30, 17, and 48 bits/joule respectively. In addition, we compare the impact of multi-IRS with the proposed cooperative IRS-relay and conventional relay-supported ITS. Simulation results show that both the proposed cooperative IRS-relay-aided ITS network and multi-IRS-aided network outperform the relay-assisted ITS with the increase in SNR.

Despite the growing interest in using deep reinforcement learning (DRL) for drone control, several challenges remain to be addressed, including issues with generalization across task variations and agent training (which requires significant computational power and time). When the agent’s input changes owing to the drone’s sensors or mission variations, significant retraining overhead is required to handle the changes in the input data pattern and the neural network architecture to accommodate the input data. These difficulties severely limit their applicability in dynamic real-world environments. In this paper, we propose an efficient DRL method that leverages the knowledge of the source agent to accelerate the training of the target agent under task variations. The proposed method consists of three phases: collecting training data for the target agent using the source agent, supervised pre-training of the target agent, and DRL-based fine-tuning. Experimental validation demonstrated a remarkable reduction in the training time (up to 94.29%), suggesting a potential avenue for the successful and efficient application of DRL in drone control.

Computing-aware networking (CAN) is introduced to unite the computing resources distributed in different platforms. This paper proposes a joint routing and resource allocation model to minimize the total operational cost in CAN. We formulate the problem as an integer linear programming problem. We introduce a polynomial-time algorithm for larger-size problems. The numerical results reveal that the introduced algorithm reduces the computation time 9.18 times, with increasing the objective no more than 4% compared to the optimal solution; the proposed model can reduce 70% of the total cost compared to a baseline adopting a two-stage strategy in our examined cases.