IET Communications最新文献

Customer service primarily involves interaction with clients through phone calls. Precise keyword extraction from customer complaint texts facilitates the implementation of intelligent task assignment and efficient response systems. However, existing keyword extraction technologies perform sub-optimally in the customer service domain of telecommunications operators and require substantial manual word segmentation. Given the pronounced clustering of customer service data, this research introduces a synonym matching approach and a few-shot learning-based method tailored for extracting keywords in this sector. This enables model training with minimal labelled data and computational resources. Using a dataset generated from the transcription of customer service calls, the proposed model demonstrates a 24.94% improvement in accuracy compared to popular existing methods.

In this paper, a decode-and-forward relay-assisted device-to-device (D2D) network is examined over novel multi-cluster fluctuating two-ray (MFTR) fading channels. All communication links are functioning in terahertz (THz) conditions. Co-channel interference (CCI) is considered as well. We assume an eavesdropper is also present near the receiver and overhears the relay's signal. With the help of characteristic functions, expressions of outage, success probability, capacity with outage, secrecy outage, probability of strictly positive secrecy capacity and intercept probability are presented. These expressions are functions of MFTR fading conditions, THz channel parameters and distances between various nodes in the system. Numerical results based on the derived expressions are discussed under various scenarios.

Nowadays, deep neural network (DNN) partition is an effective strategy to accelerate deep learning (DL) tasks. A pioneering technology, computing and network convergence (CNC), integrates dispersed computing resources and bandwidth via the network control plane to utilize them efficiently. This paper presents a novel network-cloud (NC) architecture designed for DL task inference in CNC scenario, where network devices directly participate in computation, thereby reducing extra transmission costs. Considering multi-hop computing-capable network nodes and one cloud node in a chain path, leveraging deep reinforcement learning (DRL), we develop a joint-optimization algorithm for DNN partition, subtask offloading and computing resource allocation based on deep Q network (DQN), referred to as POADQ, which invokes a subtask offloading and computing resource allocation (SORA) algorithm with low complexity, to minimize delay. DQN searches the optimal DNN partition point, and SORA identifies the next optimal offloading node for next subtask through our proposed NONPRA (next optimal node prediction with resource allocation) method, which selects the node that exhibits the smallest predicted increase in cost. We conduct some experiments and compare POADQ with other schemes. The results show that our proposed algorithm is superior to other algorithms in reducing the average delay of subtasks.

Spoofing attacks, which impersonate legitimate users, pose significant challenges to communication security by exploiting the dependence of received signal strength (RSS) on the spatial position of the transmitter. An enhanced GA_BPNNC algorithm was proposed to learn the distribution of RSS vectors to classify positions, distinguishing between attackers and legitimate users. The algorithm's performance was evaluated using real datasets which are collected in a room of the University of California, San Diego, demonstrating accuracy and robustness compared to existing neural network models. Our method achieved accuracy of over 95% and execution time of less 0.56 s. The experimental results indicate that the proposed algorithm outperforms other state-of-the-art algorithms, with the advantage of not relying on specific communication protocols, offering high throughput and fast decision-making capabilities.

Mobile devices pose several distinct challenges from a security perspective. First, they have varied and ephemeral network connections, often using a cellular provider network as a backup option when connectivity is not available via wireless local access networks. This varied network connectivity makes it difficult to comprehensively deploy in-network solutions, such as firewalls or intrusion detection systems, since they would have to be active in every network the device would use. Second, with personally owned devices, the device owner may have security goals and privacy priorities that are distinct from organizations that provide connectivity or data assets, such as employers or schools. These complex relationships may complicate efforts to protect the devices. This paper explores a technique that runs on the mobile device endpoints to learn about the usage patterns associated with the device, in order to enforce network policy. We explore sensors that examine the mobile device's user interface, using physical inputs via finger taps, and that link them with the network activity on the device. We incorporate with allow-list policies that can be provided by organizations to make on-device access control decisions. Using IP address and DNS host name allow-lists as a baseline, we explore the accuracy of interface-aware allow-lists. We find the interface-aware allow-lists can reach over 98.5% accuracy, even when user-specified destinations are used, greatly exceeding the baseline accuracy. Our performance evaluation indicates our approach introduces a median of 3.87 ms of overall delay with low CPU usage.

Security in supply chain management has become a critical concern due to the increasing complexity and interconnectivity of global supply chains. Therefore, the need for robust security measures to protect against various risks becomes paramount. The traditional supply chain management system does not ensure several parameters, such as regulatory compliance, immutability, latency, scalability, traceability, and authenticity. The main contributions of our proposed system are to integrate a supply chain management system using blockchain to ensure the above parameters, mitigate the challenges associated with blockchain-based methods, and reduce deployment and operational costs associated with the proposed blockchain-based system. The proposed model includes smart contract mechanisms to enhance security, efficiency, and transparency by recording every transaction and action on the blockchain. Its immutable behaviour minimises the risk of fraud, is tamper-resistant, and ensures security. We used a consensus mechanism to ensure integrity and security by validating the transaction within a blockchain-based supply chain management system. Proof of Work (PoW) is a consensus algorithm used in our model to prevent single points of failure and reduce the risk of manipulation or fraud. This paper describes the hash generation process, the digital signature generation process, the digital signature verification process, and the Merkle tree construction process. The security analysis ensures that our proposed model can detect all possible security threats and ensure the security of the supply chain management system.

Integrating intelligent reflecting surfaces (IRS) with unmanned aerial vehicles (UAVs) presents a promising approach for future energy-efficient wireless communications. This paper proposes an adaptive framework that dynamically balances energy harvesting (EH) efficiency and system throughput by adjusting the required EH efficiency based on the UAV's power levels and communication needs. Utilising real-coded genetic algorithm (RCGA), the framework effectively tackles challenges posed by multi-user interference (MUI) and imperfect channel estimation (CE). Our results demonstrate that the RCGA-based approach outperforms deep reinforcement learning (DRL) methods, delivering superior energy harvesting and throughput in realistic conditions. The adaptive EH strategy not only optimises throughput performance but also ensures efficient UAV energy management, particularly in dynamic and energy-constrained environments, making it a robust solution for sustained UAV operations in dynamic and energy-constrained environments.

To address the issue of obstacles in wireless channels, this paper proposes a time-division switching SR-DCSK-WPC scheme based on reconfigurable intelligent surface (RIS)-assisted communication. The scheme consists of an energy source (ES) transmitter, a RIS, and a terminal node D with an energy buffer. Firstly, the ES transmits an energy signal composed of multiple chaotic sequences. This energy signal powers the terminal node, and the proportion coefficient of the energy signal is adjusted by controlling the repetition count of the chaotic sequences. Secondly, the RIS adjusts the amount of transmitted energy and assists in bypassing obstacles, further enhancing the performance of the wireless power transfer (WPC) system. Finally, after collecting energy, the terminal node repeatedly transmits a short reference signal composed of multiple chaotic sequences using the short reference technique to improve energy efficiency and transmission rate. Both the ES and D sides employ a repeated transmission method to adjust the energy coefficient and short reference coefficient, respectively. Although their functions differ, their structures are consistent and simple. Simulations and numerical calculations verify that, compared to existing RIS-assisted DCSK-WPC models, the proposed model improves overall transmission efficiency and saves energy without compromising the bit error rate.

This paper investigates a buffer-aided cooperative non-orthogonal multiple access scheme for simultaneous wireless information and power transfer systems, where the near users are equipped with data and energy buffers. To improve the overall energy efficiency, the transmission mode of the considered system adaptively chooses the base station or near-user transmission mode. Given the target quality of service and limited resources, the long-term power consumption minimization is formulated by a mixed integer non-linear programming problem, where the model selection, user scheduling, and power consumption are jointly optimized. To efficiently tackle such a challenging problem, we transform the original problem into an instantaneous non-convex problem by employing the Lyapunov optimization framework. Then, using the successive convex approximation algorithms, we approximate the instantaneous non-convex problem as a linear programming problem, where the Karush–Kuhn–Tucker solution of the power allocation can be obtained. The optimality and complexity of the proposed scheme are theoretically analyzed. The simulation results show that the proposed scheme can achieve a near-optimal performance and significantly improves the energy efficiency compared with the conventional transmission schemes.

Cloud computing (CC) has emerged as a transformative technology, offering customers unprecedented access to extensive computing resources and the diverse services for hosting various applications. However, this environment comes with several challenges. While cloud users seek optimal resources to cater to their specific requirements, the prevalent scenario often involves trading more monetary resources for less computational time. Existing algorithms, mostly focused on optimizing individual variables, lack a holistic approach. Addressing these issues necessitates a new approach to combine these conflicting objectives. This research focuses on developing and improving a dynamic task-processing framework that can find and use the optimal resources in real-time. The focus extends to running applications of different types and levels of complexity on virtual machines (VMs) using the multi-objective adaptive particle swarm optimization (MAPSO) algorithm. The MAPSO handles the multi-objective problem using the weighted-sum approach. The system operates within predefined constraints to meet users' specific time limitations. Through comprehensive simulations on a wide range of datasets, the proposed methodology yields a set of non-dominated optimal solutions. This outcome is instrumental in improving critical quality of service (QoS) metrics, including processing time, execution costs, throughput, and task rejection ratios. The effectiveness of the MAPSO-based approach are evident in its capacity to improve these numerous QoS aspects, including processing time, execution cost, throughput, and task rejection ratio compared and clearly shows that it is superior to the existing algorithms, such as ant colony optimization (ACO), hybrid version of bat optimization algorithm and particle swarm optimization (BOA+PSO), and hybrid grey wolf optimization and artificial bee colony (GWO+ABC). The time complexity for completing the tasks of the MAPSO algorithm is reduced by 5%, executes each schedule's tasks faster by 5% to 13%, and calculated execution costs also get reduced when compared to ACO, BOA+PSO, and GWO+ABC. Moreover, the suggested methodology convincingly outperforms existing state-of-the-art methods in terms of computational performance. This study pioneers a unique solution in cloud service provisioning by integrating multi-objective optimization within a real-time resource allocation framework. The resulting combination of intelligent resource allocation and enhanced QoS metrics promises to change the way cloud-based application deployment is done. Ultimately, this work establishes a paradigm shift in balancing resource allocation and user-centric QoS optimization in cloud computing environments.