IET Communications最新文献

Scalability is one of the important parameters for mobile communication networks of the present generation and further to the future 5G and beyond networks. When a user is in motion transferring from one cell site to another, then the handover procedure becomes important in the sense that it ensures that a user gets consistent connection without interruption. Nevertheless, the classic handover process in cellular networks has some sort of drawback like causing service interruptions, affecting packet transmission, and increased latency which is highly uncongenial to the evolving applications which have stringent requirement to latency. To overcome these challenges and improve the mobile handover in 5G and future mobile networks, this article puts forth a predictive handover mechanism using reinforcement learning algorithm. The RL algorithm outperforms the ML algorithm in several aspects. Compared to ML, RL has a higher handover success rate (∼95% vs. ∼90%), lower latency (∼30 ms vs. ∼40 ms), reduced failure rate (∼5% vs. ∼10%), and shorter disconnection time (∼50 ms vs. ∼70 ms). This demonstrates the RL algorithm's superior ability to adapt to dynamic network conditions.

Reconfigurable intelligent surfaces (RISs) have emerged as propitious solution to configure random wireless channel into suitable propagation environment by adjusting a large number of low-cost passive reflecting elements. It is considered that narrowband downlink millimeter wave (mmWave) multiple-input multiple-output (MIMO) communication is aided by deploying an RIS. Large antenna arrays are used to counter the huge propagation loss suffered by the mmWave signals. Hybrid precoding in which precoding is performed in digital and analog domains is employed to reduce the number of costly and power-consuming radio frequency (RF) chains. Passive beamforming at RIS is designed together with precoder and combiner through joint optimization problem to minimize the mean square error between the transmit signal and the estimate of signal at the receiver. The optimization problem is solved by an iterative procedure in which solution to the non-convex reflecting coefficients design problem is approximated by extracting the phases of the solution to unconstrained problem without unit amplitude constraint of the reflecting elements. It is shown that the proposed design principle also applies to the wideband channel. Simulation results show that the proposed design delivers performance better than existing state-of-the-art solutions, but at lower complexity.

As organizations increasingly rely on network services, the prevalence and severity of Denial of Service (DoS) and Distributed Denial of Service (DDoS) attacks have emerged as significant threats. The cornerstone of effectively addressing these challenges lies in the timely and precise detection capabilities offered by advanced intrusion detection systems (IDS). Hence, an innovative IDS framework is introduced that seamlessly integrates the extended Berkeley Packet Filter (eBPF) with powerful machine learning algorithms—specifically Decision Tree (DT), Random Forest (RF), Support Vector Machine (SVM), and TwinSVM—enabling unparalleled real-time detection of DDoS attacks. This cutting-edge solution provides a robust and scalable IDS framework to combat DoS and DDoS threats with high efficiency, leveraging eBPF's capabilities within the Linux kernel to bypass typical user space constraints. The methodology encompasses several key steps: (a) Collection of data from the renowned CIC-IDS-2017 repository; (b) Processing the raw data through a meticulous series of steps, including transmission, cleaning, reduction, and discretization; (c) Utilizing an ANOVA F-test for the extraction of critical features from the preprocessed data; (d) Application of various ML algorithms (DT, RF, SVM, and TwinSVM) to analyze the extracted features for potential intrusion; (e) Implementing an eBPF program to capture network traffic and harness trained model parameters for efficient attack detection directly within the kernel. The experimental results reveal outstanding accuracy rates of 99.38%, 99.44%, 88.73%, and 93.82% for DT, RF, SVM, and TwinSVM, respectively, alongside remarkable precision values of 99.71%, 99.65%, 84.31%, and 98.49%. This high-speed, accurate detection model is ideally suited for high-traffic environments such as data centers. Furthermore, its foundational architecture paves the way for future advancements, including the potential integration of eBPF with XDP to achieve even lower-latency packet processing. The experimental code is available at the GitHub repository link: https://github.com/NemalikantiAnand/Project.

Advanced wireless communication is important in distribution systems for sharing information among Internet of Things (IoT) edges. Artificial intelligence (AI) analyzed the generated IoT data to make these decisions, ensuring efficient and effective operations. These technologies face significant security challenges, such as eavesdropping and adversarial attacks. Recent studies addressed this issue by using clustering analysis (CA) to uncover hidden patterns to provide AI models with clear interpretations. The high volume of overlapped samples in IoT data affects partitioning, interpretation, and reliability of CAs. Recent CA models have integrated machine learning techniques to address these issues, but struggle in the limited resources of IoT environments. These challenges are addressed by proposing a novel unsupervised lightweight distance clustering (DC) model based on data separation ($�\beta$). $�\beta$ raises the tension between samples using cannot-link relations to separate the overlap, thus DC provides the interpretations. The optimal time and space complexity enables DC-$�\beta$ to be implemented on on-edge computing, reducing data transmission overhead, and improving the robustness of the AI-IoT application. Extensive experiments were conducted across various datasets under different circumstances. The results show that the data separated by $�\beta$ improved the efficiency of the proposed solution, with DC outperforming the baseline model.

A $�K$-user cyclic shift-aided serial concatenated code (SCC) is proposed over a Gaussian multiple access channel (MAC). In this code, a cyclic shift-aided SCC, which integrates a regular repeat-accumulate code with rate-$��1�/�q�$ and cyclic shift spreading (CSS), is employed for each user. Here, CSS consists of a length-$�L$ bit spreading, a same chip-level interleaver and a user-specific cyclic shifter. This change avoids repeated complex interleaver generation operation and the main objective is to replace the random interleaver. At the receiver, iterative decoding is considered. For the design and optimization of the proposed code, a fixed point analysis (FPA) is developed to obtain the optimal $�q^{\ast }$ and $�L^{\ast }$ that achieves the maximum sum rate. Then, an analytic expression of collision probability is derived as a function of $�q$ and $�L$. Finally, the advantage is analysed by comparing the generation time of interleaver. Moreover, FPA provides lower computation complexity in optimizing system parameters than the traditional EXIT chart method. Numerical results show that the design and optimization of the proposed code is more accurate, and support distinguishing between the users. The proposed code not only outperforms the traditional IDMA scheme in the bit error rate performance, but also provides much lower generation time of interleaver, especially for multiuser scenarios.

The dual function radar and communication technology plays a crucial role in Internet of Vehicles. However, fewer studies have focused on optimizing the joint performance of sensing and communication in full-duplex Internet of Vehicles while considering fairness. Therefore, this paper considers the scenario that the full-duplex gNB employs dual function radar and communication technology to simultaneously achieve vehicle localization and communication with multiple half-duplex vehicles. By leveraging a reasonable beamforming scheme to mitigate the residual self-interference of the echo signal, an optimization problem that maximizing the joint performance metric of sensing and communication is formulated with limited power at the gNB. After problem transformation, a double descent iteration power allocation algorithm is proposed to solve the non-convex optimization problem and the complexity of the algorithm is analysed. Simulation results demonstrate that the proposed algorithm converges and could fairly improve both the sensing and communication performance.

In this work, MineStudio is introduced, a novel AI filmmaking framework designed to facilitate future creative collaborative networks. MineStudio uses a hybrid digitization approach that involves reconstructing 3D digital environments, capturing 2D live-action performances, employing AI tools to generate synthetic images and videos, and compositing with AI assistance. This method effectively addresses the main challenges in the current AI video generation, including consistency, directability, and issues with human actions and interactions. MineStudio has been utilized to create pioneering AI films, such as the love story “Next Stop Paris” and the sci-fi short film “Message in a Bot”, and has been recognized as a trailblazer in the AI filmmaking industry.

Specific emitter identification (SEI) plays an important role in enhancing physical layer transmission security. However, with the promotion of wireless technology, the environment is filled with a large number of unknown wireless signals. SEI will face a more challenging scenario referred to as “open set.” To cope with the above difficulties, an open-set recognition (OSR) model based on complex-valued convolutional neural network (CVCNN) is proposed. The CVCNN can adapt to IQ signal input and extract complex domain features. Furthermore, a novel inter-class loss is proposed to effectively improve the classification performance. Finally, the classifier is designed based on the incremental approach. It can continuously learn new classes to achieve the recognition of multiple unknown emitters. The experiments show that compared with the real-valued convolutional neural network and the single loss function, the accuracy is improved by 3.8% and 10%, respectively.

Scene text recognition has gained increasing attention in recent years, as it can connect products without an open interface in IoT. The non-local network is particularly popular in text recognition, as it can aggregate the temporal message of the input. However, existing text recognition methods based on RNN encoder-decoder structures encounter the problem of attention drift, especially in complex ship name recognition scenarios, because the features extracted by these methods are extremely similar. To address this problem, this paper proposes a novel text recognition approach named Multi Global Context-aware Transformer (MG-Cat). The proposed approach has two main properties: (1) a Global Context block that captures the global relationships among pixels inside the encoder, and (2) multiple global context-aware attention modules stacked in the encoder process. This way, the MG-Cat approach can learn a more robust intermediate feature representation in the text recognition pipeline. Moreover, the paper collected a new ship name dataset to evaluate the proposed approach. Extensive experiments were conducted on the collected dataset to verify the effectiveness of the proposed approach. The experimental results show the generalization ability of our squeeze-and-excitation global context attention module.

Deep learning techniques have shown promise in audio steganalysis, which involves detecting the presence of hidden information (steganography) in audio files. However, deep learning models are prone to overfitting, particularly when there is limited data or when the model architecture is too complex relative to the available data for VoIP steganography. To address these issues, new deep-learning approaches need to be explored. In this study, a new convolutional neural network for audio steganalysis, incorporating a multi-scale feature fusion method and an attention mechanism, was devised to enhance the detection of steganographic content in audio signals encoded with G729a. To improve the network's adaptability, a multi-scale parallel multi-branch architecture was employed, allowing characteristic signals to be sampled with varying granularities and adjusting the receptive field effectively. The attention mechanism enables weight learning on the feature information after multi-scale processing, capturing the most relevant information for steganalysis. By combining multiple feature representations using a weighted combination, the deep learning model's performance was significantly enhanced. Through rigorous experimentation, an impressive accuracy rate of 94.55% was achieved in detecting malicious steganography. This outcome demonstrates the efficacy of the proposed neural network, leveraging both the multi-scale feature fusion method and the attention mechanism.