Advanced Information and Communication
ISSN: 3106-1443 (Print)
ISSN: 3106-1451 (Online)
FBFS: fog backdoor attack based on feature similarity
Most existing backdoor attacks embed triggers in images in ways that are either conspicuous to human observers or easily detected by feature-space defenses, thereby sacrificing stealthiness or robustness. To address these issues, we propose a fog backdoor attack based on feature similarity (FBFS), which enhances the visual concealment of backdoor triggers as well as the feature-space homogeneity between poisoned and clean samples. Specifically, FBFS employs a standard optical model to simulate natural fog as a visually plausible trigger and injects it into image samples. Additionally, a feature similarity penalty term is incorporated into the loss function to enforce consistency in the feature representations of poisoned and clean samples, thereby evading defenses that rely on latent separability. Experiments conducted on Canadian Institute for Advanced Research 10-class dataset (CIFAR-10), German Traffic Sign Recognition Benchmark (GTSRB), and a subset of ImageNet demonstrate that, under a 10% poisoning rate, FBFS achieves over 90% attack success rate while maintaining clean sample accuracy above 85%. Moreover, detection rates under representative feature-space defense methods, including activation clustering and spectral signature analysis, remain below 40%, demonstrating that the proposed method effectively balances attack performance and resistance to detection, exhibiting both stealth and robustness.
Performance analysis and prediction of physical layer security over α-Beaulieu-Xie shadowed fading channels
This paper investigates the secrecy performance of α-Beaulieu-Xie shadowed fading channels based on the Wyner’s wiretap model. In particular, the lower bound expression of secure outage probability (SOP) and the exact expression of strictly positive secrecy capacity (SPSC) are derived. Monte Carlo simulation verifies the accuracy of theoretical analysis. The results indicate that the larger αD, mDX and mDY or smaller ΩDX can enhance the performance of model. Moreover, to predict the security performance of the model, the self long short-term memory (Self-LSTM) algorithm is proposed. We found that the Self-LSTM algorithm has better prediction performance by comparing with LSTM, DenseNet and convolutional neural network (CNN) network. Compared with the LSTM, the prediction accuracy of Self-LSTM is increased by 55.95%, the time complexity is decreased by 81.03%. Comparing to the DenseNet network, the precision of the Self-LSTM algorithm is improved by 64.20%. It can be concluded that the proposed Self-LSTM has higher prediction accuracy and lower time complexity.
BS-ADV: a secure black-box image steganography framework enhanced by adversarial sample attack
Mainstream image steganography approaches struggle to evade increasingly accurate steganalyzers, while many security-enhancement techniques require white-box access to the steganographic encoder, hindering deployment in black-box settings. We address this gap with a framework that achieves strong security and high image quality without access to encoder internals. Our method Black-box Steganography via Transferable Adversarial Attack (BS-ADV) adds small, transferable adversarial perturbations to stego images using gradients from a chosen steganalysis model, yet remains independent of the steganographic encoder. Building on this idea, we instantiate two variants: FGSM-adv, which applies a single-step Fast Gradient Sign Method to inject fixed-sign perturbations, and PGD-adv, which performs multi-step Projected Gradient Descent to enhance the robustness and security of the resulting adversarial stego images. Experiments on public BOSSBase 1.01 (Break Our Steganographic System Base v1.01) and BOWS (Break Our Watermarking System 2) datasets show that BS-ADV substantially outperforms baseline approaches against both feature-based and convolutional neural network (CNN)-based steganalyzers. Beyond conventional algorithms, we further validate BS-ADV with DeepSteganography, HiNet, and CRoSS (diffusion model makes controllable, robust and secure image steganography), a coverless steganography scheme built on Stable Diffusion, demonstrating broad generality and adaptability. Overall, BS-ADV improves the security and robust-ness of image steganography while preserving image quality and reliable payload recovery, making it well suited for practical black-box deployment.
A survey on insertion/deletion error detection and correction: progress and future directions
The detection and correction of insertion/deletion (indel) errors have become increasingly critical in domains such as traditional mobile communication systems, the Internet of Things (IoT), smart homes, smart healthcare, vehicular networks, and large-scale urban infrastructure, establishing it as a prominent research focus. As a typical form of synchronization error, the randomness and asymmetry of indel errors severely disrupt symbol alignment and induce significant synchronization drift, thereby imposing substantial challenges on reliable data transmission. This paper systematically reviews methodologies for detecting and correcting indel errors, tracing their evolution from model-driven to data-driven paradigms. First, we summarize the traditional technical framework, which includes synchronization markers, edit distance (ED) codes, sequence alignment, trellis/convolutional structures, and probabilistic models, with an analysis of their theoretical foundations, representative algorithms, and applicable scenarios. Next, we focus on recent advances in deep learning (DL)-based synchronization recovery methods and semantic communication-driven intelligent error correction frameworks, highlighting their distinct advantages over conventional approaches in handling complex channels and unstructured data. Finally, we outline the current research landscape and key challenges in this field and propose future directions for emerging scenarios such as 6th Generation (6G) ultra-reliable communication, satellite links, and ultra-high-density storage. This review aims to provide comprehensive insights and guidance for the design of synchronization and error correction mechanisms in next-generation communication systems.
Optical reconfigurable intelligent surfaces assisted visible light positioning and communications: performance analysis and design optimization
This paper investigates the integration of visible light positioning and communication (VLP&C) facilitated by optical reconfigurable intelligent surfaces (ORIS) to address line-of-sight (LoS) blockage challenges within indoor environments. In contrast to conventional VLP&C systems, which experience significant performance deterioration under LoS blockage, the proposed ORIS-assisted framework dynamically adjusts the reflection patterns to establish reliable non-LoS (NLoS) links. Initially, a comprehensive system model is formulated, encompassing the physical properties of ORIS, including an analysis of time delays and strategies for ORIS deployment. Subsequently, the Cramér-Rao lower bound (CRLB) for positioning accuracy is rigorously derived from the underlying signal models, thereby providing a realistic theoretical performance benchmark. Additionally, closed-form expression for the average mutual information (AMI) and bit error rate (BER) of the communication subsystem are developed, accounting for the finite-alphabet characteristics of on-off keying (OOK) modulation. The study further investigates the trade-offs between positioning accuracy and communication performance across various system parameters, such as the number of ORIS reflection units, half-power angle, and spatial distribution of users. Extensive simulation results demonstrate that the proposed ORIS-assisted system attains centimeter-level positioning accuracy alongside reliable communication performance, even in scenarios where LoS links are blocked. The theoretical findings are validated through Monte Carlo simulations, and the practical implementation challenges are discussed to inform future real-world deployments.
Energy efficiency optimization of STAR-RIS-aided ISAC systems based on user fairness
In this paper, the communication energy efficiency (EE) of simultaneously transmitting and reflecting reconfigurable intelligent surface (STAR-RIS) aided integrated sensing and communications (ISAC) systems underlying a near-field scenario is investigated, where the dual-functional base station (DFBS) serves multiple users and senses multiple targets simultaneously. To ensure user fairness, we formulate an optimization problem that maximizes the minimum (max-min) communication EE while satisfying the minimum target illumination power requirement, the maximum transmission power budget, and the hardware constraints of STAR-RIS under its three operation modes. The formulated max-min optimization problem exhibits non-convexity due to the high coupling among the optimization variables. So as to resolve this issue, the fractional programming is first leveraged to transform the objective function into a more tractable structure. Then, the original max-min problem is transformed into an equivalent maximization problem via introducing the auxiliary variable. Next, we propose an alternating optimization framework to decouple the newly reformulated maximization problem into several sub-problems, which are optimized iteratively until convergence. Finally, the outcomes from the simulations are executed to confirm the advantages and effectiveness of the schemes we have introduced.
