Traditional unimodal sensing faces limitations in accuracy and capability, and its decoupled implementation with communication systems increases latency in bandwidth-constrained environments. Additionally, single-task-oriented sensing systems fail to address users’ diverse demands. To overcome these challenges, we propose a semantic-driven integrated multimodal sensing and communication (SIMAC) framework. This framework leverages a joint source-channel coding architecture to achieve simultaneous sensing, decoding, and transmission of sensing results. Specifically, SIMAC first introduces a multimodal semantic fusion (MSF) network, which employs two extractors to extract semantic information from radar signals and images, respectively. MSF then applies cross-attention mechanisms to fuse these unimodal features and generate multimodal semantic representations. Secondly, we present a large language model (LLM)-based semantic encoder (LSE), where relevant communication parameters and multimodal semantics are mapped into a unified latent space and input to the LLM, enabling channel-adaptive semantic encoding. Thirdly, a task-oriented sensing semantic decoder (SSD) is proposed, in which different decoded heads are designed according to the specific needs of tasks. Simultaneously, a multi-task learning strategy is introduced to train the SIMAC framework, achieving diverse sensing services. Finally, experimental simulations demonstrate that the proposed framework achieves diverse and higher-accuracy sensing services.

Despite the widespread adoption of vision sensors in edge applications, such as surveillance, video transmission consumes substantial spectrum resources. Semantic communication (SC) offers a solution by extracting and compressing information at the semantic level, but traditional SC without sensing capabilities faces inefficiencies due to the repeated transmission of static frames in edge videos. To address this challenge, we propose an SC with computer vision sensing (SCCVS) framework for edge video transmission. The framework first introduces a compression ratio (CR) adaptive SC (CRSC) model, capable of adjusting CR based on whether the frames are static or dynamic, effectively conserving spectrum resources. Simultaneously, we present a knowledge distillation (KD)-based approach to ensure the efficient learning of the CRSC model. Additionally, we implement a computer vision (CV)-based sensing model (CVSM) scheme, which intelligently perceives the scene changes by detecting the movement of the sensing targets. Therefore, CVSM can assess the significance of each frame through in-context analysis and provide CR prompts to the CRSC model based on real-time sensing results. Moreover, both CRSC and CVSM are designed as lightweight models, ensuring compatibility with resource-constrained sensors commonly used in practical edge applications. Experimental results show that SCCVS improves transmission accuracy by approximately 70% and reduces transmission latency by about 89% compared with baselines. We also deploy this framework on an NVIDIA Jetson Orin NX Super, achieving an inference speed of 14 ms per frame with TensorRT acceleration and demonstrating its real-time capability and effectiveness in efficient semantic video transmission.

Reverse skyline query (RSQ) has been widely used in practice since it can pick out the data of interest to the query vector. To save storage resources and facilitate service provision, data owners usually outsource data to the cloud for RSQ services, which poses huge challenges to data security and privacy protection. Existing privacy-preserving RSQ schemes are either based on a two-cloud model or cannot fully protect privacy. To this end, we propose an efficient privacy-preserving reverse skyline query scheme over a single cloud (ePRSQ). Specifically, we first design a privacy-preserving inner product's sign determination scheme (PIPSD), which can determine whether the inner product of two vectors satisfies a specific relation with 0 without leaking the vectors’ information. Next, we propose a privacy-preserving reverse dominance checking scheme (PRDC) based on symmetric homomorphic encryption. Finally, we achieve ePRSQ based on PIPSD and PRDC. Security analysis shows that PIPSD and PRDC are both secure in the real/ideal world model, and ePRSQ can protect the security of the dataset, the privacy of query requests and query results. Extensive experiments show that ePRSQ is efficient. Specifically, for a 3-dimensional dataset of size 1000, the computational and communication overheads of ePRSQ for a query are 79.47 s and 0.0021 MB, respectively. The efficiency is improved by$3.78\times$(300.58 s) and$928.57\times$(1.95 MB) respectively compared with PPARS, and by$61.31\times$(4872.55 s) and$407309\times$(855.35 MB) respectively compared with OPPRS.

The intelligent reflecting surface (IRS) and unmanned aerial vehicle (UAV)-assisted mobile edge computing (MEC) system is widely used in temporary and emergency scenarios. Our goal is to minimize the energy consumption of the MEC system by jointly optimizing UAV locations, IRS phase shift, task offloading, and resource allocation with a variable number of UAVs. To this end, we propose a flexible resource scheduling (FRES) framework by employing a novel deep progressive reinforcement learning that includes the following innovations. First, a novel multitask agent is presented to deal with the mixed integer nonlinear programming (MINLP) problem. The multitask agent has two output heads designed for different tasks, in which a classified head is employed to make offloading decisions with integer variables while a fitting head is applied to solve resource allocation with continuous variables. Second, a progressive scheduler is introduced to adapt the agent to the varying number of UAVs by progressively adjusting a part of neurons in the agent. This structure can naturally accumulate experiences and be immune to catastrophic forgetting. Finally, a light taboo search (LTS) is introduced to enhance the global search of the FRES. The numerical results demonstrate the superiority of the FRES framework, which can make real-time and optimal resource scheduling even in dynamic MEC systems.

In fire surveillance, Industrial Internet of Things (IIoT) devices require transmitting large monitoring data frequently, which leads to huge consumption of spectrum resources. Hence, we propose an Industrial Edge Semantic Network to allow IIoT devices to send warnings through Semantic communication (SC). Thus, we should consider 1) data privacy and security; 2) SC model adaptation for heterogeneous devices; 3) explainability of semantics. Therefore, first, we present an eXplainable Semantic Federated Learning (XSFL) to train the SC model, thus ensuring data privacy and security. Then, we present an adaptive client training strategy to provide a specific SC model for each device according to its Fisher information matrix, thus overcoming the heterogeneity. Next, an Explainable SC mechanism is designed, which introduces a leakyReLU-based activation mapping to explain the relationship between the extracted semantics and monitoring data. Finally, simulation results demonstrate the effectiveness of XSFL.

This article studies a dynamic Mobile Edge Computing (MEC) system assisted by Unmanned Aerial Vehicles (UAVs) and Intelligent Reflective Surfaces (IRSs). We propose a scaleable resource scheduling algorithm to minimize the energy consumption of all UEs and UAVs in the MEC system with a variable number of UAVs. We propose a Multi-tAsk Resource Scheduling (MARS) framework based on Deep Reinforcement Learning (DRL) to solve the problem. First, we present a novel Advantage Actor-Critic (A2C) structure with the state-value critic and entropy-enhanced actor to reduce variance and enhance the policy search of DRL. Then, we present a multi-head agent with three different heads in which a classification head is applied to make offloading decisions and a regression head is presented to allocate computational resources, and a critic head is introduced to estimate the state value of the selected action. Next, we introduce a multi-task controller to adjust the agent to adapt to the varying number of UAVs by loading or unloading a part of weights in the agent. Finally, a Light Wolf Search (LWS) is introduced as the action refinement to enhance the exploration in the dynamic action space. The numerical results demonstrate the feasibility and efficiency of the MARS framework.