Large Language Models (LLMs), with advanced content creation and inference capabilities, can provide immersive intelligent services to users in mobile edge networks. However, the increasing demand for real-time artificial intelligence (AI) applications aggravates the limitations of cloud-based LLMs due to the long response time. Meanwhile, Small Language Models (SLMs), which are cost-effective and locally deployable for terminal devices, can serve as an efficient supplement to LLMs for performing latency-sensitive tasks with lower generalization capability. Due to the resource constraints of edge networks and the diverse requirements of user tasks, it is critical to design an inference framework that effectively coordinates the deployment and collaboration of LLMs and SLMs. In this paper, we propose an LLM-SLM collaborative inference (LSCI) scheme under a mobile edge computing (MEC) architecture, which jointly decides where to cache models and how to offload inference tasks to balance latency, accuracy, and resource costs. To optimize inference performance subject to resource constraints, we jointly solve the inference task offloading and model caching problem in LSCI scheme. Specifically, we employ deep reinforcement learning (DRL) to select highly popular SLMs to be cached on the edge server, and distributed belief propagation technique to solve the associated inference task offloading issue. Numerical results show that the proposed LSCI scheme can achieve significant performance gain in terms of inference performance when compared with a number of baseline solutions.

Advanced Persistent Threats (APTs) are exceptionally challenging to detect due to their high stealthiness. Audit logs, which provide detailed process-level information and record all activities before and after an APT attack, are crucial for detecting such threats. However, the sheer volume of data in audit logs also poses a significant challenge. Current methods suffer from the following issues: 1) difficulty in extracting information from the complex contextual relationships within audit logs, 2) reliance on prior knowledge for detecting APT attacks, and 3) coarse-grained detection signals. In this paper, we introduce Autumn, an APT detection method focused on processes as the primary research object. Autumn is an unsupervised learning model that does not rely on prior knowledge, making it more suitable for real-world APT detection scenarios. Autumn can swiftly identify critical information from vast audit logs and provides fine-grained detection signals by focusing on processes. We begin by constructing a graph from audit logs, segmenting it into subgraphs based on time, and applying strategies to reduce data volume. We then learn the characteristics of processes. By converting process information into input vectors using word2vec and calculating the reconstruction error for each subgraph through the encoding and decoding process of a transformer autoencoder, we train the model by associating the processes' IDF scores. Finally, we train the model on benign data and test it on a separate set of subgraphs containing attack events. Compared to other unsupervised learning methods, Autumn shows significant improvement in detection performance.

The stealth and persistence of APT attacks make investigation particularly challenging, further complicated by the diversity and volume of host logs. Existing methods, though effective, have limitations: 1) They rely heavily on manual processing and complex models that often fail to capture temporal relationships in logs; 2) These models struggle to differentiate highly similar attack behaviors from normal activities; 3) Interpretability and security challenges in deep learning models remain unresolved. This paper introduces DCASI, a sequence-based investigation method that integrates DTW similarity with contrastive learning. By constructing provenance graphs and extracting sequences through node pairs, DCASI performs semantic analysis, computes DTW similarity matrices, and employs contrastive learning to generate robust sequence representations. A lightweight random forest model is then used to identify attack behaviors. Evaluation on public datasets shows that DCASI outperforms existing methods.

By leveraging the parameter-efficient fine-tuning (PEFT) technology, large pre-trained models (LPMs) can effectively capture task-specific PEFT knowledge and generalize to downstream artificial intelligence (AI) tasks emerging in wireless networks. However, the inherent system heterogeneity of wireless networks and data across users brings difficulties in combining the PEFT knowledge of users. This in turn hinders the enhancement of LPM's global generalization capability. In this paper, we propose a distributed fine-tuning (DFT) scheme based on PEFT knowledge sharing, to capture the global PEFT knowledge of downstream users by aggregating heterogeneous PEFT models. In DFT, we use low-rank adaption (LoRA) model to capture the PEFT knowledge of users, where the LoRA models can be heterogeneous for different users with specific resource constraints. Then, we further design an aggregation and iterative training strategy for the heterogeneous LoRA models. The numerical results verify the effectiveness and advantages of the proposed DFT scheme in enhancing the global generalization capability of LPM in resource-heterogeneous wireless networks.

Executing computation offloading under the mobile edge computing (MEC) network architecture is a widely recognized way to reduce latency and energy consumption for IoT systems. However, under legacy MEC network architecture, it is hard to jointly utilize global and local information to derive the optimal offloading strategy at the distributed edge servers, which may cause the latency and energy consumption reduction to the bottleneck. To meet the ultra-low latency and ultra-low energy consumption requirements of future IoT system, researchers start to focus on involving the recent emerging open radio access network (O-RAN) architecture into IoT systems. In this paper, we design a computation offloading strategy for an O-RAN based IoT systems to reduce the energy consumption of IoT devices (IoTDs) under the premise of latency requirement. Specifically, we first adopt O-RAN architecture to deploy the IoT system. Under the O-RAN based IoT system, we formulate computation offloading problem with the aim of minimizing energy consumption by jointly deciding offloading ratio, transmit power, and local processing speed. To make the optimization problem tractable, we decouple the original problem into two sub-problems, and the original problem is solved by iteratively solving these two sub-problems. Experimental results show that under our proposed O-RAN based IoT system, the strategy we obtained can reduce the average energy consumption of all IoTDs by more than 20%.

The recently emerging blockchain technology provides a promising tool to enable endogenous security in Internet-of-Things (IoT) systems. However, when applying the legacy blockchain technology to the existing IoT systems, some technical bottlenecks due to resource constraints, such as storage resource, should be carefully addressed. In this article, we propose an optimal blockchain deployment mechanism for wireless IoT systems, where IoT devices have limited resources and the wireless links connection IoT devices are vulnerable, to improve the storage efficiency of massive blockchain data and realize the organic integration of blockchain technology and the communication process. We propose to maintain a complete blockchain by a set of proximity IoT nodes in a collaborative way on the premise of ensuring that each node can check every transaction. Through dynamically adjusting optimal block assignment, the tradeoff between the length of the blockchain to be stored and the security level provided can be well made. Moreover, a chaotic-based genetic algorithm is developed to obtain the near-optimal block assignment solution efficiently. Simulation results show that our proposed blockchain-based mechanism can effectively address the security issue in wireless IoT systems.

This paper deals with the problem of finding the robust fuzzy controller to achieve the robust stability with guaranteed cost for a class of uncertain time-delay systems. Based on the Takagi-Sugeno fuzzy model, the robust stability condition is derived to ensure the robust quadratic stability of the system by applying Lyapunov functional approach and the upper bound of considered cost is found. Moreover, a fuzzy controller is developed to stabilize the uncertain system and the guaranteed cost function is get. In addition, the effectiveness of the proposed design method is demonstrated by a numerical example

This paper studies on delay-dependent robust H∞control for uncertain linear systems with time-varying delays both in state and input. Based on an integral inequality, the sufficient condition of the existence for the state feedback controller is presented in terms of LMI, by means of Lyapunov-Krasovskii functional approach; it is proved that the new criterion does not require any assumptions about the system matrices. Finally, a numerical example is presented to demonstrate that using this integral inequality is effective