The rapid growth of connected vehicles creates new opportunities to exploit internet of vehicles (IoV) data for traffic state estimation (TSE), which is a key enabler of intelligent transportation systems (ITS). In this paper, we propose a cost-effective TSE framework that leverages sparse IoV data, which significantly reducing the data collection overhead associated with large-scale IoV datasets. We further analyze the impact of data sparsification and show that the induced estimation errors can be well approximated by Gaussian noise, thereby reformulating sparse IoV-based TSE as a denoising problem. To enhance estimation accuracy, we develop a spatial-temporal attention model, termed the convolutional retentive network (CRNet), which integrates convolutional neural networks (CNNs) for spatial correlation learning with a retentive network (RetNet) for temporal dependency modeling. Extensive experiments conducted on a large-scale real-world IoV dataset validate the feasibility of TSE under sparse IoV sensing conditions. Notably, even when only 5% of the data is available, CRNet achieves a mean absolute error (MAE) below 5 km/h, demonstrating both the high accuracy of the proposed approach and its practical applicability in real-world scenarios.

The fully-decoupled radio access network (FD-RAN) is an innovative architecture designed for next-generation mobile communication networks, featuring decoupled control and data planes as well as separated uplink and downlink transmissions. To further enhance energy efficiency, this paper explores a green approach to FD-RAN by incorporating adaptive base station (BS) sleeping and resource allocation. First, we introduce a holistic power consumption model and formulate a energy efficiency maximization problem for FD-RAN, involving joint optimization of user equipment (UE) association, BS sleeping, and power control. Subsequently, the optimization problem is decomposed into two subproblems. The first subproblem, involving UE power control, is solved using a successive lower-bound maximization approach based on Dinkelbach’s algorithm. The second subproblem, addressing UE association and BS sleeping, is tackled via a modified, low-complexity many-to-many swap matching algorithm. Extensive simulation results demonstrate the superior effectiveness of FD-RAN with our proposed algorithms, revealing the sources of energy efficiency gains.

The consumption of energy by data center networks has increased markedly in recent times, becoming a matter of concern for all sections of society. It is essential to develop comprehensive models and optimize energy consumption in data centers. As the Internet of Things (IoT) industry experiences accelerated growth, predictive modeling of data centers utilizing IoT devices to gather real-time data and incorporating datadriven deep learning techniques has demonstrated an enhanced degree of accuracy. The functionality of existing deep learning models is limited in real data center scenarios, where concept drift is a common occurrence. This results in a reduction in the predictive accuracy of the models. In this paper, we propose a time series prediction framework for data center scenarios. Under this framework, online prediction and offline training can be implemented to effectively address the frequent concept drift phenomenon in data center scenarios and enhance the stability of prediction. Extensive experiments are conducted in this study using real datasets obtained from IoT device measurements. The experiment results show that the proposed time series prediction framework can achieve good performance in data center scenarios when facing the concept drift.

Data centers (DC), as the computational core, play a pivotal role in driving the development of various industries. However, their 24/7 operational workload places immense pressure on operations staff, and traditional AI for IT Operations technologies due to their inability to communicate with operations personnel through language, are unable to effectively alleviate this pressure. In contrast, Retrieval-Augmented Generation (RAG), by integrating external knowledge bases with Large Language Models, enables effective communication with operations teams. Nevertheless, the direct application of RAG in DC currently faces various challenges, such as limited text representation, context length limitations, and poor scalability for large-scale databases. In this paper, we propose multi-modal multi-agent framework for DC facility operations. This framework is primarily composed of a manager and multiple specialized experts, which supports providing intuitive multi-modal information to assist in responses and optimizes response quality in terms of both depth and breadth, delivering more comprehensive and precise answers to offer an innovative solution for enhancing DC operations. Finally, we build a DC operations question-answering dataset and conduct comparative testing of the proposed framework using multiple evaluation metrics.

With the advancement of low-earth orbit (LEO) satellite technologies, the Large-Scale LEO Satellite Networks (LSLSNs) have become the cornerstone of next-generation wireless communication, delivering global coverage with low latency and massive throughput. However, the LSLSNs operate in an open space environment, where the effects of space environmental factors such as electromagnetic radiation, thermal conditions and solar flares can easily lead to regional correlated damage for satellite nodes, severely degrading network performance. To ensure stable and robust transmission, we propose a Robust and Intelligent Multipath QUIC Transmission (RIMT) method in LSLSNs. The RIMT method is deployed in multi-domain based satellite networks leveraging distributed Software-Defined Networking (SDN) technology, where satellites are divided into various autonomous domains managed by a local SDN controller for high efficiency and flexible management. To address regional correlated damage, RIMT employs pre-computed backup flows to seamlessly switch from compromised flows. Additionally, we introduce an innovative congestion control algorithm designed to maintain stable data transmission during the route switching process. We implement RIMT in the Kuiper K3 shell network, experiments show that RIMT can achieve more enhanced performance than other mechanisms.

Deep learning methods for accurately predicting data center status, which are essential for addressing the exponential growth of energy consumption, have gained significant attention, driven by the vast amounts of data collected through the advancement of Internet of Things (IoT) technologies. However, conventional small models often face data scarcity issues in practical deployment. While large models show promise in addressing this challenge, they encounter obstacles, such as multivariate tasks, computational intensity, and ineffective information capture. Moreover, their applications in data centers remain largely unexplored. In this article, we investigate local semantic capture empowered large model for multivariate time series forecasting in IoT-enabled data centers. We first introduce time series tasks within data centers and propose the Point Lag (Plag)-Llama framework with the Lag-Llama backbone to support zero-shot forecasting and fine-tuning for multivariate point time series forecasting. To address computational intensity and enhance the capabilities of multivariate forecasting, we propose the local semantic capture (LOSEC) for adapter fine-tuning, which captures local semantic information across time and channel dimensions alternately with low-complexity. Specifically, time series are patched into tokens, and channels are clustered together, forming local semantic information that can be captured more effectively. Extensive experiments demonstrate that Plag-Llama exhibits superior zero-shot capability and that the LOSEC empowered adapter fine-tuning achieves state-of-the-art performance on real-world datasets collected from data centers, with ablation studies further validating the effectiveness of each module within the proposed models.

By harnessing advanced Internet-of-Things (IoT) technologies, deep-learning methods have garnered significant attention for accurately predicting data center status, serving as the cornerstone for mitigating the exponential growth of energy consumption in data centers. However, these methods encounter data scarcity issues during practical deployment. While the proliferation of large models holds promise for addressing this challenge, their application in the context of data centers remains largely unexplored. Furthermore, these models encounter diverse obstacles, including multivariate tasks, immersive computation, and so on. In this paper, we investigate multivariate time series forecasting in IoT-enabled data centers by harnessing large models. Specifically, we introduce a multivariate time series forecasting framework tailored for IoT-enabled data centers. We propose the point Lag (Plag)-Llama model, which transfers univariate forecasting knowledge from the Lag-Llama for multivariate forecasting. The Plag-Llama benefits from zero-shot ability and fine-tuning facilitated by our proposed transfer block. To mitigate computational intensity and enhance the Plag-Llama’s performance, we introduce a novel Joint Channel-Time (JCT)-adapter fine-tuning technique. Extensive experiments demonstrate that the transferred Plag-Llama exhibits superior zero-shot ability, while the proposed JCT-adapter fine-tuning achieves state-of-the-art performance and remarkable few-shot ability on real-world datasets collected from data centers.

The increasingly severe energy consumption caused by exploding wireless demands attracts considerable research. Remarkably, base station (BS) sleeping is a promising technique to enable the green network. A disruptive and original fully-decoupled radio access network (FD-RAN) architecture aiming at the next-generation mobile communication networks is developed, which removes the obstacles to achieving BS sleeping, i.e. deficient cooperation between BSs, coupled data-control transmission and coupled uplink-downlink transmission. In this paper, we investigate the joint user association and uplink BS sleeping considering power control in the FD-RAN with the superiority of fully decoupled architectures. Specifically, we propose an energy consumption model for the uplink FD-RAN and tackle the mixed-integer second-order cone problem to minimize the whole network energy consumption by leveraging the many-to-many swap matching theory. Extensive simulation results validate a higher energy efficiency of the uplink FD-RAN compared to the traditional cellular network and cell-free networks and demonstrate the effectiveness of our proposed algorithm.