In this paper, we investigate the joint beamforming and power allocation problem in multi-antenna Cellular Vehicle-to-Everything (C-V2X) networks under uncertain channel state information (CSI). Our objective is to minimize the beamforming vector at the base station and the transmit powers of vehicle users while satisfying probabilistic quality-of-service (QoS) constraints. To address the uncertainty of CSI, we first develop an ellipsoid-based uncertainty set learning approach, which models the uncertain CSI as a symmetric ellipsoid. Building on this uncertainty set, we propose a robust counterpart transformation method to reformulate the joint beamforming and power allocation problem into a deterministic semi-definite problem without probabilistic constraints. Through our analysis, the ellipsoid-based uncertainty set exhibits significant conservatism when applied to uncertain CSI with asymmetric distributions. To mitigate this conservatism, we propose a support vector clustering (SVC)-based uncertainty set learning approach, which can tightly enclose the distribution of uncertain CSI. To further simplify the spatial structure of the SVC-based uncertainty set, we develop aK-Medoids-based equivalent set construction (KMSC) approach, significantly reducing the number of variables in the resulting robust equivalent problem. Finally, we conduct extensive simulations to evaluate the performance of our proposed robust approaches and compare them with non-robust methods.

Recently, scalable video coding (SVC) and scalable semantic coding (SSC) have emerged as effective solutions for supporting online video streaming in applications such as remote cockpits and telemedicine. Although the significance of SVC and SSC is well established, the underlying principles related to their performance bounds have not been thoroughly investigated. To address this issue, we study the performance bounds for online SVC/SSC in this paper. Initially, to evaluate coding performance, we propose a new metric referred to as effective coding gain (ECG), which jointly considers the entropy of the source data and the mutual information between the source and the encoded video data. Next, from the perspective of information theory, we derive a closed-form expression for the ECG bound while accounting for the impacts of diverse SVC/SSC coding structures. Our results not only ensure that the performance bounds can be efficiently evaluated but also provide valuable insights for resource allocation, cost optimization, performance evaluation and comparison, as well as for understanding the optimization space of existing systems.

Currently, in-vehicle Voice Assistants (VAs) have been widely integrated into in-vehicle infotainment (IVI) systems to enhance driver safety when performing functions such as navigation and phone calls while driving. Although various studies have demonstrated the existence of vulnerabilities in general-purpose VAs, there is a lack of research specifically targeting in-vehicle VAs, which operate in a closed and black-box environment. In this paper, we utilize fuzzing testing to analyze how speech errors in voice commands affect the recognition performance of in-vehicle VAs. First, we simulate speech errors by applying linguistic knowledge to mutate voice commands. Then, we adopt a genetic algorithm to efficiently generate additional erroneous commands. To further improve the quality of these mutated commands, we assign a risk level to each original command and prioritize the mutation of those whose misrecognition by the in-vehicle VA results in an increased risk level. We conducted comprehensive fuzzing experiments on both local (Whisper–DistilBERT-based) and cloud-based (Amazon Lex) in-vehicle VA systems. Our approach generated 59112 speech-error commands in the local VA, achieving a 60.13% misrecognition rate, significantly outperforming Baseline-Fuzzing, which had an effectiveness of 40.26%. On the cloud-based VA, the effectiveness improved from 2.83% to 33.24%. These results confirm the superiority of our method in generating high-impact erroneous commands.

Recently, scalable video coding (SVC) has gained significant recognition in mobile video streaming because it can adapt bitstreams to time-varying transmission conditions. However, the coding performance of SVC, which is determined by its coding structure, has not been thoroughly studied. To address this issue, we propose analyzing the redundancy, reduction, distortion, and mutuality of video information within the video coding processes. This analysis facilitates the development of a novel information-theoretical framework for quantifying coding performance, which includes an information theory (IT)-based quantification method and a graphical representation system. The representation system accurately delineates the coding reference structure for encoding each video frame, while the proposed method utilizes mutual information to quantify the achievable coding performance of SVC under the delineated structure. To demonstrate the significance of our research, we apply the proposed framework to encode a basic coding unit, showcasing its effectiveness in improving SVC schemes. Consequently, our framework not only provides an efficient approach for quantifying the coding performance of SVC but also serves as an invaluable tool for optimizing SVC in various applications.

In next-generation wireless networks, high-mobility unmanned aerial vehicles (UAVs) are promising to provide content coverage, where users can receive sufficient requested content within a given time. However, trajectory planning for multiple UAVs to provide content coverage is challenging since 1) UAVs cannot provide content coverage for all users due to the limited energy and caching storage, and 2) the trajectory planning of UAV is coupled with each other. Moreover, the complete information based trajectory planning methods are unusable since UAVs cannot obtain prior information on the rapidly changing environment. In this paper, we investigate the multi-UAV trajectory planning for energy-efficient content coverage. We first formulate an energy efficiency maximization problem considering recharging scheduling, which aims to reduce the total length of trajectories of UAVs under the quality of service (QoS) constraints. To settle environment uncertainty, the trajectory planning problem is modeled as two coupled multi-agent stochastic games, whose equilibrium constitute the optimal trajectory. To obtain the equilibrium, we propose a decentralized reinforcement learning algorithm, which can decouple the two games. We prove that the proposed algorithm can converge to the optimal solution of the Bellman equation with a higher rate compared to the centralized one. Moreover, simulation results show that the energy efficiency of the proposed algorithm is smaller than 5% compared the optimal, which is obtained with the prior information of environments.

In ultra-dense networks (UDNs), the dense deployment of base stations (BSs) is facing challenges due to the pronounced unbalanced traffic loads, severe inter-cell interference, and uncertain traffic demands. In this paper, we tame traffic uncertainty for the joint optimization of BS activation and user association in UDNs to mitigate interference and balance traffic loads among BSs. Specifically, we address the traffic uncertainty by using chance constraint programming with the known first- and second-order statistics of the uncertain traffic. We formulate the joint BS activation and user association problem as a mixed integer non-linear programming problem, which is then decomposed into a set of user association sub-problems by modeling the BS states (active or idle) as a Markov chain. We solve the user association sub-problem at each BS state by transforming it into a convex problem over the positive orthant. In particular, at each BS state, the candidate serving BSs that lead to the optimal load balancing performance are identified for each user and parts of the user's traffic are offloaded to the identified BSs. Based on the obtained solutions, we propose a distributed near-optimal BS activation and user association scheme. Numerical results demonstrate that our proposed scheme is more robust to traffic uncertainty and provides better load-balancing performance than the existing schemes.

Content caching has limitations on achieving cache gains (e.g., cache hit ratio) in small cell networks, due to limited storages of small base stations (SBSs) and inherent user demand patterns (i.e., initial content preferences). Two effective approaches have been proposed to exploit the potential of content caching: SBS cooperation to utilize cache storage, and proactive content recommendation to shape user demand. In this paper, we investigate cooperative content caching and recommendation to maximize cache gains, while guaranteeing users' satisfaction by recommending appealing content items. We propose a generic framework for cooperative content caching and recommendation, based on which we propose an online and distributed scheme by designing a continuous-time Markov chain (CTMC). In particular, online content caching (a.k.a., content replacement) is implemented by hopping from one cache state to another in the CTMC, while content recommendation is performed heuristically through sequential fixing at each cache state. Besides, we characterize the performance gap between our proposed scheme and the theoretical optimum in terms of cache hit ratio. Simulation results demonstrate that the proposed scheme achieves better cache hit ratios than other schemes in single-BS scenarios, and provides a competitive solution in multiple-BS scenarios.

Content caching is a promising way to overcome backhaul limitations in small cell networks. However, in such type of networks, small base stations (SBSs) are always deployed with limited cache storages. Thus, it is necessary for SBSs to adjust their contents for better caching efficiency, so as to reduce backhaul traffic. In this paper, we study the content replacement problem to minimize the traffic flowing into the costly backhaul links. However, in small cell networks where SBSs make up backhaul mesh networks, the effectiveness of reducing backhaul traffic depends on the hop distance from the content location to the requesting user. On this basis, we formulate a hop minimization problem that is inherently combinatorial. Through log-sum-exp approximation, we can solve the problem and arrive at a close-form solution with guaranteed performance gap to the optimal solution. By exploiting the properties of continuous-time Markov chain (CTMC), the solution can be implemented by designing a CTMC that can instruct the content replacement process. As a consequence, a concise, efficient, and flexible content replacement strategy is proposed. Simulation results verify our analysis and show that our proposed strategy outperforms the conventional strategies.

Route planning has received great attention from researchers with the dramatic development of mobile localization technology and the emerging location based services. It is a trend to consider users' preferences and the budget limit for the shortest path problem. The optimal route search with user's preferences focuses on finding an optimal route from a source to a target with the given keywords and the budget constraint, such that the route can maximally satisfy the user's needs on weighted preferences. We solve the NP-hard problem by proposing an A* based route search algorithm with some effective pruning strategies and present ORSUP, a website for visual query and route display. We describe the details and functions of the system interface and demonstrate the efficiency of proposed algorithm and effectiveness on solving the route search problem via a common scenario.

User association has redrawn much attention lately, due to the introduction of content caching in small base stations (SBSs). To reduce traffic burden on backhaul links, users are associated with different SBSs when requesting different contents. However, user-perceived delay increases if the serving SBSs that have the desired contents are overloaded. Moreover, the user association problem becomes complex due to the vast number of contents. In this paper, to reduce user-perceived delay as well as backhaul loads, we propose a cluster-level user association scheme where content clustering is leveraged to simplify user association and reduce its complexity. Particularly, similar contents are clustered together according to the content preferences of users and cached contents in SBSs. Thus, the dimensionality of the user problem becomes smaller. On this basis, we propose a distributed cluster-level user association scheme, where each user selects SBSs based on their traffic loads and cached contents. Simulation results show that our scheme based on clustered contents outperforms the traditional schemes.

Video object segmentation (VOS) is important for various computer vision problems, and handling it with minimal human supervision is highly desired for the large-scale applications. To bring down the supervision, existing approaches largely follow a data mining perspective by assuming the availability of multiple videos sharing the same object categories. It, however, would be problematic for the tasks that consume a single video. To address this problem, this paper proposes a novel approach that explores weakly labeled images to solve video object segmentation. Given a video labeled with a target category, images labeled with the same category are collected, from which noisy object exemplars are automatically discovered. After that the proposed approach extracts a set of region proposals on various frames and efficiently matches them with massive noisy exemplars in terms of appearance and spatial context. We then jointly select the best proposals across the video by solving a novel submodular problem that combines region voting and global region matching. Finally, the localization results are leveraged as strong supervision to guide pixel-level segmentation. Extensive experiments are conducted on two challenging public databases: Youtube-Objects and DAVIS. The results suggest that the proposed approach improves over previous weakly supervised/unsupervised approaches significantly, showing a performance even comparable with the several approaches supervised by the costly manual segmentations.

With the popularity of cloud computing, there have been increasing concerns about its security and privacy. Since the cloud computing environment is distributed and untrusted, data owners have to encrypt outsourced data to enforce confidentiality. Therefore, how to achieve practicable access control of encrypted data in an untrusted environment is an urgent issue that needs to be solved. Attribute-based encryption (ABE) is a promising scheme suitable for access control in cloud storage systems. This paper proposes a hierarchical attribute-based access control scheme with constant-size ciphertext. The scheme is efficient because the length of ciphertext and the number of bilinear pairing evaluations to a constant are fixed. Its computation cost in encryption and decryption algorithms is low. Moreover, the hierarchical authorization structure of our scheme reduces the burden and risk of a single authority scenario. We prove the scheme is of CCA2 security under the decisional q-Bilinear Diffie-Hellman Exponent assumption. In addition, we implement our scheme and analyse its performance. The analysis results show the proposed scheme is efficient, scalable, and fine-grained in dealing with access control for outsourced data in cloud computing.

Obtaining an appropriate dictionary is the key point when sparse representation is applied to computer vision or image processing problems such as image restoration. It is expected that preserving data structure during sparse coding and dictionary learning can enhance the recovery performance. However, many existing dictionary learning methods handle training samples individually, while missing relationships between samples, which result in dictionaries with redundant atoms but poor representation ability. In this paper, we propose a novel sparse representation approach called conformal and low-rank sparse representation (CLRSR) for image restoration problems. To achieve a more compact and representative dictionary, conformal property is introduced by preserving the angles of local geometry formed by neighboring samples in the feature space. Furthermore, imposing low-rank constraint on the coefficient matrix can lead more faithful subspaces and capture the global structure of data. We apply our CLRSR model to several image restoration tasks to demonstrate the effectiveness.

In this paper, an intelligent vehicle monitoring and management system (IVMMS) based on the multi-net is designed. First, its architecture and functions are introduced. It uses many technologies fusing GPS, GPRS/CDMA, GIS, RS and Internet, and is composed mainly by two major parts: the Mobile Terminal and the Monitoring Center. Especially, GPRS/CDMA technology is the reliable technical support for wireless communication smooth transition from the 2.5G to the 3G; then, the principles and the key technologies of Mobile Terminal and Monitoring Center are discussed respectively in detail, and figures of their structures are also shown below; finally, a novel vehicle scheduling model based on ant algorithms are proposed and the course of scheduling tasks is also introduced. It is indicated that, the vehicles scheduling efficiency is enhanced greatly. The system not only can be used in the vehicles scheduling, but also in the vessel and other fields, and the solid technical support will be provided for the ITS research