With the rapid development of Internet Things, more and small devices are connected into for monitoring control purposes. One such type devices, smart plugs, have been extensively deployed worldwide in millions homes home automation. These however, would pose serious security problems if their vulnerabilities were not carefully investigated. Indeed, we discovered that some popular plugs severe which could be fixed but unfortunately left open. In this paper, case study a plug system known...

Accurately predicting the future motions of surrounding traffic agents is critical for safety autonomous ve-hicles. Recently, vectorized approaches have dominated motion prediction community due to their capability capturing complex interactions in scenes. How-ever, existing methods neglect symmetries prob-lem and suffer from expensive computational cost, facing challenge making real-time multi-agent without sacrificing performance. To tackle this challenge, we propose Hierarchical Vector...

Limited energy supply is one of the major constraints in wireless sensor networks. A feasible strategy to aggressively reduce spatial sampling rate sensors, that is, density measure points a field. By properly scheduling, we want retain high fidelity data collection. In this paper, propose collection method based on careful analysis surveillance reported by sensors. exploring correlation sensing data, dynamically partition nodes into clusters so sensors same cluster have similar time series....

Sensor scheduling plays a critical role for energy efficiency of wireless sensor networks. Traditional methods use either sensing coverage or network connectivity, but rarely both. In this paper, we deal with challenging task: without accurate location information, how do schedule nodes to save and meet both constraints connectivity? Our approach utilizes an integrated method that provides statistical guaranteed connectivity. We random then turn on extra nodes, if necessary, is totally...

Sensor localization has become an essential requirement for realistic applications over wireless sensor networks (WSN). Radio propagation irregularity and the stringent constraint on hardware cost, however, make in WSN very challenging. Range-free localizations are more appealing than range-based ones, since they do not depend received signal strength to estimate distance thus need simple cheap only. In this paper, we propose a ring-overlapping, range-free approach using Ring Overlapping...

We initiate a systematic study to help distinguish special group of online users, called hidden paid posters, or termed "Internet water army" in China, from the legitimate ones. On Internet, posters represent new type job opportunities. They get for posting comments articles on different communities and websites purposes, e.g., influence opinion other people towards certain social events business markets. While being an interesting strategy marketing, may create significant negative effect...

The cloud resource management belongs to the category of combinatorial optimization problems, most which have been proven be NP-hard. In recent years, reinforcement learning (RL), as a special paradigm machine learning, has used tackle these NP-hard problems. this article, we present deep RL-based solution, called DeepRM_Plus, efficiently solve different We use convolutional neural network capture model and utilize imitation in process reduce training time optimal policy. Compared with...

Non-intrusive load monitoring (NILM) helps disaggregate a household's main electricity consumption to energy usages of individual appliances, greatly cutting down the cost fine-grained towards green home vision. To address privacy concern in NILM applications, federated learning (FL) could be leveraged for model training and sharing. When applying FL paradigm real-world however, we are faced with challenges edge resource restriction, personalization, data scarcity. We present FedNILM,...

UAV-based air-ground integrated computing networks (AGIN) have gained significant traction in remote areas for the Power Internet of Things (PIoT). This paper considers an AGIN-PIoT, where tasks generated by ground PIoT devices are offloaded to aerial UAVs that perform edge computing. Jointly optimizing task offloading and UAV trajectory poses challenges such as many decision variables, information uncertainty, long-term queue delay constraints. Due limited battery capacity UAVs, our...

Wireless sensor networks consist of a large number tiny sensors that have only limited energy supply. One the major challenges in constructing such is to maintain long network lifetime as well sufficient sensing area. To achieve this goal, broadly-used method turn off redundant sensors. In paper, problem estimating areas among neighbouring wireless analysed. We present an interesting observation concerning minimum and maximum neighbours are required provide complete redundancy introduce...

Secure in-network aggregation in wireless sensor networks (WSNs) is a necessary and challenging task. In this paper, we first propose integration of system monitoring modules intrusion detection the context WSNs. We an extended Kalman filter (EKF) based mechanism to detect false injected data. Specifically, by behaviors its neighbors using EKF predict their future states (actual aggregated values), each node aims at setting up normal range neighbors' transmitted values. This task because...

Discretionary lane change (DLC) is a basic but complex maneuver in driving, which aims at reaching faster speed or better driving conditions, e.g., further line of sight ride quality. Although modeling DLC decision-making has been studied for years, the impact human factors, crucial accurately modelling strategies, largely ignored existing literature. In this paper, we integrate factors that are represented by styles to design new model. Specifically, our proposed model takes not only...

Efficient Internet-wide scanning plays a vital role in network measurement and cybersecurity analysis. While IPv4 is solved problem, for IPv6 still mission yet to be accomplished due its vast address space. To tackle this challenge, generally needs use pre-defined seed addresses guide further directions. Under general principle, various solutions have been developed, but all suffer from two primary pitfalls, low hit rate probing speed, caused by the inherent sparse distribution of active...

Clustered federated learning (CFL) is a promising solution to address the non-IID problem in spatial domain for (FL). However, existing CFL solutions overlook issue temporal and lack consideration of time efficiency. In this work, we propose novel approach, called <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">ClusterFLADS</i> , which takes advantage false predictions inappropriate global models, together with knowledge temperature scaling...

A Mobile Ad hoc NETwork (MANET) is a collection of wireless mobile computers forming temporary network without any existing wire line infrastructure. Due to the dynamic nature topology and resource constraints, routing in MANETs challenging task. Multipath can increase end-to-end throughput provide load balancing wired networks. However, its advantage not obvious because traffic along multiple paths may interfere with each other . In addition, accurate knowledge topology, finding disjoint...

We study the routing problem for multi-hop wireless ad hoc networks based on cooperative transmission. prove that minimum energy path (MECP) problem, i.e., using radio transmission to find best route with cost from a source node destination node, is NP-complete. thus propose shortest (CSP) algorithm uses Dijkstra's as basic building block and reflects properties in relaxation procedure. Simulation results show more nodes added network, our approach achieves saving compared traditional...

In wireless sensor networks, some nodes are put in sleep mode while other active for sensing and communication tasks order to reduce energy consumption extend network lifetime. This approach is a special case (k=2) of randomized scheduling algorithm, which k subsets sensors work alternatively. this paper, we first study the algorithm via both analysis simulations terms coverage intensity, detection delay, probability. We further asymptotic properties. Finally, analyze problem maximizing...

Infrastructure as a service (IaaS) allows users to rent resources from the Cloud meet their various computing requirements.The pay-as-you-use model, however, poses nontrivial technical challenge IaaS cloud providers: how fast provision large number of virtual machines (VMs) users' dynamic requests?We address this with VMThunder, new VM provisioning tool, which downloads data blockson demand during booting process and speeds up image streaming by strategically integrating peer-to-peer (P2P)...

Due to the high-measuring cost, monitoring of power quality (PQ) is nontrivial. This paper aimed at reducing cost PQ in network. Using a real-world dataset, this adopts learn-from-data approach obtain device latent feature model, which captures behavior as transition function. With network could be modeled, analogy, data-driven network, presents opportunity use well-investigated and data estimation algorithms solve problem grid. Based on are proposed intelligently place measurement devices...

WiFi-based human activity recognition explores the correlations between body movement and reflected WiFi signals to classify different activities. State-of-the-art solutions mostly work on a single channel hence are quite sensitive quality of particular channel. Co-channel interference in an indoor environment can seriously undermine accuracy. In this paper, we for first time explore wideband information with advanced deep learning toward more accurate robust recognition. We present...

This paper tackles a recent challenge in identifying culprit actors, who try to hide confidential payload with steganography, among many innocent actors social media networks. The problem is called steganographer detection and significantly different from the traditional stego that classifies an individual object as cover or stego. To solve over large-scale networks, this proposes method uses high-order joint features clustering ensembles. It employs 250-D calculated matrices of Discrete...