Crowd counting is a very challenging task in crowded scenes due to heavy occlusions, appearance variations and perspective distortions. Current crowd methods typically operate on an image patch level with overlaps, then sum over the patches get final count. In this paper, we propose end-to-end convolutional neural network (CNN) architecture that takes whole as its input directly outputs result. While making use of sharing computations overlapping regions, our method advantages contextual...

Unmanned aerial vehicle (UAV)-aided data collection is a new and promising application in many practical scenarios. In this work, we study the age-optimal trajectory planning problem UAV-enabled wireless sensor networks, where UAV dispatched to collect from ground nodes (SNs). The age of information (AoI) collected each SN characterized by uploading time elapsed since leaves SN. We attempt design two trajectories, referred as Max-AoI-optimal Ave-AoI-optimal respectively. minimize `oldest'...

Due to the malicious attacks in wireless networks, physical layer security has attracted increasing concerns from both academia and industry. The research on mainly focuses either secrecy capacity/achievable rate/capacity-equivocation region perspective of information theory, or designs viewpoints optimization signal processing. Because its importance designs, latter direction is surveyed a comprehensive way this paper. survey begins with typical wiretap channel models cover common scenarios...

In this paper, we generally formulate the dynamics prediction problem of various network systems (e.g., mobility, traffic and topology) as temporal link task. Different from conventional techniques that ignore potential non-linear characteristics informative weights in dynamic network, introduce a novel model GCN-GAN to tackle challenging task weighted networks. The proposed leverages benefits graph convolutional (GCN), long short-term memory (LSTM) well generative adversarial (GAN). Thus,...

Space-air-ground integrated network (SAGIN) is envisioned as a promising solution to provide cost-effective, large-scale, and flexible wireless coverage communication services. Since realworld deployment for testing of SAGIN difficult prohibitive, an efficient simulation platform requisite. In this article, we present our developed which supports various mobility traces protocols space, aerial, terrestrial networks. Centralized decentrallized controllers are implemented optimize the...

To deal with the rapid growth of high-speed and/or ultra-low latency data traffic for massive mobile users, fog radio access networks (Fog-RANs) have emerged as a promising architecture next-generation wireless networks. In Fog-RANs, edge nodes and user terminals possess storage, computation communication functionalities to various degrees, which provide high flexibility network operation, i.e., from fully centralized distributed operation. this paper, we study cache placement problem in by...

The emergence of the D2D communications paradigm has transformed way in which cellular networks are operated. Previous work communications, however, mainly focused on interference management and capacity maximization both links links. Recent literature observed that QoE can be greatly enhanced by caching contents mobile devices, with a carefully designed strategy. In this article, we propose novel hypergraph framework designs based communication scheme taking social ties among users common...

In this work, we study the UAV-enabled data collection problem for high information freshness in wireless sensor networks, where one UAV is dispatched to collect of ground Sensor Nodes (SNs). The measured by Age Information (AoI) each SN, which defined as sum SN's uploading time and UAV's flight after leaving SN. Two optimization problems age-optimal are formulated minimize SNs' maximal AoI average AoI, respectively. An iterative SN association trajectory planning policy proposed seek...

In this paper, we propose a novel online multi-object tracking (MOT) framework, which exploits features from multiple convolutional layers. particular, use the top layer to formulate category-level classifier and lower identify instances one category under intuition that layers contain much more details. To avoid computational cost caused by fine-tuning, train our appearance model with an offline learning strategy using historical reserved for each object. We evaluate proposed framework on...

The heterogeneous cloud radio access network (Cloud-RAN) provides a revolutionary way to densify networks. It enables centralized coordination and signal processing for efficient interference management flexible adaptation. Thus it can resolve the main challenges next-generation wireless networks, including higher energy efficiency spectral efficiency, cost scalable connectivity, low latency. In this article we will provide an algorithmic approach new design dense Cloud-RAN based on convex...

Heterogeneous Internet of Things (IoT) and multi-access mobile edge computing (MA-MEC) are believed as supporting technologies for building a smart city. The advancement flourish IoT facilitating the entry human society into Everything era, which lay foundation To address conflict between computation capability low-cost devices in IoT, MA-MEC is available resource-limited computation-sensitive services applications by offloading distributed content delivery/caching. However, deploying cloud...

With the advance of unmanned aerial vehicles (UAVs) and low earth orbit (LEO) satellites, integration space, air ground networks has become a potential solution to beyond fifth generation (B5G) Internet remote things (IoRT) networks. However, due network heterogeneity high mobility UAVs LEOs, how design an efficient UAV-LEO integrated data collection scheme without infrastructure support is very challenging. In this paper, we investigate resource allocation problem for two-hop uplink B5G...

Due to the flexibility and low operational cost, dispatching unmanned aerial vehicles (UAVs) collect information from distributed sensors is expected be a promising solution in Internet of Things (IoT), especially for time-critical applications. How maintain freshness challenging issue. In this paper, we investigate fresh data collection problem UAV-assisted IoT networks. Particularly, UAV flies towards status update packets within given duration while maintaining non-negative residual...

With the development of deep learning techniques, combination with image compression has drawn lots attention. Recently, learned methods had exceeded their classical counterparts in terms rate-distortion performance. However, continuous rate adaptation remains an open question. Some use multiple networks for rates, while others one single model at expense computational complexity increase and performance degradation. In this paper, we propose a continuously adjustable framework, Asymmetric...

Maintaining the freshness of information in Internet Things (IoT) is a critical yet challenging problem. In this paper, we study cooperative data collection using multiple Unmanned Aerial Vehicles (UAVs) with objective minimizing total average Age Information (AoI). We consider various constraints UAVs, including kinematic, energy, trajectory, and collision avoidance, order to optimize process. Specifically, each UAV, which has limited on-board takes off from its initial location flies over...

By moving to millimeter wave (mmWave) frequencies, base stations (BSs) will be densely deployed provide seamless coverage in sixth generation (6G) mobile communication systems, which, unfortunately, leads severe cell-edge problem. In addition, with massive multiple-input-multiple-output (MIMO) antenna arrays employed at BSs, the beamspace channel is sparse for each user, and thus there no need serve all users a cell by beams therein jointly. Therefore, it of paramount importance develop...

The cloud radio access network (Cloud-RAN) has recently been proposed as one of the cost-effective and energy-efficient techniques for 5G wireless networks. By moving signal processing functionality to a single baseband unit (BBU) pool, centralized resource allocation are enabled in cloud-RAN, thereby providing promise improving energy efficiency via effective adaptation interference management. In this paper, we propose holistic sparse optimization framework design green cloud-RAN by taking...

For transmitting data in a secret and energy-efficient manner collaborative amplify-and-forward relay networks, the secure energy efficiency (EE) defined as bits transferred with unit is maximized to satisfy each node power constraint target secrecy rate requirement, based on physical security framework. The EE by joint source allocation, which nonconvex optimization problem. To cope this difficulty, solution scheme corresponding algorithms are developed jointly applying fractional...

Named data networking (NDN) enables fast and efficient content dissemination in mission-critical unmanned aerial vehicle ad hoc networks (UAANETs); however, its in-network caching mechanism brings a new security challenge: poisoning. Poisoned can contaminate the cache on routers isolate valid from network, leading to performance degradation or denial of service. To mitigate such attacks enhance network-layer trust NDN-based UAANETs, this article proposes novel systematic framework that...

Caching popular contents at the edge of wireless networks has recently emerged as a promising technology to improve quality service for mobile users, while balancing peak-to-average transmissions over backhaul links. In contrast existing works, where central coordinator is required design cache placement strategy, we consider distributed caching problem which highly relevant in dense network settings. considered scenario, each Base Station (BS) storage finite capacity, and user will be...

In this paper, energy-efficient secure communications via untrusted two-way relaying are investigated considering physical layer security to prevent the relays from intercepting confidential information of users. The performance metric energy efficiency (EE), defined as ratio secrecy sum rate total power consumption, is maximized by jointly optimizing allocation for all nodes, with constraints maximum allowed and minimum target rate. To deal intractable nonconvex optimization problem, some...

In this work, we study the age-optimal data collection problem in UAV-enabled wireless sensor networks, where a UAV is dispatched to collect from ground nodes (SNs). The age of information (AoI) collected each SN consists uploading time and elapsed since receives data. Based on definition, combinatorial optimization formulated minimize maximum AoI among all SNs. To solve problem, propose joint association trajectory planning strategy strike balance between SNs' UAV's flight various...

In the Internet of Things (IoTs), unmanned aerial vehicle (UAV) has been considered as an efficient solution to collect information from ground sensor nodes (SNs) due its controllable mobility and high maneuverability. this paper, we study a UAV-aided data collection problem for IoTs, where SNs sample with fixed or random rates cache sampled update packets under sample-and-replace policy. An energy-constrained UAV is deployed each SN when it flies over SN. The UAV's flight trajectory...

Information freshness, which is characterized by a new performance metric called age of information (AoI), significantly influences decision making in numerous applications. In wireless sensor networks, unmanned aerial vehicle (UAV) has been widely adopted for fresh data collection. The key to applying UAV lies trajectory planning. Considering several fixed waypoints trajectory, the planning an NP-hard combinatorial optimization problem, and difficult solve practice. To well balance between...

In this paper, we raise the concern on energy-efficient secure communication over a decode-and-forward relay network with potential eavesdroppers. Our objective is maximizing energy efficiency (EE), which defined as ratio of secrecy rate to total power, subject given individual decoding rate, and target constraints. The problem formulated unified mixed integer nonlinear optimization adapting different assumptions channel state information (CSI). To deal combinatorial optimization, suboptimal...