Textual Graphs (TGs) present a graph-based representation of textual data and find wide applications in real-world scenarios, such as citation networks, knowledge graphs, and social networks. While the traditional "pre-train, fine-tune" framework effectively addresses tasks requiring abundant labeled data, it falls short in scenarios with limited resource or zero-shot learning capabilities, particularly in low-resource textual graph node classification. Additionally, prevalent approaches that convert text nodes into shallow or manually engineered features fail to capture the rich semantic nuances within the text. The conventional methods often neglect the fusion of semantic and topological information, resulting in suboptimal model learning. To overcome these challenges, we proposed a novel method of low-resource textual graph node classification based on large language models, i.e., Textual graph learning with semantic and topological awareness (TGLsta), which comprehensively explores the semantic information, near neighborhood information, and the topology information in textual graphs, where these components are the most important information source contained in textual graphs. Graph prompt tuning for both zero- and few-shot textual graph node classification is further introduced.

Federated learning (FL) has emerged as the most promising distributed machine learning training framework due to its advantages in efficiency, privacy preservation, and scalability. In the practical deployment, FL usually faces the system heterogeneity, data heterogeneity, and limited communication resources. Many works attempt to address the above challenges by client selection, but seldom consider the issues of missing classes in training samples and communication resource allocation, which leads to poor training performance. In this paper, we propose an efficient FL framework with clustering-based client selection and bandwidth allocation, called FedCCS. Specifically, FedCCS first clusters clients based on the local label distribution. By ensuring clients from each cluster participate in training, a wider range of sample classes are covered to mitigate data heterogeneity effects. Furthermore, considering system heterogeneity and limited communication resources, we develop an iterative-based joint optimization algorithm for client selection and bandwidth allocation to minimize latency. Experimental results on both simulations and real-world prototypes show that, compared to other methods, FedCCS can significantly reduce latency and improve the stability of model performance during the training process.

With advancements in AI, smart home gyms are becoming increasingly popular for providing fitness assistance in indoor environments. In this research, we propose a layer-by-layer framework, called Wi-Fitness, which bridges video perception with Wi-Fi sensing for smart fitness. At the data preprocessing layer, the singular value decomposition-based channel state information denoising mechanism is leveraged to do the Wi-Fi data calibration. Diverse and high-quality training samples are generated by a random quantization-based data augmentation method. At the bimodal fusion layer, the heterogeneity between the Wi-Fi and video is mitigated by the local attention mechanism and the bimodal feature integration mechanism. For the video modality, the attention-based spatio-temporal graph convolutional network (AST-GCN Net) is proposed to refine spatial information. The spatio-temporal semantic alignment module is proposed to transfer spatial information from video to Wi-Fi and maintain temporal consistency across modalities. The fitness assessment layer provides exercise visualization. The generalization of Wi-Fitness is enhanced by layer-by-layer collaboration. Wi-Fitness demonstrates its effectiveness by achieving an average F1-Score of 92.68% in three typical indoor environments.

Behavior biometrics-based user authentication with Wi-Fi gains significant attention due to its ubiquitous and contact-free manners. An individual’s identity can be verified by analyzing activities induced signal variances, excellently balancing the security demands and user experience. However, the inherent complexity of Wi-Fi signals presents significant challenges for behavior biometrics-based user authentication. The susceptibility of Wi-Fi signals results in a poor cross-environment generalization capability, which is overlooked by the existing research. In addition, most existing works of behavior-based user authentication are based on one-off activity. This makes them vulnerable to zero-effort attacks and imitation attacks. To address these issues, we propose a cross-environment continuous gesture-based user authentication framework with Wi-Fi, dubbed Wi-CGAuth. Specifically, the cross-environment generalization capability is enhanced by the cross-layer joint optimization approach. At the lowest signal layer, the signals’ time, spatial, and frequency diversity are extended maximally, by a novel, subcarrier-level, cost-effective signal optimization strategy. At the middle layer, the multi-view fusion method, i.e., multi-transfer component analysis (TCA), is applied to refine the signals from transceiver pairs after signal preprocessing. The continuous gesture segmentation problem is modeled as the classification problem, which is solved by CNN. At the upper layer, a Convolutional Neural Network-Transformer (CNN-Transformer) model is employed to achieve the dual task of effective user authentication and accurate gesture recognition. After extensive experiments in three typical indoor scenarios, Wi-CGAuth can achieve an average authentication accuracy of 92.7%, demonstrating its robustness and effectiveness.

Recovering fine-grained urban vehicle trajectory data hold great potential for understanding mobility patterns and improving the efficiency of transportation systems. Most existing solutions rely on regularly sampled GPS points, which capture data only from participating vehicles. In this work, we exploit traffic camera observation data for city-wide vehicle trajectory recovery. Nevertheless, there are two major challenges. First, traffic cameras are sparsely deployed at road intersections, resulting in uneven vehicle observations and irregular sampling intervals. Second, substantial uncertainty arises when reconstructing trajectories between distant consecutive observations. To address the above challenges, we propose TR-ODE , a novel encoder-decoder framework based on Neural Ordinary Differential Equations for continuous-time trajectory recovery. Specifically, TR-ODE employs customized GRU-ODE-Bayes to evolve hidden states between observations and update the current hidden state to incorporate the incoming observations. By incorporating a multi-task learning block, TrajRec , our method integrates road embeddings and self-attention scores to assist in simultaneous predictions on road segments and moving ratios for trajectory recovery. Extensive experimental studies on two real-world datasets demonstrate that TR-ODE outperforms state-of-the-art methods in vehicle trajectory recovery from irregular observations, reducing MAE and RMSE by up to 17.11% and 14.47%, respectively.

In large-scale metropolis, it is critical to efficiently allocate various resources such as electricity, medical care, and transportation to meet the living demands of citizens, according to the spatio-temporal distributions of resources and demands. Previous researchers have done plentiful work on such problems by leveraging Multi-Agent Reinforcement Learning (MARL) methods, where multiple agents cooperatively regulate and allocate the resources to meet the demands. However, facing the great number of agents in large cities, existing MARL methods lack efficient parameter sharing strategies among agents to reduce computational complexity. There remain two primary challenges in efficient parameter sharing: (1) during the RL training process, the behavior of agents changes significantly, limiting the performance of group parameter sharing based on fixed role division decided before training; (2) the behavior of agents forms complicated action trajectories, where their role characteristics are implicit, adding difficulty to dynamically adjusting agent role divisions during the training process. In this paper, we propose Dynamic Parameter Sharing (DyPS) to solve the above challenges. We design self-supervised learning tasks to extract the implicit behavioral characteristics from the action trajectories of agents. Based on the obtained behavioral characteristics, we propose a hierarchical MARL framework capable of dynamically revising the agent role divisions during the training process and thus shares parameters among agents with the same role, reducing computational complexity. In addition, our framework can be combined with various typical MARL algorithms, including IPPO, MAPPO, etc. We conduct 7 experiments in 4 representative resource allocation scenarios, where extensive results demonstrate our method's superior performance, outperforming the state-of-the-art baseline methods by up to 31%. Our source codes are available at https://github.com/tsinghua-fib-lab/DyPS.

The Domain Name System (DNS) runs as a fundamental infrastructure of the mobile Internet. Various DNS protocols employed in the current network ecology can be predominantly classified as unencrypted DNS and encrypted DNS. However, existing research mainly focuses on assessing DNS performance within conventional internet structures, neglecting their evaluation in mobile contexts. In our pioneering study examining DNS within mobile networks, we developed an Android-based application to evaluate the efficiency and reliability of DNS protocols. The App issues nine domain name lookups to four cloud DNS providers supporting unencrypted and encrypted DNS protocols. Collaborating with volunteers from four countries, we collected about 52,000 test records. Our findings reveal substantial variability in the efficiency and reliability of all DNS protocols across different mobile scenarios. Overall, encrypted DNS protocols exhibit superior efficiency compared to plaintext DNS when oriented towards cloud DNS resolvers. In high-speed mobile scenarios, all DNS protocols demonstrate reduced efficiency, with encrypted DNS protocols showing relatively higher reliability. Our broad-scale measurement results indicate that the performance of DNS protocols varies across mobile contexts, but users are typically uninformed about these differences and do not realize how to break free. Intending to assist users in selecting an optimal DNS protocol, we propose a protocol choice model based on auto-encoding LSTM networks which leverages features of networking and protocols to predict the most suitable DNS protocol with reduced query time and enhanced reliability in the current scenario. Notably, we have achieved the prediction of the optimal DNS protocol for the future by foreseeing the network status ahead. Empirical results demonstrate an impressive 98.73% accuracy in prediction of DNS protocol selection.

The existing commodity Wi-Fi-based human gait recognition systems mainly focus on a single subject due to the challenges of multisubject walking monitoring. To tackle the problem, we propose Wi-Diag, the first commodity Wi-Fi-based multisubject abnormal gait diagnosis system that leverages only one pair of off-the-shelf commercial Wi-Fi transceivers to separate each subject’s gait information and maintains an excellent performance when the scenario changes. It is an intelligent multisubject gait diagnosis system that can release an experienced doctor from heavy load work. Multisubject abnormal gait diagnosis is modeled as a blind source separation (BSS) issue, and multisubject walking mixed signals are efficiently separated by IC analysis (ICA) approach. This fact is verified by comprehensive theoretical derivation and experimental validation. In addition, CycleGAN is leveraged to mitigate the environmental dependency so that Wi-Diag can be robust when the scenario changes. The excellent performance of Wi-Diag is verified by extensive experiments. The average mean diagnosis accuracy with a maximum group size of four and various scenarios is 87.77%.

Behavior-based Wi-Fi user authentication has gained popularity in user-centered smart systems. However, its wide adoption has been hindered by certain critical issues, including significant performance degradation when the environment changes, the inability to handle unknown activities, and weak security due to basing authentication on the recognition of a single, one-off activity. In this paper, we propose Wi-Dist, which authenticates a user using a behavior password, i.e. a pre-chosen sequence of activities. Wi-Dist addressed the previously mentioned technical challenges through a cross-layer joint optimization framework. In particular, we address environment dependency by incorporating adversarial learning and optimizing both the signal layer and the domain adaptation layer. This enhances the performance of the learned model across various environments. To effectively handle unknown behaviors, we utilize an adversarial learning-based network. This network establishes a pseudo-decision boundary between samples from known and unknown sources, ensuring robust authentication. Additionally, for authentication using continuous activities, we employ double-sliding windows activity monitoring. This approach, coupled with activity state correction, partitions activities for accurate recognition. We also conducted extensive experiments in indoor environments to demonstrate that Wi-Dist is effective and robust.

Mobile Edge Computing (MEC) technology can be implemented at cellular base stations, enabling flexible and configurable provisions of services for mobile users to access. Nevertheless, the conventional solutions mainly focus onstaticalservice provisioning, which ignores the dynamic nature of the arriving service requests. In this work, we first conduct comprehensive data-driven observations on over 4 million service requests throughout 9,800 base stations. Our key findings suggest that users’ demands intrinsically exhibit spatial and temporal patterns, which inevitably lead to performance degradation in statical service provisioning. Motivated by that, we design and implement MobiEdge, a predictive service provisioning system with online learning in wireless edge networks. We propose a graph embedding learning-based model for representation learning, thus to achieve accurate request prediction at different base stations. Then, based on the prediction of incoming service requests, we study the service provisioning reconfiguration problem, i.e., how to jointly optimize service placement and corresponding request scheduling across dual timescales, under constraints of network resources and the total budget. By leveraging the submodular technique, we transform the research issue into a submodular function maximization problem under the$q$-independence system constraint, where$q$is a positive constant related to the ratio of coefficients in constraint conditions. On this basis, we propose a$1/(1+q)$approximation algorithm with rigorous theoretical analysis on the bounded maximum utility. Extensive trace-driven evaluations are conducted over networks of different scales, and MobiEdge shows remarkable performance enhancements by achieving the accuracy of up to 98% in service prediction and an average utility of 92.9% to the optimal solution in service provisioning.

Considering the long charging time and limitation of available charging stations in the traditional battery recharging facilities, the construction of battery swapping station (BSS) has become a new paradigm to satisfy the energy demand timely and sufficiently. Previous solutions lack a joint consideration of battery swapping station establishment and request assignment with the long distance subsidy. In this paper, we study the Station Placement and Assignment (SPA) problem to minimize the overall operation cost. Unfortunately, it shows great difficulty due to the infinite candidate locations and the complex coupling relation for request assignments. To address these challenges, we first devise the bundle generation strategy to reduce the infinite candidate location to finite, then propose an efficient algorithm based on the greedy strategy to assign the battery swapping requests. Extensive evaluations are carried out to show the outstanding performance of our proposed algorithms.

Abstract Pulmonary lobe segmentation is an important prerequisite for accurately quantifying pulmonary damage in many pulmonary diseases and planning treatment. However, due to the incomplete lobar structures and morphological changes caused by diseases, the lobe segmentation still encounters great challenges. In this study, a Learnable Interpolation and Extrapolation Network (LIE‐Net) is proposed to form complete and consecutive fissure surfaces by learning to extract information of the fissures from existing fissure points and absent points (unsegmented points belonging to fissures) to predict the z coordinate of the absent fissure points. The completed pulmonary fissures are further used for accurate pulmonary lobe segmentation. Specifically, LIE‐Net takes the coordinate information of existing fissure points (their ( x , y , z ) coordinates) and absent fissure points (their ( x , y ) coordinates) as two independent inputs, and predicts the z coordinates of absent points. The proposed LIE‐Net makes voxel‐wise predictions based on the spatial structure characteristics of the lung fissure, and is able to provide a consecutive fissure surface in space. According to the evaluation of radiologists, the lobe segmentation performance was remarkably enhanced in approximately 76% of patients in our additional dataset after the application of LIE‐Net, especially for those cases with large‐scale missing fissures.

With the gradual improvements in COVID-19 metrics and the accelerated immunization progress, countries around the world have began to focus on reviving the economy while continuously strengthening epidemic control. POInt-of-Interest (POI) reopening, as a necessity for restoring human mobilities, has become a crucial step to recouple economic recovery and public health management. In contrast to the lock-down policy, POI reopening demands a dynamic trade-off between epidemic interventions and economic costs. In the urban scenario, there exist three key challenges in developing effective POI reopening strategies as follows. (1) During the POI reopening process, there are multiple urban factors affecting the epidemic transmission, which are difficult to simultaneously incorporate and balance in a single reopening strategy; (2) the effects of POI reopening on both economic recovery and epidemic control are long-term, which are hard to capture by static models; and (3) the dual objectives of minimizing infections and maintaining POIs' visits are conflicting, making it difficult to achieve a flexible and scalable trade-off. To tackle the above challenges, we propose Reopener, a deep reinforcement learning (RL) framework for smart POI reopening. First, we utilize a bipartite graph neural network to automatically encode all urban factors that would affect the epidemic prevention and POI visit restriction. Second, we employ a RL-based deep policy network to enable flexible updates in restrictions on POIs along with the trend of epidemic. Third, we design a novel reward function to guide the RL agent to learn smartly, thus comprehensively trading off infections and visit sustainability of POIs. Extensive experimental results demonstrate that Reopener outperforms all baseline methods with remarkable improvements, by reducing the overall economic cost by at least 6.42%. Reopener can effectively suppress infections and support a phase-based POI reopening process, which provides valuable insights for strategy design in post-COVID-19 economic recovery.

As a preliminary exploration, based on the historical order data of the battery-swap stations, this paper predicts the order quantity of the stations in the future period, which can be used as a reference for the battery scheduling of battery-swap stations. Compared with the existing studies, we establish the network of battery-swap stations based on the real large-scale order data, and build the graph model combined with multihead attention to predict. The challenges in this paper include similarity analysis, building the adjacency matrix of battery-swap stations, and the model for distributed stations fine-grained order volume prediction. To address the above challenges, we build the GAT model to accurately predict the order quantity of batteryswap stations. Our work consists of two parts. First, we construct three features and calculate the similarity among the features based on the spatio-temporal correlation data of battery-swap stations. The similarity will be used to construct the association matrix of stations and fuse it with the distance adjacency matrix to construct the topological network of battery-swap stations. Second, we build a graph neural network model and input the network structure of battery-swap stations and historical order data for prediction. In order to capture the correlation features of data, we introduce the multi-attention mechanism, which uses attention to capture the features of different graph nodes, and finally achieves better prediction effect through multi-connection. Our model performs well in hourly order prediction, with an average MAE of about 1.07. We believe that this work can provide reference and ideas for enterprises in order balance, battery scheduling and other aspects.

To control the outbreak of COVID-19, efficient individual mobility intervention for EPidemic Control (EPC) strategies are of great importance, which cut off the contact among people at epidemic risks and reduce infections by intervening the mobility of individuals. Reinforcement Learning (RL) is powerful for decision making, however, there are two major challenges in developing an RL-based EPC strategy: (1) the unobservable information about asymptomatic infections in the incubation period makes it difficult for RL's decision-making, and (2) the delayed rewards for RL causes the deficiency of RL learning. Since the results of EPC are reflected in both daily infections (including unobservable asymptomatic infections) and long-term cumulative cases of COVID-19, it is quite daunting to design an RL model for precise mobility intervention. In this paper, we propose a Variational hiErarcHICal reinforcement Learning method for Epidemic control via individual-level mobility intervention, namely Vehicle. To tackle the above challenges, Vehicle first exploits an information rebuilding module that consists of a contact-risk bipartite graph neural network and a variational LSTM to restore the unobservable information. The contact-risk bipartite graph neural network estimates the possibility of an individual being an asymptomatic infection and the risk of this individual spreading the epidemic, as the current state of RL. Then, the Variational LSTM further encodes the state sequence to model the latency of epidemic spreading caused by unobservable asymptomatic infections. Finally, a Hierarchical Reinforcement Learning framework is employed to train Vehicle, which contains dual-level agents to solve the delayed reward problem. Extensive experimental results demonstrate that Vehicle can effectively control the spread of the epidemic. Vehicle outperforms the state-of-the-art baseline methods with remarkably high-precision mobility interventions on both symptomatic and asymptomatic infections.

Indoor localization has become an essential demand driven by indoor location-based services (ILBSs) for mobile users. With the rising of Internet of Things (IoT), heterogeneous smartphones and wearables have become ubiquitous. However, the ILBSs for heterogeneous IoT devices confront significant challenges, such as received signal strength (RSS) variances caused by hardware heterogeneity, multipath reflections from complex environments, and localization time restricted by computation resources. This article proposes EdgeLoc, a robust and real-time indoor localization system toward heterogeneous IoT devices to solve the above challenges. In particular, the RSS fingerprinting data of Wi-Fi is employed for localization and tackling the heterogeneity of IoT devices in twofold. First, feature-level and signal-level solutions are presented to address the random RSS variances. At the feature level, this work proposes a novel capsule neural network model to efficiently extract incremental features from RSS fingerprinting data. At the signal level, a multistep dataflow is further devised to process RSS fingerprints into image-like data, which utilizes the feature matrix to reduce absolute sensing errors introduced by hardware heterogeneity. Second, an edge-IoT framework is designed to utilize the edge server to train the deep learning model and further supports real-time localization for heterogeneous IoT devices. Extensive field experiments with over 33 600 data points are conducted to validate the effectiveness of EdgeLoc with a large-scale Wi-Fi fingerprint data set. The results show that EdgeLoc outperforms the state-of-the-art SAE-CNN method in localization accuracy by up to 14.4%, with an average error of 0.68 m and an average positioning time of 2.05 ms.

Practicing gymnastics activities at home with online resources has become an increasingly popular choice due to its convenience and accessibility. However, without face-to-face guidance by a trainer, a major challenge is how to assess the quality of performed gymnastics activities, effectively and fairly. Existing intrusive assessing approaches usually require live cameras or wearable sensors, which usually generate privacy and feasibility concerns. There is a lacking of accurate approaches to assess the quality of the activities. To address these challenges, a gymnastics activity assessment approach is proposed in this article, and Wi-Gym, an effective first-of-its-kind gymnastics activity assessment system is developed utilizing commodity Wi-Fi. Wi-Gym is designed to compare the activity-induced channel state information (CSI) dynamics by an exerciser and that of a trainer utilizing dynamic time warping (DTW). The comparison results are provided by a fuzzy inference system (FIS). To make Wi-Gym robust to the changes in the environment, domain adaptation is leveraged to mitigate the data distribution imbalance caused by the environment changes. Extensive experimental studies have been conducted using Wi-Gym, acoustic, and video-based sensing systems. The experimental results validate the effectiveness and robustness of the proposed approach.

With the proliferation of mobile and Internet of Things (IoT) devices, there has been an unprecedented growth of data consumption and computation requests at the network edge. To support latency-sensitive and resource-intensive mobile services, cellular base stations can be integrated with Mobile Edge Computing (MEC) technologies for service provisioning. MEC prompts flexible and configurable provisions of applications or services to make more efficient responses to mobile users' demands. Nevertheless, the time-varying nature of service demands inevitably becomes a vital challenge for existing service provisioning solutions. In this work, we propose MobiEdge, a multi-frame service provisioning scheme across two distinct timescales for MEC networks under various constraints of computation, communication and storage resources. We show that the large timescale indicated by ‘frame’ is more suitable for adjusting edge server activation and service placement, while the small timescale indicated by ‘time slot’ is more feasible to schedule users' requests. By leveraging the submodular techniques, we formulate a joint optimization problem and further propose an approximation algorithm with theoretical analysis and proofs. Synthetic and trace-driven evaluation results validate that MobiEdge can benefit both service providers and mobile users in MEC with high profits (e.g., 94% of the optimal) and a relatively low complexity.

With the rapid development of smart cities, smart buildings are generating a massive amount of building sensing data by the equipped sensors. Indeed, building sensing data provides a promising way to enrich a series of data-demanding and cost-expensive urban mobile applications. In this paper, as a preliminary exploration, we study how to reuse building sensing data to predict traffic volume on nearby roads. Compared with existing studies, reusing building sensing data has considerable merits of cost-efficiency and high-reliability. Nevertheless, it is non-trivial to achieve accurate prediction on such cross-domain data with two major challenges. First, relationships between building sensing data and traffic data are not unknown as prior, and the spatio-temporal complexities impose more difficulties to uncover the underlying reasons behind the above relationships. Second, it is even more daunting to accurately predict traffic volume with dynamic building-traffic correlations, which are cross-domain, non-linear, and time-varying. To address the above challenges, we design and implement BuildSenSys, a first-of-its-kind system for nearby traffic volume prediction by reusing building sensing data. Our work consists of two parts, i.e., Correlation Analysis and Cross-domain Learning. First, we conduct a comprehensive building-traffic analysis based on multi-source datasets, disclosing how and why building sensing data is correlated with nearby traffic volume. Second, we propose a novel recurrent neural network for traffic volume prediction based on cross-domain learning with two attention mechanisms. Specifically, a cross-domain attention mechanism captures the building-traffic correlations and adaptively extracts the most relevant building sensing data at each predicting step. Then, a temporal attention mechanism is employed to model the temporal dependencies of data across historical time intervals. The extensive experimental studies demonstrate that BuildSenSys outperforms all baseline methods with up to 65.3 percent accuracy improvement (e.g., 2.2 percent MAPE) in predicting nearby traffic volume. We believe that this work can open a new gate of reusing building sensing data for urban traffic sensing, thus establishing connections between smart buildings and intelligent transportation.

With the unprecedented demand of location-based services in indoor scenarios, wireless indoor localization is emerging as an essential application for mobile users. While the line-of-sight GPS signal is not available at indoor spaces, WiFi fingerprinting using received signal strength (RSS) has become popular with its ubiquitous accessibility. Although the fingerprinting data can be easily collected by portable mobile devices, to achieve robust and efficient indoor localization remains challenging with two constraints. First, the localization accuracy will be degraded by the random fluctuation of signals that caused by multipath effects from RSS signals. Second, indoor localization algorithms are time-consuming due to the handcrafting features and complex filtering on raw dataset. To achieve high localization accuracy with WiFi fingerprinting, in this paper, we propose CapsLoc, a robust indoor localization system by using capsule networks. Specifically, the capsule network model can efficiently extract hierarchical structures from WiFi fingerprint with three main components, including a convolutional layer, a primary capsule layer and a feature capsule layer. We conduct a real-world experimental field test with over 33600 data points. The experimental results show that CapsLoc can achieve accurate indoor localization with an averaged error of 0.68 m, which outperforms conventional machine learning methods (KNN and SVM) and existing deep learning methods (CNN and SAE-CNN).

Mobile cloud computing has emerged as a pervasive paradigm to execute computing tasks for capacity- limited mobile devices. More specifically, at the network edge, the resource-rich and trusted cloudlet system is acting as a 'data center in a box' to support compute-intensive mobile applications. The mobile cloudlets can provide in-proximity services by executing the workloads for nearby devices. Nevertheless, load balancing in mobile cloudlet network is of great importance, as it has a huge impact on task response time. Existing methods for cloudlet load balancing basically rely on the strategic placement or user cooperation. However, the above solutions require the global task load information from the whole network, which is costly in both communication and computation. To achieve more efficient and low-cost load balancing, we propose 'CTOM', a Collaborative Task Offloading Mechanism for mobile cloudlet networks. Our solution is based on the balls-and-bins theory and can balance the task load only requiring limited information. Extensive simulations and evaluation based on mobility trace demonstrate that, our CTOM outperforms the conventional random and proportional allocation schemes by reducing the task gaps among mobile cloudlets by 65% and 55% respectively. Meanwhile, CTOM's performance is close to that of the greedy algorithm but with much lower computing complexity.

Abstract Flower blooming is a beautiful phenomenon in nature as flowers open in an intricate and complex manner whereas petals bend, stretch and twist under various deformations. Flower petals are typically thin structures arranged in tight configurations with heavy self‐occlusions. Thus, capturing and reconstructing spatially and temporally coherent sequences of blooming flowers is highly challenging. Early in the process only exterior petals are visible and thus interior parts will be completely missing in the captured data. Utilizing commercially available 3D scanners, we capture the visible parts of blooming flowers into a sequence of 3D point clouds. We reconstruct the flower geometry and deformation over time using a template‐based dynamic tracking algorithm. To track and model interior petals hidden in early stages of the blooming process, we employ an adaptively constrained optimization. Flower characteristics are exploited to track petals both forward and backward in time. Our methods allow us to faithfully reconstruct the flower blooming process of different species. In addition, we provide comparisons with state‐of‐the‐art physical simulation‐based approaches and evaluate our approach by using photos of captured real flowers.

Summary This paper investigates the selfish load balancing problem in mobile distributed crowdsourcing networks. Conventional methods heavily relied on cooperation among users to achieve balanced resource utilization in a platform‐centric view. In achieving fairly low communication and computational overhead, this work leverages the d‐choice method based on Ball and Bin theory for effective balancing under limited information and the Proportional Allocation scheme for selfish load balancing, maintaining good load balancing property among selfish users. Even with limited information, the balancing performance could be improved significantly. Moreover, theoretical analysis has been presented in convergence property. Extensive evaluations have been made to show that Chance‐Choice outperforms several existing algorithms. Typically, comparing with Proportional Allocation scheme, it could decrease the load gap between the maximum and the minimal in system by 50% to 80% and reduce the overhead complexity from O(n) to O(1) comparing with the Max‐weight Best Response algorithm, where n denotes the number of mobile users in a crowdsourcing system. Copyright © 2017 John Wiley & Sons, Ltd.

Abstract With the pervasive use of smart mobile devices and increasing wireless networking technologies, collaborations among mobile users are becoming deeper and ubiquitous. Appropriate task collaborations among mobile users could effectively improve the network processing ability with so called ‘mobile cloud’ or ‘cloudlet’. However, task allocations confront with the security issues. The possible collusion or re‐collaborations among the mobile users would possibly merge the allocated tasks of the specific users. Moreover, considering the delivery reliability and task execution efficiency, replications are applied for enhancement, which would also lead to more sever security threat for users. We investigate how to secure the security when task collaborations are allowed for mobile users. Our security scheme is built upon the load balancing scheme, and our intuitive solution is, if the tasks could be effectively balanced among users, the security issues could be guaranteed, because averaging the task assignment could effectively raise the threshold for collusion among potential malicious users. In this work, we propose ‘SAFE‐CROWD’: a secure task offloading and reassignment scheme among mobile users. The basic idea is simple, we leverage the ‘ball and bin’ theory for task assignment, wheredmobile users in contact range are investigated, and we select the least loaded ones among them. It has been proved that such simple cases can effectively reduce the largest queueing length from to . Inspired by this theoretical result, we develop a task reassignment policy for security issues. Simulation and trace‐driven studies have shown that our simple but effective scheme could enhance the security for mobile users, when the tasks are collaboratively executed among mobile devices. Copyright © 2015 John Wiley & Sons, Ltd.

With the increasing development of mobile networking technologies, optimization methods for efficient task assignment plays a key role for mobile crowdsourcing process. However, what hiding behind the strategies are solutions to motivate users for participation, which reveals a fundamental problem: the fairness issue of crowdsourcing system. Since the participators are human beings with intensive interest for obtaining benefits, it is reasonable to build a sustainable crowd with guaranteed fairness among users. Thus in this study, we investigate the fairness issue in mobile social network, which could be more complicated when uncontrollable mobile users are concerned. The intuitive solution is, if the tasks could be effectively assigned among users in a balanced way, the fairness could be guaranteed. Unfortunately, there is still a big challenge for this issue, because it's difficult to acquire accurate global information of task loading, which is highly dynamic and distributed. By leveraging the power of two random choices, which is based on the balls and bins theory, we develop a lightweight scheme to allocate tasks. Indeed, we proposed a heuristic algorithm to achieve balanced task allocation effectively with O(1) complexity. To the best of our knowledge, it is the first effort for incorporating fair load balancing in pure distributed mobile crowdsourcing systems. Our extensive evaluation results validate our task offloading algorithm, showing that the proposed scheme outperforms the random choice method.

Studying growth and development of plants is of central importance in botany. Current quantitative are either limited to tedious and sparse manual measurements, or coarse image-based 2D measurements. Availability of cheap and portable 3D acquisition devices has the potential to automate this process and easily provide scientists with volumes of accurate data, at a scale much beyond the realms of existing methods. However, during their development, plants grow new parts (e.g., vegetative buds) and bifurcate to different components --- violating the central incompressibility assumption made by existing acquisition algorithms, which makes these algorithms unsuited for analyzing growth. We introduce a framework to study plant growth, particularly focusing on accurate localization and tracking topological events like budding and bifurcation. This is achieved by a novel forward-backward analysis, wherein we track robustly detected plant components back in time to ensure correct spatio-temporal event detection using a locally adapting threshold. We evaluate our approach on several groups of time lapse scans, often ranging from days to weeks, on a diverse set of plant species and use the results to animate static virtual plants or directly attach them to physical simulators.