In recent years, the success of large language models (LLMs) has driven the exploration of scaling laws in recommender systems. However, models that demonstrate scaling laws are actually challenging to deploy in industrial settings for modeling long sequences of user behaviors, due to the high computational complexity of the standard self-attention mechanism. Despite various sparse self-attention mechanisms proposed in other fields, they are not fully suited for recommendation scenarios. This is because user behaviors exhibit personalization and temporal characteristics: different users have distinct behavior patterns, and these patterns change over time, with data from these users differing significantly from data in other fields in terms of distribution. To address these challenges, we propose SparseCTR, an efficient and effective model specifically designed for long-term behaviors of users. To be precise, we first segment behavior sequences into chunks in a personalized manner to avoid separating continuous behaviors and enable parallel processing of sequences. Based on these chunks, we propose a three-branch sparse self-attention mechanism to jointly identify users' global interests, interest transitions, and short-term interests. Furthermore, we design a composite relative temporal encoding via learnable, head-specific bias coefficients, better capturing sequential and periodic relationships among user behaviors. Extensive experimental results show that SparseCTR not only improves efficiency but also outperforms state-of-the-art methods. More importantly, it exhibits an obvious scaling law phenomenon, maintaining performance improvements across three orders of magnitude in FLOPs. In online A/B testing, SparseCTR increased CTR by 1.72% and CPM by 1.41%. Our source code is available at https://github.com/laiweijiang/SparseCTR.

Recent years have seen rapid advancements in ultra-wideband (UWB)-based localization systems. However, most existing solutions offer only centimeter-level accuracy and support a limited number of UWB tags, which fails to meet the growing demands of emerging sensing applications (e.g., virtual reality). This paper presents MULoc, the first system that can localize an unlimited number of UWB tags with millimeter-level accuracy. At the core of MULoc is the innovative use of UWB phase, which can provide finer-grained distance measurement than traditional time-of-f1ight (ToF) estimates. To accurately obtain phase estimates from unsynchronized devices, we introduce a novel localization scheme called anchor overhearing (AO) and eliminate raw signal errors through a signal-difference-based technique. For precise tag localization, we resolve phase ambiguity by combining a fusion-based filtering method and frequency hopping. We implement MULoc on commercial UWB modules. Extensive experiments demonstrate that our system achieves a median localization error of 0.47 mm and 90-th percentile error of 1.02 cm, reducing the error of traditional method by 91.12%.

Accurate 3D human mesh recovery from point clouds remains challenging. Most existing methods depend on full 3D supervision or complete input data, both of which are difficult to obtain in practice. %cr update This calls for robust solutions capable of handling partial point clouds in a self-supervised manner. However, the incompleteness of point clouds and the absence of supervision signals pose dual challenges. To tackle these challenges, This calls for robust solutions to handle partial point clouds in a self-supervised manner. To tackle the dual challenges of point cloud incompleteness and the absence of supervision signals, we propose a novel method named SS-HMR, which offers three key insights. First, we estimate point-wise semantics in a self-supervised manner to match partial inputs with a canonical template. The resulting correspondences serve as supervision signals for the regression network in human mesh recovery. Second, we incorporate regression-based and optimization-based paradigms into a self-improving loop: the regression network provides strong initialization for optimization, while the optimization routine generates pseudo-labels that, in turn, enhance the regression network. This mutual feedback enables more accurate and stable mesh recovery over time. Third, generating multiple initializations and selecting the best result mitigates the optimization routine's sensitivity to initialization, improving robustness to sparse and noisy data. %cr update Third, to mitigate sensitivity to initialization in the optimization routine, we generate diverse initialization candidates and transform the challenge of escaping local optima into a controllable selection task, improving robustness against sparse and noisy data. Extensive experiments are conducted on three public datasets and results demonstrate that SS-HMR outperforms existing methods. Notably, SS-HMR performs excellently on different test data, whether from original point clouds captured by depth cameras or LiDAR devices, or from noise-added ones. This shows that SS-HMR has strong generalization ability and robustness across different data sources. Codes are available at https://github.com/suchang-99/SS-HMR.

Beyond communication, wireless signals have been extensively utilized for localization, tracking, and sensing in recent years. The key information extracted for these purposes includes distance and angle. While distance measurement accuracy is mainly limited by signal bandwidth, angle accuracy depends on the number of antennas and phase noise. Conventional approaches typically improve angle estimation by boosting signal strength and increasing the number of antennas. In this paper, we propose employing a quantum receiver to substantially improve angle estimation performance. Rather than amplifying signal strength, the quantum receiver reduces the inherent hardware noise. Furthermore, we exploit the unique properties of a quantum RF receiver to construct a multi-antenna quantum system. Using only two physical quantum antennas, we generate virtual antennas by leveraging the receiver's broad frequency range, effectively increasing the number of antennas and significantly improving angle measurement performance. Our experimental results demonstrate that, with only two quantum antennas, we achieve angle estimation performance surpassing that of a conventional RF receiver equipped with 40 antennas. Furthermore, quantum antennas are not constrained by the coupling effects that typically limit the spacing between conventional RF antennas, allowing for much closer placement. This represents a significant step toward reducing the size of antenna arrays while preserving localization and tracking performance.

Click-Through Rate (CTR) models play a vital role in improving user experience and boosting business revenue in many online personalized services.However, current CTR models generally encounter bottlenecks in performance improvement.Inspired by the scaling law phenomenon of Large Language Models (LLMs), we propose a new paradigm for improving CTR predictions: first, constructing a CTR model with accuracy scalable to the model grade and data size, and then distilling the knowledge implied in this model into its lightweight model that can serve online users.To put it into practice, we construct a CTR model named SUAN (Stacked Unified Attention Network).In SUAN, we propose the unified attention block (UAB) as a behavior sequence encoder.A single UAB unifies the modeling of the sequential and non-sequential features and also measures the importance of each user behavior feature from multiple perspectives.Stacked UABs elevate the configuration to a high grade, paving the way for performance improvement.In order to benefit from the high performance of the high-grade SUAN and avoid the disadvantage of its long inference time, we modify the SUAN with sparse self-attention and parallel inference strategies to form LightSUAN, and then adopt online distillation to train the low-grade LightSUAN, taking a high-grade SUAN as a teacher.The distilled LightSUAN has superior performance but the same inference time as the LightSUAN, making it well-suited for online deployment.Experimental results show that SUAN performs exceptionally well and holds the scaling laws spanning three orders * Corresponding author.

CTR (Click-Through Rate) prediction, crucial for recommender systems and online advertising, etc., has been confirmed to benefit from modeling long-term user behaviors. Nonetheless, the vast number of behaviors and complexity of noise interference pose challenges to prediction efficiency and effectiveness. Recent solutions have evolved from single-stage models to two-stage models. However, current two-stage models often filter out significant information, resulting in an inability to capture diverse user interests and build the complete latent space of user interests. Inspired by multi-interest and generative modeling, we propose DiffuMIN (Diffusion-driven Multi-Interest Network) to model long-term user behaviors and thoroughly explore the user interest space. Specifically, we propose a target-oriented multi-interest extraction method that begins by orthogonally decomposing the target to obtain interest channels. This is followed by modeling the relationships between interest channels and user behaviors to disentangle and extract multiple user interests. We then adopt a diffusion module guided by contextual interests and interest channels, which anchor users' personalized and target-oriented interest types, enabling the generation of augmented interests that align with the latent spaces of user interests, thereby further exploring restricted interest space. Finally, we leverage contrastive learning to ensure that the generated augmented interests align with users' genuine preferences. Extensive offline experiments are conducted on two public datasets and one industrial dataset, yielding results that demonstrate the superiority of DiffuMIN. Moreover, DiffuMIN increased CTR by 1.52% and CPM by 1.10% in online A/B testing. Our source code is available at https://github.com/laiweijiang/DiffuMIN.

Watching micro-videos is becoming a part of public daily life. Usually, user watching behaviors are thought to be rooted in their multiple different interests. In the paper, we propose a model named OPAL for micro-video matching, which elicits a user’s multiple heterogeneous interests by disentangling multiple soft and hard interest embeddings from user interactions. Moreover, OPAL employs a two-stage training strategy, in which the pre-train is to generate soft interests from historical interactions under the guidance of orthogonal hyper-categories of micro-videos and the fine-tune is to reinforce the degree of disentanglement among the interests and learn the temporal evolution of each interest of each user. We conduct extensive experiments on two real-world datasets. The results show that OPAL not only returns diversified micro-videos but also outperforms six state-of-the-art models in terms of recall and hit rate.

User interests can be viewed over different time scales, mainly including stable long-term preferences and changing short-term intentions, and their combination facilitates the comprehensive sequential recommendation. However, existing work that focuses on different time scales of user modeling has ignored the negative effects of different time-scale noise, which hinders capturing actual user interests and cannot be resolved by conventional sequential denoising methods. In this paper, we propose a Long- and Short-term Interest Denoising Network (LSIDN), which employs different encoders and tailored denoising strategies to extract long- and short-term interests, respectively, achieving both comprehensive and robust user modeling. Specifically, we employ a session-level interest extraction and evolution strategy to avoid introducing inter-session behavioral noise into long-term interest modeling; we also adopt contrastive learning equipped with a homogeneous exchanging augmentation to alleviate the impact of unintentional behavioral noise on short-term interest modeling. Results of experiments on two public datasets show that LSIDN consistently outperforms state-of-the-art models and achieves significant robustness.

The respiration state during overnight sleep is an important indicator of human health. However, existing contactless solutions for sleep respiration monitoring either perform in controlled environments and have low usability in practical scenarios or only provide coarse-grained respiration rates, being unable to accurately detect abnormal events in patients. In this article, we propose Respnea, a non-intrusive sleep respiration monitoring system using an ultra-wideband device. Particularly, we propose a profiling algorithm, which can locate the sleep positions in non-controlled environments and identify different subject states. Further, we construct a deep learning model that adopts a multi-head self-attention mechanism and learns the patterns implicit in the respiration signals to distinguish sleep respiration events at a granularity of seconds. To improve the generalization of the model, we propose a contrastive learning strategy to learn a robust representation of the respiration signals. We deploy our system in hospital and home scenarios and conduct experiments on data from healthy subjects and patients with sleep disorders. The experimental results show that Respnea achieves high temporal coverage and low errors (a median error of 0.27 bpm) in respiration rate estimation and reaches an accuracy of 94.44% on diagnosing the severity of sleep apnea-hypopnea syndrome.

Recent years have witnessed a tremendous amount of interest in wireless sensing, i.e., instead of employing traditional sensors, wireless signal is utilized for sensing purposes. Contact-free wireless sensing has been successfully demonstrated using various RF signals such as WiFi, RFID, LoRa, and mmWave, enabling a large range of applications. However, limited by hardware thermal noise, the granularity of RF sensing is still relatively coarse. In this paper, instead of using the macro signal power/phase for sensing, we propose the first quantum wireless sensing system, which uses the micro energy level of atoms for sensing, improving the sensing granularity by an order of magnitude. The proposed quantum wireless sensing system is capable of utilizing a wide spectrum of frequencies (e.g., 2.4 GHz, 5 GHz and 28 GHz) for sensing. We demonstrate the superior performance of quantum wireless sensing with two widely-used signals, i.e., WiFi and 28 GHz millimeter wave. We show that quantum wireless sensing can push the sensing granularity of WiFi from millimeter level to sub-millimeter level and push the sensing granularity of millimeter wave to micrometer level.

User behavior sequence modeling plays a significant role in Click-Through Rate (CTR) prediction on e-commerce platforms. Except for the interacted items, user behaviors contain rich interaction information, such as the behavior type, time, location, etc. However, so far, the information related to user behaviors has not yet been fully exploited. In the paper, we propose the concept of a situation and situational features for distinguishing interaction behaviors and then design a CTR model named Deep Situation-Aware Interaction Network (DSAIN). DSAIN first adopts the reparameterization trick to reduce noise in the original user behavior sequences. Then it learns the embeddings of situational features by feature embedding parameterization and tri-directional correlation fusion. Finally, it obtains the embedding of behavior sequence via heterogeneous situation aggregation. We conduct extensive offline experiments on three real-world datasets. Experimental results demonstrate the superiority of the proposed DSAIN model. More importantly, DSAIN has increased the CTR by 2.70%, the CPM by 2.62%, and the GMV by 2.16% in the online A/B test. Now, DSAIN has been deployed on the Meituan food delivery platform and serves the main traffic of the Meituan takeout app. Our source code is available at https://github.com/W-void/DSAIN.

Existing contactless solutions on sleep respiration monitoring are either performed in controlled environments, having poor usability in practical scenarios, or only provide coarse-grained respiration rates, being unable to accurately detect abnormal events of patients. In this paper, we propose Respnea, a non-invasive sleep respiration monitoring system using an impulse-radio ultra-wideband (IR-UWB) radar. Particularly, we propose a profiling algorithm, which can locate the sleep positions in non-controlled environments and identify different states of subjects. Further, we construct a deep learning model which adopts a multi-headed self-attention and learn the patterns implicit in the respiration signal so as to distinguish sleep respiration events at a granularity of seconds. We conduct experiments on data collected from patients with sleep disorders and healthy subjects. The experimental results show that Respnea achieves a low error (less than 0.27 bpm) in respiration rate estimation and reaches the accuracy of 88.89% diagnosing the severity of Sleep Apnea-Hypopnea Syndrome.

Abstract This paper presents AcousticPAD, a contactless and robust handwriting recognition system that extends the input and interactions beyond the touchscreen using acoustic signals, thus very useful under the impact of the COVID-19 epidemic. To achieve this, we carefully exploit acoustic pulse signals with high accuracy of time of fight (ToF) measurements. Then we employ trilateration localization method to capture the trajectory of handwriting in air. After that, we incorporate a data augmentation module to enhance the handwriting recognition performance. Finally, we customize a back propagation neural network that leverages augmented image dataset to train a model and recognize the acoustic system generated handwriting characters. We implement AcousticPAD prototype using cheap commodity acoustic sensors, and conduct extensive real environment experiments to evaluate its performance. The results validate the robustness of AcousticPAD, and show that it supports 10 digits and 26 English letters recognition at high accuracies.

Training multiple agents to perform safe and coop-erative control in the complex scenarios of autonomous driving has been a challenge. For a small fleet of cars moving together, this paper proposes Lepus, a new approach to training multiple agents. Lepus adopts a pure cooperative manner for training multiple agents, featured with the shared parameters of policy networks and the shared reward function of multiple agents. In particular, Lepus pre-trains the policy networks via an adversarial process, improving its collaborative decision-making capability and further the stability of car driving. Moreover, for alleviating the problem of sparse rewards, Lepus learns an approximate reward function from expert trajectories by combining a random network and a distillation network. We conduct extensive experiments on the MADRaS simulation plat-form. The experimental results show that multiple agents trained by Lepus can avoid collisions as many as possible while driving simultaneously and outperform the other four methods, that is, DDPG-FDE, PSDDPG, MADDPG, and MAGAIL(DDPG) in terms of stability.

Besides the communication function, various RF signals such as WiFi and RFID have been actively exploited for sensing purposes recently. However, a missing component of existing RF sensing is sensing under device motions. This paper takes the first step to involve device mobility into the ecosystem of RF sensing. Owning to the miniaturization and low cost of ultra-wideband (UWB) chips in recent years, we propose to integrate the accuracy of UWB sensing with device mobility to support truly ubiquitous RF sensing. This is a challenging task because the motion artifacts from RF devices can easily overwhelm the target motion, such as subtle chest displacement for respiration sensing. In this demo, we propose Mobi2Sense to support sensing under device motions. We propose novel signal processing schemes to remove the effect of device motions on sensing and prototype Mobi2Sense using a commodity UWB module. Comprehensive evaluation demonstrates that Mobi2Sense is able to "hear" music and "see" human respiration at high accuracy in the presence of device motions.

Among various wireless sensing modalities, Ultra-Wideband (UWB) exhibits unique advantages such as fine granularity owing to its super large bandwidth (500 MHz - 2 GHz). Though promising, UWB sensing was only demonstrated on dedicated hardware including DW1000 and XETHRU X4 which are not available in existing consumer-level devices. In the last few years, we observed an interesting trend of UWB module being embedded into consumer-level devices such as smartphones and smart watches. However, leveraging UWB module inside consumer-level devices for sensing poses new challenges. One key challenge is that while dedicated UWB hardware can present us with raw physical-layer signal amplitude and phase, only upper-layer distance and angle information can be extracted from consumer-level devices. In this demo, we address the challenges and present the first UWB sensing system hosted on iPhone and Apple Watch without any dedicated hardware components. We show that with just the upper-layer UWB data reported from smartphones, exciting sensing applications such as fine-grained 3D handwriting and multi-target tracking can be realized, pushing RF sensing one step forward towards real-life adoption.

With the rapid increase of micro-video creators and viewers, how to make personalized recommendations from a large number of candidates to viewers begins to attract more and more attention. However, existing micro-video recommendation models rely on expensive multi-modal information and learn an overall interest embedding that cannot reflect the user's multiple interests in micro-videos. Recently, contrastive learning provides a new opportunity for refining the existing recommendation techniques. Therefore, in this paper, we propose to extract contrastive multi-interests and devise a micro-video recommendation model CMI. Specifically, CMI learns multiple interest embeddings for each user from his/her historical interaction sequence, in which the implicit orthogonal micro-video categories are used to decouple multiple user interests. Moreover, it establishes the contrastive multi-interest loss to improve the robustness of interest embeddings and the performance of recommendations. The results of experiments on two micro-video datasets demonstrate that CMI achieves state-of-the-art performance over existing baselines.

Recently, a zero-shot entity linking task is introduced to challenge the generalization ability of entity linking models. In this task, mentions must be linked to unseen entities and only the textual information is available. In order to make full use of the documents, previous work has proposed a BERT-based model which can only take fixed length of text as input. However, the key information for entity linking may exist in nearly everywhere of the documents thus the proposed model cannot capture them all. To leverage more textual information and enhance text understanding capability, we propose a bidirectional multi-paragraph reading model for the zero-shot entity linking task. Firstly, the model treats the mention context as a query and matches it with multiple paragraphs of the entity description documents. Then, the mention-aware entity representation obtained from the first step is used as a query to match multiple paragraphs in the document containing the mention through an entity-mention attention mechanism. In particular, a new pre-training strategy is employed to strengthen the representative ability. Experimental results show that our bidirectional model can capture long-range context dependencies and outperform the baseline model by 3-4% in terms of accuracy.

Hongyin Tang, Xingwu Sun, Beihong Jin, Jingang Wang, Fuzheng Zhang, Wei Wu. Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers). 2021.

In this paper, we propose a new ranking model DR-BERT, which improves the Document Retrieval (DR) task by a task-adaptive training process and a Segmented Token Recovery Mechanism (STRM).In the task-adaptive training, we first pre-train DR-BERT to be domain-adaptive and then make the two-phase fine-tuning.In the first-phase fine-tuning, the model learns query-document matching patterns regarding different query types in a pointwise way.Next, in the second-phase finetuning, the model learns document-level ranking features and ranks documents with regard to a given query in a listwise manner.Such pointwise plus listwise fine-tuning enables the model to minimize errors in the document ranking by incorporating ranking-specific supervisions.Meanwhile, the model derived from pointwise fine-tuning is also used to reduce noise in the training data of the listwise fine-tuning.On the other hand, we present STRM which can compute OOV word representation and contextualization more precisely in BERT-based models.As an effective strategy in DR-BERT, STRM improves the matching perfromance of OOV words between a query and a document.Notably, our DR-BERT model keeps in the top three on the MS MARCO leaderboard since May 20, 2020.

Currently, cardiac diseases have become one of the biggest health concerns. Existing heartbeat monitoring methods either require dedicated intrusive devices (e.g., ECG devices) that suffer high costs or leverage video camera analyses that are light-sensitive. In this paper, leveraging the acoustic signals sent by a speaker and received by a microphone array, we develop a prototype system to achieve the contactless and low-cost heartbeat monitoring. In particular, while we exploit the passive beamforming to enhance the user's heartbeat signal, we design a filtering method in frequency domain to remove the line-of-sight (LoS) impact and retain the target-reflected signals, and propose a wideband time-delay method to estimate the direction of arrival of target-reflected signal. Thus, our prototype is able to robustly estimate the human heartbeat and push the limit of acoustic sensing range. The experimental results show that our prototype achieves a heart rate monitoring at 1.7 m with the estimation error of 0.5 bpm, which is comparable to ECG or other contact-based solutions.

For autonomous driving in extremely complex scenarios, existing research utilizes deep reinforcement learning or imitation learning to obtain the decision-making capability of agents. However, due to the incomplete information nature of such driving scenarios, existing techniques usually suffer issues such as the incorrect rewards or unstable training which would impact the learning quality seriously. In this paper, we propose a new approach named Columba which trains the agent to learn from expert trajectory data and abnormal trajectory data instead of relying on any manually-set reward functions. In particular, Columba designs a positive and negative feedback regulator to reduce the dangerous or bad states of the car agent at the beginning of training. Further, Columba generates the rewards by coordinating with the discriminator, the random distillation network and the regulator, enhancing the accuracy of rewards. We conduct extensive experiments on the Torcs simulation platform. Experimental results show that the agent trained by Columba outperforms the agents trained by DDPG and GAIL, which are strong baselines in the deep reinforcement learning and the imitation learning, respectively.

Xingwu Sun, Yanling Cui, Hongyin Tang, Qiuyu Zhu, Fuzheng Zhang, Beihong Jin. Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. 2021.

Document retrieval (DR) is a crucial task in NLP. Recently, the pre-trained BERT-like language models have achieved remarkable success, obtaining a state-of-the-art result in DR. In this paper, we come up with a new BERT-based ranking model for DR task, named TABLE. In the pre-training stage of TABLE, we present a domain-adaptive strategy. More essentially, in the fine-tuning stage, we develop a two-phase task-adaptive process, i.e., type-adaptive pointwise fine-tuning and listwise fine-tuning. In the type-adaptive pointwise fine-tuning phase, the model can learn different matching patterns regarding different query types. In the listwise fine-tuning phase, the model matches documents with regard to a given query in a listwise fashion. This task-adaptive process makes the model more robust. In addition, a simple but effective exact matching feature is introduced in fine-tuning, which can effectively compute matching of out-of-vocabulary (OOV) words between a query and a document. As far as we know, we are the first who propose a listwise ranking model with BERT. This work can explore rich matching features between queries and documents. Therefore it substantially improves model performance in DR. Notably, our TABLE model shows excellent performance on the MS MARCO leaderboard.

In the deep reinforcement learning, the sparse reward problem directly impacts the quality of agent training. Existing methods have not been satisfactory, especially for the scenarios with high-dimensional state information. In this paper, we propose a new approach Monoceros to training a game agent. Monoceros can work for the scenarios with high-dimensional state information and alleviate the sparse reward problem during the agent training. Specifically, we present a composite reward function which combines both the knowledge implied in expert trajectories and manually-set reward functions. Moreover, we design a specific policy network to adapt to the high-dimensional information scenarios, and adopt the behavior clone as a pre-training strategy to accelerate the training process. Technically, Monoceros can be applied to train the agents to play First Person Shooter (FPS) games. We conduct extensive experiments on three scenarios in the VIZDoom platform. Experimental results show that in all the scenarios, the agent trained by Monoceros outperforms the agents trained by Arnold and GAIL, which are representative methods in the deep reinforcement learning and the imitation learning, respectively.

Temperature is an important indicator for agriculture irrigation, industrial manufacture, food safety, etc. While temperature measurement can be achieved via dedicated sensors, there still have a demand to sense temperature with ubiquitous computing devices. In this paper, we propose to enable the sound signal to measure the air temperature using commodity acoustic devices. Different from existing FMCW and OFDM based acoustic sensing system, we are the first to employ acoustic pulse signal and get rid of offsets to obtain the accurate sound speed. Then we precisely obtain temperature by quantifying the relation between sound speed and temperature. We build a temperature monitoring prototype named AcousticThermo, and conduct extensive experiments. Experimental results show that the proposed system can achieve an average estimation error of below 0.2°C in various temperature environments.

On-demand cinemas are a new type of offline entertainment venues which have shown the rapid expansion in the recent years. Recommending movies of interest to the potential audiences in on-demand cinemas is keen but challenging because the recommendation scenario is totally different from all the existing recommendation applications including online video recommendation, offline item recommendation and group recommendation. In this paper, we propose a novel spatio-temporal approach called Pegasus. Because of the specific characteristics of on-demand cinema recommendation, Pegasus exploits the POI (Point of Interest) information around cinemas and the content descriptions of movies, apart from the historical movie consumption records of cinemas. Pegasus explores the temporal dynamics and spatial influences rooted in audience behaviors, and captures the similarities between cinemas, the changes of audience crowds, time-varying features and regional disparities of movie popularity. It offers an effective and explainable way to recommend movies to on-demand cinemas. The corresponding Pegasus system has been deployed in some pilot on-demand cinemas. Based on the real-world data from on-demand cinemas, extensive experiments as well as pilot tests are conducted. Both experimental results and post-deployment feedback show that Pegasus is effective.

Hongyin Tang, Miao Li, Beihong Jin. Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP). 2019.

Text clustering is one of the fundamental tasks in natural language processing and text data mining. It remains challenging because texts have complex internal structure besides the sparsity in the high-dimensional representation. In the paper, we propose a new Neural Variational model with mixture-of-Gaussians prior for Text Clustering (abbr. NVTC) to reveal the underlying textual manifold structure and cluster documents effectively. NVTC is a deep latent variable model built on the basis of the neural variational inference. In NVTC, the stochastic latent variable, which is modeled as one obeying a Gaussian mixture distribution, plays an important role in establishing the association of documents and document labels. On the other hand, by joint learning, NVTC simultaneously learns text encoded representations and cluster assignments. Experimental results demonstrate that NVTC is able to learn clustering-friendly representations of texts. It significantly outperforms several baselines including VAE+GMM, VaDE, LCK-NFC, GSDPMM and LDA on four benchmark text datasets in terms of ACC, NMI, and AMI. Furthermore, NVTC learns effective latent embeddings of texts which are interpretable by topics of texts, where each dimension of latent embeddings corresponds to a specific topic.

Predicting the passenger flows in a city, especially in a metropolis, can guide traffic dispersion, and help assessing the risks of public safety and improving urban planning. However, it is challenging as passenger flows in a road network may vary with time and space, affected by weather conditions, urban activities, etc. In the paper, we propose a passenger flow prediction approach named Yildun, which constructs an encoder-decoder neural network and captures the spatial and temporal correlations inherent in passenger flows. More specifically, to predict the passenger flows at each and every station, a spatial attention mechanism is presented to adaptively extract inter-station correlations of flows by referring to the previous hidden state of the encoder at each time step. Meanwhile, a temporal attention mechanism is employed to capture time-dependent connections of flows by selecting relevant hidden states of the encoder across all time steps. Further, extra factors, such as POI (Point of Interest) data and day of the week, are fused in the decoder. With this spatio-temporal attention scheme, Yildun not only can make predictions effectively, but also is easily explainable. Extensive experiments are conducted on large-scale real-world data. The experimental results show that Yildun can predict passenger flows with small prediction errors and outperforms five baselines significantly.

In large and medium-sized cities, detecting unusual changes of crowds of people on the streets is needed for public security, transportation management, emergency control, and terrorism prevention. As public transportation has the capability to bring a large number of people to an area in a short amount of time, real-time discovery of anomalies in passenger numbers is an effective way to detect crowd anomalies. In this paper, we devise an approach called Kochab. Kochab adopts a generative model and combines the prior knowledge about passenger flows. Hence, it can detect anomalies in the numbers of incoming and outgoing passengers within a certain time and spatial area, including anomalous events along with their durations and severities. Through well-designed inference algorithms, Kochab requires only a moderate amount of historical data to be sample data. As such, Kochab shows good performance in real time and makes prompt responses to user' s interactive analysis requests. In particular, based on the recognized anomalous events, we capture event patterns which give us hints to link to activities or status in cities. In addition, for the convenience of method evaluation and comparison, we create an open Stream Anomaly Benchmark on the basis of large-scale real-world data. This benchmark will prove useful for other researchers too. Using this benchmark, we compare Kochab with four other methods. The experimental results show that Kochab is sensitive to population flow anomalies and has superior accuracy in detecting anomalies in terms of precision, recall and the F1 score.

Vehicles on the roads have high heterogeneity in vehicle types. Real-time and full-coverage vehicle classification has always been a challenge. Existing intrusive and non-intrusive methods cannot meet the requirements with satisfaction. Considering that signaling data from mobile operators have the advantages such as the wide coverage and the low cost, a new approach named Lepus, which analyzes the signaling stream to achieve the real-time multi-class classification of vehicles on highways, is proposed. Following the Lepus, the historical GPS trajectories with vehicle types and the signaling trajectories occurring at the same time and space are first examined to establish the relation among signaling trajectories, vehicles and vehicle types and then identify signaling-recognizable vehicles. Further, the driving characteristics of these labeled signaling-recognizable vehicles are analyzed so as to determine vehicle classification features. Finally, the vehicle classification model is established and used to analyze the incoming signaling stream and classify the vehicles in real time. Extensive experiments are conducted on real data and the results show that the Lepus approach is effective in real time vehicle classification.

Analyzing and mining trajectories of moving objects (such as persons or vehicles) in the cities bring a promising way to discover the potential knowledge and therefore can foster diversified applications, including personalized travel services, intelligent transportation systems (ITSs), and etc. For increasing the intelligence of current public transit systems, the paper proposes to discover and utilize the patterns in passenger trajectory streams to optimize bus scheduling. More specifically, the paper first analyzes the real world data from bus smart cards so as to fully understand the nature of passenger trajectories and bus operations. Based on it, the paper defines a new trip pattern, i.e., the frequent bus passenger trip pattern for bus scheduling (the FBPT4BS pattern in short). Then, the paper proposes an approach. The approach gives the procedure of discovering FBPT4BS patterns from passenger trajectory streams and finds the bus lines whose capacities are not enough to satisfy the passengers' travel demands. Further, the approach gives the suggestion on the corresponding scheduling adjustment strategy for bus lines. Experiments are conducted on the data from the Beijing Public Transport Group. The experimental results show that the proposed approach can efficiently decrease the travel times of passengers.

Sensory data are often of low quality, for example, data are incomplete, ambiguous, or indirect, which has become the bottleneck of many data-driven applications. Two kinds of data which are handled in the paper for estimating traffic volumes on highways are no exception. In particular, the traffic volume data obtained from the loop detectors are accurate but sparse, and the mobile signaling data for estimating relative traffic volumes are wide in coverage and low in cost, but they are indirect and inaccurate. Keeping the characteristics of data in mind, the paper proposes a data fusion approach named Polaris which extends compressive sensing to estimate traffic volumes on highways. The Polaris analyzes the sparsity of the traffic volumes reported by detectors, mines the spatial-temporal correlations between the two kinds of data, and then gives the computational steps in the light of compressive sensing. Experiments are conducted on the large-scale real signaling data and the loop detector data. The experimental results show that the Polaris has the lowest estimation errors in comparison with several other methods. The corresponding Polaris system has been built and deployed in Fujian Province, China. It can obtain real-time traffic volumes on the highways with full coverage at a very low cost.1

Compared to general vehicular systems, bus systems have advantages including wide coverage, fixed routes, and regular service. Inspired by these unique features of the bus systems, we propose to use the bus systems as routing backbones of VANETs. In this work, we present a Community-based Bus System (CBS) which consists of two components: a community-based backbone and a routing scheme over the backbone. The backbone construction is a one-off operation which is done offline while the routing is done online in individual buses. We build a community-based backbone by applying community detection techniques and propose a twolevel routing scheme which operates over the backbone. The proposed routing scheme performs sequentially in the inter-community level and the intra-community level, and is able to support message delivery to both buses and specific locations/areas. We develop a probabilistic model to analyze the message delivery latency of CBS. The average error of the analytically-derived latency is shown to be 8.9 percent of the latency derived from the real traces. Extensive experiments are conducted on real-world traces from the Beijing bus system and the Dublin bus system and the results show that CBS can significantly lower the delivery latency and improve the delivery ratio, compared to the existing solutions. CBS is a general solution which is applicable to any bus-based VANETs.

The newly emerging location-based social networks (LBSN) such as Tinder and Momo extends social interaction from friends to strangers, providing novel experiences of making new friends. Familiar strangers refer to the strangers who meet frequently in daily life and may share common interests; thus they may be good candidates for friend recommendation. In this paper, we study the problem of discovering familiar strangers, specifically, public transportation trip companions, and their common interests. We collect 5.7 million transaction records of smart cards from about 3.02 million people in the city of Beijing, China. We first analyze this dataset and reveal the temporal and spatial characteristics of passenger encounter behaviors. Then we propose a stability metric to measure hidden friend relations. This metric facilitates us to employ community detection techniques to capture the communities of trip companions. Further, we infer common interests of each community using a topic model, i.e., LDA4HFC (Latent Dirichlet Allocation for Hidden Friend Communities) model. Such topics for communities help to understand how hidden friend clusters are formed. We evaluate our method using large-scale and real-world datasets, consisting of two-week smart card records and 901,855 Point of Interests (POIs) in Beijing. The results show that our method outperforms three baseline methods with higher recommendation accuracy. Moreover, our case study demonstrates that the discovered topics interpret the communities very well.

With the aid of ubiquitous sensors/devices and pervasive networks, various types of multi-source data in the form of texts, videos, pictures, audios, etc. can be collected. They can be applied to detecting the traffic conditions. For highways, it has inevitable inherent defects in detecting traffic conditions by analyzing data from a single source or user participation. Therefore, the paper proposes a data fusion approach named Phecda. Phecda combines the signaling data of mobile phones with the data from static loop detectors. Phecda works in an unobtrusive way, it not only incorporates with the characteristics of traffic flows, but also includes the strategies of setting and optimizing parameters by learning historical data. Experiments are conducted with the large-scale real-world data as input. The experimental results show that the Phecda approach has high precisions and recalls in vehicle speed estimation. The corresponding Phecda system has been built and deployed in Fujian Province, China. It achieves the highway traffic monitoring with full road segment coverage at a very low cost.

Mobile crowdsensing (MCS) is a promising paradigm that utilizes the mobility of people and the sensing capabilities of their mobile devices to accomplish a variety of sensing tasks. In this paper, we adopt the Signaling System No.7 (SS7) as the MCS platform since SS7 can well capture trajectories and mobility patterns of the mobile users. We collect a real-world SS7 data of 1.18 million mobile users at 3512 cell towers/sites in Xiamen, China. We first analyze this dataset and reveal important characteristics of user mobility. Then, we address a Mobile User Recruitment (MUR) problem which is crucial to all MCS systems. Given SS7 data of mobile users, a set of target cells to be sensed/covered, and recruitment cost functions of the mobile users, the MUR problem is to recruit a set of mobile users such that all the target cells are covered and the total recruitment cost is minimized. Our MUR problem is general and includes the existing problems as its special cases. We prove NP-hardness of the problem. We propose an approximation algorithm to this problem and derive the approximation ratio. Extensive experiments are conducted on the real-world SS7 dataset and results show that the proposed solution outperforms two baseline algorithms by saving 22.6% and 62.9% recruitment costs, respectively, on average.

Bus networks in cities have distinctive features such as wide coverage and fixed bus routes so that they show the potential of forming the communication backbone in vehicular ad hoc networks (VANETs). This paper focuses on the geocast in bus-based VANETs and presents a geocast routing mechanism named Vela. Specifically, Vela analyzes and mines historical bus trajectories and characterizes spatial–temporal patterns (i.e., bus travel-time patterns and bus spatial encounter patterns) in a moderate granularity of road segments, which makes the mined patterns both accurate and steady. Furthermore, Vela exploits these acquired patterns to build a probabilistic spatial–temporal graph model and provides the available routing paths with the best possible quality-of-service levels for data delivery requests. Moreover, Vela also employs a two-hop aware strategy that utilizes the real-time spatial–temporal relationships between buses to increase the chances of forwarding the data. The results of the experiments on the real and synthetic trajectories show that Vela performs much better in terms of delivery ratio and delay and has stronger scalability than the other solutions.

Low delivery latency and high delivery ratio are two key goals in the design of routing schemes in Vehicular Ad Hoc Networks (VANETs). The existing routing schemes utilize real-time information (e.g., Geographical position and vehicle density) and historical information (e.g., Contacts of vehicles), which usually suffer from a long delivery latency and a low delivery ratio. Inspired by the unique features of bus systems such as wide coverage, fixed routes and regular service, we propose to use the bus systems as routing backbones of VANETs. In this work, we present a Community-based Bus System (CBS) which consists of two components: a community-based backbone and a routing scheme over the backbone. We collect real traces of 2515 buses in Beijing and build a community-based backbone by applying community detection techniques in the Beijing bus system. A two-level routing scheme is proposed to operate over the backbone. The proposed routing scheme performs sequentially in the inter-community level and the intra-community level, and is able to support message delivery to both mobile vehicles and specific locations/areas. Extensive experiments are conducted on the real trace data of the Beijing bus system and the results show that CBS can significantly lower the delivery latency and improve the delivery ratio. CBS is applicable to any bus-based VANETs.

In the infotainment applications over Vehicular Ad Hoc Networks(VANETs), the Roadside Units (RSUs) often play the roles of data switching centers. However, the RSUs tend to be the bottlenecks in the data dissemination because of the limited channel bandwidth and heavy traffic loads. In order to improve the throughput of the RSUs, we devise a new MAC protocol Mizar from a cross-layer design perspective. Mizar concentrates on improving the spatio-temporal efficiency in disseminating data from an RSU to the moving vehicles. Leveraging the space diversity of wireless signals, Mizar increases the channel utilization by the cooperative transmission. Moreover, motivated by the location distribution characteristics of the vehicles, Mizar seeks to catch the concurrent transmission opportunities generated in the process of the cooperative communication. It is shown from the experimental results that Mizar can increase the throughput and decrease the transmission delay significantly in comparison with the fixed data rate scheme of IEEE 802.11p, the variable data rate scheme and the cooperative transmission scheme in Mizar.

Ontology reuse offers great benefits by measuring and comparing ontologies. However, the state of art approaches for measuring ontologies neglects the problems of both the polymorphism of ontology representation and the addition of implicit semantic knowledge. One way to tackle these problems is to devise a mechanism for ontology measurement that is stable, the basic criteria for automatic measurement. In this paper, we present a graph derivation representation based approach (GDR) for stable semantic measurement, which captures structural semantics of ontologies and addresses those problems that cause unstable measurement of ontologies. This paper makes three original contributions. First, we introduce and define the concept of semantic measurement and the concept of stable measurement. We present the GDR based approach, a three-phase process to transform an ontology to its GDR. Second, we formally analyze important properties of GDRs based on which stable semantic measurement and comparison can be achieved successfully. Third but not the least, we compare our GDR based approach with existing graph based methods using a dozen real world exemplar ontologies. Our experimental comparison is conducted based on nine ontology measurement entities and distance metric, which stably compares the similarity of two ontologies in terms of their GDRs.

The different types of applications in VANETs impose diversified traffic loads. Although multi-channel operations are applied in the MAC layer to improve throughput and potentially reduce the latency, the existing multi-channel approaches show some limitations to meet the changing demands of applications. In this paper, we present an adaptive channel coordination mechanism named Merak to enhance IEEE 1609.4 by adjusting the length ratio of CCH interval and SCH interval dynamically. To be specific, we model the channel interval allocation with a Markov decision process, and estimate the optimal channel interval for the current traffic load by employing a fuzzy actor-critic algorithm. The extensive experiments are conducted to observe Merak, the alternating access scheme in IEEE 1609.4 and the variable channel interval scheme (VCI). The experimental results indicate that Merak can adapt to the varying traffic loads better than the original IEEE 1609.4 and VCI, especially boosting the timeliness and delivery rate of burst emergent or periodic control message loads on the CCH and maintaining the satisfactory throughput and delivery rate of fluctuating service data on the SCHs.

Abstract With the drastic development of semantic‐driven applications, assessing the quality of ontologies has received more attention. Measuring and assessing the quality of ontologies can help ontology engineers to control project management and reduce the risk of project failures. However, most of the existing ontology metrics for measuring and assessing the quality of ontologies are defined based on ontology structure, and neglect the stability of ontology measurement. In this paper, we concentrate on stable ontology measurement by using semantically derived ontology metrics. We propose four ontology cohesion metrics, which fully consider the implicitly expressed semantic information and are defined based on ontological semantics rather than ontology structure. Before measuring and assessing an ontology, we materialize a pre‐processing for stable ontology measurement by treating the ontology. The proposed ontology cohesion metrics are theoretically validated by the validation criteria of object‐oriented software. The experimental results show that we can successfully collect more semantic knowledge from the testing ontologies for stable ontology measurement by using the proposed ontology cohesion metrics. The ontology cohesion metrics proposed in this paper can be reasonably used as a cogent complementarity of existing ontology metrics. Copyright © 2010 John Wiley & Sons, Ltd.

In Pub/Sub systems, channel-based approaches to routing the subscriptions and events have many advantages such as fewer routing messages, lower costs for subscription management, etc. But a potential issue embedded in this kind of approach, i.e. loadings on different event brokers are apt to unbalancing, is ignored more or less. In this paper, we design a load balancing mechanism and integrate it into a channel-based approach in a Pub/Sub system. In particular, we define a balancing state in a Pub/Sub system, and then propose the balancing control initiation algorithm which decides not only whether to perform load balancing among event brokers but also whether to adjust the number of event brokers. Also we present the load scheduling algorithm which can achieve load balancing by channel splitting, merging and migration. We conduct the experiments by taking loads with different distributions as input to reveal the capability of dealing with changing loads. The experimental data prove that our mechanism can help balance the system loads efficiently and dynamically start or shut down event brokers when facing overloads or insufficient loads.

For constructing a Pub/Sub middleware to detect and disseminate events in the Internet of Things (IoT), we have to pay close attention to the following requirements: (1) the IoT applications usually care about the events with both spatial and temporal extents, (2) It must address the issues brought by the IoT features, such as autonomy embedded in collaboration among things, dynamics implied by running environments. In this paper, we provide a concise and expressive formulation for the events in IoT, i.e. complex spatio-temporal events. We also present the architecture of the corresponding Pub/Sub middleware, which aims at achieving the desired functionality and efficiency.

Pub/Sub systems permit users to submit subscriptions and notify interested users of the events detected in a distributed way. Moving a Pub/Sub system to a cloud infrastructure is for high performance and scalability. This paper describes how to migrate two Pub/Sub systems i.e. PADRES and Once Pub Sub to Xen Cloud Platform, especially proposes black-box method, grey-box method and white-box method so as to take full advantage of cloud mechanisms. This paper then conducts a series of experiments on the Pub/Sub systems in the cloud to evaluate benefits and costs. The experimental results indicate that the black-box method does not always take effect although it can be implemented easily, the grey-box method is more appropriate to a Pub/Sub system if its workload features and brokers' roles are known in advance. Further, the experimental results show the white-box method, combined the load balance mechanism both in the cloud and in a Pub/Sub system, can achieve satisfying performance and scalability especially facing the workload with unidentified distribution.

One of the major applications of wireless ad hoc networks is to extend the Internet coverage and support pervasive and efficient data dissemination and sharing. To reduce data access cost and delay, caching has been widely used as an important technique. The efficiency of data access in caching systems largely depends on the cost for maintaining cache consistency, which can be high in wireless ad hoc networks due to network dynamism. Therefore, to make better trade-off between cache consistency and the cost incurred, it would be highly desirable to provide users the flexibility in specifying consistency requirements for their applications. In this paper, we propose a general consistency model called Probabilistic Delta Consistency (PDC), which integrates the flexibility granted by existing consistency models, covering them as special cases. We also propose the Flexible Combination of Push and Pull (FCPP) algorithm which satisfies user-specified consistency requirements under the PDC model. The analytical model of FCPP is used to derive the balance of minimizing the consistency maintenance cost and ensuring the specified consistency requirement. Extensive simulations are conducted to evaluate whether FCPP can satisfy arbitrarily specified consistency requirements, and whether FCPP works cost-effectively in dynamic wireless ad hoc networks. The evaluation results show that FCPP can adaptively tune itself to satisfy various user-specified consistency requirements. Moreover, it can save the traffic cost by up to 50 percent and reduce the query delay by up to 40 percent, compared with the widely used Pull with TTR algorithm.

Abstract Cooperative caching is an important technique to support pervasive Internet access. In order to ensure valid data access, the cache consistency must be maintained properly. However, this problem has not been sufficiently studied in mobile computing environments, especially those with ad hoc networks. There are two essential issues in cache consistency maintenance: consistency control initiation and data update propagation. Consistency control initiation not only decides the cache consistency provided to the users, but also impacts the consistency maintenance cost. This issue becomes more challenging in asynchronous and fully distributed ad hoc networks. To this end, we propose the predictive consistency control initiation (PCCI) algorithm, which adaptively initiates consistency control based on its online predictions of forthcoming data updates and cache queries. In order to efficiently propagate data updates through multi‐hop wireless connections, the hierarchical data update propagation (HDUP) algorithm is proposed. Theoretical analysis shows that cooperation among the caching nodes facilitates data update propagation. Extensive simulations are conducted to evaluate performance of both PCCI and HDUP. Evaluation results show that PCCI cost‐effectively initiates consistency control even when faced with dynamic changes in data update rate, cache query rate, node speed, and number of caching nodes. The evaluation results also show that HDUP saves cost for data update propagation by up to 66%. Copyright © 2009 John Wiley & Sons, Ltd.

Service discovery systems over MANETs are demanded to find available services at an appropriate cost, even if nodes in MANETs frequently move, randomly join or leave, and services may be revoked by their owners. SCN4M-DL is an adaptive directory-less service discovery system for MANETs, and it meets the above challenges by deeply coupling with underlying geographic routing algorithm and setting configurable service renewal. The storage consumption and network traffic are under control due to the adopted ring-based service register and discovery mechanism. Extensive experiments are also conducted. The result data, i.e. the discovery request's success rates and the corresponding costs, show that SCN4M-DL can adapt well to MANET's characteristics and the changes of service states.

The adoption of radio frequency identification (RFID) tags and devices facilitates the observation and monitoring of event patterns, while highly temporal restrained RFID data further make events complicated. Although the existing Pub/Sub systems can help distributed applications monitor and disseminate interesting events, they are unable to express RFID-related events directly. Extending a system for new RFID requirements not only needs a lot of work, but also provides no performance guarantees. It is, therefore, necessary to design and implement an effective and efficient Pub/Sub system in order to capture and disseminate RFID-related events. This paper presents a composite subscription specification for RFID-related application scenarios. It enables the subscription of predicates for RFID code as well as various complex events. Moreover, this paper proposes the algorithms for RFID-code subscription matching and complex events detection which have been implemented in our Pub/Sub system, OncePubSub. Also, experiments were conducted to quantify the performance and overhead of the above algorithms. Performance evaluation results indicate that OncePubSub is more efficient than SIENA- and JESS-based systems.

RFID technology facilitates monitoring and managing products. In most RFID applications, users pay much attention to temporal and logical event patterns, which is further strengthened by highly temporal RFID data. If RFID events or other kinds of events are disseminated to users through Pub/Sub service, visibility of product states will be greatly increased and fast processing these events will become possible. However, the traditional Pub/Sub systems cannot support composite events involving abundant temporal constraints. Therefore, they cannot fully meet the needs of RFID applications. Aiming at RFID-related application scenarios, this paper designs a composite subscription language embedded with diversified temporal operators and proposes a matching algorithm for detecting composite events which has been implemented in our Pub/Sub system OncePubSub. Performance evaluation results indicate our algorithm is more efficient than the two systems based on Siena and JESS respectively.

Service discovery protocol (SDP) is one of the fundamental infrastructures in pervasive computing. However, various SDPs differ on service description model, system architecture, deployment network and service discovery mechanism. It is highly necessary to develop a framework that can provide interoperability for multiple SDPs. In this paper, we carry out an empirical study on building a novel interoperable framework called Service CatlogNet Interoperable Framework (SCNIF), focusing on the interoperability evaluation including the completeness, the extensibility, the transparency and the overhead. Although the empirical study is based on SCNIF, we believe the results are valuable to most interoperable frameworks because SCNIF compromises many interoperability techniques and has successfully developed plug-ins for many SDPs and legacy systems.

One of the fundamental properties of any mobile agent system is to allow efficient message delivery among mobile agents. However, the agent mobility introduces complexity in the design of a mobile agent communication protocol. To deal with this problem, we adopt the well-known mailbox based framework and propose an Adaptive Mailbox Based Protocol (AMBP) in order to best suit the network environment and the communication requirements. The experimental results show that our AMBP does behave adaptively and can achieve high performance.

Detecting event regions in a monitored environment is a canonical task of wireless sensor networks (WSNs). It is a hard problem because sensor nodes are prone to failures and have scarce energy. In this paper, we seek distributed and localized algorithms for fault-tolerant event region detection. Most existing algorithms only assume that events are spatially correlated, but we argue that events are usually both spatially and temporally correlated. By examining the temporal correlation of sensor measurements, we propose two detection algorithms by applying statistical hypothesis test (SHT). Our analyses show that SHT-based algorithm is more accurate in detecting event regions. Moreover, it is more energy efficient since it gets rid of frequent measurement exchanges. In order to improve the capability of fault recognition, we extend SHT-based algorithm by examining both spatial and temporal correlations of sensor measurements, and our analyses show that extended SHT-based algorithm can recognize almost all faults when sensor network is densely deployed.

Efficiently discovering services in terms of diversified service constraints in a dense MANET is a challenging issue. This paper proposes to build a distributed suffix tree on backbone nodes as XML-based services¿ index to provide a concise profile for service descriptions. Moreover, a content-addressable P2P overlay and corresponding fault-tolerance mechanisms are introduced to support the distributed suffix tree and deal with the changes in network topology. In such a way, the precision and recall for service discovery are guaranteed and can also be degraded gracefully in the face of node failures. Experimental results show that our approach can increase at least 30% precision in comparison with existing distributed index solutions, and our newly-added messages for every 100 queries are only 22% as many as the ones of the compared solutions.

Multiple target tracking within a monitored field usually focuses on tracking the trajectories of each individual target. However, it is incapable of identifying and locating each target when the amount of targets is large or multiple targets are closely distributed. Based on above consideration, this paper proposes a novel problem of tracking a group target, where a group target is a set of targets whose motions are correlated. A monitoring region covering all group members is proposed to represent the group target’s location. And a report node is selected to collect sensor outputs for determining this monitoring region. Based on binary sensor network which is a generic model of sensor networks, we explore the fundamental limit of the accuracy of group target localization. Furthermore, to reduce network traffic, we propose two algorithms for estimating the center and radius of monitoring region. At last, we analyze the accuracy of these algorithms through both theoretical analyses and simulation results.

Most existing algorithms for fault-tolerant event region detection only assume that events are spatially correlated, but we argue that events are usually both spatially and temporally correlated. By examining the temporal correlation of sensor measurements, we propose a detection algorithm by applying statistical hypothesis test (SHT). SHT-based algorithm is more accurate in detecting event regions, and is more energy efficient since it avoids measurement exchanges. To improve the capability of fault recognition, we extend SHT-based algorithm by examining both spatial and temporal correlations of sensor measurements. The extended SHT-based algorithm can recognize almost all faults when sensor network is densely deployed.

Caching is an important technique to support pervasive Internet access. Cache consistency measures the deviation between the cached data and the source data. In mobile computing environments, especially with ad hoc networks, users are in great need of the flexibility in tuning their consistency requirements, in order to make tradeoffs between the specified cache consistency and the cost incurred. Existing works have used delta consistency (DC) and probabilistic consistency (PC) which, to some extent, provide the users with such flexibility. In this paper, we propose a general consistency model called probabilistic delta consistency (PDC). PDC covers all existing consistency models including DC and PC, and integrates the flexibility granted by both DC and PC. Thus, PDC enables the users to flexibly specify their consistency requirements in two orthogonal dimensions, namely the deviation in time/value and the ratio of queries gaining the specified consistency. We also propose a consistency maintenance algorithm, called flexible combination of push and pull (FCPP), which can meet users' consistency requirements specified under the PDC model. An analytical model is derived to achieve the optimized combination of push and pull, so as to ensure the user-specified consistency requirements, while minimizing the consistency maintenance overhead. Extensive simulations are conducted to evaluate the performance of the FCPP algorithm. Evaluation results show that, compared with the widely used dynamic TTR algorithm, FCPP can save up to 68% of the traffic overhead and reduce the query delay by up to 84%

Peer-to-peer computing is mainly based on the virtual overlay network constructed in the application layer. Often, there is topology mismatching between the overlay network and the physical network, which may cause great traffic overhead. In this paper, we study the topology mismatching problem and its impact on communication in wireless ad hoc networks. We present an efficient, fully distributed algorithm, named D-TAOC, for constructing the overlay network. By qualitative analysis and simulation experiments, we show that D-TAOC can significantly reduce the traffic overhead while slightly sacrificing the routing efficiency. We also prove that D-TAOC works well in a dynamic peer-to-peer environment.

Web services are considered as solution for solving the interoperability problem and the challenge of integration. How to manage Web services more efficiently is a key problem to the applications based on Web services at present. This paper presents a framework for application server based Web services management named FASWSM. The main contribution of FASWSM is Web service adaptation, so that a Web service can be plugged into the application server (AS) as a Web service adapter which enables a Web service to be managed by the application server. Therefore, the Web services management efforts are greatly reduced by leveraging various services provided by the AS. The FASWSM also provides mechanisms for dynamical reconfiguration which improve the extensibility and flexibility of Web services management. The FASWSM has been applied to OnceAS 2.0 application server, and the experiments data show that FASWSM outperforms JAX-RPC in terms of Web service invocation.

Context-aware mobile computing middleware is designed to automatically adapt its behavior to changing environment. To achieve this, an important issue to be addressed is how to effectively select services for adaptation according to the user's current context. Existing work does not adequately address this issue. In this paper, we propose a fuzzy-based service adaptation model (FSAM) that can be used in context-aware middleware. We formulate the service adaptation process by using fuzzy linguistic variables and membership degrees to define the context situations and the rules for adopting the policies of implementing a service. We propose three fitness functions to calculate the fitness degree for each policy based on the distance of fuzzy status between the policy and the current context situation. The decision for service adaptation is achieved by selecting the policy with the largest fitness degree. A context-aware application scenario called Campus Assistant is used to exemplify the proposed service adaptation process and demonstrate its effectiveness.

The publish/subscribe paradigm provides a loosely coupled form of interaction that is well suitable for the large-scale distributed systems. In this paper, we introduce the semantic Web technologies into the publish/subscribe systems and propose a new type of publish/subscribe system. In the system, concept models are represented as ontologies, events are represented as RDF graphs, and subscriptions are represented as RDF graph patterns. The system can match events with subscriptions based on the semantics of events, and support events with complex data structures such as graphs

As a key component of J2EE application servers, the distributed transaction manager following JTS and JTA specifications plays an important role in the distributed transaction environment. This paper explores the conformance test of a distributed transaction service by broadening the application of protocol conformance test theory and techniques. This paper also presents the formal representation of a distributed transaction service and the corresponding test case generation not based on personal experience.

A transactional workflow is composed of traditional flat transactions, and its execution has relaxed transactional atomicity. Due to different termination characteristics of transactions, a transactional workflow may be not well-formed. Moreover, only one workflow is allowed to execute a non-compensatable transaction with the current scheduling protocol. We present a transactional workflow model and formally define the correctness criteria of transactional workflow structure (well-formedness) and execution (serializability). We then propose an efficient method for well-formedness validation through a set of computation rules. Finally a scheduling protocol based on two kinds of locking granularities is presented. The protocol is different from related research in the fine-grained locking on transaction instances and coarse-grained locking on transaction classes, multiple workflows are therefore allowed to execute non-compensatable transactions if they are not conflict in a predicated future execution.