2D-to-3D human pose lifting is an ill-posed problem due to depth ambiguity and occlusion. Existing methods relying on spatial and temporal consistency alone are insufficient to resolve these problems especially in the presence of significant occlusions or high dynamic actions. Semantic information, however, offers a complementary signal that can help disambiguate such cases. To this end, we propose LangPose, a framework that leverages action knowledge by aligning motion embeddings with text embeddings of fine-grained action labels. LangPose operates in two stages: pretraining and fine-tuning. In the pretraining stage, the model simultaneously learns to recognize actions and reconstruct 3D poses from masked and noisy 2D poses. During the fine-tuning stage, the model is further refined using real-world 3D human pose estimation datasets without action labels. Additionally, our framework incorporates masked body parts and masked time windows in motion modeling, encouraging the model to leverage semantic information when spatial and temporal consistency is unreliable. Experiments demonstrate the effectiveness of LangPose, achieving SOTA level performance in 3D pose estimation on public datasets, including Human3.6M and MPI-INF-3DHP. Specifically, LangPose achieves an MPJPE of 36.7mm on Human3.6M with detected 2D poses as input and 15.5mm on MPI-INF-3DHP with ground-truth 2D poses as input.

We conducted a scoping review to map the rapidly evolving landscape of wearable and ambient sensing technologies for monitoring people with dementia across home and institutional settings. We analyzed empirical sensing studies (2015-2025) to identify and inform future technical and human-centered design requirements. Five key implementation principles emerge: (1) human-centered design involving all stakeholders to augment rather than replace caregivers; (2) personalized, adaptable solutions that support autonomy across settings and severity levels instead of standardized approaches; (3) integration with existing workflows with adequate training and support; (4) proactive privacy and consent considerations, especially for ambient monitoring of residents and caregivers; and (5) cost-effective, ethical, equitable, scalable solutions with quantifiable outcomes. This paper identifies gaps, trends and opportunities for developing sensing systems that address the complex challenges, while enhancing automation and autonomy, in dementia care.

Capturing fine-grained 3D shapes and anthropometric features of human ears is a non-trivial and yet important task in a variety of applications. Existing approaches relying on specialized equipment such as high-definition 3D scanners or photogrammetry from RGB(-D) cameras are laborious in data acquisition and post-processing, and may perform poorly in case of occlusions from face and head covering, accessories, and the complex geometry of pinnae. Recognizing the strong dependency of acoustic anatomical transfer functions called Head-Related Transfer Functions (HRTFs) on the morphological features of pinnae, we develop HRTF2Mesh, a novel acoustic-based method for 3D human ear mesh reconstruction using commodity devices – a pair of earphone microphones and a smartphone device. HRTF2Mesh takes sparse acoustic measurement data from a subject, estimates and encodes the HRTFs as low-dimension vectors and then generates 3D mesh representation of their ear with the help of 3D morphable ear models. HRTF2Mesh has been evaluated using both public datasets and data collected from a diverse set of volunteers. Experiment results show that HRTF2Mesh achieves an average ear mesh estimation errors of 0.92 mm for test subjects in public datasets, and 1.70 mm amongst the volunteers. The data acquisition process of HRTF2Mesh can be completed under 5 minutes by a single user.

Latency-sensitive edge video analytics applications require rapid responses for real-time decision-making, driving the demand for efficient object detection pipelines. Conventional pipelines transmit and process full frames, overlooking redundancy in videos and leading to unnecessary resource consumption. Existing block-wise conditional execution methods mitigate this issue by processing only informative blocks. However, they treat all informative blocks equally and fail to further categorize these blocks. To address this limitation, we propose BlockHybrid, an edge video analytics framework designed to accelerate object detection pipelines by hybrid block-wise execution. Specifically, BlockHybrid classifies blocks into hard or easy blocks using a policy network. Hard blocks are transmitted and processed by a block-wise detector on the server, while easy blocks are handled by an efficient tracker locally on the camera, reducing redundant computation and communication. Extensive experiments demonstrate that BlockHybrid can achieve 8.8%–31.5% higher local execution speed and comparable detection accuracy compared to state-of-the-art methods, and accelerate end-to-end processing—including camera-to-server communication—by 31.5%–39.1% in a real-world testbed.

Cardiovascular diseases (CVDs) are a leading cause of mortality worldwide, highlighting the critical need for accurate and continuous heart health monitoring. Electrocardiograms (ECG), considered as the golden standard for diagnosing and monitoring heart-related conditions, offer precise measurements but require direct skin contact, limiting their practicality for long-term and everyday use. On the other hand, existing RF sensing techniques that analyze signals reflected off the skin struggle to distinguish micro cardiac motions of the heart due to weak motion amplitude and respiration interference at the chest wall. To overcome these limitations, we introduce WiCG, a novel contact-less cardiac motion monitoring system that employs 2.4 GHz Wi-Fi signals to penetrate the chest and detect subtle cardiac movements. A mix-medium Wi-Fi Fresnel zone model is developed to explain the enhanced phase sensitivity of in-body Wi-Fi signals, which is crucial for accurately detecting cardiac motions. By strategically positioning antennas near the heart, WiCG captures ventricular motions effectively. A novel cardiac Doppler method is proposed to suppress phase noise and interference from static paths and extract the time interval between the systole and diastole of the ventricular. Extensive experiments demonstrate that the proposed system can robustly estimate the R-R and Q-T intervals of human cardiac cycles across 21 subjects and different environments with an average accuracy of 99.22% and 92.8%, achieving performance comparable to ECG.

Multi-object tracking (MOT) is the “killer app” of edge video analytics. Deploying MOT pipelines for live video analytics poses a significant system challenge due to their computation-intensive nature. In this paper, we propose FastTuner, a model-agnostic framework that aims to accelerate MOT pipelines by adapting frame resolutions and backbone models. Unlike prior works that utilize a separate and time-consuming online profiling procedure to identify the optimal configuration, FastTuner incorporates multi-task learning to perform configuration selection and object tracking through a shared model. Multi-resolution training is employed to further improve the tracking accuracy across different resolutions. Furthermore, two workload placement schemes are designed for the practical deployment of FastTuner in edge video analytics systems. Extensive experiments demonstrate that FastTuner can achieve 1.1%–9.2% higher tracking accuracy and 2.5%–25.5% higher speed compared to the state-of-the-art methods, and accelerate end-to-end processing by 1.7%–22.5% in a real-world testbed consisting of an embedded device and an edge server.

In this paper, we present the novel task of estimating the extrinsic parameters of a virtual camera relative to a real camera in exercise videos with a mirror. This task poses a significant challenge in scenarios where the views from the real and mirrored cameras have no overlap or share salient features. To address this issue, prior knowledge of a human body and 2D joint locations are utilized to estimate the camera extrinsic parameters when a person is in front of a mirror. We devise a modified eight-point algorithm to obtain an initial estimation from 2D joint locations. The $2 D$ joint locations are then refined subject to human body constraints. Finally, a RANSAC algorithm is employed to remove outliers by comparing their epipolar distances to a predetermined threshold. MirrorCalib achieves a rotation error of 1.82° and a translation error of 69.51 mm on a collected real-world dataset, which out-performs the state-of-art method.

Tracking people in multi-camera surveillance systems is challenging due to disparate perspectives, large volumes of data, and high computation demands. This paper presents a distributed cooperative pipeline for pedestrian tracking that exploits the spatial and temporal redundancy within and across the video feeds from multiple synchronized cameras. It consists of three key components: 1) a lightweight policy network trained online in a self-supervised manner on each camera, 2) a sparse backbone processing unit purpose-built for parallel processing of selected regions of all cameras, and 3 an online clustering algorithm for object association. Utilizing online distributed reinforcement learning, the fully end-to-end trainable pipeline can accelerate any tracking-by-detection method by reducing detection costs across multiple perspectives. MVSparse has been evaluated using two multi-camera multi-target pedestrian tracking datasets, WildTrack and MultiviewX. It reduces the amount of processed regions by up to 52% and 39% with only moderate degradation of 1% and 0.1% in tracking accuracy on the two datasets, respectively. On a real-world testbed comprising four NVIDIA Jetson TX2 and a GPU server, MVSparse accelerates the end-to-end process and reduces the communication overheads by 1.88 and $1.60 X$ with only 2.27% and 3.17% degradation in tracking accuracy on the two datasets, respectively

Despite intensive research efforts in radio frequency noncontact sensing, capturing fine-grained geometric properties of objects, such as shape and size, remains an open problem using commodity WiFi devices. Prior attempts are incapable of characterizing object shape or size because they predominantly rely on weak signals reflected off objects in a very small number of directions. In this paper, motivated by the observation that the diffracted signals around an object between two WiFi devices carry the contour information of the object, we formulate the problem of reconstructing the 2D target profile and develop WiProfile, the first WiFi-based system that unlocks the diffraction effects for target profiling. We introduce a CSI-Profile model to characterize the relationship between the CSI measured at different target positions and the target profile in the diffraction zone. With suitable approximations, the inverse problem of deriving the target profile from CSI can be solved by the inverse Fresnel transform. To mitigate CSI measurement errors on commodity WiFi devices, we propose a novel antenna placement strategy. Comprehensive experiments demonstrate that WiProfile can accurately reconstruct profiles with median absolute errors of less than 1 cm under various conditions, and effectively estimate the profiles of everyday objects of diverse shapes, sizes, and materials. We believe this work opens up new directions for fine-grained target imaging using commodity WiFi devices.

Multi-source sound localization can find applications in many domains including auditory scene analysis, fault detection and diagnosis in manufacturing, augmented reality, etc. In far fields, 3D sound source localization is equivalent to finding the direction of arrival (DOA), namely, the azimuth and elevation angles of sound sources. Recent DOA estimation pipelines take multichannel audio inputs, extract spectral features from each channel and then feed them into a deep neural network. Unfortunately, the spectral features contain only the time-frequency information of the audio signals, while spatial information is only implicitly captured in the signals across different channels, which is highly dependent on the acoustic array geometry. To embed the spatial information of the sound source into the spectral feature representation, we propose a DSB-based spatial mapping method encode sound source location information. It can be combined with different feature extraction methods and machine learning models for DOA estimation. Furthermore, a redundancy removal procedure is proposed to accelerate DSB computation so that the pipeline can run in real-time on embedded GPUs, such as NVidia Jeston Nano. We conduct extensive experiments using two neural network models along with the DSB method on two datasets. The experiments demonstrate that the DOA errors can be effectively reduced using the DSB method. When combining DSB for feature extraction, the DOA errors are reduced by up to 19.24%. In addition, the feature extraction process is accelerated by up to 30.42% after the application of redundancy removal.

With the prevalence of wearable devices, inertial measurement unit (IMU) data has been utilized in monitoring and assessing human mobility such as human activity recognition (HAR) and human pose estimation (HPE). Training deep neural network (DNN) models for these tasks require a large amount of labelled data, which are hard to acquire in uncontrolled environments. To mitigate the data scarcity problem, we design CROMOSim, a cross-modality sensor simulator that simulates high fidelity virtual IMU sensor data from motion capture systems or monocular RGB cameras. It utilizes a skinned multi-person linear model (SMPL) for 3D body pose and shape representations to enable simulation from arbitrary on-body positions. Then a DNN model is trained to learn the functional mapping from imperfect trajectory estimations in a 3D SMPL body tri-mesh due to measurement noise, calibration errors, occlusion and other modelling artifacts, to IMU data. We evaluate the fidelity of CROMOSim simulated data and its utility in data augmentation on various HAR and HPE datasets. Extensive empirical results show that the proposed model achieves a 6.7% improvement over baseline methods in a HAR task.

Most existing RF-based vital sign monitoring systems either assume that a human subject is stationary or discard measurements when motion is detected in order to output reliable respiration rates and heart rates. Such an assumption greatly limits the usability of these systems in practice. Even during sleep, one can undergo various body states including turns and involuntary twitches in light sleep, motionlessness during deep sleep, or abnormal limb movements due to sleep disorders such as restless legs syndrome. In this work, we develop SlpRoF, a low-cost contact-free system using a commercial-off-the-shelf UWB radar that achieves high temporal coverage and high accuracy in vital sign monitoring during sleep. By classifying body states into the motionless state, limb movement state, and torso movement state, and extracting vital signs during the first two states, it directly increases effective reporting periods over nights. By analyzing high-order harmonics and leveraging spatial diversity in captured signals from multiple on-body areas, it improves the accuracy of heart rate estimations and thus indirectly increases temporal coverage through reliable assessments. Experiment results show that SlpRoF is able to achieve an average median absolute error (MAE) of 0.44 beats per minute (bpm) in respiration rates, 1.55s in respiration intervals, and 0.9 bpm for heart rates, respectively.

Multi-object tracking (MOT) is a vital component of intelligent video analytics applications such as surveillance and autonomous driving. The time and storage complexity required to execute deep learning models for visual object tracking hinder their adoption on embedded devices with limited computing power. In this paper, we aim to accelerate MOT by transferring the knowledge from high-level features of a complex network (teacher) to a lightweight network (student) at both training and inference times. The proposed AttTrack framework has three key components: 1) cross-model feature learning to align intermediate representations from the teacher and student models, 2) interleaving the execution of the two models at inference time, and 3) incorporating the updated predictions from the teacher model as prior knowledge to assist the student model. Experiments on pedestrian tracking tasks are conducted on the MOT17 and MOT15 datasets using two different object detection backbones YOLOv5 and DLA34 show that AttTrack can significantly improve student model tracking performance while sacrificing only minor degradation of tracking speed.

The prevalence of earable devices, such as earbuds and headphones, allows people to converse or listen to audio recordings on the move but often at the cost of reduced alertness of imminent health and safety threats in their surroundings. 3-D binaural sound localization (BSL), which aims to locate sound sources in space, plays a key role in improving one’s situation awareness. BSL on earable devices is inherently challenging due to the limited number of microphones available as well as subject- and location-dependent filtration effects of a person’s pinna, head and torso, described by head-related transfer functions (HRTFs). In this work, we develop DeepBSL, a deep neural network model to estimate azimuth and elevation angles of sound sources relative to a person’s head. For a new subject, an efficient procedure is developed to collect HRTFs at sparse locations using in-ear microphones and a mobile phone, which are then utilized to synthesize sounds of any type at arbitrary locations to train personalized DeepBSL models. Extensive evaluations using synthetic data from a public data set and through real-world experiments demonstrate that the personalized DeepBSL models are data-efficient and can achieve better-than-human performances in BSL while significantly outperforming a state-of-the-art model that can only predict azimuth angles of sound sources. Our best performing model has an average azimuth prediction error of$2.9^{\circ } (4.1^{\circ })$and elevation prediction error of$1.4^{\circ } (2.9^{\circ })$in an indoor (outdoor) environment.

Rotating Synthetic Aperture Radar (ROSAR) can generate a 360° image of its surrounding environment using the collected data from a single moving track. Due to its non-linear track, the Back-Projection Algorithm (BPA) is commonly used to generate SAR images in ROSAR. Despite its superior imaging performance, BPA suffers from high computation complexity, restricting its application in real-time systems. In this paper, we propose an efficient imaging method based on robust sparse array synthesis. It first conducts range-dimension matched filtering, followed by azimuth-dimension matched filtering using a selected sparse aperture and filtering weights. The aperture and weights are computed offline in advance to ensure robustness to array manifold errors induced by the imperfect radar rotation. We introduce robust constraints on the main-lobe and sidelobe levels of filter design. The resultant robust sparse array synthesis problem is a non-convex optimization problem with quadratic constraints. An algorithm based on feasible point pursuit and successive convex approximation is devised to solve the optimization problem. Extensive simulation study and experimental evaluations using a real-world hardware platform demonstrate that the proposed algorithm can achieve image quality comparable to that of BPA, but with a substantial reduction in computational time up to 90%.

Continuous monitoring of human vital signs using non-contact mmWave radars is attractive due to their ability to penetrate garments and operate under different lighting conditions. Unfortunately, most prior research requires subjects to stay at a fixed distance from radar sensors and to remain still during monitoring. These restrictions limit the applications of radar vital sign monitoring in real life scenarios. In this article, we address these limitations and present Pi-ViMo, a non-contact P hysiology- i nspired Robust Vi tal Sign Mo nitoring system, using mmWave radars. We first derive a multi-scattering point model for the human body, and introduce a coherent combining of multiple scatterings to enhance the quality of estimated chest-wall movements. It enables vital sign estimations of subjects at any location in a radar’s field of view (FoV). We then propose a template matching method to extract human vital signs by adopting physical models of respiration and cardiac activities. The proposed method is capable to separate respiration and heartbeat in the presence of micro-level random body movements (RBM) when a subject is at any location within the field of view of a radar. Experiments in a radar testbed show average respiration rate errors of 6% and heart rate errors of 11.9% for the stationary subjects, and average errors of 13.5% for respiration rate and 13.6% for heart rate for subjects under different RBMs.

Acoustic sensing has attracted significant attention recently, thanks to the pervasive availability of device support. However, adopting consumer-grade devices (e.g., smartphones) to deploy acoustic sensing applications faces the challenge of device/OS heterogeneity. Researchers have to pay tremendous efforts in tackling platform-dependent details even in simply accessing raw audio samples, thus losing focus on innovating sensing algorithms. To this end, this paper presents the first Acoustic Software Defined Platform (ASDP): a versatile sensing and general benchmarking platform. ASDP encompasses several customized acoustic modules running on a ubiquitous computing board, backed by a dedicated software framework. It is superior to commodity devices in controlling and reconfiguring physical layer settings, thus offering much better usability. The tailored software framework abstracts platform details and provides user-friendly interface for fast prototyping, while maintaining adequate programmability. To demonstrate the usefulness of ASDP, we showcase several relevant applications based on it. The promising outcomes make us believe that the release of our ASDP could greatly advance acoustic sensing research.

To address the problems of node deployment schemes in existing underwater sensor networks that lack consideration of network connectivity and high deployment costs, this article constructs an optimization model that maximizes network coverage and minimizes deployment costs while ensuring full connectivity. For the NP-hard property of this optimization model, an improved moth flame optimization node deployment algorithm based on fuzzy operators (IMF 2 O 2 ) is proposed. First, comprehensively considering the two performance metrics of network coverage and network connectivity, a multi-objective selection mechanism based on fuzzy operators is proposed to improve network coverage while ensuring full connectivity. Second, a fixed number of nodes are used to monitor the target event points, transforming the node deployment of sensors into an optimal problem and proposing an improved moth flame optimization algorithm to solve this problem. Finally, the two metrics of coverage and deployment cost are measured and the fuzzy operator is used to select the optimal number of nodes to be deployed. Numerical results showed that the proposed algorithm improved network coverage rate by 10%, 22%, and 25%, and improved network connectivity rate by 12%, 20%, and 8% as compared to PSSD, RAWS, and VODA, respectively, while ensuring full connectivity.

Modern neural networks are powerful predictive models. However, when it comes to recognizing that they may be wrong about their predictions, they perform poorly. For example, for one of the most common activation functions, the ReLU and its variants, even a well-calibrated model can produce incorrect but high confidence predictions. Most current action recognition methods are based on clip-level classifiers that densely sample a given video for non-overlapping, same-sized clips and aggregate the results using an aggregation function - typically averaging - to achieve video level predictions. While this approach has shown to be effective, it is sub-optimal in recognition accuracy and has a high computational overhead. To mitigate both these issues, we propose the confidence distillation framework to teach a student model how to select less ambiguous clips for the teacher, and divide the task of prediction between the two. We conduct extensive experiments on three action recognition datasets and demonstrate that our framework achieves significant improvements in action recognition accuracy (up to 20%) and computational efficiency (more than 40%).

A Head Related Transfer Function (HRTF) characterizes how a hu-man ear receives sounds from a point in space, and depends on the shapes of one's head, pinna, and torso. Accurate estimations of HRTFs for human subjects are crucial in enabling binaural acoustic applications such as sound localization and 3D sound spatialization. Unfortunately, conventional approaches for HRTF estimation rely on specialized devices or lengthy measurement processes. This work proposes a novel lightweight method for HRTF individual-ization that can be implemented using commercial-off-the-shelf components and performed by average users in home settings. The proposed method has two key components: a generative neural network model that can be individualized to predict HRTFs of new subjects from sparse measurements, and a lightweight measurement procedure that collects HRTF data from spatial locations. Exten-sive experiments using a public dataset and in house measurement data from 10 subjects of different ages and genders, show that the individualized models significantly outperform a baseline model in the accuracy of predicted HRTFs. To further demonstrate the advantages of individualized HRTFs, we implement two prototype applications for binaural localization and acoustic spatialization. We find that the performance of a localization model is improved by 15° after trained with individualized HRTFs. Furthermore, in hearing tests, the success rate of correctly identifying the azimuth direction of incoming sounds increases by 183% after individualization.

We consider the problem of localizing multiple targets by millimeter wave (mmWave) radars with irregular antenna placement, i.e., trapezoid virtual antenna array. The goal is to estimate both the number of targets and their 3-D locations. While many well-known algorithms have been developed for either problems, they still suffer from several limitations, such as the need for a large amount of sampled radar data and high computation complexity. In this work, we develop an efficient solution by exploring the received signal structure in two steps: 1) estimating the number of targets and their ranges by extending Barone’s method to handle data from multiple antennas and 2) estimating the angle of arrival of each target by a Least-Square algorithm optimization. The proposed algorithm has been evaluated through Monte-Carlo simulations and an indoor testbed. By comparing with baseline algorithms, including 2D-FFT and multiple signal classification (MUSIC), we find that the proposed algorithm has the best performance in high signal-to-noise ratio regimes.

Enabling pervasive WiFi devices with non-contact sensing capability is an important topic in the field of integrated sensing and communication. Doppler effect has been widely exploited to estimate targets’ velocity from wireless signals. However, the separation of signal sources and receivers complicates the relationship between Doppler frequency shift (DFS) and target velocity in WiFi-based non-contact sensing systems. In contrast to existing works that rely on either approximated relations or coarse-grained information such as whether a target is moving toward or away from WiFi transceivers, this paper investigates rigorously the dependency of velocity estimation accuracy on target locations and headings in WiFi sensing systems. The theoretical insights allow us to derive a closed-form solution and understand the fundamental limitation of velocity estimation. To optimize velocity estimation performance, we devise a receiving device selection scheme that dynamically chooses the optimal set of receivers among multiple available WiFi devices. A prototype real-time target tracking system has been implemented using commodity WiFi devices. Extensive experimental results show that the proposed system outperforms state-of-the-art approaches in velocity estimation and tracking, and is able to achieve$9.38cm/s$, 13.42°,$31.08cm$median errors in speed, heading and location estimation amongst experiments conducted in three indoor environments with three device placements and eight human subjects over 15 trajectories.

Aerial acoustic communication attracts substantial attention for its simplicity and cost-effectiveness. Unfortunately, the preferred inaudible transmission has to strike a balance between the transmission rate and communication range, when the Bit-Error-Rate (BER) is under a certain threshold. Additionally, the performance of previous proposals can be deteriorated by dynamic channel conditions including near-far problem, device heterogeneity, and multipath fading. To this end, we propose a High-speed, long-range, and Robust Chirp Spread Spectrum (HRCSS) scheme for inaudible aerial acoustic communication under dynamic channels. HRCSS innovates in the definition of a loose orthogonality condition, and it leverages this orthogonality to overlap multiple chirp carriers in a single time duration to form a data symbol representing multiple bits, thereby substantially promoting the data rate. To further enhance system robustness in long communication ranges and dynamic channel conditions, we construct a lightweight rate adaptation algorithm and design a simple yet efficient normalization method. Experiment results reveal that HRCSS achieves a significant improvement in data rate over existing methods: it delivers 500 bps data rate with a BER of 0.24 percent at 10 m, and achieves 125 bps with zero BER at 20 m. Meanwhile, HRCSS can work adaptively under dynamic channel conditions while still retaining a BER below 3 percent.

Wearable sensor-based human activity recognition (HAR) has been a research focus in the field of ubiquitous and mobile computing for years. In recent years, many deep models have been applied to HAR problems. However, deep learning methods typically require a large amount of data for models to generalize well. Significant variances caused by different participants or diverse sensor devices limit the direct application of a pre-trained model to a subject or device that has not been seen before. To address these problems, we present an invariant feature learning framework (IFLF) that extracts common information shared across subjects and devices. IFLF incorporates two learning paradigms: 1) meta-learning to capture robust features across seen domains and adapt to an unseen one with similarity-based data selection; 2) multi-task learning to deal with data shortage and enhance overall performance via knowledge sharing among different subjects. Experiments demonstrated that IFLF is effective in handling both subject and device diversion across popular open datasets and an in-house dataset. It outperforms a baseline model of up to 40 percent in test accuracy.

Large-scale spatial data can be collected using mobile robots with sensing and navigation capabilities. Due to limited battery lifetime and scarcity of charging stations, it is important to plan informative paths so as to maximize the utility of data given a limited travel budget, which is known as the informative path planning (IPP) problem. IPP is NP-hard, and existing solutions suffer from high complexity or low optimality. In this paper, we present a novel IPP solution based on reinforcement learning (RL). The basic idea is to learn the structural characteristics of informative paths, so informative paths can be predicted. As such, when budgets change, we avoid solving the problem from scratch and thus path planning efficiency can be improved dramatically. Among the 20 path planning experiments in two areas, the proposed RL based solution achieves the best path utility in 15 experiments, compared with state-of-the-art algorithms. More importantly, the inference complexity is linear with respect to the budget (equivalently, the maximum number of steps in RL), which is lower than other solutions. Despite the NP-hardness, the path planning process can be finished within a few seconds in our experiments on two graphs of different sizes.

Mobile sensing is an effective way to collect environmental data such as air quality, humidity and temperature at low costs. However, mobile robots are typically battery powered and have limited travel distances. To accelerate data collection in large geographical areas, it is beneficial to deploy multiple robots to perform tasks in parallel. In this paper, we investigate the Multi-Robot Informative Path Planning (MIPP) problem, namely, to plan the most informative paths in a target area subject to the budget constraints of multiple robots. We develop two deep reinforcement learning (RL) based cooperative strategies: independent learning through credit assignment and sequential rollout based learning for MIPP. Both strategies are highly scalable with the number of robots. Extensive experiments are conducted to evaluate the performance of the proposed and baseline approaches using real-world WiFi Received Signal Strength (RSS) data. In most cases, the RL based solutions achieve superior or similar performance as a baseline genetic algorithm (GA)-based solution but at only a fraction of running time during inference. Furthermore, when the budgets and initial positions of the robots change, the pre-trained policies can be applied directly.

Human Activity Recognition (HAR) refers to recognizing a human's ongoing actions through sensor data. At present, one of the main problems faced by Human Activity Recognition is that different subjects, devices and wearing positions can cause inconsistent sensor data distribution. When a classification model trained using some labeled dataset is used to classify a new unlabeled data with different distributions, there will be a significant performance loss. However, it is difficult to annotate manually sensor data for new subjects. Prior works applying unsupervised domain adaptation methods to solve this problem only used a single source domain. However, in practice, it is common to have multiple labeled source domains. Inspired by a work in the field of computer vision, we propose an unsupervised domain adaptation method for human activity recognition using multiple source domains. Experimental results on a commonly used public HAR dataset show that our model can effectively alleviate the performance loss caused by inconsistent distributions. Moreover, compared with the single-source domain adaptation, the multi-source domain adaptation method can improve the accuracy further.

Fitness games recently attract much attention these years due to its combination of playability and athleticism. However, most fitness games are entertainment for a single person, with only a few deliver distinctive experiences for multiplayers. Enabling interaction between multiple players would be more enjoyable due to exciting cooperation among players. Considering a Big Stomach Challenge for two players, a certain amount of food can be only eaten when a mouth size, represented by the distance between players is reached collaboratively. Similarly, a certain type of food can only be picked up when the food grabbing speed, denoted by the approaching speed between players, is fast enough. Such games require accurate ranging and speed estimation in a relatively long distance (1-15 m) to deliver a good gaming experience. However, existing ranging schemes cannot meet the above requirements. They either cannot work under Doppler channels or have to strike a balance between accuracy and operational range, prohibiting a heuristic implementation for the above games. To this end, we design MotionBeep, a novel acoustic ranging scheme that achieves centimeter-level ranging and dm/s-level speed estimation accuracy under representative indoor and outdoor scenes within 15 m. In MotionBeep, we design a new working paradigm and incorporates a state-space model to maintain accurate ranging in both static and dynamic channels. We have implemented a system prototype and evaluate its performance in representative environments. Evaluation results demonstrate that MotionBeep achieves a median of centimeter accuracy with up to 15 m even under Doppler effect.

Temperature is an important data source for weather forecasting, agriculture irrigation, anomaly detection, etc. While temperature measurement can be achieved via low-cost yet standalone hardware with reasonable accuracy, integrating thermal sensing into ubiquitous computing devices is highly non-trivial due to the design requirement for specific heat isolation and proper device layout. In this paper, we present the first integrated thermometer using commercial-off-the-shelf acoustic-enabled devices. Our software sonic thermometer (SST) utilizes on-board dual microphones on commodity mobile devices to estimate sound speed, which has a known relation with temperature. To precisely measure temperature via sound speed, we propose a chirp mixing approach to circumvent low sampling rates on commodity hardware and design a pipeline of signal processing blocks to handle channel distortions. SST, for the first time, empowers ubiquitous computing devices with thermal sensing capability. It is portable and cost-effective, making it competitive with current thermometers using dedicated hardware. SST is potential to facilitate many interesting applications such as large-scale distributed thermal sensing, yielding high temporal/spatial resolutions with unimaginable low costs. We implement SST on a commodity platform and results show that SST achieves a median accuracy of${0.5^\circ \mathrm{C}}$even at varying humidity levels.

Environmental monitoring using wireless sensors plays a key role in detecting hotspots or over-cooling conditions in data centers (DCs). However, monitoring a large enterprise or cloud DCs requires the deployment of thousands of sensors distributively with an operational time over months or years. Low Energy Monitoring Network (LEMoNet) is a two-tier Bluetooth Low Energy (BLE) based protocol for DC monitoring that leverages multi-gateway packet reception in its top tier to mitigate the unreliable BLE communication in the low tier. In this paper, we develop an analytical model to study the scalability and energy efficiency of LeMoNet in large-scale DCs. The accuracy of the model is validated through extensive event-driven simulations. Evaluation results show that LEMoNet can achieve high reliability in a network of 4800 nodes at a duty cycle of 15 sec (or equivalently, at an aggregated traffic load of 66Kbps per advertisement channel).

Large-scale spatial data such as air quality, thermal conditions and location signatures play a vital role in a variety of applications. Collecting such data manually can be tedious and labour intensive. With the advancement of robotic technologies, it is feasible to automate such tasks using mobile robots with sensing and navigation capabilities. However, due to limited battery lifetime and scarcity of charging stations, it is important to plan paths for the robots that maximize the utility of data collection, also known as the informative path planning (IPP) problem. In this paper, we propose a novel IPP algorithm using reinforcement learning (RL). A constrained exploration and exploitation strategy is designed to address the unique challenges of IPP, and is shown to have fast convergence and better optimality than a classical reinforcement learning approach. Extensive experiments using real-world measurement data demonstrate that the proposed algorithm outperforms state-of-the-art algorithms in most test cases. Interestingly, unlike existing solutions that have to be re-executed when any input parameter changes, our RL-based solution allows a degree of transferability across different problem instances.

Device-free hand gesture is one of the most natural ways to interact with everyday objects. However, existing WiFi-based gesture recognition solutions are typically restricted to indoor environments due to limited outdoor coverage. Furthermore, to achieve high sampling rates, they may interfere with normal data transmissions. In this paper, we aim to develop a robust dynamic gesture interaction system that can be ubiquitously deployed using Long-term Evolution (LTE) mobile terminals. Through both empirical studies and in-depth analysis using the Fresnel zone model, we reveal the key factors that contribute to the repeatability and discernibility of gestures. We show that the optimal location and orientation to perform gestures indeed exist and can be identified without prior knowledge of the position of LTE base stations (BSs) relative to a terminal. Guided by the design principles derived from Fresnel zone characteristics around a 4G terminal, we design highly repeatable and discernible gestures with salient received signal profiles. A gesture interaction system has been developed and implemented to achieve robust recognition with this careful design. Extensive experiments have been conducted in both indoor and outdoor environments, for different relative placements of mobile terminal and BS, and with different users. The proposed system can automatically identify the direction of BSs with a median error of less than 15 degrees and achieve gesture recognition accuracy as high as 98% in all scenarios without the need to acquire any training data.

Good user experiences in Mobile Augmented Reality (MAR) applications require timely processing and rendering of virtual objects on user devices. Today's wearable AR devices are limited in computation, storage, and battery lifetime. Edge computing, where edge devices are employed to offload part or all computation tasks, allows an acceleration of computation without incurring excessive network latency. In this paper, we use acyclic data flow graphs to model the computation and data flow in MAR applications and aim to minimize the makespan of processing input frames. Due to task dependencies and variable resource availability, makespan minimization is proven to be NP-hard in general. We design DPA, a polynomial-time heuristic algorithm for this problem. For special data flow graphs including chain or star, the algorithm can provide optimal solutions or solutions with a constant approximation ratio. The effectiveness of DPA has been evaluated using extensive simulations with realistic workloads and resource availability measured from a prototype implementation.

Recently, acoustic-based indoor localization has attracted much attention due to its affordable infrastructure costs and high localization accuracy. However, previous work is infeasible in mobile target tracking for its long latency in obtaining sufficient beacon messages. In addition, the performance can further deteriorate due to device diversity, varying channel gains, and background noises. To this end, we propose an asynchronous acoustic localization and tracking system (AALTS), which utilizes distributed acoustic anchor nodes to locate passive off-the-shelf mobile devices. In AALTS, we propose an orthogonal chirp spread spectrum (OCSS) modulation technique, which doubles the data rate and thus mitigates the latency. We design a more robust method to capture acoustic signals which embody timestamps for localization, accounting for device diversity, varying channel gains, and the multipath effect. Finally, we incorporate an acoustic Doppler speed estimation module with a path-based particle filter framework to accurately track the moving targets. We have evaluated AALTS in an indoor testbed of size 8×12 m2with commodity mobile phones and customized acoustic anchors. Our evaluation results demonstrate remarkable performance: AALTS achieves 90-percentile tracking errors of 0.49 m for mobile targets and a median of 0.12 m for stationary ones with only four anchor nodes.

Monitoring ambient air temperature is one of the important operations to ensure resilience and efficiency in large-scale data centers. However, deployment of a data center monitoring system requires recording the location of thousands of sensors which is a labor-intensive task if is done manually. Since Radio-Frequency (RF) based localization solutions in literature are inadequate in the multipath rich environment of data centers, we investigated the possibility of utilizing the measurements of the sensors in localizing themselves. The idea of thermal piloting is to correlate sensor measurements with the expected temperature values at their locations. It can be treated as a classification problem, in which the feature vector is formed by the temperature values at each sensor location across different cooling configurations. The training set is provided by Computational Fluid Dynamic (CFD) simulations. Since classical supervised learning techniques fail to account for the bijective relation between sensor indices and locations, we formulated an extra step based on the Maximum Weighted Bi-partite Matching (MWBM) problem. Experimental results show that the proposed methods can achieve an average localization error of 0.64 meters.

In this paper, the impact of redundant packet reception at multiple gateways on data reliability is studied under the LoRaWAN architecture. Given a successful transmission could be the result of either a first attempt after a data packet's generation, or a retry after several transmission failures, the Average Successful Transmission Probability (ASTP) is introduced to qualify LoRaWAN's reliability performance. To calculate the probability of a successful reception without retransmission, we consider all the possible causes for a packet collision. The number of potential interferers, which is vital for the collision analyses and directly determined by the relative locations of the relevant multiple gateways, is determined by geometric arguments. Similarly, the probability for achieving a successful retransmission is also obtained. Finally, ASTP is rigorously modeled as a function of end device density, gateway density, and traffic intensity. The analytical results have been verified by extensive simulation experiments. We believe that the analytical model can provide useful insights into the scalability of LoRaWANs and provide guidelines for their deployments.

With significant increases in wireless link capacity, edge devices are more connected than ever, which makes possible forming artificial neural network (ANN) federations on the connected edge devices. Partition is the key to the success of distributed ANN inference while unsolved because of the unclear knowledge representation in most of the ANN models. We propose a novel partition approach (TeamNet) based on the psychologically-plausible competitive and selective learning schemes while evaluating its performance carefully with thorough comparisons to other existing distributed machine learning approaches. Our experiments demonstrate that TeamNet with sockets and transmission control protocol (TCP) significantly outperforms sophisticated message passing interface (MPI) approaches and the state-of-the-art mixture of experts (MoE) approaches. The response time of ANN inference is shortened by as much as 53% without compromising predictive accuracy. TeamNet is promising for having distributed ANN inference on connected edge devices and forming edge intelligence for future applications.

Mobile crowdsensing (MCS) has recently found many applications in environmental monitoring and large-scale surveillance by recruiting crowd workers for data collection and labeling. The quality of labelled data from unknown crowd workers, however, is hard to guarantee. Therefore, it is critical to design a mechanism that can automatically make correct decisions from diverse and even conflicting labels from the crowd. To tackle the challenge, we propose a new algorithm, EFusion, which infuses knowledge from domain experts by asking them to check a small number of labels from the crowd. Taking advantage of cheaper but unreliable crowd workers as well as expensive but reliable experts, EFusion can greatly improve the accuracy in discovering the ground truth of classification-based mobile crowdsensing tasks. EFusion utilizes a probabilistic graphical model and the expectation maximization (EM) algorithm to infer the most likely expertise level for each crowd worker, the difficulty level of tasks, and the ground truth answers. EFusion has been evaluated using real-world case study as well as simulations. Evaluation results demonstrate that EFusion can return more accurate and stable classification results than the majority voting method and state-of-the-art methods.

With technological advances in mobile devices, mobile crowdsensing (MCS), a special crowd sourcing paradigm, has attracted unprecedented interest. Traditionally, MCS recruits a crowd of mobile users to capture data of interest and input their own judgment (i.e., intelligence) to facilitate the processing of big data. Recently, there is a surge of interest in another form of MCS where the sensing crowd consists of “smart” devices/programs that possess machine intelligence. One example is the smart cameras that recognize human faces or detect urgent events such as a car collision.

Today's data centers (DCs) consume up to 3% of the energy produced worldwide, much of which is wasted due to over-cooling and under utilization of IT equipment. This wastage in part stems from the lack of real-time visibility of fine-grained thermal distribution in DCs. Wireless sensing is an ideal candidate for DC monitoring as it is cost-effective, facility-friendly, and can be easily re-purposed. In this paper, we develop LEMoNet, a novel low-energy battery operated wireless sensor network design for monitoring DCs. It employs a two-tier network architecture and a multi-mode data exchange protocol to balance the trade-offs between low power consumption and high data reliability. We have evaluated the performance of LEMoNet by deploying custom-designed sensor and gateway nodes in a production DC as well as through extensive simulation studies in networks of various sizes. We show experimentally that LEMoNet achieves an average data yield over 98 % in the production DC. It scales well in large and dense networks in large-scale simulations. Under normal operations with one temperature and one humidity reading every thirty seconds, the battery lifetime of LEMoN et sensor nodes is projected to be 14.9 years on a single lithium coin battery.

Data synchronization is crucial in ubiquitous computing systems, where heterogeneous sensor devices, modalities, and different communication capabilities and protocols are the norm. A common notion of time among devices is required to make sense of their sensing data. Traditional synchronization methods rely on wireless communication between devices to synchronize, potentially incurring computational and power costs. Furthermore, they are unsuitable for synchronizing data streams that have already been collected. We present CRONOS: a post-hoc, data-driven framework for sensor data synchronization for wearable and Internet-of-Things devices that takes advantage of independent, omni-present motion events in the data streams of two or more sensors. Experimental results on pairwise and multi-sensor synchronization show a drift improvement as high as 98% and a mean absolute synchronization error of approximately 6ms for multi-sensor synchronization with sensors sampling at 100Hz.

An important category of mobile crowdsensing applications involve collecting sensor measurements from mobile devices and querying mobile users for annotations to build machine learning models for inference and prediction. Trade-offs between inference performance and the costs of data acquisition (both unlabeled and labeled) are not yet well understood. In this paper, we develop, ALSense, a distributed active learning framework for mobile crowdsensing. The goal is to minimize prediction errors for classification-based mobile crowdsensing tasks subject to upload and query cost constraints. Novel stream-based active learning strategies are developed to orchestrate queries of annotation data and the upload of unlabeled data from mobile devices. We evaluate the effectiveness of ALSense through two applications that can benefit from mobile crowdsensing, namely, WiFi fingerprint-based indoor localization and IMU-based human activity recognition. Extensive experiments demonstrate that ALSense can indeed achieve higher classification accuracy given fixed data acquisition budgets for both applications.

Many infrastructure-free indoor positioning systems rely on fine-grained location-dependent fingerprints to train models for localization. The site survey process to collect fingerprints is laborious and is considered one of the major obstacles to deploying such systems. In this paper, we propose trajectory radio fingerprint (TuRF), a fast path-based fingerprint collection mechanism for site survey. We demonstrate the feasibility to collect fingerprints for indoor localization during walking along predefined paths. A step counter is utilized to accommodate the variations in walking speed. Approximate location labels inferred from the steps are then used to train a Gaussian Process regression model. Extensive experiments show that TuRF can significantly reduce the required time for site survey, without compromising the localization performance.

Acoustic ranging based indoor positioning solutions have the advantage of higher ranging accuracy and better compatibility with commercial-off-the-self consumer devices. However, similar to other time-domain based approaches using Time-of-Arrival and Time-Difference-of-Arrival, they suffer from performance degradation in presence of multi-path propagation and low received signal-to-noise ratio (SNR) in indoor environments. In this paper, we improve upon our previous work on asynchronous acoustic indoor positioning and develop ARABIS, a robust and low-cost acoustic positioning system (IPS) for mobile devices. We develop a low-cost acoustic board custom-designed to support large operational ranges and extensibility. To mitigate the effects of low SNR and multi-path propagation, we devise a robust algorithm that iteratively removes possible outliers by taking advantage of redundant TDoA estimates. Experiments have been carried in two testbeds of sizes 10.67m × 7.76m and 15m × 15m, one in an academic building and one in a convention center. The proposed system achieves average and 95% quantile localization errors of 7.4cm and 16.0cm in the first testbed with 8 anchor nodes and average and 95% quantile localization errors of 20.4cm and 40.0cm in the second testbed with 4 anchor nodes only.

Underwater sensor networks (UWSNs) have been investigated in a variety of applications such as sea resources reconnaissance, pollution monitoring and tactical monitoring. In 3D underwater environments, it is a key topic to judge the localizability of sensor nodes given known locations of a small set of anchor nodes. In this paper, a novel localizability judgment method for UWSNs is proposed based on rigidity theory. A UWSN is modelled as an undirected graph based on acoustic connectivity. The graph is then reduced to a subgraph with global rigidity, called skeleton, from which the set of localizable sensors can determined. Furthermore, the Analytic Hierarchy Process (AHP) is used to evaluate the localization confidence of localizable sensors. Extensive simulations demonstrate that the proposed localizability judgment method can achieve low false negative rate and high efficiency networks of different sensor numbers and sensor densities. It is also shown to perform well in dynamic networks with relatively low waterflow speed.

To maintain productivity and alertness, individuals must be thermally comfortable in the space they occupy (whether it is a cubicle, a room, a car, etc.). However, it is often difficult to non-intrusively assess an occupant’s “thermal comfort,” and hence most HVAC engineers adopt fixed temperature settings to “err on the safe side.” These set temperatures can be too hot or too cold for individuals wearing different clothing, and as a result lead to feelings of discomfort as well as wastage of energy. Since humans dress to target a comfortable thermal sensation, it is reasonable to assume that clothing is an important measure of current thermal sensation. To this end, we develop SiCILIA, a platform that extracts physical and personal variables of an occupant’s thermal environment to infer the amount of clothing insulation without human intervention. The proposed inference algorithm builds upon theories of body heat transfer and is corroborated by empirical data. SiCILIA was tested in a vehicle with a passenger-controlled HVAC system. Experimental results show that the algorithm is capable of accurately predicting an occupant’s thermal insulation with a mean prediction error of 0.07clo.

In this paper, we consider the problem of optimal opportunistic spectrum access using full-duplex (FD) radios in presence of uncertain primary user (PU) channel statistics and propose a Sensing-and-Selectively-Transmit protocol (SaST). To optimize its throughput, the SU sensing period has to be carefully tuned. However, in absence of the exact knowledge of PU activity statistics, under SaST, the PU's performance may be adversely affected. A learning strategy is devised to update the estimated statistics based on spectrum sensing observations. Simulation studies demonstrate that the resulting robust solution provides a good trade-off between optimizing the SU's throughput and protecting the PU.

In this paper, we investigate the scheduling problem with multi-resource allocation in cloud computing environments. In contrast to existing work that focuses on flow-level scheduling, which treats flows in isolation, we consider dependency among subtasks of applications that imposes a partial order relationship in execution. We formulate the problem of Multi-Resource Partial-Ordered Task Scheduling (MR-POTS) to minimize the makespan. In the first stage, the proposed Dominant Resource Priority (DRP) algorithm decides the collection of subtasks for resource allocation by taking into account the partial order relationship and characteristics of subtasks. In the second stage, the proposed Maximum Utilization Allocation (MUA) algorithm partitions multiple resources among selected subtasks with the objective to maximize the overall utilization. Both theoretical analysis and experimental evaluation demonstrate the proposed algorithms can approximately achieve the minimal makespan with high resource utilization. Specifically, a reduction of 50% in makespan can be achieved compared with existing scheduling schemes.

Multimedia streaming applications with stringent QoS requirements in 60 GHz mmWave wireless personal area networks (WPANs) demand high rate and low latency data transfer as well as little service disruption. In this paper, we consider the problem of robust relay placement in 60 GHz WPANs with directional antenna. Relays forward traffic from transmitter devices to receiver devices facilitating i) the primary communication path for non-line-of-sight (NLOS) transceiver pairs, and ii) secondary (backup) communication path for line-of-sight (LOS) or NLOS transceiver pairs. By incorporating a classic directional antenna model and characterizing the link contention, we formulate the robust minimum relay placement problem and the robust maximum utility relay placement problem with the objective to minimize the number of relays deployed and maximize the network utility, respectively. Efficient algorithms are developed to solve both problems and have been shown to incur less service disruption in presence of moving subjects that may block the LOS paths in the environment.

In this paper, we consider the problem of concurrent transmissions in a wireless network consisting of multiple radio access technologies (multi-RATs). That is, a single flow of packets is dispatched over multiple RATs so that the complementary advantages of different RATs can be exploited. One of the challenging issues arising in concurrent transmissions is the packet out-of-order problem due to diverse wireless channel states and scheduling policies of different RATs, leading to substantial performance degradation to delay sensitive applications. To address this problem, we first propose a state-independent packet dispatching (SIPD) policy, which attempts to find the traffic dispatching ratios over multiple RATs to minimize the maximum average delay across different RATs in the long run. We further propose a state-dependent packet dispatching (SDPD) policy, which achieves fine-grained packet dispatching in the short-term. We use the value function as a measure of the admittance cost for packet dispatching given the current queueing states, and formulate the SDPD problem as a convex optimization problem. We derive the closed-form solutions for both problems for the special case of two RATs, and adopt the dual decomposition technique as the solution for the general cases. Simulation results are presented to compare the performance of the proposed schemes with existing solutions.

In order to maintain productivity and alertness, individuals must be thermally comfortable in the space they occupy (whether it is a cubicle, a room, a car, etc.). However, it is often difficult to non-intrusively assess an occupant's"thermal comfort" and hence most heating, ventilation, and air conditioning (HVAC) engineers adopt fixed temperature settings to "err on the safe side". These set temperatures can be too hot or too cold for individuals wearing different clothing, and as a result lead to feelings of discomfort as well as wastage of energy. To address these challenges, we develop SiCILIA, a platform that extracts physical and personal variables of an occupant's thermal environment to infer the amount of clothing insulation without human intervention. Clothing insulation is one of the most influential factors in determining thermal comfort. The proposed inference algorithm builds upon theories of body heat transfer, and is corroborated by empirical data. Experimental results show that the algorithm is capable of accurately predicting an occupant's thermal insulation with a mean prediction error of approximately 0.2 clo.

Location-based services have experienced substantial growth in the last decade. However, despite extensive research efforts, sub-meter location accuracy with low-cost infrastructure continues to be elusive. In this paper, we propose IDyLL -- an indoor localization system using inertial measurement units (IMU) and photodiode sensors on smartphones. Using a novel illumination peak detection algorithm, IDyLL augments IMU-based pedestrian dead reckoning with location fixes. We devise a robust particle filter framework to mitigate identity ambiguity due to the lack of communication capability of conventional luminaries and sensing errors. Experimental study using data collected from smartphones shows that IDyLL is able to achieve high localization accuracy at low costs. Mean location errors of 0.38 m, 0.42 m, and 0.74 m are reported from multiple walks in three buildings with different luminary arrangements, respectively.

In this paper, we present a framework of resource allocations for multicast device-to-device (D2D) communications underlaying a cellular network. The objective is to maximize the sum throughput of active cellular users (CUs) and feasible D2D groups in a cell, while guaranteeing a certain level of the signal-to-interference-plus-noise ratio (SINR) for both the CUs and D2D groups. We formulate the problem of power and channel allocations as a mixed integer nonlinear programming (MINLP) problem where each D2D group can reuse the channel of at most one CU and each CU can share their resources with at most one D2D group. A maximum weight bipartite matching based scheme is developed to assign the optimal channel for each feasible D2D group to reuse. A heuristic algorithm is then proposed which has less complexity compared to the matching algorithm. The performance of both schemes is evaluated through simulations. Numerical results demonstrate that the proposed heuristic scheme outperforms other heuristic schemes in the literature and can achieve close-to-optimal performance.

In this paper, we consider the emerging deployment of WiFi networks in sports and entertainment venues characterized by high-density, large capacity, and real-time service delivery. Due to extremely high user density, channel allocation and user association should be carefully managed so that cochannel inference can be mitigated. To this end, we propose a channel selection and user association (CSUA) solution based on the Adversarial Multi-armed Bandit (AMAB) framework, which captures not only the uncertainty of channel states, but also the selfishness of individual stations (STAs) and access points (APs). An exponentially weighted average strategy is adopted to design an online algorithm for this problem, which is guaranteed to converge to a set of correlated equilibria with vanishing regrets. Simulation results show the convergence of the proposed algorithm and its performance under different settings.

Mobile participatory sensing (MPS), a paradigm that utilizes pervasive mobile devices to efficiently collect data, has gained much interest in a variety of applications in outdoor spaces. In this paper, we demonstrate the potential of MPS in indoor environments through one specific application - burned-out luminary detection. To mitigate the inherent inaccuracy of indoor positioning systems (IPS), we devise a Dynamic Time Warping-based approach and leverage majority votes across multiple runs of measurements. Experimental study using data collected from off-the-shelf smartphones and a basic IPS shows that the proposed approach can indeed detect the presence of burnout luminaries with high accuracy, and can significantly outperform a baseline method. Interestingly, we also demonstrate that the proposed mechanism can be leveraged to improve an existing luminary-assisted solution, and thus leads to zero-configuration and robust IPS.

We present SiCILIA, a hardware platform that extracts physical and personal variables of an individual's thermal environment to infer the amount of clothing insulation and thermal sensation without human intervention. The proposed inference algorithms build upon theories of body heat transfer, and are corroborated by empirical data. Experimental results show the algorithm is capable of accurately predicting an occupant's thermal insulation with a confidence interval of approximately ±0.3 and a mean prediction error of 0.2.

We present SiCILIA, a hardware platform that exploits the physical and environmental variables of an individual's thermal environment to infer the amount of clothing insulation and thermal sensation without intervention or feedback. In this demo, we present the inference algorithms that build upon theories of body heat transfer. Using a single Micro-controller (an Arduino Uno), a dual-axis motor, an IR sensor, and an Ultrasonic Range Finder (URF) we will demonstrate SiCILIA's capability of inferring an individual's clothing insulation level correctly and effectively.

Regenerating codes provide an efficient way to recover data at failed nodes in distributed storage systems. It has been shown that regenerating codes can be designed to minimize the per-node storage (called MSR) or minimize the communication overhead for regeneration (called MBR). In this work, we propose new encoding schemes for error-correcting MSR and MBR codes that generalize our earlier results on error-correcting regenerating codes. General encoding schemes for product-matrix MSR and MBR codes are derived such that the encoder based on Reed-Solomon (RS) codes is no longer limited to the Vandermonde matrix proposed earlier. Furthermore, MSR codes and MBR codes with the least update complexity can be found. A decoding scheme is proposed that utilizes RS codes to perform data reconstruction for MSR codes. The proposed decoding scheme has better error correction capability and incurs least number of node accesses when errors are present. A new decoding scheme is also proposed for MBR codes that is more capable and can correct more error-patterns. Simulation results are presented that exhibit the superior performance of the proposed schemes.

Understanding indoor mobility patterns of people is important in applications such as targeted advertisement, microclimate control, and delivery of anticipatory notifications. In this article, we devise GreenLocs, a nonparametric, profiling-free, yet lightweight and energy-efficient inference framework, to identify recurring and new places that mobile users visit indoor. Combining WiFi scans and accelerometer readings, GreenLocs can accurately decide a new place and a revisited place with just a few radio signal strength (RSS) samples. GreenLocs consists of three major building blocks, namely, missing data handling algorithms, a nonparametric Bayesian inference model, and a stopping rule, which significantly increases the energy efficiency of the system. GreenLocs is shown to be robust to signal variations and missing data through experimental evaluations using traces collected from mobile phones of different brands/models.

This paper presents a quantitative study of Twitter, one of the most popular micro-blogging services, from the perspective of user influence. We crawl several datasets from the most active communities on Twitter and obtain 20.5 million user profiles, along with 420.2 million directed relations and 105 million tweets among the users. User influence scores are obtained from influence measurement services, Klout and PeerIndex. Our analysis reveals interesting findings of the structural properties of Twitter communities. Most importantly, we observe that whether a user retweets a message is strongly influenced by the first of his followees who posted that message. To capture such an effect, we propose the first influencer (FI) information diffusion model and show through extensive evaluation that compared to the widely adopted independent cascade model, the FI model is more stable and more accurate in predicting influence spreads in Twitter communities.

In this paper, the problem of uplink user association in small cell networks, which involves interactions between users, small cell base stations, and macro-cell stations, having often conflicting objectives, is considered. The problem is formulated as a college admissions game with transfers in which a number of colleges, i.e., small cell and macro-cell stations seek to recruit a number of students, i.e., users. In this game, the users and access points (small cells and macro-cells) rank one another based on preference functions that capture the users' need to optimize their utilities which are functions of packet success rate (PSR) and delay as well as the small cells' incentive to extend the macro-cell coverage (e.g., via cell biasing/range expansion) while maintaining the users' quality-of-service. A distributed algorithm that combines notions from matching theory and coalitional games is proposed to solve the game. The convergence of the algorithm is shown and the properties of the resulting assignments are discussed. Simulation results show that the proposed approach yields a performance improvement, in terms of the average utility per user, reaching up to 23% relative to a conventional, best-PSR algorithm.

Today's large-scale distributed storage systems are commonly built using commodity software and hardware. As a result, crash-stop and Byzantine failures in such systems become more and more prevalent. In the literature, regenerating codes have been shown to be a more efficient way to disperse information across multiple storage nodes and recover from crash-stop failures. In this paper, we propose a novel decoding design of product-matrix constructed regenerating codes in conjunction with integrity check that allows exact regeneration of failed nodes and data reconstruction in the presence of Byzantine failures. A progressive decoding mechanism is incorporated in both procedures to leverage computation performed thus far. Unlike previous works, our new regenerating code decoding has the advantage that its building blocks, such as Reed-Solomon codes and standard cryptographic hash functions, are relatively well-understood because of their widespread applications. The fault tolerance and security properties of the proposed schemes are also analyzed. In addition, the performance of the proposed schemes, in terms of the average number of access nodes and the reconstruction failure probability versus the node failure probability, are also evaluated by Monte Carlo simulations.

Passive monitoring utilizing distributed wireless sniffers is an effective technique to monitor activities in wireless infrastructure networks for fault diagnosis, resource management, and critical path analysis. In this paper, we introduce a quality of monitoring (QoM) metric defined by the expected number of active users monitored, and investigate the problem of maximizing QoM by judiciously assigning sniffers to channels based on the knowledge of user activities in a multichannel wireless network. Two types of capture models are considered. The user-centric model assumes the frame-level capturing capability of sniffers such that the activities of different users can be distinguished while the sniffer-centric model only utilizes the binary channel information (active or not) at a sniffer. For the user-centric model, we show that the implied optimization problem is NP-hard, but a constant approximation ratio can be attained via polynomial complexity algorithms. For the sniffer-centric model, we devise stochastic inference schemes to transform the problem into the user-centric domain, where we are able to apply our polynomial approximation algorithms. The effectiveness of our proposed schemes and algorithms is further evaluated using both synthetic data as well as real-world traces from an operational WLAN.

ZigBee and other wireless technologies operating in the (2.4GHz) ISM band are being applied in Wireless Personal Area Networks (WPAN) for many medical applications. However, these low duty cycle, low power, and low data rate medical WPANs suffer from WiFi co-channel interferences. WiFi interference can lead to longer latency and higher packet losses in WPANs, which can be particularly harmful to safety-critical applications with stringent temporal requirements, such as ElectroCardioGraphy (ECG). This paper exploits the Clear Channel Assessment (CCA) mechanism in WiFi devices and proposes a novel policing framework, WiCop, that can effectively control the temporal white-spaces between WiFi transmissions. Such temporal white-spaces can be utilized for delivering low duty cycle WPAN traffic. We have implemented and validated WiCop on SORA, a software-defined radio platform. Experimental results show that with the assistance of the proposed WiCop policing schemes, the packet reception rate of a ZigBee-based WPAN can increase by up to 116% in the presence of a heavy WiFi interferer. A case study on the medical application of WPAN ECG monitoring demonstrates that WiCop can bound ECG signal distortion within 2% even under heavy WiFi interference. An analytical framework is devised to model the CCA behavior of WiFi interferers and the performance of WPANs under WiFi interference with or without WiCop protection. The analytical results are corroborated by experiments.

We consider the problem of optimally selecting m out of M sniffers and assigning each sniffer one of the K channels to monitor the transmission activities in a multi-channel wireless network. The activity of users is initially unknown to the sniffers and is to be learned along with channel assignment decisions. Even with the full knowledge of user activity statistics, the offline optimization problem is known to be NP-hard. In this paper, we first propose a centralized online approximation algorithm and show that it incurs sub-linear regret bounds over time. A distributed algorithm is then proposed with moderate message complexity. We demonstrate both analytically and empirically the trade-offs between the computation cost and the rate of learning.

Cyber-Physical Systems (CPS) integrate discrete-time computing and continuous-time physical-world entities, which are often wirelessly interlinked. The use of wireless safety critical CPS (control, healthcare etc.) requires safety guarantees despite communication faults. This paper focuses on one important set of such safety rules: Proper-Temporal-Embedding (PTE). Our solution introduces hybrid automata to formally describe and analyze CPS design patterns. We propose a novel lease based design pattern, along with closed-form configuration constraints, to guarantee PTE safety rules under arbitrary wireless communication faults. We propose a formal methodology to transform the design pattern hybrid automata into specific wireless CPS designs. This methodology can effectively isolate physical world parameters from affecting the PTE safety of the resultant specific designs. We conduct a case study on laser tracheotomy wireless CPS to show that the resulting system is safe and can withstand communication disruptions.

Given a set of points of interest (POIs), the spatial skyline query for a set of locations returns the POIs that are close to all locations. Answering spatial skyline query can find many applications in Geographical Information Systems. In this paper, we consider the problem of spatial skyline query with uncertainty. Two types of uncertainty are investigated. First, location uncertainty arises when query point (user) locations are not known exactly either due to privacy concerns or measurement limitations. Second, error margins can be used to model tolerance to distance measurement errors between POIs and query points. We devise efficient polynomial-time algorithms to address both types of uncertainty, and rigorously prove their correctness.

Aging power industries together with increase in the demand from industrial and residential customers are the main incentive for policy makers to define a road map to the next generation power system called smart grid. In smart grid, the overall monitoring costs will be decreased but at the same time, the risk of cyber attacks might be increased. Recently a new type of attacks (called the stealth attack) has been introduced, which cannot be detected by the bad data detection using state estimation. In this paper, we show how normal operations of power networks can be statistically distinguished from the case under stealthy attacks. We devise two machine learning based techniques for stealthy attack detection. The first method utilizes supervised learning over labeled data and trains a support vector machine. The second method requires no training data and detects the deviation in measurement In both methods, principle component analysis is used to reduce the dimensionality of the data to be processed, and thus leads to lower computation complexities. The results of the proposed detection methods on the IEEE standard test systems demonstrate effectiveness of both schemes.

We consider the problem of optimally assigning p sniffers to K channels to monitor the transmission activities in a multi-channel wireless network. The activity of users is initially unknown to the sniffers and is to be learned along with channel assignment decisions. Previously proposed online learning algorithms face high computational costs due to the NPhardness of the decision problem. In this paper, we propose two approximate online learning algorithms, ϵ-GREEDY-APPROX and EXP3-APPROX, which are shown to have better scalability, and achieve sub-linear regret bounds over time compared to a greedy offline algorithm with complete information. We demonstrate both analytically and empirically the trade-offs between the computation cost and rate of learning.

Regenerating codes provide an efficient way to recover data at failed nodes in distributed storage systems. It has been shown that regenerating codes can be designed to minimize the per-node storage (called MSR) or minimize the communication overhead for regeneration (called MBR). In this work, we propose a new encoding scheme for [n, d] error-correcting MSR codes that generalizes our earlier work on error-correcting regenerating codes. We show that by choosing a suitable diagonal matrix, any generator matrix of the [n, α] Reed-Solomon (RS) code can be integrated into the encoding matrix. Hence, MSR codes with the least update complexity can be found. An efficient decoding scheme is also proposed that utilizes the [n, α] RS code to perform data reconstruction. The proposed decoding scheme has better error correction capability and incurs the least number of node accesses when errors are present.

Location extraction in an indoor environment is a great challenge, and yet, it is of great interest to retrieve locations information without manually labeling them. Indoor location information, e.g. which room a user is located, is precious for applications such as location based services, mobility prediction, personal health care, network resource allocation, etc. Since the GPS signal is missing, another form of identification for each location is needed. WiFi is a potential candidate due to its easy availability. However, it is very noisy and missing excessively due to the limited range of access points. We propose a two-layer clustering method that is able to i) classify the rooms in an unsupervised manner; ii) handle missing data effectively. Experiment results using the real traces show UMLI can achieves an identification rate of 99.84%.

Given a fixed budget and an arbitrary cost for selecting each node, the budgeted influence maximization (BIM) problem concerns selecting a set of seed nodes to disseminate some information that maximizes the total number of nodes influenced (termed as influence spread) in social networks at a total cost no more than the budget. Our proposed seed selection algorithm for the BIM problem guarantees an approximation ratio of (1-1/√e). The seed selection algorithm needs to calculate the influence spread of candidate seed sets, which is known to be #P-complex. Identifying the linkage between the computation of marginal probabilities in Bayesian networks and the influence spread, we devise efficient heuristic algorithms for the latter problem. Experiments using both large-scale social networks and synthetically generated networks demonstrate superior performance of the proposed algorithm with moderate computation costs. Moreover, synthetic datasets allow us to vary the network parameters and gain important insights on the impact of graph structures on the performance of different algorithms.

Spectrum sensing problem, which focuses on detecting the presence of primary users (PUs) in the cognitive radio (CR) network receives much attention recently. In this paper, we introduce the PU separation problem, which concerns with the issue of distinguishing and characterizing the activities of PUs in the context of collaborative spectrum sensing and monitor selection. Observations of secondary users (SUs) are modeled as boolean OR mixtures of underlying binary PU sources. We devise a binary inference algorithm for PU separation. With binary inference, not only PU-SU relationship are revealed, but PUs' transmission statistics and activities at each time slot can also be inferred. Simulation results show that without any prior knowledge regarding PUs' activities, the algorithm achieves high inference accuracy even in the presence of noisy measurements.

Due to the use of commodity software and hardware, crash-stop and Byzantine failures are likely to be more prevalent in today's large-scale distributed storage systems. Regenerating codes have been shown to be a more efficient way to disperse information across multiple nodes and recover crash-stop failures in the literature. In this paper, we present the design of regeneration codes in conjunction with integrity check that allows exact regeneration of failed nodes and data reconstruction in the presence of Byzantine failures. A progressive decoding mechanism is incorporated in both procedures to leverage computation performed thus far. The fault tolerance and security properties of the schemes are also analyzed.

Since the onset of wireless sensor networks, target localization and tracking have received much attention in a wide range of applications including battle field surveillance, wildlife monitoring and border security. However, little work has been done that addresses the realistic considerations of uncertainties in sensor locations and evaluates their impacts on the accuracy of target localization and tracking. In this paper, we carry out a rigorous study of these problems using a computational geometry approach. We introduce the geometric structures of order-k max and min Voronoi Diagrams (VDs) and propose an algorithm to construct these diagrams. Based on order-k max and min VDs, efficient algorithms are developed to evaluate the likelihood of noisy sensor readings and kNN queries, which serve as building blocks in target localization and tracking under sensor location uncertainties.

Location privacy is an important concern in participatory sensing applications, where users can both contribute valuable information (data reporting) as well as retrieve (location-dependent) information (query) regarding their surroundings. K-anonymity is an important measure for privacy to prevent the disclosure of personal data. In this paper, we propose a mechanism based on locality-sensitive hashing (LSH) to partition user locations into groups each containing at least K users (called spatial cloaks). The mechanism is shown to preserve both locality and K-anonymity. We then devise an efficient algorithm to answer kNN queries for any point in the spatial cloaks of arbitrary polygonal shape. Extensive simulation study shows that both algorithms have superior performance with moderate computation complexity.

To integrate the nowadays rapidly expanding distributed real-time systems, we need multi-hop real-time switched networks. A (if not "the") widely recognized/adopted real-time switch architecture is the TDMA crossbar real-time (TCRT) switch architecture. However, the original TCRT switch architecture assumes per-flow queueing. To support scalability, however, queue sharing (i.e. flow aggregation), must be allowed. With simple flow aggregation, flow burstiness can grow and infect, making schedulability and end-to-end delay bound analysis an open problem. To deal with this, we propose the real-time aggregate scheme. The scheme complies with the existing TCRT switch architecture, and deploys spatial-temporal isolation and over-provisioning to curb aggregate member flows' burstiness. This allows us to derive the closed-form end-to-end delay bound, and give the corresponding resource planning and admission control strategies. Simulations are carried out to show the effectiveness of the design.

Each node in a wireless sensor network has some data storage capability that preserves gathered data until that data is either requested by a data collector, or utilized by the network itself. Local storage, where every node stores its own data locally, is disadvantageous because a software or mechanical failure may cause permanent data loss of some or all of the data. Robust distributed networked storage, where each node stores its data amongst other nodes in a redundant manner, offers higher levels of data persistence than local storage. Using a distributed storage scheme, a data collector can reconstruct any node's data by downloading a portion of data from a previously defined threshold number of nodes. Consider a set of nodes storing some data distributively. The failure of any node in that set, reduces the data's level of redundancy. However, if a newcomer joins the network, it can be used to restore the data's level of redundancy, by downloading some data from a threshold number of live nodes. This recovery process is termed regeneration. In the case of a software failure, the node that failed may be reassigned to be the newcomer, if it is able to return to its previous working state. In this paper, we generalize the terms reconstruction and regeneration to the term recovery. We propose a framework that serves as a measure of performance for algorithms targeting data persistence. We then propose one such algorithm and show that it achieves a constant approximation ratio of the optimal, with respect to the framework. Finally, we conclude with simulations analyzing the proposed data recovery framework.

Topology engineering concerns with the problem of automatic determination of physical layer parameters to form a network with desired properties. In this paper, we investigate the joint power control, channel assignment, and radio interface selection for robust provisioning of link bandwidth in infrastructure multiradio multichannel wireless networks in presence of channel variability and external interference. To characterize the logical relationship between spatial contention constraints and transmit power, we formulate the joint power control and radio-channel assignment as a generalized disjunctive programming problem. The generalized Benders decomposition technique is applied for decomposing the radio-channel assignment (combinatorial constraints) and network resource allocation (continuous constraints) so that the problem can be solved efficiently. The proposed algorithm is guaranteed to converge to the optimal solution within a finite number of iterations. We have evaluated our scheme using traces collected from two wireless testbeds and simulation studies in Qualnet. Experiments show that the proposed algorithm is superior to existing schemes in providing larger interference margin, and reducing outage and packet loss probabilities.

We propose a decentralized progressive data retrieval (PDR) mechanism for data reconstruction in a network of Byzantine and crash-stop nodes. The scheme progressively retrieves stored data, such that it achieves the minimum communication cost possible. In particular, PDR gracefully adapts the cost of successful data retrieval to the number of Byzantine and crash-stop storage nodes. At the core of PDR is an incremental Reed-Solomon decoding (IRD) procedure that is highly computation efficient for data reconstruction. IRD's computation efficiency arises from its ability to utilize intermediate computation results. In addition, we provide an in-depth analysis of PDR and compare it to decentralized erasure coding and decentralized fountain coding algorithms for distributed storage systems. Moreover, our implementation results show that PDR has up to 35 times lower computation time over the state-of-the-art error-erasure decoding scheme for distributed storage systems. In our analysis, we also show that the code structure of PDR and the number of available storage nodes are independent of each other, and they can be used to control both the data dissemination and retrieval complexity.

Wireless monitoring employing distributed sniffers has been shown to complement wire side monitoring using SNMP and base station logs since it reveals detailed PHY (e.g., signal strength, spectrum density) and MAC behaviors (e.g, collision, retransmissions), as well as timing information (e.g., back-off time), which are often essential for network diagnosis. Due to hardware limitations, wireless sniffers typically can only collect information on one channel at a time. Thus, it is important to determine the optimal channel allocation of sniffer nodes to maximize the information collected. In this paper, we propose a Gibbs sampler approach for optimal distributed monitoring of multi-channel wireless networks with provable convergence. Simulation studies show that in general the proposed method has low computation complexity while achieving optimal or near optimal solutions.

Spectrum monitoring is important to ensure the safe operation of mission critical systems as well as the satisfactory performance of non-critical applications over wireless. In this paper, we present a novel blind technology identification (BTI) approach that utilizes only binary representation of spectrum activities to identify transmission technologies that are present in the radio spectrum. Spectrum observations are modeled as Boolean OR mixtures on the underlying signal sources, and the binary independent component analysis technique is applied. Not only can we reveal the latent independent wireless technologies, but their transmission statistics and activities at each time slot can also be inferred. Evaluation results on both synthetic and real spectrum traces show that without any high-level features, the proposed methodology achieves high inference accuracy in noisy measurements.

Erasure codes have been employed to achieve persistent storage in distributed storage networks. Recent work has shown that, in addition to reduction in storage space requirements, the communication bandwidth in the data regeneration process can be further reduced by using Regenerating Codes. In this paper, we consider the issue of energy- efficient data regeneration in wireless sensor networks with the objective of minimizing energy expenditure and thereby maximizing network lifetime. We formally prove the NP-hardness of finding the optimal set of source nodes and corresponding routes for data regeneration in general networks, and devise an optimal polynomial algorithm, TROY, for acyclic networks; the cardinality of the set is pre- defined. Building upon TROY, we devise a heuristic algorithm for general networks and show, through extensive simulation studies, that this heuristic is near-optimal.

We consider the problem of optimally assigning p sniffers to K channels to monitor the transmission activities in a multi-channel wireless network. The activity of users is initially unknown to the sniffers and is to be learned along with channel assignment decisions while maximizing the benefits of this assignment, resulting in the fundamental trade-off between exploration versus exploitation. We formulate it as the linear partial monitoring problem, a super-class of multi-armed bandits. As the number of arms (sniffer-channel assignments) is exponential, novel techniques are called for, to allow efficient learning. We use the linear bandit model to capture the dependency amongst the arms and develop two policies that take advantage of this dependency. Both policies enjoy logarithmic regret bound of time-slots with a term that is sub-linear in the number of arms.

Each wireless device has its unique fingerprint, which can be utilized for device identification and intrusion detection. Most existing literature employs supervised learning techniques and assumes the number of devices is known. In this paper, based on device-dependent channel-invariant radio-metrics, we propose a non-parametric Bayesian method to detect the number of devices as well as classify multiple devices in a unsupervised passive manner. Specifically, the infinite Gaussian mixture model is used and a modified collapsed Gibbs sampling method is proposed. Sybil attacks and Masquerade attacks are investigated. We have proven the effectiveness of the proposed method by both simulation data and experimental measurements obtained by USRP2 and Zigbee devices.

Uncertainty in sensor locations is a norm in both planned and unplanned deployments. Even carefully positioned in the deployment phase, sensors may be displaced due to environmental or human factors during the course of operation. In this paper, we present a systematic study of the impact of location uncertainty on the coverage properties of wireless sensor networks. The uncertainty is modeled as disks of possibly different radius around the nominal positions. We introduce the concept of order-k (k ≥ 1) max Voronoi Diagram (VD) and devise an efficient polynomial algorithm to construct order-k VDs. Order-k max VD is critical in determining the minimum sensing radius needed to ensure worst-case k-coverage, call k-exposure. Simulation studies validate the correctness of the proposed algorithms and demonstrate their superiority over a naive approach.

ZigBee and other wireless technologies operating in the (2.4GHz) ISM band are being applied in Wireless Body Area Networks (WBAN) for many medical applications. However, these low duty cycle, low power, and low data rate medical WBANs suffer from WiFi co-channel interferences. WiFi interference can lead to longer latency and higher packet losses in WBANs, which can be particularly harmful to safety-critical applications with stringent temporal requirements. Existing solutions to WiFi-WBAN coexistence either require modifications to WiFi or WBAN devices, or have limited applicability. In this paper, by exploiting the Clear Channel Assessment (CCA) mechanisms in WiFi devices, we propose a novel policing framework, WiCop, that can effectively control the temporal white-spaces between WiFi transmissions. Specifically, the WiCop Fake-PHY-Header policing strategy uses a fake WiFi PHY preamble-header broadcast to mute other WiFi interferers for the duration of WBAN active interval, while the WiCop DSSS-Nulling policing strategy uses repeated WiFi PHY preamble (with its spectrum side lobe nulled by a band-pass filter) to mute other WiFi interferers throughout the duration of WBAN active interval. The resulted WiFi temporal white-spaces can be utilized for delivering low duty cycle WBAN traffic. We have implemented and validated WiCop on SORA, a software defined radio platform. Experiments show that with the assistance of the proposed WiCop policing schemes, the packet reception rate of a ZigBee-based WBAN can increase by up to 43.8% in presence of a busy WiFi interferer.

In this paper, spectrum access in cognitive radio networks is modeled as a repeated auction game subject to monitoring and entry costs. For secondary users, sensing costs are incurred as the result of primary users' activity. Furthermore, each secondary user pays the cost of transmission upon successful bidding for a channel. Knowledge regarding other secondary users' activity is limited due to the distributed nature of the network. The resulting formulation is thus a dynamic game with incomplete information. To solve such a problem, a Bayesian nonparametric belief update scheme is constructed based on the Dirichlet process. Efficient bidding learning algorithms are proposed via which users can decide whether or not to participate in the bidding according to the belief update. Properties of optimal bidding and initial bidding are proved. As demonstrated through extensive simulations, the proposed distributed scheme outperforms a myopic one-stage algorithm, and can achieve a good trade-off between long-term efficiency and fairness.

Distributed active sensing is a new sensing paradigm, where active sensors and passive sensors are distributed in a field, and collaboratively detect and track the objects. "Exposure" of distributed active sensing networks (DASNs) quantifies the dimension limitations in detectability. It is important to deploy the sensors such that the exposure is minimized. Exposure minimization is shown to be NP-hard, and thus efficient heuristic algorithms are needed. In this paper, we propose a Genetic Algorithm (GA)-based solution that aims at achieving low exposure, scalability, and fast convergence. A novel flat binary chromosome encoding scheme and corresponding crossover and mutation operators are devised. Geometric knowledge is incorporated to significantly improve the convergence rate. Through extensive simulations, we demonstrate that the proposed algorithm outperforms a simple heuristic algorithm by up to 75%. The simulation results show that this algorithm is robust, self-adaptive and efficient under irregular boundary conditions.

We propose a storage-optimal and computation efficient primitive to spread information from a single data source to a set of storage nodes, to allow recovery from both crash-stop and Byzantine failures. A progressive data retrieval scheme is employed, which retrieves minimal amount of data from live storage nodes. The scheme adapts the cost of successful data retrieval to the degree of errors in the system. Implementation and evaluation studies demonstrate comparable performance to that of a genie-aid decoding process.

Distributed active sensing is a new sensing paradigm, where active sensors (aka actuators) as illuminating sources and passive sensors as receivers are distributed in a field, and collaboratively detect events of interest. In this paper, we study the fundamental properties of distributed actuator and sensor networks (DASNs) in detecting and localizing obstacles. A novel notion of “exposure” is defined, which quantifies the dimension limitations in detectability. Using simple geometric constructs, we propose polynomial-time algorithms to compute the exposure and bounding regions where the center of the obstacles may lie.

In this paper, we investigate the problem of link-level resource provision in multi-radio multi-channel (MR-MC) wireless networks. To quantify robustness of resource provision schemes, we propose the novel concept of interference margin. Using the notion of interference margin, a robust radio and channel assignment problem is formulated that explicitly takes into consideration link-level traffic demands. The key advantage of the proposed formulation is its robustness to channel variability and co-existence of external interference sources. We utilize the generalized Benders decomposition techniques to decouple the radio and channel assignment (combinatorial constraints) and network resource allocation (continuous constraints) so that the problem can be solved efficiently. The proposed algorithm is guaranteed to converge to the optimal solution within a finite number of iterations. We have evaluated our scheme using traces collected from a wireless mesh testbed and simulation studies in Qualnet. Experiments show that the proposed algorithm is superior to existing schemes in providing larger interference margin, and reducing outage and packet loss probabilities.

With the growing number of spontaneously deployed WiFi hotspots and home networks, end-users often experience significant performance degradation or even starvation. However, we observe that tuning individual system parameters (channel, Tx power, carrier sense (CS) threshold, and transmit rate etc.) is insufficient and in some cases may lead to starvation. In this paper, we develop a comprehensive analytical model to characterize throughput of individual flows in dense IEEE 802.11 wireless community networks. The proposed model subsumes existing models for the IEEE 802.11 MAC in multihop wireless networks by accounting for heterogeneous transmission power levels and CS thresholds, as well as various sources of packet collisions. Based on the insight from the theoretical analysis and simulation results, we propose a simple identification mechanism that determines the sources of starvation using local measurements. Both the theoretical model and the identification algorithm are validated using ns-2 simulations.

In this paper, BFLY-a practical localized network coding protocol for wireless mesh networks-is proposed. To supplement forwarding packets in classical networks, intermediate wireless nodes code packets from different sources, so that each transmission's information content is increased. Prior work allowed intermediate nodes to code (i.e, XOR) packets such that the recipient of the coded message must decode the message before forwarding. BFLY, however, allows intermediate recipients to, in addition to XOR-ing, forward coded packets; and thus further exploits network coding opportunities in multihop wireless networks. BFLY utilizes knowledge of local topologies and source route information in packet headers. We have developed network coding modules in ns-2 that facilitate simulations with large networks. Simulation studies show that BFLY can increase overall network throughput by a factor of 1.2 - 2 and reduce end-to-end latency.

We present the Lightning Protocol, a hard real-time, fast, and lightweight protocol to elect the sensor closest to an impulsive sound source. This protocol can serve proximity-based localization or leader election for sensor collaboration. It utilizes the fact that electromagnetic waves propagate much faster than acoustic waves to efficiently reduce the number of contending sensors in the election. With simple RF bursts, most basic comparison operations, no need of clock synchronization, and a memory footprint as small as 5,330 bytes of ROM and 187 bytes of RAM, the protocol incurs O(1) transmissions, irrespective of the sensor density, and guarantees hard real-time (O(1)) localization time cost. Experiment results using UC Berkeley Motes in a common office environment demonstrate that the time delay for the Lightning Protocol is on the order of milliseconds. The simplicity of the protocol reduces memory cost, computation complexity, and programming difficulty, making it desirable for low-end wireless sensors.

In this paper, we consider the problem of designing optimal asynchronous wake-up schedules to facilitate distributed power management and neighbor discovery in multihop wireless networks. We first formulate it as a block design problem and derive the fundamental trade-offs between wake-up latency and the average duty cycle of a node. After the theoretical foundation is laid, we then devise a neighbor discovery and schedule bookkeeping protocol that can operate on the optimal wake-up schedule derived. To demonstrate the usefulness of asynchronous wake-up, we investigate the efficiency of neighbor discovery and the application of on-demand power management, which overlays a desirable communication schedule over the wake-up schedule mandated by the asynchronous wake-up mechanism. Simulation studies demonstrate that the proposed asynchronous wake-up protocol has short discovery time which scales with the density of the network; it can accommodate various traffic characteristics and loads to achieve an energy savings that can be as high as 70 percent, while the packet delivery ratio is comparable to that without power management

It has long been recognized that energy conservation usually comes at the cost of degraded performance such as longer delay and lower throughput in stand-alone systems and communication networks. However, there have been very few research efforts in quantifying such trade-offs. In this paper, we develop analytical models to characterize the relationships among energy, delay and throughput for different power management policies in wireless communication. Based on the decision when to put nodes to low-power states, we divide power management policies into two categories, i.e., 1) time-out driven and 2) polling-based. M/G/1/K queues with multiple vacations and an attention span are used to model time-out driven policies while transient analysis is applied to derive the state transition probability in polling-based systems. We find that For time-out driven power management policies, the "optimal" policy exhibits a threshold structure, i.e., when the traffic load is below certain threshold, a node should switch to the low-power state whenever possible and always remain active otherwise. From our analysis, contrary to general beliefs, polling-based policies such as the IEEE 802.11 PSM are not energy efficient for light traffic load.

In this paper, we investigate the problem of providing efficient communication primitives across domains of wireless sensor network (WSN) applications. We argue both qualitatively and quantitatively that group communication among sensors of geographic proximity is one of the basic building blocks of many WSN applications. Furthermore, group communication awareness needs to be embedded and implemented at the MAC layer due to the broadcast nature of wireless medium. We devise a MAC protocol, called LGC-MAC to enable efficient single-hop one-to-many and many-to-one communication. We present case studies of two example applications, acoustic target tracking and propagation of information with feedback using LGC-MAC and demonstrate that LGC-MAC can improve the response time, alleviate channel contention and provide better fault tolerance to packet collisions and wireless errors.

In this article, we define the time-indexing problem as the in-network storage and querying of sensor network data based solely on the time attribute. We argue qualitatively why existing storage schemes may be insufficient as solutions. We then present, analyze, and evaluate novel and lightweight solutions to both the storage and the querying subproblems for time indexing. First, the time-indexed storage problem is formally defined and two formulations are presented seeking to optimize generic utility functions that are derived from concerns about energy, bandwidth usage, and storage balancing. We present and analyze decentralized protocols to solve these formulations and prove the optimality of some of our solutions. Secondly, maintenance and use of simple overlays among rendezvous point nodes in order to enable fault-tolerant and efficient time-indexed queries are discussed. Finally, simulation results are presented to quantify performance characteristics of the protocols, and we find that our proposed scheme has low query overhead that scales with system size and density while exhibiting very good load-balancing and fault-tolerance properties. The use of time-indexed structure is shown to achieve more than double the lifetime of sensor networks compared to existing approaches in some scenarios.

Switching the devices to low-power states in prolonged periods of inactivity is a widely used technique to conserve energy for battery-powered wireless devices. In this paper, we present a mathematical abstraction of time-out driven power management policies together with different wakeup mechanisms in wireless networks to characterize the energy-performance trade-offs. The time-out driven power management is modeled as a M/G/1/K queue with multiple vacations and an attention span. We then derive the steady state behaviors of such systems, and present a closed-form solution for systems with large buffers. The analysis reveals that the "best" power management policy to minimize the energy-delay product exhibits a threshold structure, i.e., when the traffic load is below a certain threshold, a node should switch to the low-power state whenever possible and always remain active otherwise, and suggests a threshold-based power management protocol.

We define the time indexing problem as the in-network storage and querying of sensor network data based solely on the time attribute. We argue qualitatively why existing storage schemes may be insufficient as solutions. We then present, analyze, and evaluate novel and lightweight solutions to both the storage and the querying sub-problems for time indexing. First, the time-indexed storage problem is formally defined, and two formulations are presented, seeking to optimize generic utility functions that are derived from concerns about energy, bandwidth usage, and storage balancing. We present and analyze decentralized protocols to solve these formulations, and prove the optimality of some of our solutions. Secondly, maintenance and use of simple overlays among rendezvous point nodes, in order to enable fault-tolerant and efficient time-indexed queries, are discussed. Finally, simulation results are presented to quantify performance characteristics of the protocols, and we find that our proposed scheme has low query overhead that scales with system size and density while exhibiting very good load balancing and fault tolerance properties.

Acoustic awareness is an important service in ubiquitous computing environments. This paper presents a fast lightweight acoustic event localization protocol, the Lightning protocol, to locate impulsive sound sources using arrays of wireless micro-sensors. This protocol utilizes domain-invariant knowledge of acoustic and electromagnetic wave propagation to efficiently reduce the number of contending sensors in the localization process. It incurs O(1) transmissions irrespective of the sensor density and guarantees O(1) time delay in localization. Experiment results using UC Berkeley Motes demonstrate that the time delay for Lightning Protocol to locate hand clap sounds is in terms of milliseconds.

Due to the slow advancement of battery technology, power management in wireless networks remains to be a critical issue. Asynchronous wakeup has the merits of not requiring global clock synchronization and being resilient to network dynamics. This paper presents a systematic approach to designing and implementing asynchronous wakeup mechanisms in ad hoc networks. The optimal wakeup schedule design can be formulated as a block design problem in combinatorics. We propose a neighbor discovery and schedule bookkeeping protocol that can operate on the optimal wakeup schedule derived. Two power management policies, i.e. slot-based power management and on-demand power management, are studied to overlay desirable communication schedule over the wakeup schedule mandated by the asynchronous wakeup mechanism. Simulation studies indicate that the proposed asynchronous wakeup protocol is quite effective under various traffic characteristics and loads: energy saving can be as high as 70%, while the packet delivery ratio is comparable to that without power management.

The Mobile IP standard for mobility management on the Internet enables transparent communication between mobile hosts (MH) and their correspondent hosts (CH). However, it suffers from triangular routing and prolonged handoff latency problems. In this paper, we propose to use a temporary home agent (TA) to address both problems. TA exploits the locality of user movement observed in recent studies for personal communication services (PCS) and wireless data networks. It dynamically selects a Mobile IP-based HA based on the location of the user. The TA allocates a temporary home address, THAddr to the MH, which the MH may use as its source address. The underlying objective is to shorten the distance between a MH and its home agent, which is a critical factor in reducing handoff latency and improving routing efficiency. Through both quantitative analysis and ns-2 simulation, we show that the TA approach outperforms Mobile IP and achieves comparable performance to route optimization (RO). While the TA approach focuses on improving the performance of traffic for sessions that are initiated by the MH, methods for optimizing traffic handling for sessions initiated by CH are discussed.

We propose a new active queue management algorithm, called average rate early detection (ARED). An ARED gateway measures/uses the average packet enqueue rate as a congestion indicator, and judiciously signals end hosts of incipient congestion, with the objective of reducing packet loss ratio and improving link utilization. We show (via simulation in ns-2) that the performance of ARED is better than that of RED and comparable to that of BLUE in terms of packet loss rate and link utilization. We also explore the use of ARED in the context of the differentiated services architecture. We show analytically that the widely referenced queue management mechanism, RED with in and out (RIO) cannot achieve throughput assurance and proportional bandwidth sharing. We then extend ARED and propose a new queue management mechanism, called ARED with in and out (AIO), in the assured services architecture. To share surplus bandwidth in a rate-proportional manner, we incorporate into AIO the derived analytic results, and propose an enhanced version of AIO, called the differentiated-rate AIO (DAIO) mechanism.

In this paper, we study why the current assurance services (AS) architecture with profilers/markers at edge routers and with queue management mechanisms (RIO) at core routers cannot achieve throughput assurance and fairness. Based on the findings of the simulation study, we propose an enhanced version of the time sliding window (TSW) profiler, called the enhanced TSW (ETSW). We also design two enhanced versions of the RIO queue management mechanism, based on rigorous, analytical reasoning, called respectively, the (r,RTT)-adaptive algorithm and the dynamic RIO (DRIO) algorithm. To validate the proposed design, we implement the proposed mechanisms, along with the 2-window TCP scheme, the three color-marker scheme, and the CSFQ scheme in ns-2, and examine their behavior under a variety of network topologies and traffic sources. The simulation results indicate that both DRIO and (r,RTT)-adaptive algorithms, when combined with ETSW, do fulfill better throughput assurance and fairness, especially under the case that AS flows require different target rates, incur different round trip times, or co-exist with non-responsive UDP flows.