Understanding human patterns-of-life (PoL) is essential towards ensuring safe and secure indoor facility environment as well as outdoor urban environment. Prediction of human movement in between places of interest is vital in understanding human PoL. Movement between spaces maybe represented and detected in one of the two forms: 1) trajectories: locations measured at regular time intervals by mobile sensors, bluetooth or GPS sensors; or 2) stay transitions: semantic PoI (points of interest) and stay duration data measurable by eventbased sensors that collect data when a check-in or check-out event is detected. Stay transition data provides a more compressed data format compared to trajectories data, especially in situations with longer stay durations, while preserving the information necessary for PoL analysis. Now as introduced briefly in the paper, our deployed end application (Digital Twin of a facility with non-player characters, besides the interactive user in virtual reality) needed a well-performing and validated AI/ML model for simulating high quality stay transitions behavior. In this study we thus primarily present our findings with developing and validating that model, which is a multi-task neural network for stay transition prediction. The neural network consists of two heads, for corresponding two tasks of stay category prediction and stay duration prediction. We evaluated gated recurrent units and multi-layer perceptrons of varying network sizes for stay category prediction; while mixture density networks, noisy generator-only networks, and generative adversarial networks of varying network sizes for stay duration prediction. We have then evaluated four multi-task models, constructed by combining these specialized models, on their ability to predict stay transition data. We tested our models on datasets from two different cases: 1) a simulation-generated dataset of indoor movement within the HFIR (high flux isotope reactor) nuclear reactor facility at Oak Ridge National Laboratory (ORNL); and 2) the GeoLife human mobility dataset of outdoor urban movement available in literature. Our results indicate that GRUMDN, which combines gated recurrent units (GRU) for stay category prediction task, and mixture density networks (MDN) for stay duration prediction task, did overall outperform other multitask models and the current state-of-the-art.

Spatiotemporal mobility data are available in massive quantities, but large quantities of data typically include fewer variables or data fields. Often, the only available fields are User ID, Longitude, Latitude, Timestamp ($U L L T$). This raises an important question: how much can we infer about human mobility patterns using only these four fields? With$U L L T$data, we do not know individuals' socioeconomic status information or when they are visiting their anchor points (AP) or locations (such as homes, places of employment, or schools), and it is a modern challenge to use this data to infer these characteristics. When detecting anchor locations with limited input information, verification and validation ($\mathrm{V} \quad \mathrm{V}$) are significant challenges. This paper addresses the problem of identifying individuals' anchor locations using only temporal information from spatiotemporal datasets with limited attributes. Our approach does not explicitly use latitude and longitude during analysis. Locationbased information is only employed in the preprocessing stage to identify periods of movement (trips) and stops (dwelling). Beyond this step, all analysis is based on temporal patterns. In theory, if stops and dwell times could be detected through alternative means, our method could function entirely without location-based input. We demonstrate this methodology on the 2017 National Household Travel Survey (NHTS) data, because it includes a carefully designed and collected time use survey with representative sampling and labeled ground truth. The high-quality survey data allows us to test the accuracy of our methods because NHTS contains intended place labels and agent/user characteristics. We have also applied our validated AP identification algorithm on very large-scale GPS based trajectory data for Patterns-of-Life (PoL) assessment and other applications, but due to space limit that could not be presented here.

Understanding, analyzing, and predicting human mobility and dynamics are valuable to solving pressing problems, developing effective plans, and prescribing timely remedies. As a computational approach, realistic human mobility simulations allow us to understand, analyze, and predict complex systems, including human societies. Accurate simulations rely on (1) the model that captures interactions and behaviors of myriad entities in our society and (2) the mapping of model instances to real-world entities. Taking this into account, this paper introduces the Human Mobility Network simulation framework (HumoNet), an integrated patterns of life (POL) simulation framework that leverages real-world data layers including transportation networks, points of interest, populations, popularity, and human trajectories. HumoNet is a data informed model in which agents are equipped with activities, locomotion, and planning capabilities. To simulate realistic kinematic maneuvers of individuals in transportation networks, HumoNet harnesses a microscopic traffic simulator that provides interaction among vehicles and traffic objects. In this paper, we describe the framework, outline our methodologies, and discuss the data processing and challenges of each data layer. Through experiments, we demonstrate that our simulations capture key features of human mobility by comparing them to the literature and real data using standard measures of human mobility (i.e., the radius of gyration, number of locations visited, level of exploration) and metrics scoring (i.e., Jensen-Shannon divergence). We envision that the synthetic data produced by HumoNet will serve as a benchmark for analyzing epidemics, deploying EV charging networks, and validating AI/ML tasks such as location prediction.

With the significant increase in sources and volume of human mobility data through commercial data vendors as well as microsimulation of cities, the scale of geospatial-temporal data to analyze and assess for mobility characterization has grown to the level of Big Data. There are mobility related commercial organizations deploying scalable computing, but often the system architecture, workflow, and intermediate processing components are not fully disclosed in relevant scope. Current research literature has a notable lack of studies demonstrating architectures and workflows for human mobility analytics that are implemented on a TeraByte scale of geospatial-temporal data. In this context, this paper presents a hyperscale-level system solution named DICER (Data Intensive Computing Environment and Runtime) for processing and analytics of geospatial-temporal data at big data scale. Although the cluster computing architecture of DICER with Apache Spark job running on Kubernetes cluster is not new, there are innovations in the workflow, hierarchical processing logic, and a wide range of intermediate preprocessing and mobility metrics calculation. We have performed case studies to validate the effectiveness of DICER system solution by performing detailed analytics and assessment of human mobility microsimulation output at three different scopes and scale, including a usecase with 16.97 TeraByte and 259.2 Billion rows of data. In addition, we have presented another case study of utilizing DICER to perform the same mobility processing and comparative analytics on large-scale commercially available geospatial-temporal data. All these case studies validate the efficiency and usefulness of DICER in computing population mobility characteristics from geospatial-temporal trajectory data at an unprecedented scale (not only just data volume, but also combination of: number of user entities, temporal frequency, spatial resolution, data duration).

Insider threats are perhaps the most serious challenges that nuclear and radiological security systems face. Insiders pose such a great threat due to their access, authority, and knowledge, granting them opportunities to bypass dedicated nuclear and radiological security elements. For example, in one of the latest major insider threat incidents to nuclear security, the Doel-4 nuclear powerplant in Belgium suffered a shutdown, the threat of nuclear materials diversion, and long-term loss of tens of millions of dollars. Seven years of investigation concluded that it was an inside job and attempted sabotage. In this regard, there is an immediate need for R&D and technology integration in the domain of modeling indoor Patterns-of-Life (PoL) and anomaly detection. This can be achieved by using datasets of facility users’ mobility and activity, which can support the design of algorithms for insider threat modeling and detection. However, due to classification, privacy, sensitivity, and safety protocols, such datasets from real physical nuclear reactor facilities are not only hard to share, but also not always feasible to deploy and collect. Aiming to find an alternate solution, our proposed demonstration work - MetaPoL, is the first-ever (for the application space) immersive VR (virtual reality) environment of a real-world secure facility and allows users to move-and-stay through the designed indoor physical layout and also encounter NPCs (non-player characters) that emulate other facility users. In the MetaPoL an interactive user performs realistic spatio-temporal movement, dwelling and activities using a Meta Quest Pro VR headset, and that generates high-frequency (in time) high-resolution (in space) indoor spatial-temporal datasets that are valuable for PoL modeling and anomaly detection research specifically for insider threat modeling and detection mission. Such generated realistic, rich in context, and mission specific datasets can boost AI/Machine Learning based research for modeling and detecting insider threats in nuclear security and nonproliferation.

Relief and recovery from disasters (both natural and human-made) require a coordinated approach across several federal and state government agencies. In order to achieve optimal resource allocation and deployment of first responders, accurate and timely assessment of the impact and extent of destruction are the cornerstones to any recovery effort. Ideally, this knowledge should be gathered and shared within the first 0-24 hours (termed as "Acute Phase" by the U.S. CDC guideline) for informed decision-making. But achieving this poses significant challenges for the data collection and data harmonization processes, particularly when voluminous data are being generated from diverse and distributed sources during the disaster responses. To this end, this work developed a scalable and efficient workflow to dynamically collect and harmonize crowd-sourced geographic multi-modal data, and then assess critical infrastructure (CI) damaged during disaster events. We demonstrate the application of our framework with two real-world experiences in addressing post-disaster recovery efforts - for the Bahamas (Natural - due to Hurricane Dorian, 2019) and Beirut (Human-made - due to explosion caused by the ammonium nitrate stored in a warehouse, 2020). We have illustrated that a coordinated effort is needed for planning as well as for execution to achieve informed decision making.

It is time-critical to administer pre-clinical cognitive health screening of patients to detect dementia and differentiate from mild cognitive impairment (MCI), and ensure early intervention, treatment and caregiving plans. Current medical standard and practice for pre-clinical screening follows standardized cognitive questionnaire test and patients’ verbal response based scoring. However, such practice misses sensing valuable signals of non-verbal stress response relevant to deciding dementia vs. MCI. To this end, we propose a novel approach, Alz-Sense, that integrates non-invasive passive sensing (during the standard cognitive questionnaire test) with an intelligent algorithm based assessment of dementia vs. MCI. The contributions of our Alz-Sense approach are: (i) innovative non-invasive passive sensing of non-verbal stress response from a custom made Smart Chair Cover; (ii) identifying and quantifying a novel indicator of stress response from chair cover multi-sensors data; and (iii) intelligent integration of stress indicator into a revised scoring mechanism. We deployed the Alz-Sense system for patients study in a hospital clinic, and validated its performance through 50 patients dataset. We compare the ROC (Receiver Operating Characteristic) performance of Alz-Sense approach with widely used standardized SLUMS questionnaire based scoring. We also compare performance in the contextual region of ROC curve relevant to pre-clinical cognitive health screening to show advantages of our approach. Further analysis selects optimal model parameters and compares SLUMS performance using medical clinic recommended threshold.

Current state-of-the-art systems in the literature using wearables are not capable of distinguishing a large number of fine-grained and/or complex human activities, which may appear similar but with vital differences in context, such as lying on floor versus lying on bed versus lying on sofa. This paper fills the gap by proposing a novel system, called HuMAn, that recognizes and classifies complex at-home activities of humans with wearable sensing. Specifically, HuMAn makes such classifications feasible by leveraging selective multi-modal sensor suites from wearable devices, and enhances the richness of sensed information for activity classification by carefully leveraging placement of the wearable devices across multiple positions on the human body. The HuMAn system consists of the following components: (a) a practical feature set extraction method from selected multi-modal sensor suites; and (b) a novel two-level structured classification algorithm that improves accuracy by leveraging sensors in multiple body positions; and (c) improved refinement in classification of complex activities with minimal external infrastructure support (e.g., only a few Bluetooth beacons used for location context). The proposed system is evaluated with 10 users in real home environments. Experimental results demonstrate that the HuMAn system can detect 21 complex at-home activities with high degree of accuracy. For same-user evaluation strategy, the average activity classification accuracy is as high as 95 percent over all of the 21 activities. For the case of 10-fold cross-validation evaluation strategy, the average classification accuracy is 92 percent, and for the case of leave-one-out cross-validation strategy, the average classification accuracy is 75 percent.

A large majority of population spend substantial amount of time each day in sedentary positions, which make the chairs potentially rich source of information and insight about people daily activities and behavior patterns. These information, which often gets under-utilized, can reveal valuable knowledge about the users's wellness level and quality of life. Considering this we have designed Care-Chair, a simple and cost effective smart sensing system with just four pressure sensors on the backrest of a chair, equipped with intelligent data analytics. Our proposed Care-Chair system has been able to classify among large number of 19 fine-grained and complex user sedentary activities. To best of our knowledge this is the first work to detect user functional activities and user emotion based activities (in addition to static and movement based sedentary activities and postures) with just 4 pressure sensors on chair and sensor data analytics. The performance is validated with 5 users combined dataset (thus considering user specific variations) with 86% overall accuracy. Our system is also validated to achieve precise measurement in user breathing rate during relatively static sedentary postures.

In this work we present A-Wristocracy, a novel framework for recognizing very fine-grained and complex inhome activities of human users (particularly elderly people) with wrist-worn device sensing. Our designed A-Wristocracy system improves upon the state-of-the-art works on in-home activity recognition using wearables. These works are mostly able to detect coarse-grained ADLs (Activities of Daily Living) but not large number of fine-grained and complex IADLs (Instrumental Activities of Daily Living). These are also not able to distinguish similar activities but with different context (such as sit on floor vs. sit on bed vs. sit on sofa). Our solution helps accurate detection of in-home ADLs/ IADLs and contextual activities, which are all critically important for remote elderly care in tracking their physical and cognitive capabilities. A-Wristocracy makes it feasible to classify large number of fine-grained and complex activities, through Deep Learning based data analytics and exploiting multi-modal sensing on wrist-worn device. It exploits minimal functionality from very light additional infrastructure (through only few Bluetooth beacons), for coarse level location context. A-Wristocracy preserves direct user privacy by excluding camera/ video imaging on wearable or infrastructure. The classification procedure consists of practical feature set extraction from multi-modal wearable sensor suites, followed by Deep Learning based supervised fine-level classification algorithm. We have collected exhaustive home-based ADLs and IADLs data from multiple users. Our designed classifier is validated to be able to recognize very fine-grained complex 22 daily activities (much larger number than 6-12 activities detected by state-of-the-art works using wearable and no camera/ video) with high average test accuracies of 90% or more for two users in two different home environments.

Sensor networks have particularly important applications in challenging environments. However, those challenging environments also pose significant challenges to network sustainability and reliability. In such environments, the network often becomes disruptive and even unavailable during downtime. This results in undesired loss of valuable spatial-temporal sensor data. Data persistence can be achieved by using in-situ encoding and caching of data through distributed mechanisms. However, the existing methods in the literature are mainly based on network random walks, which not only incur significant communication overhead, but also are prone to network or node failures. In this article, we presentECPC, a distributedErasureCoding with randomizedPowerControl protocol for preserving data in disruptive sensor networks.ECPConly requires each sensor node to perform several rounds of broadcast in its neighborhood at some randomly chosen radio transmission power levels, and thus it incurs low communication overhead. We proved thatECPCachieves the expected code degree distribution and pseudo-global randomness of erasure coding principles. We have also evaluated the performance ofECPCby comparing it with several key related approaches in the literature (such as EDFC and RCDS). The performance comparisons validate that our proposedECPCprotocol can reach higher data reliability under varying node failure probabilities. In addition,ECPCprotocol is also shown to be scalable with different network sizes.

In this paper we have proposed and designed FindingHuMo (Finding Human Motion), a real-time user tracking system for Smart Environments. FindingHuMo can perform device-free tracking of multiple (unknown and variable number of) users in the Hallway Environments, just from non-invasive and anonymous (not user specific) binary motion sensor data stream. The significance of our designed system are as follows: (a) fast tracking of individual targets from binary motion data stream from a static wireless sensor network in the infrastructure. This needs to resolve unreliable node sequences, system noise and path ambiguity, (b) Scaling for multi-user tracking where user motion trajectories may crossover with each other in all possible ways. This needs to resolve path ambiguity to isolate overlapping trajectories, FindingHumo applies the following techniques on the collected motion data stream: (i) a proposed motion data driven adaptive order Hidden Markov Model with Viterbi decoding (called Adaptive-HMM), and then (ii) an innovative path disambiguation algorithm (called CPDA). Using this methodology the system accurately detects and isolates motion trajectories of individual users. The system performance is illustrated with results from real-time system deployment experience in a Smart Environment.

A sensor network, unlike a traditional communication network, is deeply embedded in physical environments and its operation is mainly driven by the event activities in the environment. In long-term operations, the event activities usually show certain patterns that can be learned and exploited to optimize the network design. However, this has been underexplored in the literature. One work related to this is using Activity Transition Probability Graph (ATPG) for radio duty cycling [Tang et al. (2011) ActSee: Activity-Aware Radio Duty-Cycling for Sensor Networks in Smart Environments. Proc. IEEE INSS 2011, Penghu, Taiwan, June 12–15. IEEE Press]. In this paper, we present a novel Energy and Activity-aware Routing (EAR) protocol for sensor networks. As a case study, we have evaluated EAR with the data trace of real Smart Environments. In EAR an ATPG is learned and built from the event activity patterns. EAR is an online routing protocol, that chooses the next-hop relay node by utilizing: activity pattern information in the ATPG graph and a novel index of energy balance in the network. EAR extends the network lifetime by maintaining an energy balance across the nodes in the network, while meeting the application performance with desired throughput and low data delivery latency. We theoretically prove that: (i) the network throughput with EAR achieves a competitive ratio (i.e. the ratio of the performance of any offline algorithm that has knowledge of all past and future packet arrivals to the performance of our online algorithm) that is asymptotically optimal, and (ii) EAR achieves a lower bound in the network lifetime. Extensive experimental results from: (i) a 82 node Motelab sensor network testbed [Werner-Allen et al. (2005) MoteLab: A Wireless Sensor Network Testbed. Proc. ACM IPSN 2005, Los Angeles, CA, USA, April 25–27, pp. 483–488. IEEE Press, NJ, USA] and (ii) a varying size network (20–100) in sensor network simulator TOSSIM, validate that EAR outperforms the existing methods both in terms of network performance (network lifetime, network energy consumption) and application performance (low latency, desired throughput) for an energy-constrained sensor network.

A real-time monitor and rescue system must be able to both quickly and reliably detect the event happening in its monitoring region. Furthermore, it is required to fulfill certain rescue mission, e.g., navigate victims to exit through safe path in case of emergency. Current monitor and rescue approaches generally rely on either teleoperated robots, or teams of wireless robots. Typically the robots used in these systems tend to have high cost which make them unpractical in large scale deployment and applications. In this work, we present a realtime monitor and rescue system, TelosW-Bot Net, utilizing integrated networks. The integrated network is an integration of stationary sensor networks and robots: static sensor networks comprised of large numbers of small, simple, and inexpensive wireless sensors, and the robots which can communicate and controlled by sensor nodes. We demonstrate the efficacy of our system in real test bed composed of 46 sensors, which is one of the largest robotic sensor network to our knowledge, providing empirical results.

This paper present a wake-on sensor network formed with the wake-on motes, TelosW. Our wake-on hardware and software design enable lower power operations and longer network lifetime.

In near future, wireless sensors networks (WSN) are expected to be deployed for a vast variety of applications. For many of these applications, sensed data individually have no meaning without proper knowledge of position where the data was sensed. Although there have been proposed algorithms for localization, many of them fall short with respect to complexity, cost, security and accuracy. Some issues that still hinder the efficiency of available localization techniques are multi-path propagation, security attack (internal and external attack by faulty or malicious nodes), and sometimes specific network topologies. For example, the Long-Thin (LT) topology of WSN is highly prone to localization error due to its special distribution pattern. In many real world WSN applications, the localization is also needed to be Global Positioning Systems (GPS) free. Apart from efficient localization, another important operation in WSN application is the need to have in-network detection of faulty sensor readings, without compromising detection of important events. Looking at these critically important operations of WSN, our first contribution is proposing a self-organized distributed GPS free localization error detection and correction algorithm. Our next contribution is utilizing reliability gained in the proposed localization to construct an improved algorithm for in-network detection of faulty readings. We have also proposed a fault tolerant structure for Long-Thin (LT) WSN topology, that can well utilize both of the proposed algorithms.

The advances in multicore technology and modern interconnects is rapidly accelerating the number of cores deployed in today's commodity clusters. A majority of parallel applications written in MPI employ collective operations in their communication kernels. Optimization of these operations on the multicore platforms is the key to obtaining good performance speed-ups. However, designing these operations on the modern multicores is a non-trivial task. Modern multicores such as Intel's Clovertown and AMD's Opteron feature various architectural attributes resulting in interesting ramifications. For example, Clovertown deploys shared 12 caches for a pair of cores whereas in Opteron, 12 caches are exclusive to a core. Understanding the impact of these architectures on communication performance is crucial to designing efficient collective algorithms. In this paper, we systematically evaluate these architectures and use these insights to develop efficient collective operations such as MPI_Bcast, MPI_Allgather, MPI_Allreduce and MPI_Alltoall. Further, we characterize the behavior of these collective algorithms on multicores especially when concurrent network and intra-node communications occur. We also evaluate the benefits of the proposed intra-node MPI_Allreduce over Opteron multicores and compare it with Intel Clovertown systems. The optimizations proposed in this paper reduce the latency of MPI-Bcast and MPI_Allgather by 1.9 and 4.0 times, respectively on 512 cores. For MPI_Allreduce, our optimizations improve the performance by as much as 33% on the multicores. Further, we observe up to three times improvement in performance for matrix multiplication benchmark on 512 cores.