Convenient tracking of food intake is essential for linking diet to health, enabling personalised nutrition guidance, early metabolic risk detection, and prevention of chronic disease. Recent wearable sensing advances have begun to automate eating monitoring. However, these systems largely focus on detecting when users eat and only weakly address what they eat. In particular, state-of-the-art solutions typically cover only a narrow range of foods or textures and rely on strong assumptions about individual eating behaviour. Moreover, they overlook the nutritional implications most relevant to end users, limiting their usefulness for real-world dietary guidance. In this paper, we present NutriEar, an in-ear audio sensing system for nutrition-aware classification of food intake from chewing sounds. Rather than recognising arbitrary food types, NutriEar maps in-ear acoustics to an eight-class nutrition-texture taxonomy grounded in food science, capturing both dominant macronutrient role and mechanical texture. NutriEar records in-ear audio during eating, segments chewing events, and derives a hybrid representation combining engineered acoustic features with learned embeddings from supervised contrastive learning, enabling a compact nutrition-aware classification pipeline. Evaluation on a dataset collected from 15 users consuming over 30 food types under varied eating conditions shows that NutriEar achieves 80.18% average leave-one-subject-out (LOSO) accuracy and outperforms state-of-the-art baselines. These results highlight the untapped potential of earable audio sensing as a practical pathway toward everyday dietary monitoring with meaningful nutritional insights.

Pervasive sensing enables diverse wearable event detection (WED) applications, but deploying machine learning models on resource-constrained microcontrollers (MCUs) poses significant challenges, particularly in ensuring prediction reliability under data shifts or out-of-distribution (OOD) inputs. While Uncertainty quantification methods offer a way to assess this reliability, many are computationally prohibitive for MCUs, and detecting multiple events concurrently further exacerbates resource constraints. Addressing these combined challenges, this paper presents an uncertainty and resource-aware framework designed for reliable and efficient multi-event WED on MCUs, significantly extending our preliminary work. The proposed framework achieves this by integrating Evidential Deep Learning (EDL) for efficient, single-pass uncertainty estimation with a novel cascade learning architecture. This architecture promotes resource efficiency via: (i) intra-event sharing using uncertainty-aware early exits within a staged model (shallow, medium, deep), allowing simpler samples to terminate inference earlier; and (ii) inter-event sharing using a multi-head design where multiple event detectors share a common backbone, minimizing overhead. System efficiency is further enhanced through MCU-specific optimizations, including targeted architecture search, quantization, efficient uncertainty operator implementation using standard TensorFlow Lite Micro (TFLM) operations, and library footprint reduction. We conducted extensive experiments on four distinct wearable datasets (Oesense, KWS, ECG5000, and HHAR) and two MCU platforms (STM32F446ZE, STM32H747XI), comparing the proposed framework against strong baselines including Deep Ensembles and Vanilla EDL. Results demonstrate the proposed framework’s effectiveness, achieving competitive accuracy and uncertainty performance (e.g., up to 22% lower NLL than data augmentation) while drastically reducing resource consumption, offering up to 8.64 × faster inference, up to 8.57 × lower energy use, and 55% smaller memory footprint compared to ensemble methods. The proposed framework enables the deployment of reliable, uncertainty-aware multi-event detection on a wider range of low-power MCUs.

Electrocardiogram (ECG) recording systems are increasingly being integrated into consumer wearable systems such as smartwatches, providing users with access to clinically-relevant information about their heart activity anytime, anywhere. The increasing adoption of in-ear wearables, known as earables, as well as their stable position on the body, makes them an attractive prospect for ECG integration. However, this comes with several challenges. Other biosignals, including those from the brain and surrounding muscles, are detectable at the ear in the same frequency bands with much higher amplitudes. This means that the ECG signal-to-noise ratio (SNR) can be extremely low at this location. The few existing denoising approaches mostly rely on autoencoders. In some cases they fail to recover the ECG morphology, and their black-box nature does not allow for explainability or understanding of limitations. To address these issues, we introduce a novel system to record and denoise ear-ECG signals, leveraging open-source hardware and the Extended Kalman Filter. In-ear audio recording of heart sounds is used to accurately determine timings of cardiac cycles. From these timings, a short-term ensemble average ECG signal is calculated, which is used to fit the parameters of a dynamical ECG model to an individual user. The Kalman filter is then applied to the full time series ECG for denoising, using the dynamical model for its state prediction steps, and heart sounds as phase measurements. We have evaluated the system with data collected from 18 participants. The results report a mean SNR of 6.4 dB, mean absolute QT interval error of 54 ms, and heart rate error of 3 BPM, demonstrating the system's potential for continuous, non-invasive, user-friendly ECG monitoring.

Hypertension, a major risk factor for cardiovascular diseases, often goes undetected due to its asymptomatic nature. This study explores a novel approach to detecting elevated blood pressure using heart sounds, aiming to provide a non-invasive, potentially continuous monitoring solution. We evaluated our approach on a new dataset of 260 participants, employing patient-independent cross-validation to ensure generalisability. Our methodology utilises a convolutional neural network-based Hidden Semi-Markov Model for heart sound segmentation, followed by extraction of hand-crafted amplitude, duration, and frequency features. A random forest model was implemented for the binary classification of hypertension, achieving a promising 70% accuracy with 72% sensitivity. We conducted comprehensive analyses, including auscultation location and feature importance evaluation, and the investigation of the heart rate – blood pressure relationship. Our findings demonstrate the feasibility of this approach, providing a robust foundation for further research and development in this domain.

Interpreting electrocardiograms (ECGs) and generating comprehensive reports remain challenging tasks in cardiology, often requiring specialized expertise and significant time investment. To address these critical issues, we propose ECG-ReGen, a retrieval-based approach for ECG-to-text report generation and question answering. Our method leverages self-supervised learning for the ECG encoder, enabling efficient similarity searches and report retrieval. By combining pre-training with dynamic retrieval and Large Language Model (LLM)-based refinement, ECG-ReGen effectively analyzes ECG data and answers related queries, with the potential of improving patient care. Experiments conducted on the PTB-XL and MIMIC-IV-ECG datasets demonstrate superior performance in both in-domain and cross-domain scenarios for report generation. Furthermore, our approach exhibits competitive performance on the ECG-QA dataset compared to fully supervised methods when utilizing off-the-shelf LLMs for zero-shot question answering. This approach, effectively combining self-supervised encoder and LLMs, offers a scalable and efficient solution for accurate ECG interpretation, holding significant potential to enhance clinical decision-making.

We present SensorLM, a family of sensor-language foundation models that enable wearable sensor data understanding with natural language. Despite its pervasive nature, aligning and interpreting sensor data with language remains challenging due to the lack of paired, richly annotated sensor-text descriptions in uncurated, real-world wearable data. We introduce a hierarchical caption generation pipeline designed to capture statistical, structural, and semantic information from sensor data. This approach enabled the curation of the largest sensor-language dataset to date, comprising over 59.7 million hours of data from more than 103,000 people. Furthermore, SensorLM extends prominent multimodal pretraining architectures (e.g., CLIP, CoCa) and recovers them as specific variants within a generic architecture. Extensive experiments on real-world tasks in human activity analysis and healthcare verify the superior performance of SensorLM over state-of-the-art in zero-shot recognition, few-shot learning, and cross-modal retrieval. SensorLM also demonstrates intriguing capabilities including scaling behaviors, label efficiency, sensor captioning, and zero-shot generalization to unseen tasks. Code is available at https://github.com/Google-Health/consumer-health-research/tree/main/sensorlm.

Continuous respiratory rate (RR) monitoring is essential for understanding physical and mental health, as well as tracking fitness. However, performing reliable and non-obtrusive RR monitoring across diverse daily routines and activities is still an open research problem. In this work, we present RespEar, a pipeline for robust RR monitoring across various sedentary and active scenarios using earphones. RespEar relies solely on in-ear microphones, repurposing them for continuous RR monitoring purposes. Specifically, leveraging the unique properties of in-ear audio, RespEar enables the use of respiratory sinus arrhythmia (RSA) and locomotor respiratory coupling (LRC), physiological couplings between cardiovascular activity, gait and respiration, to determine the RR. This effectively addresses the challenges posed by the almost imperceptible breathing signals encountered during common daily activities. Additionally, RespEar uniquely identifies and addresses three key practical issues for the RSA and LRC-based solutions and introduces a suite of meticulously crafted signal processing techniques to enhance the accuracy of RR measurements. With data collected from 18 subjects over 8 activities, RespEar measures RR with a mean absolute error (MAE) of 1.48 breaths per minute (BPM) and a mean absolute percent error (MAPE) of 9.12% in sedentary conditions, and a MAE of 2.28 BPM and a MAPE of 11.04% in active conditions, respectively. To the best of our knowledge, RespEar is the first earable-based system capable of accurately determining RR in a variety of realistic settings.

Gait behaviour is a key health metric. Temporal, spatial and kinetic walking gait parameters are valuable in enhancing sport performance and early health diagnostics Full gait assessment requires a gait clinic and existing wearable gait tracking systems typically measure isolated subsets of parameters tailored to specific applications. This is useful when the condition to be monitored is known, but fails to offer a comprehensive view of an individual’s gait traits when their pathology is unknown or changing, or a general assessment is required. To support holistic walking gait tracking, we introduce WalkEar, a novel sensing platform designed to simultaneously track gait parameters using commodity earbuds. WalkEar operates by detecting gait events to derive temporal gait parameters and segment the IMU data. WalkEar then progresses earable gait assessment by, for the first time, estimating kinetic gait parameters and reconstructing the vGRF curve using machine learning. Each parameter is calculated on a step-to-step basis for gait variability and asymmetry. We developed an earbud prototype and collected data from 13 participants using gold standard force plates and instrumented treadmill ground truth. Extensive experiments demonstrate the promising performance of WalkEar, achieving an overall MAPE of 5.1% in estimating gait, 2.0% MAPE on kinetic gait parameters, and an NRMSE of 5.3% for vGRF curve reconstruction.

Machine Learning models typically assume that time series are regularly spaced; however, this is often unrealistic in healthcare, where missing data recordings are common. In this context, uncertainty estimates play a pivotal role, as they can enable confident and non-confident predictions to be distinguished. We propose SQUIREDL, a novel uncertainty-aware sequence-to-sequence prediction method for sparse healthcare time series. Specifically, we enhance the state-of-the-art evidential regression framework, widely used for uncertainty estimation, to handle missing data. Following data imputation with an Akima spline-based method, we modify the loss function of evidential regression by assigning different weights to imputed and observed data points, to offer more reliable uncertainty estimates. Additionally, we examine a variety of metrics for assessing the success of uncertainty estimations on sequence-to-sequence predictions, providing a reliable way to evaluate the models in a medical setting. Our proposal is demonstrated in two clinical applications. In continuous glucose monitoring, we use sequence-to-sequence prediction to obtain the hypoglycaemia risk from glucose sensor readings. Our approach captures the ground truth risk values 30% more accurately, bringing consistent improvements in both uncertainty-aware and accuracy-based metrics. Similarly, in COVID-19 hospital admissions data, we achieve a 22% improvement in the accuracy of uncertainty-aware predictions, enabling better resource planning.

Contrastive learning (CL) has emerged as a promising approach for representation learning in time series data by embedding similar pairs closely while distancing dissimilar ones. However, existing CL methods often introduce false negative pairs (FNPs) by neglecting inherent characteristics and then randomly selecting distinct segments as dissimilar pairs, leading to erroneous representation learning, reduced model performance, and overall inefficiency. To address these issues, we systematically define and categorize FNPs in time series into semantic false negative pairs and temporal false negative pairs for the first time: the former arising from overlooking similarities in label categories, which correlates with similarities in non-stationarity and the latter from neglecting temporal proximity. Moreover, we introduce StatioCL, a novel CL framework that captures non-stationarity and temporal dependency to mitigate both FNPs and rectify the inaccuracies in learned representations. By interpreting and differentiating non-stationary states, which reflect the correlation between trends or temporal dynamics with underlying data patterns, StatioCL effectively captures the semantic characteristics and eliminates semantic FNPs. Simultaneously, StatioCL establishes fine-grained similarity levels based on temporal dependencies to capture varying temporal proximity between segments and to mitigate temporal FNPs. Evaluated on real-world benchmark time series classification datasets, StatioCL demonstrates a substantial improvement over state-of-the-art CL methods, achieving a 2.9% increase in Recall and a 19.2% reduction in FNPs. Most importantly, StatioCL also shows enhanced data efficiency and robustness against label scarcity.

On-device training is essential for user personalisation and privacy. With the pervasiveness of IoT devices and microcontroller units (MCUs), this task becomes more challenging due to the constrained memory and compute resources, and the limited availability of labelled user data. Nonetheless, prior works neglect the data scarcity issue, require excessively long training time ($\textit{e.g.}$ a few hours), or induce substantial accuracy loss ($\geq$10%). In this paper, we propose TinyTrain, an on-device training approach that drastically reduces training time by selectively updating parts of the model and explicitly coping with data scarcity. TinyTrain introduces a task-adaptive sparse-update method that $\textit{dynamically}$ selects the layer/channel to update based on a multi-objective criterion that jointly captures user data, the memory, and the compute capabilities of the target device, leading to high accuracy on unseen tasks with reduced computation and memory footprint. TinyTrain outperforms vanilla fine-tuning of the entire network by 3.6-5.0% in accuracy, while reducing the backward-pass memory and computation cost by up to 1,098$\times$ and 7.68$\times$, respectively. Targeting broadly used real-world edge devices, TinyTrain achieves 9.5$\times$ faster and 3.5$\times$ more energy-efficient training over status-quo approaches, and 2.23$\times$ smaller memory footprint than SOTA methods, while remaining within the 1 MB memory envelope of MCU-grade platforms.

Federated Learning (FL) enables model development by leveraging data distributed across numerous edge devices without transferring local data to a central server. However, existing FL methods still face challenges when dealing with scarce and label-skewed data across devices, resulting in local model overfitting and drift, consequently hindering the performance of the global model. In response to these challenges, we propose a pioneering framework called FLea, incorporating the following key components: i) A global feature buffer that stores activation-target pairs shared from multiple clients to support local training. This design mitigates local model drift caused by the absence of certain classes; ii) A feature augmentation approach based on local and global activation mix-ups for local training. This strategy enlarges the training samples, thereby reducing the risk of local overfitting; iii) An obfuscation method to minimize the correlation between intermediate activations and the source data, enhancing the privacy of shared features. To verify the superiority of FLea, we conduct extensive experiments using a wide range of data modalities, simulating different levels of local data scarcity and label skew. The results demonstrate that FLea consistently outperforms state-of-the-art FL counterparts (among 13 of the experimented 18 settings, the improvement is over 5%) while concurrently mitigating the privacy vulnerabilities associated with shared features.

The increased adoption of Internet of Things (IoT) devices has led to the generation of large data streams with applications in healthcare, sustainability, and robotics. In some cases, deep neural networks have been deployed directly on these resource-constrained units to limit communication overhead, increase efficiency and privacy, and enable real-time applications. However, a common challenge in this setting is the continuous adaptation of models necessary to accommodate changing environments, i.e., data distribution shifts. Test-time adaptation (TTA) has emerged as one potential solution, but its validity has yet to be explored in resource-constrained hardware settings, such as those involving microcontroller units (MCUs). TTA on constrained devices generally suffers from i) memory overhead due to the full backpropagation of a large pre-trained network, ii) lack of support for normalization layers on MCUs, and iii) either memory exhaustion with large batch sizes required for updating or poor performance with small batch sizes. In this paper, we propose TinyTTA, to enable, for the first time, efficient TTA on constrained devices with limited memory. To address the limited memory constraints, we introduce a novel self-ensemble and batch-agnostic early-exit strategy for TTA, which enables continuous adaptation with small batch sizes for reduced memory usage, handles distribution shifts, and improves latency efficiency. Moreover, we develop the TinyTTA Engine, a first-of-its-kind MCU library that enables on-device TTA. We validate TinyTTA on a Raspberry Pi Zero 2W and an STM32H747 MCU. Experimental results demonstrate that TinyTTA improves TTA accuracy by up to 57.6\%, reduces memory usage by up to six times, and achieves faster and more energy-efficient TTA. Notably, TinyTTA is the only framework able to run TTA on MCU STM32H747 with a 512 KB memory constraint while maintaining high performance.

Respiratory audio, such as coughing and breathing sounds, has predictive power for a wide range of healthcare applications, yet is currently under-explored. The main problem for those applications arises from the difficulty in collecting large labeled task-specific data for model development. Generalizable respiratory acoustic foundation models pretrained with unlabeled data would offer appealing advantages and possibly unlock this impasse. However, given the safety-critical nature of healthcare applications, it is pivotal to also ensure openness and replicability for any proposed foundation model solution. To this end, we introduce OPERA, an OPEn Respiratory Acoustic foundation model pretraining and benchmarking system, as the first approach answering this need. We curate large-scale respiratory audio datasets ($\sim$136K samples, over 400 hours), pretrain three pioneering foundation models, and build a benchmark consisting of 19 downstream respiratory health tasks for evaluation. Our pretrained models demonstrate superior performance (against existing acoustic models pretrained with general audio on 16 out of 19 tasks) and generalizability (to unseen datasets and new respiratory audio modalities). This highlights the great promise of respiratory acoustic foundation models and encourages more studies using OPERA as an open resource to accelerate research on respiratory audio for health. The system is accessible from https://github.com/evelyn0414/OPERA.

Traditional machine learning techniques are prone to generating inaccurate predictions when confronted with shifts in the distribution of data between the training and testing phases. This vulnerability can lead to severe consequences, especially in applications such as mobile healthcare. Uncertainty estimation has the potential to mitigate this issue by assessing the reliability of a model's output. However, existing uncertainty estimation techniques often require substantial computational resources and memory, making them impractical for implementation on microcontrollers (MCUs). This limitation hinders the feasibility of many important on-device wearable event detection (WED) applications, such as heart attack detection. In this paper, we present UR2M, a novel Uncertainty and Resource-aware event detection framework for MCUs. Specifically, we (i) develop an uncertainty-aware WED based on evidential theory for accurate event detection and reliable uncertainty estimation; (ii) introduce a cascade ML framework to achieve efficient model inference via early exits, by sharing shallower model layers among different event models; (iii) optimize the deployment of the model and MCU library for system efficiency. We conducted extensive experiments and compared UR2M to traditional uncertainty baselines using three wearable datasets. Our results demonstrate that UR2M achieves up to 864% faster inference speed, 857% energy-saving for uncertainty estimation, 55% memory saving on two popular MCUs, and a 22% improvement in uncertainty quantification performance. UR2M can be deployed on a wide range of MCUs, significantly expanding real-time and reliable WED applications.

Self-supervised learning (SSL) has shown remarkable performance in computer vision tasks when trained offline. However, in a Continual Learning (CL) scenario where new data is introduced progressively, models still suffer from catastrophic forgetting. Retraining a model from scratch to adapt to newly generated data is time-consuming and inefficient. Previous approaches suggested re-purposing self-supervised objectives with knowledge distillation to mitigate forgetting across tasks, assuming that labels from all tasks are available during fine-tuning. In this paper, we generalize self-supervised continual learning in a practical setting where available labels can be leveraged in any step of the SSL process. With an increasing number of continual tasks, this offers more flexibility in the pre-training and fine-tuning phases. With Kaizen1, we introduce a training architecture that is able to mitigate catastrophic forgetting for both the feature extractor and classifier with a carefully designed loss function. By using a set of comprehensive evaluation metrics reflecting different aspects of continual learning, we demonstrated that Kaizen significantly outperforms previous SSL models in competitive vision benchmarks, with up to 16.5% accuracy improvement on split CIFAR-100. Kaizen is able to balance the trade-off between knowledge retention and learning from new data with an end-to-end model, paving the way for practical deployment of continual learning systems.

With the soaring adoption of in-ear wearables, the research community has started investigating suitable in-ear heart rate detection systems. Heart rate is a key physiological marker of cardiovascular health and physical fitness. Continuous and reliable heart rate monitoring with wearable devices has therefore gained increasing attention in recent years. Existing heart rate detection systems in wearables mainly rely on photoplethysmography (PPG) sensors, however, these are notorious for poor performance in the presence of human motion. In this work, leveraging the occlusion effect that enhances low-frequency bone-conducted sounds in the ear canal, we investigate for the first time in-ear audio-based motion-resilient heart rate monitoring. We first collected heart rate-induced sounds in the ear canal using an in-ear microphone under seven stationary activities and two full-body motion activities (i.e., walking, and running). Then, we devised a novel deep learning based motion artefact (MA) mitigation framework to denoise the in-ear audio signals, followed by a heart rate estimation algorithm to extract heart rate. With data collected from 15 subjects over nine activities, we demonstrate that hEARt, our end-to-end approach, achieves a mean absolute error (MAE) of 1.88 ± 2.89 BPM, 6.83 ± 5.05 BPM, and 13.19 ± 11.37 BPM for stationary, walking, and running, respectively, opening the door to a new non-invasive and affordable heart rate monitoring with usable performance for daily activities. Not only does hEARt outperform previous in-ear heart rate monitoring work, but it outperforms reported in-ear PPG performance.

Uncertainty quantification is critical for ensuring the safety of deep learning-enabled health diagnostics, as it helps the model account for unknown factors and reduces the risk of misdiagnosis. However, existing uncertainty quantification studies often overlook the significant issue of class imbalance, which is common in medical data. In this paper, we propose a class-balanced evidential deep learning framework to achieve fair and reliable uncertainty estimates for health diagnostic models. This framework advances the state-of-the-art uncertainty quantification method of evidential deep learning with two novel mechanisms to address the challenges posed by class imbalance. Specifically, we introduce a pooling loss that enables the model to learn less biased evidence among classes and a learnable prior to regularize the posterior distribution that accounts for the quality of uncertainty estimates. Extensive experiments using benchmark data with varying degrees of imbalance and various naturally imbalanced health data demonstrate the effectiveness and superiority of our method. Our work pushes the envelope of uncertainty quantification from theoretical studies to realistic healthcare application scenarios. By enhancing uncertainty estimation for class-imbalanced data, we contribute to the development of more reliable and practical deep learning-enabled health diagnostic systems.

Abdominal sounds (ABS) have been traditionally used for assessing gastrointestinal (GI) disorders. However, the assessment requires a trained medical professional to perform multiple abdominal auscultation sessions, which is resource-intense and may fail to provide an accurate picture of patients’ continuous GI wellbeing. This has generated a technological interest in developing wearables for continuous capture of ABS, which enables a fuller picture of patient’s GI status to be obtained at reduced cost. This paper seeks to evaluate the feasibility of extracting heart rate (HR) from such ABS monitoring devices. The collection of HR directly from these devices would enable gathering vital signs alongside GI data without the need for additional wearable devices, providing further cost benefits and improving general usability. We utilised a dataset containing 104 hours of ABS audio, collected from the abdomen using an e-stethoscope, and electrocardiogram as ground truth. Our evaluation shows for the first time that we can successfully extract HR from audio collected from a wearable on the abdomen. As heart sounds collected from the abdomen suffer from significant noise from GI and respiratory tracts, we leverage wavelet denoising for improved heart beat detection. The mean absolute error of the algorithm for average HR is 3.4BPM with mean directional error of -1.2BPM over the whole dataset. A comparison to photoplethysmography-based wearable HR sensors shows that our approach exhibits comparable accuracy to consumer wrist-worn wearables for average and instantaneous heart rate.

Federated learning (FL) aided health diagnostic models can incorporate data from a large number of personal edge devices (e.g., mobile phones) while keeping the data local to the originating devices, largely ensuring privacy. However, such a cross-device FL approach for health diagnostics still imposes many challenges due to both local data imbalance (as extreme as local data consists of a single disease class) and global data imbalance (the disease prevalence is generally low in a population). Since the federated server has no access to data distribution information, it is not trivial to solve the imbalance issue towards an unbiased model. In this paper, we propose FedLoss, a novel cross-device FL framework for health diagnostics. Here the federated server averages the models trained on edge devices according to the predictive loss on the local data, rather than using only the number of samples as weights. As the predictive loss better quantifies the data distribution at a device, FedLoss alleviates the impact of data imbalance. Through a real-world dataset on respiratory sound and symptom-based COVID-19 detection task, we validate the superiority of FedLoss. It achieves competitive COVID-19 detection performance compared to a centralised model with an AUC-ROC of 79%. It also outperforms the state-of-the-art FL baselines in sensitivity and convergence speed. Our work not only demonstrates the promise of federated COVID-19 detection but also paves the way to a plethora of mobile health model development in a privacy-preserving fashion.

Training deep learning models on data distributed and local to edge devices such as mobile phones is a prominent recent research direction. In a Gossip Learning (GL) system, each participating device maintains a model trained on its local data and iteratively aggregates it with the models from its neighbours in a communication network. While the fully distributed operation in GL comes with natural advantages over the centralized orchestration in Federated Learning (FL), its convergence becomes particularly slow when the data distribution is heterogeneous and aligns with the clustered structure of the communication network. These characteristics are pervasive across practical applications as people with similar interests (thus producing similar data) tend to create communities. This paper proposes a data-driven neighbor weighting strategy for aggregating the models: this enables faster diffusion of knowledge across the communities in the network and leads to quicker convergence. We augment the method to make it computationally efficient and fair: the devices quickly converge to the same model. We evaluate our model on real and synthetic datasets that we generate using a novel generative model for communication networks with heterogeneous data. Our exhaustive empirical evaluation verifies that our proposed method attains a faster convergence rate than the baselines. For example, the median test accuracy for a decentralized bird image classifier application reaches 81% with our proposed method within 80 rounds, whereas the baseline only reaches 46%.

Time series forecasting, as one of the fundamental machine learning areas, has attracted tremendous attentions over recent years. The solutions have evolved from statistical machine learning (ML) methods to deep learning techniques. One emerging sub-field of time series forecasting is individual disease progression forecasting, e.g., predicting individuals' disease development over a few days (e.g., deteriorating trends, recovery speed) based on few past observations. Despite the promises in the existing ML techniques, a variety of unique challenges emerge for disease progression forecasting, such as irregularly-sampled time series, data sparsity, and individual heterogeneity in disease progression. To tackle these challenges, we propose novel Conditional Neural Ordinary Differential Equations Processes (CNDPs), and validate it in a COVID-19 disease progression forecasting task using audio data. CNDPs allow for irregularly-sampled time series modelling, enable accurate forecasting with sparse past observations, and achieve individual-level progression forecasting. CNDPs show strong performance with an Unweighted Average Recall (UAR) of 78.1%, outperforming a variety of commonly used Recurrent Neural Networks based models. With the proposed label-enhancing mechanism (i.e., including the initial health status as input) and the customised individual-level loss, CNDPs further boost the performance reaching a UAR of 93.6%. Additional analysis also reveals the model's capability in tracking individual-specific recovery trend, implying the potential usage of the model for remote disease progression monitoring. In general, CNDPs pave new pathways for time series forecasting, and provide considerable advantages for disease progression monitoring.

With the soaring adoption of in-ear wearables, the research community has started investigating suitable in-ear heart rate (HR) detection systems. HR is a key physiological marker of cardiovascular health and physical fitness. Continuous and reliable HR monitoring with wearable devices has therefore gained increasing attention in recent years. Existing HR detection systems in wearables mainly rely on photoplethysmography (PPG) sensors, however, these are notorious for poor performance in the presence of human motion. In this work, leveraging the occlusion effect that enhances low-frequency bone-conducted sounds in the ear canal, we investigate for the first time in-ear audio-based motion-resilient HR monitoring. We first collected HR-induced sounds in the ear canal leveraging an in-ear microphone under stationary and three different activities (i.e., walking, running, and speaking). Then, we devised a novel deep learning based motion artefact (MA) mitigation framework to denoise the in-ear audio signals, followed by an HR estimation algorithm to extract HR. With data collected from 20 subjects over four activities, we demonstrate that hEARt, our end-to-end approach, achieves a mean absolute error (MAE) of 3.02$\pm\ \boldsymbol{ 2.97}$BPM, 8.12$\pm\ \boldsymbol{6.74}$BPM, 11.23$\pm\ \boldsymbol{9.20}$BPM and 9.39$\pm\ \boldsymbol{6.97}$BPM for stationary, walking, running and speaking, respectively, opening the door to a new non-invasive and affordable HR monitoring with usable performance for daily activities. Not only does hEARt outperform previous in-ear HR monitoring work, but it outperforms reported in-ear PPG performance.

Continual Learning (CL) allows applications such as user personalization and household robots to learn on the fly and adapt to context. This is an important feature when context, actions, and users change. However, enabling CL on resource-constrained embedded systems is challenging due to the limited labeled data, memory, and computing capacity.

Ensembles of deep learning models can be used for estimating predictive uncertainty. Existing ensemble approaches, however, introduce a high computational and memory cost limiting their applicability to real-time biosignal applications (e.g. ECG, EEG). To address these issues, we propose early exit ensembles (EEEs) for estimating predictive uncertainty via an implicit ensemble of early exits. In particular, EEEs are a collection of weight sharing sub-networks created by adding exit branches to any backbone neural network architecture. Empirical evaluation of EEEs demonstrates strong performance in accuracy and uncertainty metrics as well as computation gain highlighting the benefit of combining multiple structurally diverse models that can be jointly trained. Compared to state-of-the-art baselines (with an ensemble size of 5), EEEs can improve uncertainty metrics up to 2× while providing test-time speed-up and memory reduction of approx. 5×. Additionally, EEEs can improve accuracy up to 3.8 percentage points compared to single model baselines.

Many deep learning applications, like keyword spotting [1], [2], require the incorporation of new concepts (classes) over time, referred to as Class Incremental Learning (CIL). The major challenge in CIL is catastrophic forgetting, i.e., preserving as much of the old knowledge as possible while learning new tasks. Various techniques, such as regularization, knowledge distillation, and the use of exemplars, have been proposed to resolve this issue. However, prior works primarily focus on the incremental learning step, while ignoring the optimization during the base model training. We hypothesise that a more transferable and generalizable feature representation from the base model would be beneficial to incremental learning.In this work, we adopt multitask learning during base model training to improve the feature generalizability. Specifically, instead of training a single model with all the base classes, we decompose the base classes into multiple subsets and regard each of them as a task. These tasks are trained concurrently and a shared feature extractor is obtained for incremental learning. We evaluate our approach on two datasets under various configurations. The results show that our approach enhances the average incremental learning accuracy by up to 5.5%, which enables more reliable and accurate keyword spotting over time. Moreover, the proposed approach can be combined with many existing techniques and provides additional performance gain.

Internet of Things (IoT) systems provide large amounts of data on all aspects of human behavior. Machine learning techniques, especially deep neural networks (DNN), have shown promise in making sense of this data at a large scale. Also, the research community has worked to reduce the computational and resource demands of DNN to compute on low-resourced micro controllers (MCUs). However, most of the current work in embedded deep learning focuses on solving a single task efficiently, while the multi-tasking nature and applications of IoT devices demand systems that can handle a diverse range of tasks (such as activity, gesture, voice, and context recognition) with input from a variety of sensors, simultaneously. In this paper, we propose YONO, a product quantization (PQ) based approach that compresses multiple heterogeneous models and enables in-memory model execution and model switching for dissimilar multi-task learning on MCUs. We first adopt PQ to learn codebooks that store weights of different models. Also, we propose a novel network optimization and heuristics to maximize the com-pression rate and minimize the accuracy loss. Then, we develop an online component of YONO for efficient model execution and switching between multiple tasks on an MCU at run time without relying on an external storage device. YONO shows remarkable performance as it can compress multiple heterogeneous models with negligible or no loss of accuracy up to 12.37x. Furthermore, YONO's online component enables an efficient execution (latency of 16–159 ms and energy consumption of 3.8-37.9 mJ per operation) and reduces modelloading/switching la-tency and energy consumption by 93.3-94.5% and 93.9-95.0%, respectively, compared to external storage access. Interestingly, YONO can compress various architectures trained with datasets that were not shown during YONO's offline codebook learning phase showing the generalizability of our method. To summarize, YONO shows great potential and opens further doors to enable multi-task learning systems on extremely resource-constrained devices.

While the global healthcare market of wearable devices has been growing significantly in recent years and is predicted to reach $60 billion by 2028, many important healthcare applications such as seizure monitoring, drowsiness detection, etc. have not been deployed due to the limited battery lifetime, slow response rate, and inadequate biosignal quality.

The development of fast and accurate screening tools, which could facilitate testing and prevent more costly clinical tests, is key to the current pandemic of COVID-19. In this context, some initial work shows promise in detecting diagnostic signals of COVID-19 from audio sounds. In this paper, we propose a voice-based framework to automatically detect individuals who have tested positive for COVID-19. We evaluate the performance of the proposed framework on a subset of data crowdsourced from our app, containing 828 samples from 343 participants. By combining voice signals and reported symptoms, an AUC of 0.79 has been attained, with a sensitivity of 0.68 and a specificity of 0.82. We hope that this study opens the door to rapid, low-cost, and convenient pre-screening tools to automatically detect the disease.

Continual learning approaches help deep neural network models adapt and learn incrementally by trying to solve catastrophic forgetting. However, whether these existing approaches, applied traditionally to image-based tasks, work with the same efficacy to the sequential time series data generated by mobile or embedded sensing systems remains an unanswered question. To address this void, we conduct the first comprehensive empirical study that quantifies the performance of three predominant continual learning schemes (i.e., regularization, replay, and replay with examples) on six datasets from three mobile and embedded sensing applications in a range of scenarios having different learning complexities. More specifically, we implement an end-to-end continual learning framework on edge devices. Then we investigate the generalizability, trade-offs between performance, storage, computational costs, and memory footprint of different continual learning methods. Our findings suggest that replay with exemplars-based schemes such as iCaRL has the best performance trade-offs, even in complex scenarios, at the expense of some storage space (few MBs) for training examples (1% to 5%). We also demonstrate for the first time that it is feasible and practical to run continual learning on-device with a limited memory budget. In particular, the latency on two types of mobile and embedded devices suggests that both incremental learning time (few seconds - 4 minutes) and training time (1 - 75 minutes) across datasets are acceptable, as training could happen on the device when the embedded device is charging thereby ensuring complete data privacy. Finally, we present some guidelines for practitioners who want to apply a continual learning paradigm for mobile sensing tasks.

Various incremental learning (IL) approaches have been proposed to help deep learning models learn new tasks/classes continuously without forgetting what was learned previously (i.e., avoid catastrophic forgetting). With the growing number of deployed audio sensing applications that need to dynamically incorporate new tasks and changing input distribution from users, the ability of IL on-device becomes essential for both efficiency and user privacy. However, prior works suffer from high computational costs and storage demands which hinders the deployment of IL on-device. In this work, to overcome these limitations, we develop an end-to-end and on-device IL framework, FastICARL, that incorporates an exemplar-based IL and quantization in the context of audio-based applications. We first employ k-nearest-neighbor to reduce the latency of IL. Then, we jointly utilize a quantization technique to decrease the storage requirements of IL. We implement FastICARL on two types of mobile devices and demonstrate that FastICARL remarkably decreases the IL time up to 78-92% and the storage requirements by 2-4 times without sacrificing its performance. FastICARL enables complete on-device IL, ensuring user privacy as the user data does not need to leave the device.

The INTERSPEECH 2021 Computational Paralinguistics Challenge addresses four different problems for the first time in a research competition under well-defined conditions: In the COVID-19 Cough and COVID-19 Speech Sub-Challenges, a binary classification on COVID-19 infection has to be made based on coughing sounds and speech; in the Escalation SubChallenge, a three-way assessment of the level of escalation in a dialogue is featured; and in the Primates Sub-Challenge, four species vs background need to be classified. We describe the Sub-Challenges, baseline feature extraction, and classifiers based on the 'usual' COMPARE and BoAW features as well as deep unsupervised representation learning using the AuDeep toolkit, and deep feature extraction from pre-trained CNNs using the Deep Spectrum toolkit; in addition, we add deep end-to-end sequential modelling, and partially linguistic analysis.

Recently, sound-based COVID-19 detection studies have shown great promise to achieve scalable and prompt digital prescreening.However, there are still two unsolved issues hindering the practice.First, collected datasets for model training are often imbalanced, with a considerably smaller proportion of users tested positive, making it harder to learn representative and robust features.Second, deep learning models are generally overconfident in their predictions.Clinically, false predictions aggravate healthcare costs.Estimation of the uncertainty of screening would aid this.To handle these issues, we propose an ensemble framework where multiple deep learning models for sound-based COVID-19 detection are developed from different but balanced subsets from original data.As such, data are utilized more effectively compared to traditional up-sampling and down-sampling approaches: an AUC of 0.74 with a sensitivity of 0.68 and a specificity of 0.69 is achieved.Simultaneously, we estimate uncertainty from the disagreement across multiple models.It is shown that false predictions often yield higher uncertainty, enabling us to suggest the users with certainty higher than a threshold to repeat the audio test on their phones or to take clinical tests if digital diagnosis still fails.This study paves the way for a more robust sound-based COVID-19 automated screening system.

Body-focused repetitive behaviors (BFRBs), like face-touching or skin-picking, are hand-driven behaviors which can damage one’s appearance, if not identified early and treated. Technology for automatic detection is still under-explored, with few previous works being limited to wearables with single modalities (e.g., motion). Here, we propose a multi-sensory approach combining motion, orientation, and heart rate sensors to detect BFRBs. We conducted a feasibility study in which participants (N=10) were exposed to BFRBs-inducing tasks, and analyzed 380 mins of signals1 under an extensive evaluation of sensing modalities, cross-validation methods, and observation windows. Our models achieved an AUC > 0.90 in distinguishing BFRBs, which were more evident in observation windows 5 mins prior to the behavior as opposed to 1-min ones. In a follow-up qualitative survey, we found that not only the timing of detection matters but also models need to be context-aware, when designing just-in-time interventions to prevent BFRBs.

Human gait is a widely used biometric trait for user identification and recognition. Given the wide-spreading, steady diffusion of ear-worn wearables (Earables) as the new frontier of wearable devices, we investigate the feasibility of earable-based gait identification. Specifically, we look at gait-based identification from the sounds induced by walking and propagated through the musculoskeletal system in the body. Our system, EarGate, leverages an in-ear facing microphone which exploits the earable's occlusion effect to reliably detect the user's gait from inside the ear canal, without impairing the general usage of earphones. With data collected from 31 subjects, we show that EarGate achieves up to 97.26% Balanced Accuracy (BAC) with very low False Acceptance Rate (FAR) and False Rejection Rate (FRR) of 3.23% and 2.25%, respectively. Further, our measurement of power consumption and latency investigates how this gait identification model could live both as a stand-alone or cloud-coupled earable system.

Smart earbuds are recognized as a new wearable platform for personal-scale human motion sensing. However, due to the interference from head movement or background noise, commonly-used modalities (e.g. accelerometer and microphone) fail to reliably detect both intense and light motions. To obviate this, we propose OESense, an acoustic-based in-ear system for general human motion sensing. The core idea behind OESense is the joint use of the occlusion effect (i.e., the enhancement of low-frequency components of bone-conducted sounds in an occluded ear canal) and inward-facing microphone, which naturally boosts the sensing signal and suppresses external interference. We prototype OESense as an earbud and evaluate its performance on three representative applications, i.e., step counting, activity recognition, and hand-to-face gesture interaction. With data collected from 31 subjects, we show that OESense achieves 99.3% step counting recall, 98.3% recognition recall for 5 activities, and 97.0% recall for five tapping gestures on human face, respectively. We also demonstrate that OESense is compatible with earbuds' fundamental functionalities (e.g. music playback and phone calls). In terms of energy, OESense consumes 746 mW during data recording and recognition and it has a response latency of 40.85 ms for gesture recognition. Our analysis indicates such overhead is acceptable and OESense is potential to be integrated into future earbuds.

Earables (in-ear wearables) are a new frontier in wearables. Acting both as leisure devices, providing personal audio, as well as sensing platforms, earables could collect sensor data for the upper part of the body, subject to fewer vibrations and random movement variations than the lower parts of the body, due to inherent damping in the musculoskeletal system. These data may enable application domains such as augmented/virtual reality, medical rehabilitation, and health condition screening. Unfortunately, earables have inherent size, shape, and weight constraints limiting the type and position of the sensors on such platforms. For instance, lacking a magnetometer in all earables reference platforms, earables lack reference points. Thus, it becomes harder to work with absolute orientations. Embedding magnetometers in earables is challenging, as these rely heavily on radio (mostly Bluetooth) communication (RF) and contain magnets for magnetic-driven speakers and docking. We explore the feasibility of adding a built-in magnetometer in an earbud, presenting the first comprehensive study of the magnetic interference impacting the magnetometer when placed in an earable: both that caused by the speaker and by RF (music streaming and voice calls) are considered. We find that appropriate calibration of the magnetometer removes the offsets induced by the magnets, the speaker, and the variable interference due to BT. Further, we present an automatic, user-transparent adaptive calibration that obviates the need for alternative, expensive, and error-prone manual, or robotics, calibration procedures. Our evaluation shows how our calibration approach performs under different conditions, achieving convincing results with errors below 3° for the majority of the experiments.

Modern machine learning applications should be able to address the intrinsic challenges arising over inference on massive real-world datasets, including scalability and robustness to outliers. Despite the multiple benefits of Bayesian methods (such as uncertainty-aware predictions, incorporation of experts knowledge, and hierarchical modeling), the quality of classic Bayesian inference depends critically on whether observations conform with the assumed data generating model, which is impossible to guarantee in practice. In this work, we propose a variational inference method that, in a principled way, can simultaneously scale to large datasets, and robustify the inferred posterior with respect to the existence of outliers in the observed data. Reformulating Bayes theorem via the β-divergence, we posit a robustified generalized Bayesian posterior as the target of inference. Moreover, relying on the recent formulations of Riemannian coresets for scalable Bayesian inference, we propose a sparse variational approximation of the robustified posterior and an efficient stochastic black-box algorithm to construct it. Overall our method allows releasing cleansed data summaries that can be applied broadly in scenarios involving structured and unstructured data contamination. We illustrate the applicability of our approach in diverse simulated and real datasets, and various statistical models, including Gaussian mean inference, logistic and neural linear regression, demonstrating its superiority to existing Bayesian summarization methods in the presence of outliers.

Audio signals generated by the human body (e.g., sighs, breathing, heart, digestion, vibration sounds) have routinely been used by clinicians as indicators to diagnose disease or assess disease progression. Until recently, such signals were usually collected through manual auscultation at scheduled visits. Research has now started to use digital technology to gather bodily sounds (e.g., from digital stethoscopes) for cardiovascular or respiratory examination, which could then be used for automatic analysis. Some initial work shows promise in detecting diagnostic signals of COVID-19 from voice and coughs. In this paper we describe our data analysis over a large-scale crowdsourced dataset of respiratory sounds collected to aid diagnosis of COVID-19. We use coughs and breathing to understand how discernible COVID-19 sounds are from those in asthma or healthy controls. Our results show that even a simple binary machine learning classifier is able to classify correctly healthy and COVID-19 sounds. We also show how we distinguish a user who tested positive for COVID-19 and has a cough from a healthy user with a cough, and users who tested positive for COVID-19 and have a cough from users with asthma and a cough. Our models achieve an AUC of above 80% across all tasks. These results are preliminary and only scratch the surface of the potential of this type of data and audio-based machine learning. This work opens the door to further investigation of how automatically analysed respiratory patterns could be used as pre-screening signals to aid COVID-19 diagnosis.

We present a federated, asynchronous, and $(\varepsilon, \delta)$-differentially private algorithm for $\PCA$ in the memory-limited setting. % Our algorithm incrementally computes local model updates using a streaming procedure and adaptively estimates its $r$ leading principal components when only $\mathcal{O}(dr)$ memory is available with $d$ being the dimensionality of the data. % We guarantee differential privacy via an input-perturbation scheme in which the covariance matrix of a dataset $\B{X} \in \R^{d \times n}$ is perturbed with a non-symmetric random Gaussian matrix with variance in $\mathcal{O}\left(\left(\frac{d}{n}\right)^2 \log d \right)$, thus improving upon the state-of-the-art. % Furthermore, contrary to previous federated or distributed algorithms for $\PCA$, our algorithm is also invariant to permutations in the incoming data, which provides robustness against straggler or failed nodes. % Numerical simulations show that, while using limited-memory, our algorithm exhibits performance that closely matches or outperforms traditional non-federated algorithms, and in the absence of communication latency, it exhibits attractive horizontal scalability.

Standard Bayesian inference algorithms are prohibitively expensive in the regime of modern large-scale data. Recent work has found that a small, weighted subset of data (a coreset) may be used in place of the full dataset during inference, taking advantage of data redundancy to reduce computational cost. However, this approach has limitations in the increasingly common setting of sensitive, high-dimensional data. Indeed, we prove that there are situations in which the Kullback-Leibler divergence between the optimal coreset and the true posterior grows with data dimension; and as coresets include a subset of the original data, they cannot be constructed in a manner that preserves individual privacy. We address both of these issues with a single unified solution, Bayesian pseudocoresets --- a small weighted collection of synthetic "pseudodata"---along with a variational optimization method to select both pseudodata and weights. The use of pseudodata (as opposed to the original datapoints) enables both the summarization of high-dimensional data and the differentially private summarization of sensitive data. Real and synthetic experiments on high-dimensional data demonstrate that Bayesian pseudocoresets achieve significant improvements in posterior approximation error compared to traditional coresets, and that pseudocoresets provide privacy without a significant loss in approximation quality.

Smartphones have started to be used as self reporting tools for mental health state as they accompany individuals during their days and can therefore gather temporally fine grained data. However, the analysis of self reported mood data offers challenges related to non-homogeneity of mood assessment among individuals due to the complexity of the feeling and the reporting scales, as well as the noise and sparseness of the reports when collected in the wild. In this paper, we propose a new end-to-end ML model inspired by video frame prediction and machine translation, that forecasts future sequences of mood from previous self-reported moods collected in the real world using mobile devices. Contrary to traditional time series forecasting algorithms, our multi-task encoder-decoder recurrent neural network learns patterns from different users, allowing and improving the prediction for users with limited number of self-reports. Unlike traditional feature-based machine learning algorithms, the encoder-decoder architecture enables to forecast a sequence of future moods rather than one single step. Meanwhile, multi-task learning exploits some unique characteristics of the data (mood is bi-dimensional), achieving better results than when training single-task networks or other classifiers.

Point-of-Interest (POI) recommender systems play a vital role in people's lives by recommending unexplored POIs to users and have drawn extensive attention from both academia and industry. Despite their value, however, they still suffer from the challenges of capturing complicated user preferences and fine-grained user-POI relationship for spatio-temporal sensitive POI recommendation. Existing recommendation algorithms, including both shallow and deep approaches, usually embed the visiting records of a user into a single latent vector to model user preferences: this has limited power of representation and interpretability. In this paper, we propose a novel topic-enhanced memory network (TEMN), a deep architecture to integrate the topic model and memory network capitalising on the strengths of both the global structure of latent patterns and local neighbourhood-based features in a nonlinear fashion. We further incorporate a geographical module to exploit user-specific spatial preference and POI-specific spatial influence to enhance recommendations. The proposed unified hybrid model is widely applicable to various POI recommendation scenarios. Extensive experiments on real-world WeChat datasets demonstrate its effectiveness (improvement ratio of 3.25% and 29.95% for context-aware and sequential recommendation, respectively). Also, qualitative analysis of the attention weights and topic modeling provides insight into the model's recommendation process and results.

The investigation of the effect of the built environment in a neighbourhood and how it impacts residents' health is of value to researchers from public health policy to social science. The traditional methods to assess this impact is through surveys which lead to temporally and spatially coarse grained data and are often not cost effective. Here we propose an approach to link the effects of neighbourhood services over citizen health using a technique that attempts to highlight the cause-effect aspects of these relationships. The method is based on the theory of propensity score matching with multiple 'doses' and it leverages existing fine grained open web data. To demonstrate the method, we study the effect of sport venue presence on the prevalence of antidepressant prescriptions in over 600 neighbourhoods in London over a period of three years. We find the distribution of effects is approximately normal, centred on a small negative effect on prescriptions with increases in the availability of sporting facilities, on average. We assess the procedure through some standard quantitative metrics as well as matching on synthetic data generated by modelling the real data. This approach opens the door to fast and inexpensive alternatives to quantify and continuously monitor effects of the neighborhood built environment on population health.

As modern transportation systems become more complex, there is need for mobile applications that allow travelers to navigate efficiently in cities. In taxi transport the recent proliferation of Uber has introduced new norms including a flexible pricing scheme where journey costs can change rapidly depending on passenger demand and driver supply. To make informed choices on the most appropriate provider for their journeys, travelers need access to knowledge about provider pricing in real time. To this end, we developed OpenStreetCab a mobile application that offers advice on taxi transport comparing provider prices. We describe its development and deployment in two cities, London and New York, and analyse thousands of user journey queries to compare the price patterns of Uber against major local taxi providers. We have observed large heterogeneity across the taxi transport markets in the two cities. This motivated us to perform a price validation and measurement experiment on the ground comparing Uber and Black Cabs in London. The experimental results reveal interesting insights: not only they confirm feedback on pricing and service quality received by professional driver users, but also they reveal the tradeoffs between prices and journey times between taxi providers. With respect to journey times in particular, we show how experienced taxi drivers, in the majority of the cases, are able to navigate faster to a destination compared to drivers who rely on modern navigation systems. We provide evidence that this advantage becomes stronger in the centre of a city where urban density is high.

Chronic diseases like cancer and diabetes are major threats to human life. Understanding the distribution and progression of chronic diseases of a population is important in assisting the allocation of medical resources as well as the design of policies in preemptive healthcare. Traditional methods to obtain large scale indicators on population health, e.g., surveys and statistical analysis, can be costly and time-consuming and often lead to a coarse spatio-temporal picture. In this paper, we leverage a dataset describing the human mobility patterns of citizens in a large metropolitan area. By viewing local human lifestyles we predict the evolution rate of several chronic diseases at the level of a city neighborhood. We apply the combination of a collaborative topic modeling (CTM) and a Gaussian mixture method (GMM) to tackle the data sparsity challenge and achieve robust predictions on health conditions simultaneously. Our method enables the analysis and prediction of disease rate evolution at fine spatio-temporal scales and demonstrates the potential of incorporating datasets from mobile web sources to improve population health monitoring. Evaluations using real-world check-in and chronic disease morbidity datasets in the city of London show that the proposed CTM+GMM model outperforms various baseline methods.

Cultural activity is an inherent aspect of urban life and the success of a modern city is largely determined by its capacity to offer generous cultural entertainment to its citizens. To this end, the optimal allocation of cultural establishments and related resources across urban regions becomes of vital importance, as it can reduce financial costs in terms of planning and improve quality of life in the city, more generally. In this paper, we make use of a large longitudinal dataset of user location check-ins from the online social network WeChat to develop a data-driven framework for cultural planning in the city of Beijing. We exploit rich spatio-temporal representations on user activity at cultural venues and use a novel extended version of the traditional latent Dirichlet allocation model that incorporates temporal information to identify latent patterns of urban cultural interactions. Using the characteristic typologies of mobile user cultural activities emitted by the model, we determine the levels of demand for different types of cultural resources across urban areas. We then compare those with the corresponding levels of supply as driven by the presence and spatial reach of cultural venues in local areas to obtain high resolution maps that indicate urban regions with lack of cultural resources, and thus give suggestions for further urban cultural planning and investment optimisation.

Abstract Human mobility is often represented as a mobility network, or graph, with nodes representing places of significance which an individual visits, such as their home, work, places of social amenity, etc., and edge weights corresponding to probability estimates of movements between these places. Previous research has shown that individuals can be identified by a small number of geolocated nodes in their mobility network, rendering mobility trace anonymization a hard task. In this paper we build on prior work and demonstrate that even when all location and timestamp information is removed from nodes, the graph topology of an individual mobility network itself is often uniquely identifying. Further, we observe that a mobility network is often unique, even when only a small number of the most popular nodes and edges are considered. We evaluate our approach using a large dataset of cell-tower location traces from 1 500 smartphone handsets with a mean duration of 430 days. We process the data to derive the top−N places visited by the device in the trace, and find that 93% of traces have a unique top−10 mobility network, and all traces are unique when considering top−15 mobility networks. Since mobility patterns, and therefore mobility networks for an individual, vary over time, we use graph kernel distance functions, to determine whether two mobility networks, taken at different points in time, represent the same individual. We then show that our distance metrics, while imperfect predictors, perform significantly better than a random strategy and therefore our approach represents a significant loss in privacy.

Estimating revenue and business demand of a newly opened venue is paramount as these early stages often involve critical decisions such as first rounds of staffing and resource allocation. Traditionally, this estimation has been performed through coarse measures such as observing numbers in local venues. The advent of crowdsourced data from devices and services has opened the door to better predictions of temporal visitation patterns for locations and venues. In this paper, using mobility data from the location-based service Foursquare, we treat venue categories as proxies for urban activities and analyze how they become popular over time. The main contribution of this work is a prediction framework able to use characteristic temporal signatures of places together with k-nearest neighbor metrics capturing similarities among urban regions to forecast weekly popularity dynamics of a new venue establishment. Our evaluation shows that temporally similar areas of a city can be valuable predictors, decreasing error by 41%. Our findings have the potential to impact the design of location-based technologies and decisions made by new business owners.

GPUs have recently enjoyed increased popularity as general purpose software accelerators in multiple application domains including computer vision and natural language processing. However, there has been little exploration into the performance and energy trade-offs mobile GPUs can deliver for the increasingly popular workload of deep-inference audio sensing tasks, such as, spoken keyword spotting in energy-constrained smartphones and wearables. In this paper, we study these trade-offs and introduce an optimization engine that leverages a series of structural and memory access optimization techniques that allow audio algorithm performance to be automatically tuned as a function of GPU device specifications and model semantics. We find that parameter optimized audio routines obtain inferences an order of magnitude faster than sequential CPU implementations, and up to 6.5x times faster than cloud offloading with good connectivity, while critically consuming 3-4x less energy than the CPU. Under our optimized GPU, conventional wisdom about how to use the cloud and low power chips is broken. Unless the network has a throughput of at least 20Mbps (and a RTT of 25 ms or less), with only about 10 to 20 seconds of buffering audio data for batched execution, the optimized GPU audio sensing apps begin to consume less energy than cloud offloading. Under such conditions we find the optimized GPU can provide energy benefits comparable to low-power reference DSP implementations with some preliminary level of optimization; in addition to the GPU always winning with lower latency.

The ability to detect and distinguish interactions in the workplace can shed light over productivity, team work and on employees' use of space. Questionnaires and direct observations have often been used as mechanisms to identify office based interactions, however, these are either very time consuming, yield coarse grained information or do not scale to large numbers of people. Technology has been recently employed to cut costs and improve output, however precise interaction dynamics gathering often requires individuals to wear custom hardware. In this paper, we present an extensive evaluation of Bluetooth Low Energy (BLE) as a technology to monitor people proximity in the workplace. We examine the key parameters that affect the accuracy of the detected contacts and their impact on power consumption. We study how this system can be implemented on popular wearable devices (i.e., Android Wear and Tizen) and the resulting limitations. Through a real world deployment in a commercial organisation with 25 participants we evaluate the performances of a BLE-based proximity detection technique. Our results show the suitability of BLE for workplace interaction detection and give guidance to vendors and Operating System (OS) developers on the impact of the restrictions regarding the use of BLE on commodity wearables.

Measuring mental well-being with mobile sensing has been an increasingly active research topic. Pervasiveness of smartphones combined with the convenience of mobile app distribution platforms (e.g., Google Play) provide a tremendous opportunity to reach out to millions of users. However, the studies at the confluence of mental health and mobile sensing have been longitudinally limited, controlled, or confined to a small number of participants. In this paper we report on what we believe is the largest longitudinal in-the-wild study of mood through smartphones. We describe an Android app to collect participants' self-reported moods and system triggered experience sampling data while passively measuring their physical activity, sociability, and mobility via their device's sensors. We report the results of a large-scale analysis of the data collected for about three years from 18,000 users.

The papers in this special section focuses on the concept of urban computing. This is a process of acquisition, integration, and analysis of big and heterogeneous data generated by a diversity of sources in urban spaces, such as sensors, devices, vehicles, buildings, and human, to tackle the major issues that cities face, e.g., air pollution, increased energy consumption and traffic congestion. Urban computing connects unobtrusive and ubiquitous sensing technologies, advanced data management and analytics models, and novel visualization methods, to create win-win-win solutions that improve urban environment, human life quality, and city operation systems. Urban computing also helps us understand the nature of urban phenomena and even predict the future of cities. Urban computing is an interdisciplinary field fusing the computing science with traditional fields, like transportation, civil engineering, economy, ecology, and sociology, in the context of urban spaces.

Summary form only given. With the advent of powerful and inexpensive sensing technology the ability to study human behaviour and activity at large scale and for long periods is becoming a firm reality. Wearables and mobile devices further allow the continuous physical colocation with the users. This reality generates new challenges but also opens the door to potentially innovative ways of understanding our daily lives. In this talk we will discuss our experience in large mobile sensor deployments. We will discuss the issues raised by mobile sensing big data in terms of data crowdsourcing, continuous sensing challenges, data analysis, privacy, user feedback. Examples will be drawn from our healthcare, transport, urban planning and organization analytics studies.

Mobile apps that use sensors to monitor user behavior often employ resource heavy inference algorithms that make computational offloading a common practice. However, existing schedulers/offloaders typically emphasize one primary offloading aspect without fully exploring complementary goals (e.g., heterogeneous resource management with only partial visibility into underlying algorithms, or concurrent sensor app execution on a single resource) and as a result, may overlook performance benefits pertinent to sensor processing.

Summary form only given. With the advent of powerful and inexpensive sensing technology the ability to study human behaviour and activity at large scale and for long periods is becoming a firm reality. Wearables and mobile devices further allow the continuous physical colocation with the users. This reality generates new challenges but also opens the door to potentially innovative ways of understanding our daily lives. In this talk we will discuss our experience in large mobile sensor deployments and in using complex network science for the analysis of mobile sensing data. We will discuss the issues raised by mobile sensing big data in terms of data crowdsourcing, continuous sensing challenges, data analysis, privacy, user feedback. Examples will be drawn from our healthcare, transport, urban planning and organization analytics studies.

Large metropolitan cities bring together diverse individuals, creating opportunities for cultural and intellectual exchanges, which can ultimately lead to social and economic enrichment. In this work, we present a novel network perspective on the interconnected nature of people and places, allowing us to capture the social diversity of urban locations through the social network and mobility patterns of their visitors. We use a dataset of approximately 37K users and 42K venues in London to build a network of Foursquare places and the parallel Twitter social network of visitors through check-ins. We define four metrics of the social diversity of places which relate to their social brokerage role, their entropy, the homogeneity of their visitors and the amount of serendipitous encounters they are able to induce. This allows us to distinguish between places that bring together strangers versus those which tend to bring together friends, as well as places that attract diverse individuals as opposed to those which attract regulars. We correlate these properties with wellbeing indicators for London neighbourhoods and discover signals of gentrification in deprived areas with high entropy and brokerage, where an influx of more affluent and diverse visitors points to an overall improvement of their rank according to the UK Index of Multiple Deprivation for the area over the five-year census period. Our analysis sheds light on the relationship between the prosperity of people and places, distinguishing between different categories and urban geographies of consequence to the development of urban policy and the next generation of socially-aware location-based applications.

The wearable revolution, as a mass-market phenomenon, has finally arrived. As a result, the question of how wearables should evolve over the next 5 to 10 years is assuming an increasing level of societal and commercial importance. A range of open design and system questions are emerging, for instance: How can wearables shift from being largely health and fitness focused to tracking a wider range of life events? What will become the dominant methods through which users interact with wearables and consume the data collected? Are wearables destined to be cloud and/or smartphone dependent for their operation?

Smartphone sensing research has been advancing at a brisk pace. Yet, current social networking services often only take advantage of location sensing: applications like Foursquare use the phone's GPS and Wi-Fi radios to infer the user's location to simplify checking-in to a place. However, smartphone sensing could be exploited to considerably expand the spectrum of information a user can share with a few clicks with friends: not only the location of an event but activities such as “cooking dinner” or “waiting for a bus” can be predicted and suggested to the user to ease the check-in process. In this paper we show how mobile phone sensing can be used in this sense. For this prediction process to be accurate however, sensors need to be sampled often, with a considerable impact on the phone battery. To alleviate this issue, we explore streams of phone usage data (soft sensors), such as application usage, messages, and phone calls for predicting the user's activity in a more efficient fashion for augmenting mobile social check-in apps. We have deployed our application and collected a dataset of over 2700 check-ins to 48 activities from 20 users. Our analysis shows a prediction accuracy of 75% when offering 5 check-in suggestions to users. Furthermore, we show that when using only soft sensors we can achieve very similar performance to that obtained with real sensors, thereby significantly reducing the impact on the phone battery. This finding might have a potentially high impact on smartphone based activity check-in apps.

In many work environments, serendipitous interactions between members of different groups may lead to enhanced productivity, collaboration and knowledge dissemination. Two factors that may have an influence on such interactions are cultural differences between individuals in highly multicultural workplaces, and the layout and physical spaces of the workplace itself. In this work, we investigate how these two factors may facilitate or hinder inter-group interactions in the workplace. We analyze traces collected using wearable electronic badges to capture face-to-face interactions and mobility patterns of employees in a research laboratory in the UK. We observe that those who interact with people of different roles tend to come from collectivist cultures that value relationships and where people tend to be comfortable with social hierarchies, and that some locations in particular are more likely to host serendipitous interactions, knowledge that could be used by organizations to enhance communication and productivity.

The layouts of the buildings we live in shape our everyday lives. In office environments, building spaces affect employees' communication, which is crucial for productivity and innovation. However, accurate measurement of how spatial layouts affect interactions is a major challenge and traditional techniques may not give an objective view.

The rapidly growing adoption of sensor-enabled smartphones has greatly fueled the proliferation of applications that use phone sensors to monitor user behavior. A central sensor among these is the microphone which enables, for instance, the detection of valence in speech, or the identification of speakers. Deploying multiple of these applications on a mobile device to continuously monitor the audio environment allows for the acquisition of a diverse range of sound-related contextual inferences. However, the cumulative processing burden critically impacts the phone battery.

While there is significant work on sensing and recognition of significant places for users, little attention has been given to users' significant routes. Recognizing these routine journeys, can open doors for the development of novel applications, like personalized travel alerts, and enhancement of user's travel experience. However, the high energy consumption of traditional location sensing technologies, such as GPS or WiFi based localization, is a barrier to passive and ubiquitous route sensing through smartphones.

The Experience Sampling Method (ESM) has been widely used to collect longitudinal survey data from participants; in this domain, smartphone sensors are now used to augment the context-awareness of sampling strategies. In this paper, we study the effect of ESM design choices on the inferences that can be made from participants' sensor data, and on the variance in survey responses that can be collected. In particular, we answer the question: are the behavioural inferences that a researcher makes with a trigger-defined subsample of sensor data biased by the sampling strategy's design? We demonstrate that different single-sensor sampling strategies will result in what we refer to as contextual dissonance: a disagreement in how much different behaviours are represented in the aggregated sensor data. These results are not only relevant to researchers who use the ESM, but call for future work into strategies that may alleviate the biases that we measure.

The problem of identifying the optimal location for a new retail store has been the focus of past research, especially in the field of land economy, due to its importance in the success of a business. Traditional approaches to the problem have factored in demographics, revenue and aggregated human flow statistics from nearby or remote areas. However, the acquisition of relevant data is usually expensive. With the growth of location-based social networks, fine grained data describing user mobility and popularity of places has recently become attainable.

Inferring the type of activities in neighborhoods of urban centers may be helpful in a number of contexts including urban planning, content delivery and activity recommendations for mobile web users or may even yield to a deeper understanding of the geographical evolution of social life in the city . During the past few years, the analysis of mobile phone usage patterns, or of social media with longitudinal attributes, have aided the automatic characterization of the dynamics of the urban environment. In this work, we combine a dataset sourced from a telecommunication provider in Spain with a database of millions of geotagged venues from Foursquare and we formulate the problem of urban activity inference in a supervised learning framework. In particular, we exploit user communication patterns observed at the base station level in order to predict the activity of Foursquare users who checkin-in at nearby venues. First, we mine a set of machine learning features that allow us to encode the input telecommunication signal of a tower. Subsequently, we evaluate a diverse set of supervised learning algorithms using labels extracted from Foursquare place categories and we consider two application scenarios. Initially, we assess how hard it is to predict specific urban activity of an area, showing that Nightlife and Entertainment spots are those easier to infer, whereas College and Shopping areas are those featuring the lowest accuracy rates. Then, considering a candidate set of activity types in a geographic area, we aim to elect the most prominent one. We demonstrate how the difficulty of the problem increases with the number of classes incorporated in the prediction task, yet the classifiers achieve a considerably better performance compared to a random guess even when the set of candidate classes increases.

Studies of automotive traffic have shown that on average 30% of traffic in congested urban areas is due to cruising drivers looking for parking. While we have witnessed a push towards sensing technologies to monitor real-time parking availability, instrumenting on-street parking throughout a city is a considerable investment. In this paper, we present ParkSense, a smartphone based sensing system that detects if a driver has vacated a parking spot. ParkSense leverages the ubiquitous Wi-Fi beacons in urban areas for sensing unparking events. It utilizes a robust Wi-Fi signature matching approach to detect driver's return to the parked vehicle. Moreover, it uses a novel approach based on the rate of change of Wi-Fi beacons to sense if the user has started driving. We show that the rate of change of the observed beacons is highly correlated with actual user speed and is a good indicator of whether a user is in a vehicle. Through empirical evaluation, we demonstrate that our approach has a significantly smaller energy footprint than traditional location sensors like GPS and Wi-Fi based positioning while still maintaining sufficient accuracy.

Mobile phones play a pivotal role in supporting ubiquitous and unobtrusive sensing of human activities. However, maintaining a highly accurate record of a user's behavior throughout the day imposes significant energy demands on the phone's battery. In this paper, we present the design, implementation, and evaluation of METIS: an adaptive mobile sensing platform that efficiently supports social sensing applications. The platform implements a novel sensor task distribution scheme that dynamically decides whether to perform sensing on the phone or in the infrastructure, considering the energy consumption, accuracy, and mobility patterns of the user. By comparing the sensing distribution scheme with sensing performed solely on the phone or exclusively on the fixed remote sensors, we show, through benchmarks using real traces, that the opportunistic sensing distribution achieves over 60% and 40% energy savings, respectively. This is confirmed through a real world deployment in an office environment for over a month: we developed a social application over our frameworks, that is able to infer the collaborations and meetings of the users. In this setting the system preserves over 35% more battery life over pure phone sensing.

The application of complex network models to communication systems has led to several important results: nonetheless, previous research has often neglected to take into account their temporal properties, which in many real scenarios play a pivotal role. At the same time, network robustness has come extensively under scrutiny. Understanding whether networked systems can undergo structural damage and yet perform efficiently is crucial to both their protection against failures and to the design of new applications. In spite of this, it is still unclear what type of resilience we may expect in a network which continuously changes over time. In this work, we present the first attempt to define the concept of temporal network robustness: we describe a measure of network robustness for time-varying networks and we show how it performs on different classes of random models by means of analytical and numerical evaluation. Finally, we report a case study on a real-world scenario, an opportunistic vehicular system of about 500 taxicabs, highlighting the importance of time in the evaluation of robustness. Particularly, we show how static approximation can wrongly indicate high robustness of fragile networks when adopted in mobile time-varying networks, while a temporal approach captures more accurately the system performance.

Mobile location-based services are thriving, providing an unprecedented opportunity to collect fine grained spatio-temporal data about the places users visit. This multi-dimensional source of data offers new possibilities to tackle established research problems on human mobility, but it also opens avenues for the development of novel mobile applications and services. In this work we study the problem of predicting the next venue a mobile user will visit, by exploring the predictive power offered by different facets of user behavior. We first analyze about 35 million check-ins made by about 1 million Foursquare users in over 5 million venues across the globe, spanning a period of five months. We then propose a set of features that aim to capture the factors that may drive users' movements. Our features exploit information on transitions between types of places, mobility flows between venues, and spatio-temporal characteristics of user check-in patterns. We further extend our study combining all individual features in two supervised learning models, based on linear regression and M5 model trees, resulting in a higher overall prediction accuracy. We find that the supervised methodology based on the combination of multiple features offers the highest levels of prediction accuracy: M5 model trees are able to rank in the top fifty venues one in two user check-ins, amongst thousands of candidate items in the prediction list.

Connections established by users of online social networks are influenced by mechanisms such as preferential attachment and triadic closure. Yet, recent research has found that geographic factors also constrain users: spatial proximity fosters the creation of online social ties. While the effect of space might need to be incorporated to these social mechanisms, it is not clear to which extent this is true and in which way this is best achieved.

Mobile phones are integral to everyday life with emails, social networking, online banking and other applications; however, the wealth of private information accessible increases economic incentives for attackers. Compared with fixed networks, mobile malware can replicate through both long range messaging and short range radio technologies; the former can be filtered by the network operator but determining the best method of containing short range malware is an open problem. While global software updates are sometimes possible, they are often not practical. An alternative and more efficient strategy is to distribute the patch to the key nodes so that they can opportunistically disseminate it to the rest of the network via short range encounters; but how can these key nodes be identified in a highly dynamic network topology? In this paper, we address these questions by presenting Socio- Temporal Opportunistic Patching (STOP), a two-tier predictive mobile malware containment system: devices collect co-location data in a decentralized manner and report to a central server which processes and targets delivery of hot fixes to a small subset of k devices at runtime; in turn mobile devices spread the patch opportunistically. The STOP system is underpinned by a recent theoretical framework for analysing dynamic networks that takes into account temporal information of links. Using empirical contact traces, we find firstly, the top-k ranking temporal centrality nodes are highly correlated with past time windows; and secondly, simple prediction functions can be designed to select the set of top-k nodes that are optimal for patch spreading.

The increasing adoption of wireless sensor network technology in a variety of applications, from agricultural to volcanic monitoring, has demonstrated their ability to gather data with unprecedented sensing capabilities and deliver it to a remote user. However, a key issue remains how to maintain these sensor network deployments over increasingly prolonged deployments. In this article, we present the challenges that were faced in maintaining continual operation of an automated wildlife monitoring system over a one-year period. This system analyzed the social colocation patterns of European badgers ( Meles meles ) residing in a dense woodland environment using a hybrid RFID-WSN approach. We describe the stages of the evolutionary development, from implementation, deployment, and testing, to various iterations of software optimization, followed by hardware enhancements, which in turn triggered the need for further software optimization. We highlight the main lessons learned: the need to factor in the maintenance costs while designing the system; to consider carefully software and hardware interactions; the importance of rapid prototyping for initial deployment (this was key to our success); and the need for continuous interaction with domain scientists which allows for unexpected optimizations.

Nowadays companies increasingly aggregate location data from different sources on the Internet to offer location-based services such as estimating current road traffic conditions, and finding the best nightlife locations in a city. However, these services have also caused outcries over privacy issues. As the volume of location data being aggregated expands, the comfort of sharing one's whereabouts with the public at large will unavoidably decrease. Existing ways of aggregating location data in the privacy literature are largely centralized in that they rely on a trusted location-based service. Instead, we propose a piece of software (SpotMe) that can run on a mobile phone and is able to estimate the number of people in geographic locations in a privacy-preserving way: accurate estimations are made possible in the presence of privacy-conscious users who report, in addition to their actual locations, a very large number of erroneous locations. The erroneous locations are selected by a randomized response algorithm. We evaluate the accuracy of SpotMe in estimating the number of people upon two very different realistic mobility traces: the mobility of vehicles in urban, suburban and rural areas, and the mobility of subway train passengers in Greater London. We find that erroneous locations have little effect on the estimations (in both traces, the error is below 18% for a situation in which more than 99% of the locations are erroneous), yet they guarantee that users cannot be localized with high probability. Also, the computational and storage overheads for a mobile phone running Spot Me are negligible, and the communication overhead is limited.

The application of complex network theory to communication systems has led to several important results. Nonetheless, previous research has often neglected to take into account their temporal properties, which in many real scenarios play a pivotal role. Mainly because of mobility, transmission delays or protocol design, a communication network should not be considered only as a static entity. At the same time, network robustness has come extensively under scrutiny. Understanding whether networked systems can undergo structural damage and yet perform efficiently is crucial to both their protection against failures and to the design of new applications. In spite of this, it is still unclear what type of resilience we may expect in a network that continuously changes over time. In this work we present the first attempt to define the concept of temporal network robustness: we describe a measure of network robustness for time-varying networks and we show how it performs on different classes of random models by means of analytical and numerical evaluation. Particularly, we show how static approximation can wrongly indicate high robustness of fragile networks when adopted in mobile time-varying networks, while a temporal approach captures more accurately the system performance.

Link prediction systems have been largely adopted to recommend new friends in online social networks using data about social interactions. With the soaring adoption of location-based social services it becomes possible to take advantage of an additional source of information: the places people visit. In this paper we study the problem of designing a link prediction system for online location-based social networks. We have gathered extensive data about one of these services, Gowalla, with periodic snapshots to capture its temporal evolution. We study the link prediction space, finding that about 30% of new links are added among "place-friends", i.e., among users who visit the same places. We show how this prediction space can be made 15 times smaller, while still 66% of future connections can be discovered. Thus, we define new prediction features based on the properties of the places visited by users which are able to discriminate potential future links among them.

The interactions and social relations among users in workplaces have been studied by many generations of social psychologists. There is evidence that groups of users that interact more in workplaces are more productive. However, it is still hard for social scientists to capture fine-grained data about phenomena of this kind and to find the right means to facilitate interaction. It is also difficult for users to keep track of their level of sociability with colleagues. While mobile phones offer a fantastic platform for harvesting long term and fine grained data, they also pose challenges: battery power is limited and needs to be traded-off for sensor reading accuracy and data transmission, while energy costs in processing computationally intensive tasks are high.

As content that users access on their mobile devices becomes bulkier, opportunistic networking is becoming a potential complement to centralised and infrastructure based downloads. We study how users share items of mutual interest with each other with a simple model based on a `networked urn process'. We investigate the effect of different content sharing policies upon a multi-category set of items. We find that the process of sharing mutual interests inherently disproportionately reinforces category replication disparity, i.e., the most popular categories become proportionally even more numerous. These findings uncover a major hurdle in the creation of automatic opportunistic file sharing between users. Even if users altruistically sacrifice battery power and network resources to share content not relevant to them, overall, the system may not be able to fairly distribute items that belong to niche categories.

Malicious mobile phone worms spread between devices via short-range Bluetooth contacts, similar to the propagation of human and other biological viruses. Recent work has employed models from epidemiology and complex networks to analyse the spread of malware and the effect of patching specific nodes. These approaches have adopted a static view of the mobile networks, i.e., by aggregating all the edges that appear over time, which leads to an approximate representation of the real interactions: instead, these networks are inherently dynamic and the edge appearance and disappearance are highly influenced by the ordering of the human contacts, something which is not captured at all by existing complex network measures. In this paper we first study how the blocking of malware propagation through immunisation of key nodes (even if carefully chosen through static or temporal betweenness centrality metrics) is ineffective: this is due to the richness of alternative paths in these networks. Then we introduce a time-aware containment strategy that spreads a patch message starting from nodes with high temporal closeness centrality and show its effectiveness using three real-world datasets. Temporal closeness allows the identification of nodes able to reach most nodes quickly: we show that this scheme reduces the cellular network resource consumption and associated costs, achieving, at the same time, complete containment of malware in a limited amount of time.

Providers such as YouTube offer easy access to multimedia content to millions, generating high bandwidth and storage demand on the Content Delivery Networks they rely upon. More and more, the diffusion of this content happens on online social networks such as Facebook and Twitter, where social cascades can be observed when users increasingly repost links they have received from others. In this paper we describe how geographic information extracted from social cascades can be exploited to improve caching of multimedia files in a Content Delivery Network. We take advantage of the fact that social cascades can propagate in a geographically limited area to discern whether an item is spreading locally or globally. This informs cache replacement policies, which utilize this information to ensure that content relevant to a cascade is kept close to the users who may be interested in it. We validate our approach by using a novel dataset which combines social interaction data with geographic information: we track social cascades of YouTube links over Twitter and build a proof-of-concept geographic model of a realistic distributed Content Delivery Network. Our performance evaluation shows that we are able to improve cache hits with respect to cache policies without geographic and social information.

Road congestion results in a huge waste of time and productivity for millions of people. A possible way to deal with this problem is to have transportation authorities distribute traffic information to drivers, which, in turn, can decide (or be aided by a navigator) to route around congested areas. Such traffic information can be gathered by relying on static sensors placed at specific road locations (e.g., induction loops and video cameras) or by having single vehicles report their location, speed, and travel time. While the former approach has been widely exploited, the latter has come about only more recently; consequently, its potential is less understood. For this reason, in this paper, we study a realistic test case that allows the evaluation of the effectiveness of such a solution. As part of this process, (a) we designed a system that allows vehicles to crowd-source traffic information in an ad hoc manner, allowing them to dynamically reroute based on individually collected traffic information; (b) we implemented a realistic network-mobility simulator that allowed us to evaluate such a model; and (c) we performed a case study that evaluates whether such a decentralized system can help drivers to minimize trip times, which is the main focus of this paper. This study is based on traffic survey data from Portland, OR, and our results indicate that such navigation systems can indeed greatly improve traffic flow. Finally, to test the feasibility of our approach, we implemented our system and ran some real experiments at UCLA's C-Vet test bed.

Today's mobile phones represent a rich and powerful computing platform, given their sensing, processing and communication capabilities. Phones are also part of the everyday life of billions of people, and therefore represent an exceptionally suitable tool for conducting social and psychological experiments in an unobtrusive way.

Nowadays, the navigation systems available on cars are becoming more and more sophisticated. They greatly improve the experience of drivers and passengers by enabling them to receive map and traffic updates, news feeds, advertisements, media files, etc. Unfortunately, the bandwidth available to each vehicle with the current technology is severely limited. There have been many reports on the inability of 3G networks to cope with large size file downloads, especially in dense and mobile settings. A possible alternative is provided by WiFi access points (APs) that are being installed in several countries along the main routes and in popular areas. Although this approach significantly increases the available bandwidth, it still does not provide a fully satisfactory solution due to the limited transmission range (usually a few hundred meters). In this paper we present a novel routing protocol, based on opportunistic vehicle to vehicle communication, to enable efficient multi-hop routing capabilities between mobile vehicles and APs. Unlike prior work, this protocol fully supports two- way communication, i.e., the traditional vehicle-to-AP as well as the more challenging AP-to-vehicle. We leverage the information offered by the navigation system in terms of final destination and path, to i) route packets to the closest AP and ii) to route replies back to the moving vehicle efficiently.

As sensor network technologies become more mature, they are increasingly being applied to a wide variety of applications, ranging from agricultural sensing to cattle, oceanic and volcanic monitoring. Significant efforts have been made in deploying and testing sensor networks resulting in unprecedented sensing capabilities. A key challenge has become how to make these emerging wireless sensor networks more sustainable and easier to maintain over increasingly prolonged deployments.

An increasing number of sensor networks have been deployed to monitor a variety of conditions and situations. At the same time, more and more applications are starting to rely on the data from sensor networks to provide users with (near) real-time information and conditions. This increasing demand of users for accurate information about natural and surrounding phoenomena is creating a business case for application providers.

Content-based information dissemination has a potential number of applications in vehicular networking, including advertising, traffic and parking notifications and emergency announcements. In this paper we describe a protocol for content based information dissemination in hybrid (i.e., partially structureless) vehicular networks. The protocol allows content to ldquostickrdquo to areas where vehicles need to receive it. The vehicle's subscriptions indicate the driver's interests about types of content and are used to filter and route information to affected vehicles. The publications, generated by other vehicles or by central servers, are first routed into the area, then continuously propagated for a specified time interval. The protocol takes advantage of both the infrastructure (i.e., wireless base stations), if this exists, and the decentralized vehicle-to-vehicle communication technologies. We evaluate our approach by simulation over a number of realistic vehicular traces based scenarios. Results show that our protocol achieves high message delivery while introducing low overhead, even in scenarios where no infrastructure is available.

Wireless Sensor Networks enable scientists to collect information about the environment with a granularity unseen before, while providing numerous challenges to software designers. Since sensor devices are often powered by small batteries, which take considerable effort to replace, it is of major importance to use energy carefully. We present two efficient ways of extending the lifetime of such systems: 1. an adaptive duty cycling protocol and 2. an adaptive data management protocol. Further, we present some details of our deployed sensor network in Wytham Woods, Oxfordshire.

Most of the existing research work in mobile ad hoc networking is based on the assumption that a path exists between the sender and the receiver. On the other hand, applications of decentralised mobile systems are often characterised by network partitions. As a consequence delay tolerant networking research has received considerable attention in the recent years as a means to obviate to the gap between ad hoc network research and real applications. In this paper we present the design, implementation and evaluation of the context-aware adaptive routing (CAR) protocol for delay tolerant unicast communication in intermittently connected mobile ad hoc networks. The protocol is based on the idea of exploiting nodes as carriers of messages among network partitions to achieve delivery. The choice of the best carrier is made using Kalman filter based prediction techniques and utility theory. We discuss the implementation of CAR over an opportunistic networking framework, outlining possible applications of the general principles at the basis of the proposed approach. The large scale performance of the CAR protocol are evaluated using simulations based on a social network founded mobility model, a purely random one and real traces from Dartmouth College.

Sensor networks are now enabling the monitoring of various environmental phenomena with more accuracy than the previous labour intensive and less technological solutions. This paper is concerned with the application of opportunistic networking techniques to wildlife monitoring, where the sensors are attached to animals moving in their habitat. We present seal-2-seal, a novel protocol for logging of node (i.e., animal) contacts in mobile networks and for dissemination of that information to sinks for further analysis. The protocol utilises an efficient data summary mechanism to reduce the amount of information that needs to be transmitted, thus reducing energy consumption. To evaluate the performance of the protocol, we implemented it for the Contiki operating system on sensor devices and ran simulations based on real-life mobility traces using the Cooja emulator.

People living in urban areas spend a considerable amount of time on public transport, for example, commuting to/from work. During these periods, opportunities for inter-personal networking present themselves, as many members of the public now carry electronic devices equipped with Bluetooth or other wireless technology. Using these devices, individuals can share content (e.g., music, news and video clips) with fellow travellers that are on the same train or bus. Transferring media content takes time; in order to maximise the chances of successful downloads, users should identify neighbours that possess desirable content and who will travel with them for long-enough periods. In this paper, we propose a user-centric prediction scheme that collects historical colocation information to determine the best content sources. The scheme works on the assumption that people have a high degree of regularity in their movements. We first validate this assumption on a real dataset, that consists of traces of people moving in a large city's mass transit system. We then demonstrate experimentally on these traces that our prediction scheme significantly improves communication efficiency, when compared to a memory(history)-less source selection scheme.

Developing protocols and applications for opportunistic networking can represent a daunting task given the many aspects that must be taken into consideration, such as intermittent connectivity, smart choice among multiple interfaces and intelligent data storage. The implementation of these protocols can be based on generic layer-less communication frameworks that provide programming abstractions for the extraction and analysis of social, colocation and mobility information and allows data exchange by means of heterogeneous devices. We propose Gently, a novel fully implemented solution which combines techniques of context awareness and social knowledge to concretely solve issues related to opportunistic forwarding. More precisely, Gently is born as the combination of the Context-aware Adaptive Routing (CAR) and the socially aware LABEL protocol. We discuss the implementation of our solution on top of the layer-less Haggle framework presenting the key design choices and the lessons learnt.

Applications involving the dissemination of information directly relevant to humans (e.g., service advertising, news spreading, environmental alerts) often rely on publish-subscribe, in which the network delivers a published message only to the nodes whose subscribed interests match it. In principle, publish- subscribe is particularly useful in mobile environments, since it minimizes the coupling among communication parties. However, to the best of our knowledge, none of the (few) works that tackled publish-subscribe in mobile environments has yet addressed intermittently-connected human networks. Socially-related people tend to be co-located quite regularly. This characteristic can be exploited to drive forwarding decisions in the interest-based routing layer supporting the publish-subscribe network, yielding not only improved performance but also the ability to overcome high rates of mobility and long-lasting disconnections. In this paper we propose SocialCast, a routing framework for publish-subscribe that exploits predictions based on metrics of social interaction (e.g., patterns of movements among communities) to identify the best information carriers. We highlight the principles underlying our protocol, illustrate its operation, and evaluate its performance using a mobility model based on a social network validated with real human mobility traces. The evaluation shows that prediction of colocation and node mobility allow for maintaining a very high and steady event delivery with low overhead and latency, despite the variation in density, number of replicas per message or speed.

The integration of clusters of computers into computational grids has recently gained the attention of many computational scientists. While considerable progress has been made in building middleware and workflow tools that facilitate the sharing of compute resources, little attention has been paid to grid scheduling and load balancing techniques to reduce job waiting time. Based on a detailed analysis of usage characteristics of an existing grid that involves a large CPU cluster, we observe that grid scheduling decisions can be significantly improved if the characteristics of current usage patterns are understood and extrapolated into the future. The paper describes an architecture and an implementation for a predictive grid scheduling framework which relies on Kalman filter theory to predict future CPU resource utilisation. By way of replicated experiments we demonstrate that the prediction achieves a precision within 15-20% of the utilisation later observed and can significantly improve scheduling quality, compared to approaches that only take into account current load indicators.

Sensors are now embedded in all sorts of devices (such as phones and PDAs) and attached to many moving things such as robots, vehicles and animals. The collection of data from these mobile sensors presents challenges related to the variability of the topology of the sensor network and the need to limit communication (for energy or bandwidth saving). Fortunately, the data collected, despite considerable, is often delay tolerant and its delivery to the sinks is, in most cases, not time critical. We have devised SCAR, a context aware opportunistic routing protocol which allows efficient routing of sensor data to sinks, through selection of best paths by prediction over movement patterns and current battery level of nodes. In this paper we present the implementation of the protocol in Contiki and validate the approach through the use of the COOJA simulator with mobility traces provided by the ZebraNet Project. We compare the performance with respect to random choice based dissemination.

Large scale market penetration of electronic devices equipped with Bluetooth technology now gives the ability to share content (such as music or video clips) between members of the public in a decentralised manner. Achieved using opportunistic connections, formed when they are colocated, in environments where Internet connectivity is expensive or unreliable, such as urban buses, train rides and coffee shops. Most people have a high degree of regularity in their movements (such as a daily commute), including repeated contacts with others possessing similar seasonal movement patterns. We argue that this behaviour can be exploited in connection selection, and outline a system for the identification of long-term companions and sources that have previously provided quality content, in order to maximise the successful receipt of content flies. We utilise actual traces and existing mobility models to validate our approach, and show how consideration of the colocation history and the quality of previous data transfers leads to more successful sharing of content in realistic scenarios.

Due to the inherent nature of their heterogeneity, resource scarcity and dynamism, the provision of middleware for future networked embedded environments is a challenging task. In this paper we present a middleware approach that addresses these key challenges; we also discuss its application in a realistic networked embedded environment. Our application scenario involves fire management in a road tunnel that is instrumented with networked sensor and actuator devices. These devices are able to reconfigure their behaviour and their information dissemination strategies as they become damaged under emergency conditions, and firefighters are able to coordinate their operations and manage sensors and actuators through dynamic reprogramming. Our supporting middleware is based on a two-level architecture: the foundation is a language-independent, component-based programming model that is sufficiently minimal to run on any of the devices typically found in networked embedded environments. Above this is a layer of software components that offer the necessary middleware functionality. Rather than providing a monolithic middleware 'layer', we separate orthogonal areas of middleware functionality into self-contained components that can be selectively and individually deployed according to current resource constraints and application needs. Crucially, the set of such components can be updated at runtime to provide the basis of a highly dynamic and reconfigurable system

The testing of the performance of opportunistic communication protocols and applications is usually done through simulation as i) deployments are expensive and should be left to the final stage of the development process, and ii) the number of varying parameters in thesesystems is so high that it would be very hard to conduct thorough testing of all the functionality within a single deployment. Therefore, protocols and applications are often plugged into mobility simulators to test their performance; however, until recently, most of the testing has been conducted with random mobility models which do not mirror reality. Furthermore, despite disconnections playing a veryprominent role in the performance of any opportunistic mobile system, most models do not really account for it. A different approach to testing is the use of real traces of movement collected in specific domains as test cases. These cases, however, do not allow for flexible performance testing, as they are specific for a given scenario withfixed connectivity properties.

Vehicular networks can be seen as an example of hybrid delay tolerant network where a mixture of infostations and vehicles can be used to geographically route the information messages to the right location. In this paper we present a forwarding protocol which exploits both the opportunistic nature and the inherent characteristics of the vehicular network in terms of mobility patterns and encounters, and the geographical information present in navigator systems of vehicles. We also report about our evaluation of the protocol over a simulator using realistic vehicular traces and in comparison with other geographical routing protocols.

Sensor devices are being embedded in all sorts of items including vehicles, furniture but also animal and human bodies through health monitors and tagging techniques. The collection of the information generated by these devices is a challenging task as the data results in enormous amounts and the sensors have scarce resources (especially in terms of energy for the forwarding of the data). Fortunately, the data is often delay tolerant and its delivery to the sinks is, in most cases, not time critical.This paper tackles the problem of the delivery of mobile sensor data to sinks. We devise a Sensor Context-Aware Routing protocol (SCAR), which exploits movement and resource prediction techniques to smartly forward data towards the right direction at any point in time. In order to cope with the possibly frequent sensor faults, we also adopt a multi-path routing approach which increases the reliability.

Next generation embedded systems will be composed of large numbers of heterogeneous devices. These will typically be resource-constrained, will use different operating systems, and will be connected through different types of network interfaces. Additionally, they may be mobile and/or form ad-hoc networks with their peers, and will need to be adaptive to changing conditions based on context-awareness. As an example of these system we consider disaster recovery scenarios where large numbers of different devices need to interconnect in an ad-hoc manner. In this respect, our goal is the provisioning of a middleware framework for such system environments. Our approach is based on a small and efficient middleware kernel supporting highly modularised and customisable component-based middleware services. These services can be tailored for specific embedded environments, and are runtime-reconfigurable to support adaptivity. This paper describes a demonstration that highlights some of the features available in our middleware. In particular, we focus on heterogeneity handling by showing our middleware running on resource-rich as well as resource-constrained devices, and on adaptivity features by demonstrating runtime reprogramming and on-the-fly component deployment

Traditional middleware primitives offer very elementary information dissemination mechanisms, which, in the case of a decentralized and dynamic network such as a mobile ad hoc network, do not offer the ability to control the information spreading. Control over information dissemination could instead be very critical especially in terms of lifetime of the network. Gossip-based communication and epidemic-style algorithms, which are based on a store and forward approach, have been proposed to obtain message dissemination with probabilistic guarantees and lower overheads. However, epidemic algorithms have never been used to allow designers to control the spreading of the information depending on the desired reliability and the network structure. In this paper, we present a middleware for ad hoc networking, which uses epidemic-style information dissemination techniques to tune the reliability of the communication in mobile ad hoc networks. The approach is based on recent results of complex networks theory; the novelty of our idea resides in the evaluation and the exploitation of the structure of the underlying network for the automatic tuning of the dissemination process and its use in the design of the API offered by the middleware. We present a detailed analytical model supported by several simulation results

No abstract available.

Delay tolerant and mobile ad hoc networks present considerable challenges to the development of protocols and systems. In particular, the challenge of being able to cope with their variability is an important one: sometimes the rate at which these systems change in terms of context (such as topology, collocation duration and availability and quality of the local resources) is very high and these changes are unpredictable. Knowledge of context could be used to improve the performance of such systems. For example, context information may be extremely useful to make routing decisions. Some recent approaches have successfully exploited context and prediction on future context condition to improve performance, for instance in terms of delivery ratio and delay. In this paper, we present a model of predictability of context information and the design of a generic component implementing it. The component can be used to decide if (or in which measure) context is predictable. The model is based on the analysis of the time series representing the context information. In order to show how the component can be used in practice, we describe its integration in our context-aware adaptive routing (CAR) protocol

Mobile computing devices, such as personal digital assistants and mobile phones, are becoming increasingly popular, smaller, and more capable. We argue that mobile systems should be able to adapt to changing requirements and execution environments. Adaptation requires the ability-to reconfigure the deployed code base on a mobile device. Such reconfiguration is considerably simplified if mobile applications are component-oriented rather than monolithic blocks of code. We present the SATIN (system adaptation targeting integrated networks) component metamodel, a lightweight local component metamodel that offers the flexible use of logical mobility primitives to reconfigure the software system by dynamically transferring code. The metamodel is implemented in the SATIN middleware system, a component-based mobile computing middleware that uses the mobility primitives defined in the metamodel to reconfigure both itself and applications that it hosts. We demonstrate the suitability of SATIN in terms of lightweightedness, flexibility, and reusability for the creation of adaptable mobile systems by using it to implement, port, and evaluate a number of existing and new applications, including an active network platform developed for satellite communication at the European space agency. These applications exhibit different aspects of adaptation and demonstrate the flexibility of the approach and the advantages gained

In this paper the RUNES approach to the development of software for networked embedded systems is described. There is a need for a program platform with abstractions that are able to span the full range of heterogeneous embedded systems, and which also offers consistent mechanisms with which to configure, deploy, and dynamically reconfigure networked embedded systems software. This paper discusses the need of such a programming platform. The work is being carried out in the context of the RUNES project (reconfigurable, ubiquitous, and networked embedded systems) which has the general goal of developing an architecture for networked embedded systems that encompasses dedicated radio layers, networks

The vast majority of mobile ad hoc networking research makes a very large assumption - that communication can only take place between nodes that are simultaneously accessible within the same connected cloud (i.e., that communication is synchronous). In reality, this assumption is likely to be a poor one, particularly for sparsely or irregularly populated environments. We present the context-aware routing (CAR) algorithm. CAR is a novel approach to the provision of asynchronous communication in partially-connected mobile ad hoc networks, based on the intelligent placement of messages. We discuss the details of the algorithm, and then present simulation results demonstrating that it is possible for nodes to exploit context information in making local decisions that lead to good delivery ratios and latencies with small overheads.

Almost all work on mobile ad hoc networks relies on simulations, which, in turn, rely on realistic movement models for their credibility. Since there is a total absence of realistic data in the public domain, synthetic models for movement pattern generation must be used and the most widely used models are currently very simplistic, the focus being ease of implementation rather than soundness of foundation. Whilst it would be preferable to have models that better reflect the movement of real users, it is currently impossible to validate any movement model against real data. However, it is lazy to conclude from this that all models are equally likely to be invalid so any will do.We note that movement is strongly affected by the needs of humans to socialise in one form or another. Fortunately, humans are known to associate in particular ways that can be mathematically modelled, and that are likely to bias their movement patterns. Thus, we propose a new mobility model that is founded on social network theory, because this has empirically been shown to be useful as a means of describing human relationships. In particular, the model allows collections of hosts to be grouped together in a way that is based on social relationships among the individuals. This grouping is only then mapped to a topographical space, with topography biased by the strength of social tie.We discuss the implementation of this mobility model and we evaluate emergent properties of the generated networks. In particular, we show that grouping mechanism strongly influences the probability distribution of the average degree (i.e., the average number of neighbours of a host) in the simulated network.

Mobile devices, such as mobile phones and personal digital assistants, have gained wide-spread popularity. These devices will increasingly be networked, thus enabling the construction of distributed applications that have to adapt to changes in context, such as variations in network bandwidth, battery power, connectivity, reachability of services and hosts, etc. In this paper, we describe CARISMA, a mobile computing middleware which exploits the principle of reflection to enhance the construction of adaptive and context-aware mobile applications. The middleware provides software engineers with primitives to describe how context changes should be handled using policies. These policies may conflict. We classify the different types of conflicts that may arise in mobile computing and argue that conflicts cannot be resolved statically at the time applications are designed, but, rather, need to be resolved at execution time. We demonstrate a method by which policy conflicts can be handled; this method uses a microeconomic approach that relies on a particular type of sealed-bid auction. We describe how this method is implemented in the CARISMA middleware architecture and sketch a distributed context-aware application for mobile devices to illustrate how the method works in practice. We show, by way of a systematic performance evaluation, that conflict resolution does not imply undue overheads, before comparing our research to related work and concluding the paper.

No abstract available.

Mobile devices, such as mobile phones and personal digital assistants, have gained wide-spread popularity. These devices will increasingly be networked, thus enabling the construction of distributed mobile applications. These have to adapt to changes in context, such as variations in network bandwidth, exhaustion of battery power or reachability of services on other devices. We show how the construction of adaptive and context-aware mobile applications can be supported using a reflective middleware. The middleware provides software engineers with primitives to describe how context changes are handled using policies. These policies may conflict. In this paper, we classify the different types of conflicts that may arise in mobile computing. We argue that conflicts cannot be resolved statically at the time applications are designed, but, rather, need to be resolved at execution time. We demonstrate a method by which these policy conflicts can be treated. This method uses a micro-economic approach that relies on a particular type of sealed-bid auction.

Summary form only given. Middleware solutions for wired distributed systems cannot be used in a mobile setting, as mobile applications impose new requirements that run counter to the principle of transparency on which current middleware systems have been built. We propose the use of reflection capabilities and meta-data to pave the way for a new generation of middleware platforms designed to support mobility.

An increasing number of distributed applications will be written for mobile hosts, such as laptop computers, third generation mobile phones, personal digital assistants, watches and the like, with focus on peer-to-peer collaboration. Application engineers have to deal with a new set of problems caused by mobility, such as low bandwidth, context changes or loss of connectivity. During disconnection, independently from each others, users will typically update local replicas of shared data, possibly generated by peers. The resulting inconsistent replicas need to be reconciled upon re-connection. To support building mobile applications that use both replication and reconciliation over ad-hoc networks, we have designed XMIDDLE, a peer-to-peer middleware that targets mobile computing settings. In this paper we describe XMIDDLE and show how reflection capabilities are used to allow application engineers to influence replication and reconciliation techniques. XMIDDLE enables the transparent sharing of XML documents across heterogeneous mobile peers, allowing online and off-line access to data.

We demonstrate how XML and related technologies can be used for code mobility at any granularity, thus overcoming the restrictions of existing approaches. By not fixing a particular granularity for mobile code, we enable complete programs as well as individual lines of code to be sent across the network. We define the concept of incremental code mobility as the ability to migrate and add, remove, or replace code fragments (i.e., increments) in a remote program. The combination of fine-grained and incremental migration achieves a previously unavailable degree of flexibility. We examine the application of incremental and fine-grained code migration to a variety of domains, including user interface management, application management on mobile thin clients, for example PDAs, and management of distributed documents.

New computing paradigms for network-aware applications need specification languages able to deal with the features of mobile code-based systems. A coordination language provides a formal framework in which the interaction of active entities can be expressed. A coordination language deals with the creation and destruction of code or complex agents, their communication activites, as well as their distribution and mobility in space. We show how the coordination language PoliS offers a flexible basis for the description and the automatic analysis of architectures of systems including mobile entities. Polis is based on multiple tuple spaces and offers a basis for defining, studying, and controlling mobility as it allows decoupling mobile entities from their environments both in space and in time. The pattern-matching mechanism adopted for communication helps in abstracting from addressing issues. We have developed a model-checking technique for the automatic analysis of PoliS specifications. In the article we show how this technique can be applied to mobile code-based systems

No abstract available.

The use of the World Wide Web as a communication medium for knowledge engineers and software designers is limited by the lack of tools for writing, sharing, and verifying documents written with design notations. For instance, the Z language has a rich set of mathematical characters, and requires graphic-rich boxes and schemas for structuring a specification document. It is difficult to integrate Z specifications and text on WWW pages written with HTML, and traditional tools are not suited for the task. On the other hand, a newly proposed standard for markup languages, namely XML, allows one to define any set of markup elements; hence, it is suitable for describing any kind of notation. Unfortunately, the proposed standard for rendering XML documents, namely XSL, provides for text-only (although sophisticated) rendering of XML documents, and thus it cannot be used for more complex notations. We present a Java-based tool for applying any notation to elements of XML documents. These XML documents can thus be shown on current-generation WWW browsers with Java capabilities. A complete package for displaying Z specifications has been implemented and integrated with standard text parts. Being a complete rendering engine, text parts and Z specifications can be freely intermixed, and all the standard features of XML (including HTML links and form elements) are available outside and inside Z specifications. Furthermore, the extensibility of our engine allows any additional notations to be supported and integrated with the ones we describe.