Showing papers presented at "Static Analysis Symposium in 2019"

Analyzing Deep Neural Networks with Symbolic Propagation: Towards Higher Precision and Faster Verification

Abstract: Deep neural networks (DNNs) have been shown lack of robustness, as they are vulnerable to small perturbations on the inputs, which has led to safety concerns on applying DNNs to safety-critical domains. Several verification approaches have been developed to automatically prove or disprove safety properties for DNNs. However, these approaches suffer from either the scalability problem, i.e., only small DNNs can be handled, or the precision problem, i.e., the obtained bounds are loose. This paper improves on a recent proposal of analyzing DNNs through the classic abstract interpretation technique, by a novel symbolic propagation technique. More specifically, the activation values of neurons are represented symbolically and propagated forwardly from the input layer to the output layer, on top of abstract domains. We show that our approach can achieve significantly higher precision and thus can prove more properties than using only abstract domains. Moreover, we show that the bounds derived from our approach on the hidden neurons, when applied to a state-of-the-art SMT based verification tool, can improve its performance. We implement our approach into a software tool and validate it over a few DNNs trained on benchmark datasets such as MNIST, etc.

LoRa-based Visual Monitoring Scheme for Agriculture IoT

Abstract: LoRa is a low-power wide-range wireless networking technology suitable for low-rate long-range applications in the Internet of Things (IoT). For example in the agriculture industry, LoRa-based environmental sensing system enables farmers to remotely monitor the status of a large farm in near real-time. However, there had been only a few explorations to transfer multimedia data such as images or video using LoRa because of its low data rate and restricted bandwidth. To this end, we introduce a novel system to transmit continuous images taken from a camera on a static environment through LoRa. The key challenge is to reduce the amount of transmitted data while preserving the image quality and the quality of service delivered to the application. We develop a technique that splits image to grid patches, and transmits only the modified area of an image based on their dissimilarity measure. We implement and evaluate our scheme on a real LoRa device to show its performance and image quality.

IEEE 1451 Smart Sensor Digital Twin Federation for IoT/CPS Research

Abstract: Cyber-physical systems (CPS) are smart systems that include engineered, interacting networks of physical and cyber components. CPS are concerned with the collaborative and interactive activities between cyber and physical components through sensing and actuation. The Institute of Electrical and Electronics Engineers (IEEE) 1451, a family of smart transducer interface standards, defines a set of open, common, network- independent communication interfaces for smart transducers (sensors or actuators) to achieve sensor data interoperability between cyber and physical components of CPS. The IEEE 1516 high-level architecture (HLA) is a standard for the modeling and simulation of distributed, heterogeneous processes. CPS experimentation often requires the integration of different domain-specific simulation tools into a common platform using a method called co-simulation. In order to facilitate a wide range of CPS simulations and experiments, the National Institute of Standards and Technology (NIST) has developed a universal CPS environment for federation (UCEF), which is a toolset for designing and implementing federated, collaborative, and interactive experiments for CPS using HLA.This paper introduces an IEEE 1451 smart sensor digital twin federation, which has been developed using UCEF. The digital twin is a digital simulator or digital replica of a real IEEE 1451 smart sensor. The digital twin emulates both desired, non-linear behaviors and failure modes to simulate a real sensor in the field. This federation consists of three federates: an IEEE 1451 smart sensor digital twin federate (DTF), an IEEE 1451 digital twin tester federate (DTTF), and a federation experiment manager. In this federation, the DTTF can communicate with the DTF via HLA interactions that represent IEEE P1451.1 request and response messages. One federated experiment using three instances of IEEE 1451 digital twins is described in detail and the experimental results are provided in this paper. This federation provides a foundation for CPS research and experiments in the NIST CPS testbed.

Challenges in aggregation of heterogeneous sensors for Autonomous Driving Systems

Abstract: This paper is highlighting the complexity associated with sensors aggregation in an autonomous driving system regarding aspects of sensing technologies, space coverage around ego vehicle, detection issues and software architecture. Building Advanced Driving Assistance systems (ADAS) and autonomous driving functions of Level 3 ([1]) and above mandates that these sources of complexity are properly addressed by novel approaches in dealing with sensors raw data and produced features. The paper tries to depict a structured view of the sources of complexity related to the driving scene that must be apprehended, then goes through situational examples for each theme and pitches key directions for improvement.

A LoRa-based IoT Sensor Node for Waste Management Based on a Customized Ultrasonic Transceiver

Abstract: This paper discusses the use of an integrated ultrasonic transducer for the realization of a LoRa-based Smart Bin architecture, to be employed for waste management in the Smart Cities context. In particular, the paper analyses the usability of waterproof ultrasonic sensors for the measurement of waste level inside a trash bin: in this context, a solution is described to increase the directionality of the sensor. Indeed, ultrasonic sensors used in common off-the-shelf waterproof ranging systems are characterized by a wide radiation lobe, corresponding to an aperture angle up to 70°: this means that, when employed in trash bins, these sensors detect the presence of the lateral surface of the bin and are not able to detect the garbage layer top or its position in the bin. In this paper we propose a tailored cap, designed to be mounted on a commercial ultrasonic sensor, able to reduce the detection angle down to around 28° degrees. The modified sensor can thus be exploited in this application. The proposed sensing structure is integrated into a sensor node provided with a LoRa transmission module, that allows to employ it for the development of city-scale monitoring infrastructures.

Collaborative Mapping with IoE-based Heterogeneous Vehicles for Enhanced Situational Awareness

Abstract: The development of autonomous vehicles or advanced driving assistance platforms has had a great leap forward to get closer to human daily life over the last decade. Nevertheless, it is still challenging to achieve an efficient and fully autonomous vehicle or driving assistance platform due to many strict requirements and complex situations or unknown environments. One of the main remaining challenges is a robust situation awareness in autonomous vehicles when the environment is unknoen. An autonomous system with a poor situation awareness due to low quantity or quality of data may directly or indirectly cause serious consequences. For instance, a person’s life might be at risk due to a delay caused by a long or incorrect path planning of an autonomous ambulance. Internet of Everything (IoE) is currently becoming a prominent technology for many applications such as automation. In this paper, we propose an IoE-based architecture consisting of a heterogeneous team of cars and drones for enhancing situational awareness in autonomous cars, especially when dealing with critical cases of natural disasters. In particular, we show how an autonomous car can plan in advance the possible paths to a given destination, and send orders to other vehicles. These, in turn, perform terrain reconnaissance for avoiding obstacles and dealing with difficult situations. Together with a map merging algorithm deployed into the team autonomous vehicles, the proposed architecture can help to save traveling distance and time significantly in case of complex scenarios.

Temporal Verification of Programs via First-Order Fixpoint Logic

Abstract: This paper presents a novel program verification method based on Mu-Arithmetic, a first-order logic with integer arithmetic and predicate-level least/greatest fixpoints. We first show that linear-time temporal property verification of first-order recursive programs can be reduced to the validity checking of a Mu-Arithmetic formula. We also propose a method for checking the validity of Mu-Arithmetic formulas. The method generalizes a reduction from termination verification to safety property verification and reduces validity of a Mu-Arithmetic formula to satisfiability of CHC, which can then be solved by using off-the-shelf CHC solvers. We have implemented an automated prover for Mu-Arithmetic based on the proposed method. By combining the automated prover with a known reduction and the reduction from first-order recursive programs above, we obtain: (i) for while-programs, an automated verification method for arbitrary properties expressible in the modal \(\mu \)-calculus, and (ii) for first-order recursive programs, an automated verification method for arbitrary linear-time properties expressible using Buchi automata. We have applied our Mu-Arithmetic prover to formulas obtained from various verification problems and obtained promising experimental results.

Green Inertial Sensors based on Bacterial Cellulose

Abstract: In this paper, an environmentally friendly inertial sensor is proposed and characterized. The sensor has been realized by using bacterial cellulose, impregnated by using ionic liquids, and covered by conducting polymers, as the electrodes. The composite has mechano-electric transduction capabilities, which are exploited for realizing a generating sensor. More specifically, the system is mounted in cantilever configuration and used for sensing the anchor acceleration. Results of the chemical and transduction characterization of the composite as an accelerometer are reported in the paper.

LoRa-based Measurement Station for Water Quality Monitoring: Case of Botanical Garden Pool

Abstract: The advent of Internet of Things (IoT) has made easier to build number of applications. In fact, the remote monitoring or sensing has been facilitated by the IoT. A number of sensor nodes with a networking capability can be deployed in order to have an ad hoc or continuous monitoring system. However, physicists at UCAD’s Faculty of Science still use traditional means to collect their water quality data from a pool in the faculty’s Botanical Garden with on-site measurements. This pool is used for aquaculture and to study some aquatic species. In this paper, we present a water quality monitoring system through LoRa transmission. It’s a low cost infrastructure composed of a remote station for real-time data collection and a web platform for visualization and exploitation. To evaluate the reliability and efficiency of the system, we perform some performance tests and the results are also presented.

Broken-Rotor-Bar Detection Through STFT and Windowing Functions

Abstract: Monitoring systems for induction motor fault detection have been of great interest for industry in recent years. Those based on the analysis of the electric current supply are the most common; since the current signal provides relevant information regarding the motor condition. The Fourier transform is a popular method for performing this analysis; however, it does not allow seeing the evolution of the signal through time. On the other hand, the Short-Time Fourier Transform (STFT) has proven to be useful for analyzing the signal-frequency evolution through time; unfortunately, its timefrequency representation heavily depends on the used-window size, compromising its resolution. In this work, a method based on window functions and STFT is proposed for enhancing the resolution of the time-frequency representation. The proposed approach is applied on the electric current signal supplied to an induction motor, improving the visualization of characteristicfrequency components of broken rotor bars.

Performance Evaluation of IoT Encryption Algorithms: Memory, Timing, and Energy

Abstract: Security in the Internet of Things is a crucial aspect and a lot of studies are focused on modular and scalable encryption algorithms. Resource constraints at the edge nodes of an IoT system require lightweight encryption algorithms. A comparative study of AES with and without hardware accelerators and XTEA is performed to analyze the performance of the algorithms in terms of memory, power and execution time and assess the feasibility of using XTEA in low resource embedded platforms. Although the hardware accelerated AES was fastest (0.5 ms) and consequently required the least amount of energy (0.01 mJ) out of the three, the execution time (1.25 ms) and energy consumed (0.024 mJ) by XTEA was comparatively close and can be a feasible encryption algorithm for low resource microcontrollers that do not have the resources to support AES implementation in software or lack a hardware accelerator. Software implementation of AES on 8-bit PIC architecture required 7538 bytes whereas XTEA required only 1184 bytes of program memory, leaving enough space for application firmware.

Revisiting Polyhedral Analysis for Hybrid Systems

Abstract: Thanks to significant progress in the adopted implementation techniques, the recent years have witnessed a renewed interest in the development of analysis tools based on the domain of convex polyhedra. In this paper we revisit the application of this abstract domain to the case of reachability analysis for hybrid systems, focusing on the lesson learned during the development of the tool PHAVerLite. In particular, we motivate the implementation of specialized versions of several well known abstract operators, as well as the adoption of a heuristic technique (boxed polyhedra) for the handling of finite collections of polyhedra, showing their impact on the efficiency of the analysis tool.

Multiphase-Linear Ranking Functions and Their Relation to Recurrent Sets

Abstract: Multiphase ranking functions (M\(\varPhi \)RFs) are used to prove termination of loops in which the computation progresses through a number of phases. They consist of linear functions \(\langle f_1,\ldots ,f_d \rangle \) where \(f_i\) decreases during the ith phase. This work provides new insights regarding M\(\varPhi \)RFs for loops described by a conjunction of linear constraints (\( SLC \) loops). In particular, we consider the existence problem (does a given \( SLC \) loop admit a M\(\varPhi \)RF). The decidability and complexity of the problem, in the case that d is restricted by an input parameter, have been settled in recent work, while in this paper we make progress regarding the existence problem without a given depth bound. Our new approach, while falling short of a decision procedure for the general case, reveals some important insights into the structure of these functions. Interestingly, it relates the problem of seeking M\(\varPhi \)RFs to that of seeking recurrent sets (used to prove nontermination). It also helps in identifying classes of loops for which M\(\varPhi \)RFs are sufficient, and thus have linear runtime bounds. For the depth-bounded existence problem, we obtain a new polynomial-time procedure that can provide witnesses for negative answers as well. To obtain this procedure we introduce a new representation for \( SLC \) loops, the difference polyhedron replacing the customary transition polyhedron. We find that this representation yields new insights on M\(\varPhi \)RFs and \( SLC \) loops in general, and some results on termination and nontermination of bounded \( SLC \) loops become straightforward.

Fast and Precise Handling of Positive Weight Cycles for Field-Sensitive Pointer Analysis

Abstract: By distinguishing the fields of an object, Andersen’s field-sensitive pointer analysis yields better precision than its field-insensitive counterpart. A typical field-sensitive solution to inclusion-based pointer analysis for C/C++ is to add positive weights to the edges in Andersen’s constraint graph to model field access. However, the precise modeling is at the cost of introducing a new type of constraint cycles, called positive weight cycles (PWCs). A PWC, which contains at least one positive weight constraint, can cause infinite and redundant field derivations of an object unless the number of its fields is bounded by a pre-defined value. PWCs significantly affect analysis performance when analyzing large C/C++ programs with heavy use of structs and classes.para This paper presents Dea, a fast and precise approach to handling of PWCs that significantly accelerates existing field-sensitive pointer analyses by using a new field collapsing technique that captures the derivation equivalence of fields derived from the same object when resolving a PWC.para Two fields are derivation equivalent in a PWC if they are always pointed to by the same variables (nodes) in this PWC. A stride-based field representation is proposed to identify and collapse derivation equivalent fields into one, avoiding redundant field derivations with significantly fewer field objects during points-to propagation. We have conducted experiments using 11 open-source C/C++ programs. The evaluation shows that Dea is on average 7.1X faster than Pearce et al.’s field-sensitive analysis (Pkh), obtaining the best speedup of 11.0X while maintaining the same precision.

Initial Attempt on Outdoor Human Detection using IEEE 802.11ac WLAN Signal

Abstract: In recent years, the functionality of wireless communications has not only been limited to communication between users, but has also been extended to sensing applications. In fact, several WLAN-based sensing technologies, which rely on changes in multipath radio propagation derived from a channel state information (CSI) in indoor environments, have already been developed. In this paper, we present a WLAN-based outdoor human detection system – the first in IEEE 802.11ac WLAN CSI-based outdoor sensing attempts. The number of multipaths in an outdoor environment is limited, thus, making sensing difficult. To realize outdoor CSI sensing, we present an IEEE 802.11ac WLAN-based sensing system. Specifically, this paper tries to answer the question, "How do we install a transmitter-receiver pair?" based on an analysis of the experimental results. Moreover, we provide open discussions regarding the optimum installation of WLAN sensing devices.

An Energy-Efficient IoT node for HMI applications based on an ultra-low power Multicore Processor

Abstract: Developing wearable sensing technologies and unobtrusive devices is paving the way to the design of compelling applications for the next generation of systems for a smart IoT node for Human Machine Interaction (HMI). In this paper we present a smart sensor node for IoT and HMI based on a programmable Parallel Ultra-Low-Power (PULP) platform. We tested the system on a hand gesture recognition application, which is a preferred way of interaction in HMI design. A wearable armband with 8 EMG sensors is controlled by our IoT node, running a machine learning algorithm in real-time, recognizing up to 11 gestures with a power envelope of 11.84 mW. As a result, the proposed approach is capable to 35 hours of continuous operation and 1000 hours in standby. The resulting platform minimizes effectively the power required to run the software application and thus, it allows more power budget for high-quality AFE.

Hydrodynamic Imaging using an all-optical 2D Artificial Lateral Line

Abstract: Fish and amphibians can sense their hydrodynamic environment via fluid flow sensing organs, called lateral lines. Using this lateral line they are able to detect disturbances in the hydrodynamic near field which enables hydrodynamic imaging, i.e. obstacle detection. Via two experiments we demonstrate a novel artificial lateral line of four bio-inspired 2D fluid flow sensors and show that the measurements of the enacted sensors agree with an established hydrodynamic model. These measurements from the array are then used to localize both vibrating and unidirectionally moving objects using an artificial neural network in a bounded area of 36 by 11 cm which extends beyond the area directly in front of the sensor array. In this area, the average Euclidean localization error is 1.3 cm for a vibrating object, while for moving a object it is on average 3.3 cm.

Robustness Verification of Support Vector Machines

Abstract: We study the problem of formally verifying the robustness to adversarial examples of support vector machines (SVMs), a major machine learning model for classification and regression tasks. Following a recent stream of works on formal robustness verification of (deep) neural networks, our approach relies on a sound abstract version of a given SVM classifier to be used for checking its robustness. This methodology is parametric on a given numerical abstraction of real values and, analogously to the case of neural networks, needs neither abstract least upper bounds nor widening operators on this abstraction. The standard interval domain provides a simple instantiation of our abstraction technique, which is enhanced with the domain of reduced affine forms, an efficient abstraction of the zonotope abstract domain. This robustness verification technique has been fully implemented and experimentally evaluated on SVMs based on linear and nonlinear (polynomial and radial basis function) kernels, which have been trained on the popular MNIST dataset of images and on the recent and more challenging Fashion-MNIST dataset. The experimental results of our prototype SVM robustness verifier appear to be encouraging: this automated verification is fast, scalable and shows significantly high percentages of provable robustness on the test set of MNIST, in particular compared to the analogous provable robustness of neural networks.

Learning Action Images Using Deep Convolutional Neural Networks For 3D Action Recognition

Abstract: Recently, 3D action recognition has received more attention of research and industrial communities thanks to the popularity of depth sensors and the efficiency of skeleton estimation algorithms. Accordingly, a large number of methods have been studied by using either handcrafted features with traditional classifiers or recurrent neural networks. However, they cannot learn high-level spatial and temporal features of a whole skeleton sequence exhaustively. In this paper, we proposed a novel encoding technique to transform the pose features of joint-joint distance and joint-joint orientation to color pixels. By concatenating the features of all frames in a sequence, the spatial joint correlations and temporal pose dynamics of action appearance are depicted by a color image. For learning action models, we adopt the strategy of end-to-end fine-tuning a pre-trained deep convolutional neural networks to completely capture multiple high-level features at multi-scale action representation. The proposed method achieves the state-of-the-art performance on NTU RGB+D, the largest and most challenging 3D action recognition dataset, for both the cross-subject and cross-view evaluation protocols.

Conditioning Circuit for Simultaneous Sensing and Actuation in Piezoelectric MEMS Resonators

Abstract: This paper presents a conditioning circuit able to simultaneously actuate and measure the output of an integrated resonator with embedded aluminum nitride (AlN) layer. The basic idea is to use this active layer as actuation transducer and as sensor to detect the resonant frequency of a MEMS oscillator. In this context a suitable conditioning system has been developed and the working principle has been demonstrated with a MEMS cantilever fabricated in PiezoMUMPs technology, used as a mass sensor. The sensor was characterized by means of known masses of Al 2 O 3 nano-particles, deposited at the free end of the beam. The proposed architecture has been studied and modeled, and a prototype of the MEMS has been characterized demonstrating the validity of the proposed technique.

Dual channel Electrodermal activity sensor for motion artifact removal in car drivers’ stress detection

Abstract: In this paper we present a dual channel sensor for electrodermal activity measurement, with particular attention to the drivers’ stress detection. The sensor captures the elec-trodermal signals that are present on the hands of the driver, transmits them via WiFi to a laptop and then the data are processed. In particular, we developed a novel algorithm for the removal of motion artifacts that arise when the driver moves the hands on the steering wheel. We performed several kinds of tests: first in laboratory, then on a professional driving simulator and finally in a real car in city traffic. The algorithm has been compared to several well known algorithms for signal separation. We identified, as an indicator of performances, the spectral flatness of the outputs. In this application, the proposed method outperformed the benchmark algorithms.

A Hidden Markov Model for Indoor Trajectory Tracking of Elderly People

Abstract: Tracking of elderly people is indispensable to assist them as fast as possible. In this paper, we propose a new trajectory tracking technique to localize elderly people in real time in indoor environments. A mobility model is constructed, based on the hidden Markov models, to estimate the trajectory followed by each person. However, mobility models can not be used as standalone tracking techniques due to accumulation of error with time. For that reason, the proposed mobility model is combined with measurements from the network. Here, we use the power of the WiFi signals received from surrounding Access Points installed in the building. The combination between the mobility model and the measurements result in tracking of elderly people. Real experiments are realized to evaluate the performance of the proposed approach.

IEEE P21451-1-7: Providing More Efficient Network Services over MQTT-SN

Abstract: In the last 20 years, the IEEE 1451 standards family has been presented throughout the changes in the world of smart sensors. One of the most important evolution in this field was the Internet of Things, mainly because of the popularity of light-weight and simple methods to implement communication proocols among human, machines, and sensors. In particular, one of its protocols is very important to enable this interconnection: the MQTT. It has become synonymous with large cloud service providers for the Internet of Things, such as Amazon AWS, IBM Watson and Microsoft Azure. Despite MQTT protocol being traditionally used in controlled networks within server centers, nowadays this protocol is largely used on IoT and It is part of IEEE 1451 family, having the number IEEE 1451.1.6. Following the motivation to adopt IoT’ standards protocols in the IEEE 1451 standard family, this paper proposes the use of a specific and potential variant of MQTT, the MQTT for Sensors Networks, or MQTT-SN. It introduces a more efficient protocol for the Common Architectures and Network services found in the IEEE 1451 Family of Standard, and shows how services and applications based on MQTT can be integrated.

Lightweight Machine Learning-Based Approach for Supervision of Fitness Workout

Abstract: It is widely known that physical activity helps preventing several diseases. However, unsupervised training often results in low exercise quality, ineffective training, and, in worst cases, injuries. Automatic tracking and quantification of exercises by means of wearable devices could be an effective mean for the monitoring of exercise correctness. As a consequence, such devices could help motivating people, thus improving the quantity of performed physical exercise, with positive effects on users’ health conditions. However, despite the availability of several commercial devices, the performance and effectiveness are not well documented. This work proposes a new solution for fitness workout supervision exploiting machine learning techniques, in particular Linear Discriminant Analysis for analyzing data coming from wearable Inertial Measurement Units. Efforts have been done in order to reduce the computational requirements, thus assuring compatibility in perspective of embedded implementation. The experimental tests carried out to assess the proposed approach performance showed an accuracy in exercise detection over 93% and error in exercise counting less than 6%.

Design and Optimization of a MEMS Triboelectric Energy Harvester for Nano-sensor Applications

Abstract: A novel optimized design of a triboelectric vibrational energy harvester having small MEMS scale size and low mass is presented in this work. The triboelectric energy harvester is designed to harvest energy from high frequency mechanical vibrations of machineries such as the skin of an aircraft. The novelty of this work lies in the integrated design and modeling of the energy harvester which enables it to be fabricated and packaged as a MEMS device in a traditional cleanroom environment. Dynamic optimization has been implemented on its geometric model to maximize the output power and power density. Simulation shows that at an acceleration magnitude of 9.8 ms-2 and operating frequency of 800 Hz, the device can generate an average power of 196.91 nW with surface and volume power densities of 13.1 mWm-2 and 1544.4 Wm-3, respectively. Due to its small size, low mass and comparatively high power density, this triboelectric energy harvester can have a significant impact in expanding the applications of the nano-sensors in wireless sensor nodes, in automobile industry, in space exploration programs, in micro- robotics and in prosthetics.

Heart rate estimation of a moving person using 79GHz-Band UWB radar

Abstract: In recent years, human health monitoring technology attracts attention and it is considered to measure biological information such as disorder of breathing and variation of heart rate and to utilize acquired data for health management. On the other hand, 79GHz ultra-wideband radar has attracted big attention because of the particular advantage of having high spatial resolution and good penetration ability which makes them suitable in medical applications. One of these applications is a wireless detection of heart rate and respiration rate. Two hypotheses of a static environment and fixed patient are considered in the method presented in a previous literature which is not valid for long-term monitoring of ambulant patients. In this paper, a new method to estimate heart rate of a moving person is presented with a recursive batch processing is proposed to estimate the heart rate of a moving person.

Static Analysis of Data Science Software

Abstract: Data science software is playing an increasingly important role in every aspect of our daily lives and is even slowly creeping into mission critical scenarios, despite being often opaque and unpredictable. In this paper, we will discuss some key challenges and a number of research questions that we are currently addressing in developing static analysis methods and tools for data science software.

Development of a Non-Contact ECG Application Unobtrusively Embedded into a Bed

Abstract: In this paper, the authors publish a novel designed capacitive ECG system, which can be unobtrusively implemented into a bed. To avoid user rejection, the capacitive electrodes were designed flexibly and with a large surface. In addition, the electrode surface is interfaced by a wire to the circuit boards, which increases the noise sensitivity. Therefore, the proposed solution focuses on noise filtering and shielding for the novel designed electrodes. Already considered in the design of these electrodes is a special guard and ground design, which reduces the impact of noise. Furthermore, different settings of analog filters improve the signal quality, which is proved by test measurements. These test measurements were compared to each other regarding the different filter settings in order to identify the most suitable filter for future developments. Through using analog filters, an almost real-time processing of data is made possible. By the proposed application, the home security for heart diseases and for the elderly can be improved, as well as offers different opportunities for sensor fusion in future.

The Challenges of Connecting Smart Home Health Sensors to Cloud Analytics

Abstract: There is great potential to assess the well-being of older adults in their homes, using sensors. The data derived from these sensors can be used to create solutions that can improve the lives of the users and their family by providing knowledge of health and enable independence. The system architecture commonly proposed is based on sensors deployed in the residence to collect ambient information or information through interactive use. These sensors are then connected through the Internet to cloud services for archiving and processing that is typically based on data analytics and artificial intelligence. This paper specifically focuses on modeling these flows and presents the challenges associated with the demands they place on Internet connectivity services. The paper presents an architecture for the resulting implementation models required for these solutions to be able to be scaled in communities.

Neo-SENSE: a non-invasive smart sensing mattress for cardiac monitoring of babies

Abstract: Within the first minute of life a newborn must take its first breath to make the transition from life inside the womb to the outside world. If a baby does not start breathing, its heart rate will drop and the circulation of blood carrying oxygen to the organs will be seriously affected. The damage done to a newborn who is deprived of oxygen happens so quickly that rapid response is imperative. During birth, the attending neonatal staff manually listen to the baby´s heart and count the heart rate; however, this has proven inaccurate and inefficient. Nowadays, there is not a reliable method to monitor newborn heart rate promptly throughout birth. In this paper, we report the design and development of a novel smart mattress device to measure the babies’ electrocardiogram and respiration non-invasively. The device is based on electrometer-based amplifier sensors combined with novel screen-printing techniques. Proof of concept tests are carried out to demonstrate the suitability of the smart-mattress for new born ECG monitoring. We perform tests with a young infant and demonstrate the potential of this sensing technology to provide a quick and reliable application as ECG readings were displayed within a time < 30 seconds. This will aid the neonatal staff to assess the success of the resuscitation technology aiming to lower the incidence of long-term consequences of poor adaptation to life outside the womb.