Jakub Hlavica

Energy Availability Forecasting for Harvesting-aware Wireless Sensor Networks: Analysis of Energy Demand of a Predictor Based on Evolutionary Fuzzy Rules

Environmental monitoring sensor networks often operate in remote locations and thus must be designed for energy-efficiency and reliability. The first goal of energy efficiency can be achieved through low-power design of the monitoring hardware, often supplemented by energy management schemes of varying complexity and sophistication. In case of sensor nodes endowed with energy-harvesting capabilities, the energy management systems can take advantage of harvesting outlook that can take the form of prediction of energy available for harvest in the near future. When designing such sophisticated prediction and management schemes, it is important to consider the energy cost of gathering the predictor data and executing the forecasting algorithm. This contribution describes the results of experiments designed to assess the energy required to run a recently introduced energy-availability forecasting algorithm based on evolutionary fuzzy rules. In particular, it presents an analysis of experiments conducted using several hardware platforms typically used to implement the computational core of environmental sensor nodes. The results clearly show the advantages of using modern low-power, 32-bit hardware platforms. In addition, they demonstrate the advantages of support for floating point operations, as well the importance of embedded code optimization to maximize the benefits of the modern microcontroller units.

Indoor LiDAR Scan Matching Simulation Framework for Intelligent Algorithms Evaluation

Localization in mobile robotics is one of the most challenging concerns, taking into account the demand on perfect accuracy and quick response. However, high-performance approaches in conjunction with cutting-edge technologies are not necessarily applicable in every case, and thus an optimized localization algorithms suitable for implementation in low-end hardware applications are to be favorable to fill the market niche. Simulation framework, introduced in this contribution, is capable of performing simulations of systems with LiDAR and model an ambient environment by means of user-defined vector maps. Modeled laser sensor is SICK LMS 100. The framework, developed in C# language, enables the user to generate laser scans from user-defined vector maps and trajectories. Scans can subsequently be used for simulations. Computational method considered in this study is particularly Scan Matching.

Energy Harvesting Sources, Storage Devices and System Topologies for Environmental Wireless Sensor Networks: A Review

The operational efficiency of remote environmental wireless sensor networks (EWSNs) has improved tremendously with the advent of Internet of Things (IoT) technologies over the past few years. EWSNs require elaborate device composition and advanced control to attain long-term operation with minimal maintenance. This article is focused on power supplies that provide energy to run the wireless sensor nodes in environmental applications. In this context, EWSNs have two distinct features that set them apart from monitoring systems in other application domains. They are often deployed in remote areas, preventing the use of mains power and precluding regular visits to exchange batteries. At the same time, their surroundings usually provide opportunities to harvest ambient energy and use it to (partially) power the sensor nodes. This review provides a comprehensive account of energy harvesting sources, energy storage devices, and corresponding topologies of energy harvesting systems, focusing on studies published within the last 10 years. Current trends and future directions in these areas are also covered.

Electrical Impedance Distribution in Human Torax: A Modeling Framework

Electrical impedance tomography (EIT) is an imaging system suitable for long-term monitoring. To extend current uses of EIT, improvements in the image reconstruction algorithms are essential. New image reconstruction methods for EIT can be tested on an impedance model of human body. Moreover, accurate anatomical impedance distribution models of human body are used to generate training data used in machine learning algorithms.

SLAM Algorithm Based on Cross-Correlation Scan Matching

Mobile robotics, and particularly robot localization, has undergone tremendous development over the years. Conventional approaches typically employ a large set of sensors to determine robots position. This contribution focuses on Simultaneous localization and mapping (SLAM) based on cross-correlation scan matching with input data being acquired through one laser sensor. The method does not involve additional sensors, such as odometers, thought it maintains to provide satisfactory and robust convergence towards an accurate robot position determination. The method is implemented and tested in a real indoor environment and, as experimental results show, is capable of operation in real time.

Optimization of the Radial Basis Function Neural Network Spread Factor for Electrical Impedance Tomography Image Reconstruction

Electrical impedance tomography (EIT) is a low cost, non-invasive imaging technique where the inner resistivity distribution of the investigated object, corresponding to different tissue resistivity, is estimated from voltage measured on the boundary of the this object. The Electrical impedance tomography main problem is to get the resistivity distribution image of a given cross-sectional area based on the boundary voltage measurement. We used Radial basis function (RBF) neural network for image reconstruction in EIT and focused on examining the impact changing spread factor of the RBF to the results of the image reconstruction with the RBF neural network.

Assessment of Parkinson's disease progression using neural network and ANFIS models

Patients suffering from Parkinson’s disease must periodically undergo a series of tests, usually performed at medical facilities, to diagnose the current state of the disease. Parkinson’s disease progression assessment is an important set of procedures that supports the clinical diagnosis. A common part of the diagnostic train is analysis of speech signal to identify the disease-specific communication issues. This contribution describes two types of computational models that map speech signal measurements to clinical outputs. Speech signal samples were acquired through measurements from patients suffering from Parkinson’s disease. In addition to direct mapping, the developed systems must be able of generalization so that correct clinical scale values can be predicted from future, previously unseen speech signals. Computational methods considered in this paper are artificial neural networks, particularly feedforward networks with several variants of backpropagation learning algorithm, and adaptive network-based fuzzy inference system (ANFIS). In order to speed up the learning process, some of the algorithms were parallelized. Resulting diagnostic system could be implemented in an embedded form to support individual assessment of Parkinson’s disease progression from patients’ homes.

Automatic Fuzzy Classification System for Metabolic Types Detection

Patients suffering from obesity have different demands for medical treatment regarding the causes of their metabolic disorders. To propose new medical solutions to weight reduction, it is desirable to group patients exhibiting similar characteristics. This contribution describes an automatic fuzzy classification system capable of dividing obese patients into groups of diverse metabolic types. Metabolic data were acquired through energometry tests and bioimpedance measurements. Methods considered in this paper are particularly Principal Component Analysis used for data set’s reduction and fuzzy clustering method dividing patients into groups called clusters. Newly tested patients are then classified into designed clusters. A set of statistical hypothesis testing methods is eventually applied to verify the performed classification. The designed classification system could be applied in hospitals to help the doctors with design of an individual treatment for obese patients’ groups.