Research Areas
Our research activities are divided in three major areas. Each link above includes a description of the research areas and examples of related PhD theses.
Additional funding
The COEL ERA-Chair project has enabled us to acquire funding for additional research projects, contributing to the sustainability of our activities.
Sensor Signal Processing
Cognitive electronics can be described as hardware/software electronic systems that embed some form of intelligence. In this context, the cognitive cycle includes sensing, collecting, processing, and analyzing data, as well as learning patterns and taking decisions.
Such cognitive functions can be implemented on a wide range of technological platforms that include micro and nano-sensors followed by signal processing elements such as microprocessors (GPP), digital signal processors (DSP), field programmable gate arrays (FPGA), and customized circuits (ASIC). Very often, such platforms also include wireless connectivity, as for example in Wireless Body Area Networks.
In practice, such platforms are severely constrained in terms of e.g. execution time, area, and energy consumption. At the same time, the complexity of the applications that are expected to run on such platforms is increasing. This calls for new architectures, methods and tools that enable energy-efficient sensor signal processing.
Examples of (state-of-the-art) approaches that allows dealing with the above contradiction at the algorithmic level include:
- Compressed sensing;
- Sparse digital signal processing;
- Approximate computing;
- Transient computing;
- Embedded deep learning.
This, in turn, calls for matching hardware, for example:
- Customized and optimized data-paths, control units, …
- Embedded multi-processor architectures;
- Reconfigurable architectures (e.g. FPGAs and nano-FPGAs);
- Non-volatile architectures (e.g. FRAM- and Flash-based).
Finally, it is also necessary to consider the design methods and supporting tools, e.g. approaches such as HW/SW co-design, efficient mapping of algorithms onto HW/SW platforms and their joint optimization.
This topic also supports implementation aspects of our NB-IoT activities
Examples of defended or ongoing PhD theses related to this research area:
Algorithms and methods for wireless indoor tracking, positioning and object locating systems
Industry PhD student: Taavi Laadung
Supervisors: Muhammad Mahtab Alam, Yannick Le Moullec, Sander Ulp (Eliko)
PhD thesis defence expected in 2023
Wireless indoor tracking, positioning and object locating systems have become a very popular research topic due to the increase of practical applications and demand from the industry as well as the spread of IoT devices. Although there are already techniques and many methods for indoor positioning they lack the robustness, scalability, reliability and adaptiveness to different applications, situations and environments. Industrial applications are difficult to solve in terms of obstructions, non-line of sight, RF propagation and require higher robustness and adaptability while maintaining positioning accuracy to meet the requirements of the applications.
This industry PhD thesis aims to tackle the problems related to indoor positioning systems and improve and develop new methods and algorithms to increase the adaptability, robustness and accuracy of the positioning systems. The focus of the PhD is to deal with the NLOS situations, changing environments and adaptability of the positioning system. In collaboration with Eliko OÜ the PhD thesis will be based on real-life problems and data, practical environments, applications using existing UWB KIO RTLS positioning system.
Energy-Efficient Techniques and Architectures for the Green Internet of Things (IoT)
PhD student: Sikandar Muhammad Zulqarnain Khan
Supervisors: Yannick Le Moullec, Muhammad Mahtab Alam
PhD thesis expected to be defended in 2022.
With the tremendous growth of the internet of things (IoT), millions of connected electronic devices will be scattered in our living and working environments. Powering such vast amounts of devices brings many challenges in terms of sustainability. This has spurred a new wave of research to address them; this trend is also highlighted by recent efforts such as the IEEE Initiative on Green Information and Communication Technologies (Green ICT).
This PhD work seeks to explore, propose and test new methods, algorithms and architectures for enabling the so-called “battery-less IoT”. Activities include to research, develop and publish on various aspects related to the selection, design, and optimization of: Efficient energy harvesting techniques (e.g. solar, back-scattering) to power IoT nodes with no or little energy storage; Software and hardware techniques for approximate computing to reduce the computational requirements, and hence the energy consumption of the IoT nodes; Transient computing algorithms and their implementation on non-volatile devices (e.g. microcontrollers based on FRAM or other emerging memristive technologies) to deal with the intermittent and unbalanced nature of energy harvesting; Methods for the joint design space exploration and optimization of the above together with communication protocols such as NB-IoT and/or LTE Cat M1.
NB-IoT infrastructure at TalTech
SNR and RSSI at different elevations
Approximation and predictions on top of transient computing
PhD thesis expected to be defended in 2022.
Algorithms for Learning and Adaptation Over Networks – Distributed Leader Selection
PhD student: Sander Ulp
Supervisors: Muhammad Mahtab Alam; Yannick Le Moullec; Tõnu Trump
PhD thesis defended in 2019. Link to fulltext.
Learning and adaptation over networks is a rapidly evolving topic with many possible applications and fields. There are several unanswered and unresearched aspects of these algorithms and their applications. In particular, analysing the performance and developing algorithms for distributed estimation are essential for future smart and self-organizing networks. Indeed, enabling complex and sophisticated behaviour through cooperation of simpler units in the network allows to accomplish more demanding tasks that are unattainable to single units and current solutions. Estimating from different sensors and cooperating to achieve better performance is a challenging task when taking into account the limitations and constraints presented by applications.
Classification of different communication strategies used in estimation
The algorithms should enable the networks to learn and adapt to different situations under constraints, such as limited bandwidth, limited battery capacity, physical and communication restrictions, security etc. In this work the motivation for and the background to the learning and adaptation are presented and an overview of different methods for distributed estimation as well as the author’s contributions to the existing work are given.
This PhD work proposes 1) an improvement to the existing diffusion algorithm weight calculation as well as 2) a novel algorithm for distributed leader selection (DLS). The method using MDL (minimum description length) subspace algorithm to estimate the SNR (signal to noise ratio) of the estimated signal allows the weight calculation to incorporate the values and improve the performance of the diffusion algorithm in comparison to the equally weighed diffusion algorithm.
The proposed DLS algorithm is able to select the best performing node as the leader node in a fully distributed manner and achieve its performance across the network. The algorithm outperforms both the equally weighed diffusion algorithm and the non-cooperating network. The theoretical and simulation results show that the DLS algorithm is also able to attain similar performance than that of the diffusion algorithm using optimal weights under certain conditions and overall is less complex and more robust compared to the diffusion algorithm.
The work also analyses the energy consumption and computational complexity of the diffusion algorithm and the DLS algorithm. The DLS algorithm is modified to improve its energy efficiency. The proposed energy-efficient distributed leader selection algorithm is able to reduce the energy consumption of the network by 32 − 53% and extend its lifetime by 14 − 46%.
Clockwise: Network convergence MSE with Rayleigh fading; Network MSD performances in both scenarios for the non-cooperating, the relative variance rule, the relative degree variance and the leader selection; (a) Radio communication energy consumption at different nodes for the diffusion algorithm, DLS algorithm and EEDLS algorithm, (b) Network radio communication energy consumption for the diffusion algorithm, DLS algorithm and EEDLS algorithm. Source: Sander Ulp, Algorithms for Learning and Adaptation Over Networks – Distributed Leader Selection, PhD thesis, Tallinn University of Technology, 2019
Parameter Estimation by Sparse Reconstruction with Wideband dictionaries
PhD student: Maksim Butsenko
Supervisors: Olev Märtens; Yannick Le Moullec; Tõnu Trump
PhD thesis defended in 2018. Link to fulltext.
Parameter estimation in general and spectral analysis in particular have been fruitful research areas for a long time and still rightfully remain so. Many algorithms and methods for parameter estimation were proposed during decades of research, but as technology evolves, so evolve our requirements to estimators. Non-parametric estimators were for a long time the most popular method of spectral analysis; however, their major drawback is their resolution limitation and high variance. Parametric estimators can provide high-resolution estimates, however this requires considered signal to correspond well to underlying signal model and they perform much worse than non-parametric estimators if this is not the case. Semi-parametric estimators can often provide high-resolution estimates without big dependency on the signal model as their only assumption is that the signal should be sparse. In fact, a wide range of common applications considers signals that are well approximated by sparse reconstruction framework, and this area has attracted noteworthy interest in the recent literature. Considerable number of these work focuses on formulating convex optimization algorithms that make use of different sparsity inducing penalties, thereby encouraging solutions that are well represented using only a few elements from some dictionary matrix.
It can be shown that when the dictionary is chosen properly, even limited number of measurements allows for an accurate signal reconstruction. In this work a novel procedure for constructing dictionaries for parameter estimation by sparse reconstruction methods is considered. Instead of forming the dictionary as a finite set of discrete narrowband components for evaluation of continuous parameter space, this work considers wideband dictionary elements, such that continuous parameter space is divided into B subsets. During the estimation procedure, the activated subsets are selected for further refinement and non-activated subsets are discarded from further optimization. Afterwards, a new dictionary is formed for each of the activated subsets, resulting in the zoomed dictionary for that particular region of considered parameter space. An iterative procedure may then be repeated further until required resolution is reached. The initial problem statement and the plausibility of the approach are validated by showing that the method is suitable for one-dimensional data.
The formulation of the wideband framework for multi-dimensional data and validation on different estimators (LASSO, SPICE, IAA), sampling scenarios and data from different sources are considered. An efficient implementation of the algorithm by alternative direction method of multipliers is formulated and corresponding complexity reduction calculations considered. A comprehensive overview of the best approaches to selection of the parameters for the framework is provided. The proposed approach is tested mainly by using corresponding signal model and conducting series of numerical experiments by running multiple Monte-Carlo simulations. However, the proposed method is also tested on real-life signals by considering NMR data and by investigating the possibility of employing similar sparse reconstruction framework for the separation of cardiac and respiratory signal components from the electrical bioimpedance measurements. The wideband framework allows for considerable reduction in computational complexity and decreases probability of missing off-grid components.
For situations where the number of samples is considerably smaller than the number of dictionary elements, the percentage of correct model order estimation for the proposed wideband dictionary is 40 − 50% higher than for the conventional method (90 − 100% vs 50 − 60%). Most of the errors in this situation come from missing off-grid components. Similar methods for grid-selection issues exist; however, they often impose increasing complexity of the problem formulation and therefore limiting size of considered problems as for example in the case of atomic-norm minimization; or loosing the benefits of convexity of a solution as in the case of adaptive grid approach. By employing iterative zooming procedure and decreasing risk of missing off-grid components we show that it is possible to formulate problem with smaller initial dictionary and therefore reduce amount of computations that are needed for the required resolution.
For the same resolution, the computational complexity of the proposed method can be 20 − 30 times lower, which results in considerable reduction in the time required to make an estimation. The proposed wideband dictionary method has the additional benefit of its adaptability and intuitive implementation. This can help to encourage those working in the area of signal estimation to consider applying the proposed method for their problems as wideband dictionary successfully replaces the classical narrowband dictionary for considered problems by providing at least similar performance and often outperforming the classical framework.
Clockwise: Mean-square error curves for different SNR levels for the single-stage narrowband dictionary, using L = 1000, as compared to the two-stage dictionary, using B integrated wide-band elements in the first stage, followed by Q narrowband elements in the second stage; The resulting estimates using a dictionary with 2000 narrowband elements (top), and a two-stage zooming approach using wideband elements, using B1 = 40 in the first stage and B2 = 50 for each activated bands in a second stage. The signal is a measured NMR signal of length N = 256; Cardiac component frequency estimate compared to the frequency estimate as measured from ECG signal; The peak resolving ability of the estimators; the proposed speed-up does not reduce the resolution of the resulting estimates. As expected, the use of a wideband dictionary is even yielding a somewhat improved performance. Source: Maksim Butsenko, Parameter Estimation by Sparse Reconstruction with Wideband Dictionaries, PhD thesis, Tallinn University of Technology, 2018
Distributed Signal Processing in Cognitive Radio Networks
PhD student: Ahti Ainomäe
Supervisors: Tõnu Trump; Mats Bengtsson (KTH, Sweden); Yannick Le Moullec
PhD thesis defended in 2018. Link to fulltext.
The lack of available radio frequencies is seen to be an increasing problem for implementing new modern radio communication solutions. Recent studies have shown that, while the available licensed radio spectrum becomes more occupied, the assigned spectrum is significantly underutilized.
To alleviate the situation, cognitive radio (CR) technology has been proposed to provide an opportunistic access to the licensed spectrum areas. Secondary CR systems need to cyclically detect the presence of a primary user (PU) by continuously sensing the spectrum area of interest. Radiowave propagation effects like fading and shadowing often
complicate sensing of spectrum holes. When spectrum sensing is performed in a cooperative manner, then the resulting sensing performance can be improved and stabilized.
Basic layout of CR network
In this work, three fully distributed and adaptive cooperative PU detection solutions for CR networks are studied.
First, we study a distributed energy detection scheme without using any fusion center. Due to reduced communication such a topology is more energy efficient. We propose the usage of distributed, diffusion least mean square (LMS) type of power estimation algorithms with different network topologies. We analyze the resulting energy detection performance by using a common framework and verify the theoretical findings through simulations.
Second, we propose a fully distributed detection scheme, based on the largest eigenvalue of adaptively estimated correlation matrices, assuming that the primary user signal is temporally correlated. Different forms of diffusion LMS algorithms are used for estimating and averaging the correlation matrices over the CR network. The resulting detection performance is analyzed
using a common framework. In order to obtain analytic results on the detection performance, the adaptive correlation matrix estimates are approximated by a Wishart distribution. The theoretical findings are verified through simulations.
Third, we propose a fully distributed largest eigenvalue detection scheme, where the observations of the elements of correlation matrices are weighted by independently estimated local SNR values. The resulting detection performance is analysed by using a common framework. The theoretical findings are verified through MATLAB simulations.
Clockwise: Local power estimation, fixed step , for the recursive ring-round topology; Probability of detection, CTA topology; Proposed diffusion method; ATC, DoF |H1 values with perturbations 0 dB, -1 dB and 2 dB. Source: Ahti Ainomae, Distributed Signal Processing in Cognitive Radio Networks, PhD thesis, Tallinn University of Technology, 2018
PhD thesis defended in 2018. Link to fulltext.
Classification and Denoising of Objects in TEM and CT Images Using Deep Neural Networks
PhD student: Anindya Gupta
Supervisors:Olev Märtens; Yannick Le Moullec; Ida-Maria Sintorn (Uppsala University, Sweden); TõnisSaar
PhD thesis defended in 2018, link to the thesis.
The digitization of biomedical and medical images has benefited the clinicians in comprehending (or detecting) obscure abnormalities. However, manual analysis is labor-intensive and time-consuming. Since the last few decades, computer-aided detection (CAD) systems employing learning-based methods and conventional image analysis-based methods have successfully paved the landscape for the detection (and/or classification) of deadly abnormalities. Lately, the inception of deep neural networks (DNN) (often synonymized as deep learning) as a powerful recognition module has shifted the research interest from problem-specific solutions to increasingly problem-agnostic methods that rely on learning from data.
In particular, convolutional neural networks (CNNs) have rapidly become a primary choice for many CAD systems due to their astonishing results. This impulse has been sparked by increased computational power (graphical processing units) and the evolution of learning-based methods.
This work comprises a total of five solutions: four DNN-based solutions for classification of structures in biomedical and medical images, and one solution for denoising of biomedical images to improve the image quality. The work is focused on the applications of two variants of DNN: the CNN, and the multi-layer perceptrons (MLP).
Infographical overview of the PhD work
From a biomedical image analysis perspective, the first solution is associated with improving the performance of automated workflow for primary ciliary dyskinesia (PCD) analysis. To classify cilia and non-cilia structures in low-magnification (LM) transmission electron microscopy (TEM) images,a CNN-based classifier is developed as a false positives (FP) reduction module. Although computing discriminative features of cilia structures at very low magnification is challenging, the developed CNN classifier substantially improves the F-score from 0.47 to 0.59.
The second solution takes a side step from classification and focuses on denoising. Denoising is often considered as a preprocessing step to improve the image quality for automated analysis. Given this, the second solution is associated with enhancing the structural information in short exposure high-magnification TEM images. A novel multi-stream CNN-based model is developed to denoise 100 short exposure HM images acquired at the same spatial location in the cell section. Three different strategies for combining denoising and image merging are investigated to determine the optimal structure enhancing strategy. The CNN denoising model is only trained for one strategy and used as it is for other two strategies, thus presenting the transfer learning perspective of DNN as a potential add-on to automated analysis. The presented model achieves an improved PSNR of 40.84 dB.
From a medical image analysis perspective, the third solution is associated with improving the performance of a CAD system for the early detection of multiple sizes of nodules (3 - 30 mm) in computed tomography (CT) scans. To classify nodules and non-nodules, an MLP-based classifier is developed as an FP reduction module. The CAD is extensively tested on four publically available CT datasets; this makes it the only system to be successfully validated on such large scale. The developed CAD system achieves a high sensitivity of 85.6% with only 8 FPs/scan.
Until recently, conventional CAD systems employing learning-based methods depended on handcrafted representations (features). Designing features by hand is challenging and often result in limited discriminative power; thus, this is insufficient to classify micronodules (≤ 4 mm) and cross-sectional vessels. The fourth solution is associated with developing a CAD system for the detection of micronodules in CT scans. To classify micronodules and small cross-sectional vessels, a novel 3D CNN classifier is developed as an FP reduction module. Using the largest publically available CT dataset, the developed CAD system achieves a high sensitivity of 86.7% with only 8 FPs/scan.
The fifth solution is associated with improving the performance and efficiency of automated workflow for detecting multiple sizes of vascular nodes in CT angiography (CTA) scans. To classify cross-sections of different sizes of vessel and non-vessel nodes, a patch-based CNN classifier is developed as an FP reduction module. On the given 25 CTA volumes from the clinical routine, the presented classifier substantially improves the F-score from 0.43 to 0.82.
Clockwise: overview of the overall workflow consisting of the CNN model; noisy and denoised close ups of a cilium instance obtained with the methods; overview of the developed CAD pipeline; Examples of (a) lung segmentation refinement method; (b) different types of nodules detected by CAD system in SPIE-AAPM dataset; c nodules not marked in the ground-truth list of PCF subset but detected by the proposed CAD. Source: Anindya Gupta, Classification and Denoising of Objects in TEM and CT Images Using Deep Neural Networks, PhD thesis, Tallinn University of Technology, 2018.
Modeling and Implementation of Linear Energy Prediction for Energy Harvesting in Intermittently Powered Wireless Sensor Nodes
PhD student: Faisal Ahmed.
Supervisors: Yannick Le Moullec; Paul Annus; Gert Tamberg
PhD thesis defended in 2018, link to thesis.
The advent and growth of the IoT has opened new directions and challenges for the scientific community. In particular, IoT enabling devices such as wireless sensor nodes are powered by energy-limited batteries, which affects their life-time and reliability in case of intensive utilization, and eventually leads to increased maintenance requirements and related cost. Thus, researchers have investigated and proposed various solutions under the so-called energy harvesting concept.
Such solutions help overcoming the limited batteries’ capacities by providing a supplementary or alternative source of energy to operate e.g. smart devices, wireless sensor nodes, home appliances, industrial machine etc. The positive impact of energy harvesting in IoT enables innovative applications that are no longer hindered by the batteries limits. However, energy harvesting poses several challenges both at the hardware and software levels when designing energy-autonomous wireless sensor nodes.
Indeed, energy harvesting from the environment such as from solar, wind, thermal, RF etc sources typically exhibits intermittent characteristics. This means that the wireless sensor nodes may be left without power, which in turn impacts the application’s performance in terms of e.g. connectivity and reliability.
Firstly, this work proposed a system-level framework that uses coarse-grain models of various single and hybrid energy harvesting technologies for wireless sensor nodes. Experimental results illustrate how the framework can be used to evaluate various energy harvesting sources for powering WSN nodes.
Then the work assessed the practical feasibility of powering a wireless sensor node from an energy harvesting source without energy storage. A salient feature of the work is the implementation of a transient computing mechanism on a non-volatile (FRAMbased) node. The experimental results illustrate that energy harvesting, combined with transient computing, is indeed feasible.
Next the work proposed an energy prediction model named LINE-P (Linear Energy Prediction). It builds upon sampling and approximation theory. LINE-P is more suitable for dual EH sources and various data time intervals than state-of-the-art models. The simulation results show that LINE-P’s prediction accuracy is up to ca. 98% for solar energy and up to ca. 96% for wind-based prediction.
Thereafter, the work deployed a transient computing mechanism for bidirectional communication where energy harvesting is used in combination with transient computing and the LINE-P energy prediction model. This allows firing communication tasks only if sufficient and stable energy is predicted. The results for a peer-to-peer wireless setup illustrate that the combined two modalities require only 15% of the node’s memory, and this proposed approach (combined) yields an average receiving rate up to 94.6%.
Finally, the work designed the Adaptive LINE-P model that addresses the fixed weighting parameter issue by calculating adaptive weighting parameters based on the stored energy profiles. In addition, a profile compression method has been proposed to reduce the memory requirements. The results illustrate that Adaptive LINE-P’s accuracy is up to 90-94% and compression method can 50% reduce memory overheads.
Diagram and photograph of the experimental transient-computing peer-to-peer wireless setup. Source: Ahmed, F.; Kervadec, C.; Le Moullec, Y.; Tamberg, G.; Annus, P. (2018). Autonomous Wireless Sensor Networks: Implementation of Transient Computing and Energy Prediction for Improved Node Performance and Link Quality. The Computer Journal, 1−18
Internet of Everything
In internet-of-things (IoT), there are massive numbers of miniaturized, low-power, low throughput and low latency constrained communicating devices (such as sensors, actuators, coordinators, gateway etc.,) to collect and communicate real-time information. With the continuous rise of IoT applications use cases and requirements, various standards have been proposed. On the one hand, 3GPP has recently standardized the narrow-band IoT. The key features include deployment flexibility, low device complexity, significant coverage extension, long battery life time and so on. On the other hand, open standard solutions such as LoRaWAN are getting more and more attention. These are low-power long range wide area networks which exploits the sub-giga spectrum for low data rate IoT applications. The physical layer specifications of both above-mentioned standards are thoroughly defined in respective technical specifications
However, with regards to medium access control (MAC), there is a lot of room for investigating which options could be more suitable. A baseline MAC layer is proposed in LoRaWAN; however, to provide robust connectivity, high performance and reliability in packet reception, distributed MAC protocols (i.e., CSMA/CA) may not be very suitable. It is expected that “Collaborative Communication” can significantly help to improve the lifetime of dense low power devices. This requires adaptive and dynamic communication algorithm and protocols which can adapt based on the given context and constraints. In this research direction, we will investigate new MAC protocols and techniques which would be cognitive and adaptable to the dynamic variations of the applications use-cases of IoT.
In another research direction, in-line with the framework of 5G and beyond wireless communication systems, recently, with a wider acceptability of using public safety long term evolution (PS-LTE) based radio communication systems, it is expected that in the next couple of years, classical next generation PSNs will be significantly improved and interoperable. Typically, in mission critical machine type communication (MC-MTC), often the infrastructure is either completely unavailable or only partially available. Then, it is important to deploy mobile and portable command center(s) near the disaster site to provide network connectivity to on-scene available (OS-A) devices by exploiting multi-hop device-to-device (D2D) communication. Among various such recent studies for indoor deployment and enabling D2D communication, a measurement campaign through ORACENET routing protocol shows that over multi-hop networks the devices remain disconnected for more than 90% of the time which severely impact on the resource utilization and management.
To address these concerns, few fundamental objectives would be achieved in this particular directions: i) under the above mentioned scenario, how the cooperative D2D communication could improve and extend the network coverage/connectivity and optimize radio resource allocation in indoor and outdoor dense environments. To address this objective, an emerging new paradigm having "mobile drones" would be considered as deployed small cell for wider coverage and stable connectivity. Such heterogeneous network architecture will pose an interesting setup for efficient resource management and will be investigated, ii) to propose new metrics for effective interference measurements evaluation and transmission, iii) to investigate and validate cooperative coexistence strategies, based also on interference information and management, among heterogeneous devices and networks.
Finally, another research direction is planned based on Wearable Wireless Networks (WWN) which is a self-organized network at the human body scale. It consists of heterogeneous smart devices which are low-power, miniaturized, hardware-constrained (with limited processing and storage capabilities), and attached to (or implanted inside) a human body. These devices can be sensors (to sense, transmit and receive data), actuators (to react according to the perceived data) or coordinators (to act as a gateway for the external network). Typically, sensors are connected to monitor physiological signs (e.g. heartbeat, temperature, etc.), movement and activity (e.g. acceleration, orientation etc.) and surrounding environments (e.g. temperature, toxic gases, etc.). WWN have gained significant attention in daily life applications. In health-care sector, remote and mobile monitoring of patients from physician or hospitals is a reality; self-monitoring and early diagnosis is also possible. Athletes and players uses various wearable devices to maintain their fitness. Further, the concept of augmented reality is getting mature due to convergence of technologies, data and computing. With the recent advent of body-to-body (B2B) communication, old problems with new set of dimensions have arrived. In WWN, accurate channel models pave the way for meagre resource optimizations (including power optimizations and consequently the energy efficiency, throughput optimization and bandwidth utilizations).
To obtain various system level optimizations, in this research directions the following topics will be covered:
Firstly, to propose accurate channel models using bio-mechanical mobility modelling for on-body and body-to-body communication. Secondly, energy efficient cooperative communication techniques to improve the life time of the network would be achieved. Application-specific optimizations would be targeted as case studies. Finally, novel coexistence schemes for inter-BAN and intra-BAN interference in both coordinated and un-coordinated environment would be proposed.
Our activities include collaborations with the industry, for example through the technology transfer workshop on NB-IoT.
Examples of PhD theses related to this research area:
Development and Exploitation of a Smart Wearable-Assistive Neuromuscular Stimulation System Using Data Analytics
PhD student: Abdul Saboor
Supervisors: Muhammad Mahtab Alam, Alar Kuusik
Thesis defence expected: Studied interrupted
Existing wearable assistive actuators (e.g. neuromuscular stimulators) lack context and situational awareness, and thus increase the patients’ safety risks and reduce their quality of life. We propose an innovative closed-loop wireless communication system that adds i) intelligent monitoring, ii) automated neuromuscular stimulation, iii) feedback from the actuator-to-coordinator for adaptation and decision-making.
The objective of this PhD project is to develop smart learning techniques (at the edge) which are low-complex yet efficient in precision under low-latency constraints.
The PhD work will exploit detailed application requirements for functional electrical stimulation (FES) systems for partially disabled neurodegenerative disease patients (from clinical side). In particular, in closed-loop setting, focus mainly on practical and realistic latency requirements (or maximum latency-bound to avoid fall-down) evaluation; the other requirements are associated with latency. Map the above requirements onto emerging technologies;
The PhD work will also develop and test innovative, event-based strategies for control and signal processing. Focus on the design of a hybrid event based/predictive (low computational) communication scheme for fall prediction and test neuromuscular stimulators, jointly with clinical personnel at West-Tallinn Central Hospital, Estonia.
Cooperative Device to Device Communication for Emergency and Critical Wireless Communication System
PhD student: Ali Masood
Supervisors: Muhammad Mahtab Alam, Yannick Le Moullec
Related to the NATO SPS Counter Terror project.
PhD thesis defence expected in 2022
Recently, with a wider acceptability of using public safety long term evolution (PS-LTE) based radio communication systems, it is expected that in the next couple of years the classical next generation PSNs will be significantly improved and interoperable. Typically, in mission-critical machine type communication (MC-MTC), often the infrastructure is either completely unavailable or only partially available. Then, it is important to deploy mobile and portable command center(s) near the disaster site to provide network connectivity to on-scene available (OS-A) devices by exploiting multi-hop deviceto-device (D2D) communication.
Among various such recent studies for indoor deployment and enabling D2D communication, a measurement campaign through the ORACENET routing protocol has been presented in. One of the important conclusions derived from this study is that over multi-hop networks the devices remain disconnected for more than 90% of the time, which severely affects the resource utilization and management.
To address these concerns, three fundamental objectives should be achieved in this PhD work: i) Under the above-mentioned scenario, how the cooperative D2D communication could improve and extend the network coverage/connectivity and optimize radio resource allocation in indoor and outdoor dense environments. To address this objective, an emerging new paradigm having "mobile drones" would be considered as a deployed small cell for wider coverage and stable connectivity. Such heterogeneous network architecture will pose an interesting setup for proposing new efficient resource and interference management techniques, ii) Propose a dynamic solution to improve and maintain the connectivity/throughput in an energy-efficient manner in emergency scenarios.
Packet Error Rate for the Unmanned Aerial Vehicle to User Equipment. Source: Masood, A.; Sharma, N.; Alam, M. M.; Moullec, Y. Le; Scazzoli, D.; Reggiani, L.; Magarini, M.; Ahmad, R. (2019). Device-to-Device Discovery and Localization Assisted by UAVs in Pervasive Public Safety Networks. Proceedings of Workshop on innovative aerial communication solutions for First Responders network in emergency scenarios, MobiHoc 2019: Workshop on innovative aerial communication solutions for First Responders network in emergency scenarios, MobiHoc 2019, Catania, Italy, July 2-5, 2019. Catania, Italy: ACM, 1−6
PhD thesis expected to be defended in 2022
Related to the NATO SPS Counter Terror project.
Interference Management in the Cellular Driven Heterogeneous Internet of Things (IoT) Networks
PhD student: Collins Burton Mwakwata
Supervisors: Muhammad Mahtab Alam, Hassan Malik
PhD thesis defence expected in 2022
Narrowband Internet of things (NB-IoT) and CAT-M1 are two cellular-driven IoT standards proposed by 3GPP release 13. These low power wide area (LPWA) technologies have tremendous penetration potential in many applications including smart cities, smart buildings, smart metering, consumer, agriculture, healthcare and so on.
However, with the future 5G heterogeneous deployments, massive interference is expected not only from neighboring LPWAN users but also from traditional LTE/5G users (both from small cells and maco cells). In our preliminary investigation on interference and radio resource management, the performance has been evaluated under the impact from repetition factor, time offset, control channel overhead, radio environment, interference.
For NB-IoT the maximum information rates that can be achieved are 20 Kbps, 32 Kbps and 60 Kbps for in-band, standalone and uplink (15 kHz) scenarios, respectively. However, these rates are significantly lower than the peak data rate of 226.7 Kbps and 250 Kbps in downlink and uplink, respectively. For different IoT use-cases, one has to jointly optimize throughput, delay, latency as well as security requirements and allocate resource accordingly.
To achieve such a goal, in this PhD work the following areas are included: i) to propose new and effective interference measurements models for heterogeneous 5G/IoT networks ii) to propose new techniques suitable for interference cancelation iii) to propose efficient radio resource management strategies under interference constraints. iv) to propose lightweight privacy and security protocols to ensure minimum data rates.
The geographical representation of countries with the ongoing NB-IoT real-life deployments for diverse use cases (May 2019)
Summary of NB-IoT deployment strategies. For example, when NB-IoT is deployed in macrocell and LTE in small cell, when LTE is in macrocell and NB-IoT is in small cells, when NB-IoT is in macrocell and small cells support both NB-IoT and LTE, and when LTE is in macrocell and LTE/NB-IoT is in small cells. Source: Mwakwata, C. B. Malik, H.; Alam, M. M.; Le Moullec, Y.; Pärand, S.; Mumtaz, S. (2019). Narrowband Internet of Things (NB-IoT): From Physical (PHY) and Media Access Control (MAC) Layers Perspectives. Sensors, 19 (11), 1−34
PhD thesis defence expected in 2022
Clinical assistive technology for patients with neurodegenerative disease
PhD student: Triin Kask
Supervisors: Muhammad Mahtab Alam, Alar Kuusik
Thesis defence expected: Studied interrupted
Check our Triin's video here: https://youtu.be/jrn7hqUDUmk
Wearables, or devices that can be worn has been a thriving global market since the beginning of 21th century. According to Shanhong Liu report of March 2019 it is predicted that by the year 2020, there will be 1.1 billion connected wearable devices worldwide (https://www.statista.com/topics/1556/wearable-technology/).
Wearable motion sensors are widely also used in a clinical setting to monitor activities of daily living (ADL) and make important health decisions. Increasing number of usages includes assessing gait and motor performance in a medical setting.
Neurodegenerative diseases (such as multiple sclerosis) worldwide prevalence is approximately 2 million people who benefit highly on wearable devices on remote health monitoring.
This PhD project is about creating a novel closed-loop communication system wearable device that provides intellectual neuromuscular stimulation (functional electrical stimulation) with context awareness for home use in order to increase patient’s quality of life and assist in ADL.
The system is part of precision medicine meaning that patient will be his/her own reference and it provides feedback in the communication system which is a novel concept for medical devices.
Clockwise: Comparison of current state-of-the-art and envisioned system; System measurements with 2 IMU-s and 3 muscles in order to analyse muscular activity in 3 muscles; Walk phase detection using piecewise lineaar regression. Source: Closed-loop-communication-system-to-support-highly-responsive-neuromuscular-assistive-stimulation project and Triin Kask personal collection
PhD thesis defence expected in 2021
System Level Optimization for Wearable Wireless Networks
PhD student: Rida Khan
Supervisor: Muhammad Mahtab Alam
A Wearable Wireless Network (WWN) is a heterogeneous network at the human body scale. It consists of heterogeneous smart devices which are low power, miniaturized, hardware-constrained (with limited processing and storage capabilities), and attached to (or implanted inside) a human body. The human-body sensors are responsible for monitoring the vital signs, the central coordinator or hub manages the network operations while the (optional) actuators provide some feedback (in case of close loop communication). WWN realizes a number of applications in both healthcare and consumer electronics sectors.
The sensors placed on or implanted inside the human body generally have varying traffic rates and the wireless link conditions are mainly dominated by human body shadowing and mobility. Also, the WWN communication takes place in the unlicensed spectrum which could result in interference. Therefore, accurate channel modeling is necessary, at first, to understand the channel statistics and measure the degree of interference. Dynamic MAC schemes help achieving the power optimizations (and consequently the energy efficiency), throughput optimization and bandwidth utilization, given the space-time varying channels and time-varying traffic. Finally, interference management and coexistence schemes enable better WWN operation under the interfered WWN channels.
To obtain various system level optimizations, in this PhD thesis, following topics will be covered: i) proposal of accurate channel models using biomechanical mobility modeling for on-body and body-to-body communication, ii) proposal of energy efficient medium access control techniques to improve the lifetime of the network and iii) proposal of novel coexistence schemes for inter-BAN and intra-BAN interference in both coordinated and un-coordinated environment.
Clockwise: Enhanced Traffic and Channel Aware (TCA) algorithm for SmartBAN; Packet reception rate (PRR) (%) under walking, sit-stand and running mobility profiles (TCA and enhanced TCA); Energy consumption per successful transmission (mJ) under walking, sit-stand and running mobility profiles (TCA and enhanced TCA); Hub energy consumption per successful transmission (mJ) under walking, sit-stand and running mobility profiles (TCA and enhanced TCA). Source: Khan, Rida; Alam, Muhammad Mahtab; Paso, Tuomas; Haapola, Jussi (2019). Throughput and Channel Aware MAC Scheduling for SmartBAN Standard. IEEE Access, 7, 63133−63145
PhD thesis defence expected in 2021
Radio Spectrum and Power Optimization Cognitive Techniques for Wireless Body Area Networks
PhD student: Tauseef Ahmed
Supervisor: Yannick Le Moullec
PhD thesis defended in 2017. Link to fulltext.
The recent growth in wireless technologies has opened new horizons and research challenges; in particular, the radio spectrum has become a scarce resource. Radio spectrum is the main ingredient required for wireless communication to exist, and with so many licensed wireless technologies already existing, this key resource has already been used up. Scientists and researchers have gathered to solve the spectrum scarcity problem. It has been observed that spectrum scarcity is in fact a spectrum under-utilization issue. With careful planning and advanced spectrum management techniques, this issue can be resolved. Conventional wireless technologies where dynamic network operations and maintenance is not possible, such spectrum management features are not possible.
The concept of environment-aware intelligent radio, named cognitive radio, was introduced in the beginning of the 21st century to meet the requirements of future wireless technologies and services. There has been a lot of research going on various aspects of cognitive radio networks since their introduction. Yet, as of today, realizing many of the cognitive features of such networks still poses research challenges, including spectrum management, power assignments, quality of service, interference management, etc.
In this PhD work, tasks related to the radio environment and heterogeneity have been investigated for cognitive radio networks. The work presents the unsupervised spectrum access and sharing technique based on machine learning sub-domain called reinforcement learning. The approach presented in this work takes into account the current conditions of the spectrum band (e.g. signal to noise ratio, primary network existence, secondary inter-cell interference, etc.) and it decides the fate of the spectrum. Since cognitive radios operate in the primary network environment, they must operate under the precautions and not produce any hindrance toward the primary network operations.
To cope with such restrictions, a convex optimization approach to assign transmission powers to the cognitive radios has been presented in this PhD work. The approach reduces the required transmission power required, it avoids any interference generated towards the primary network and it minimizes the inter-cell interference, this is reflected by e.g. up to x10 SINR gains and a 10% gain in user satisfaction in high traffic loads conditions. Furthermore, this work describes the synergy between the cognitive radio’s concept and the emerging technology of wireless body area networks by presenting cognitive approaches in their operations.
The concept of cognitive body area network (C-BAN) is presented in this PhD work. Similar to a cognitive radio network, a cognitive body area network can be context aware and reconfigurable in its transmission features. Wireless body area networks for health care and monitoring applications mostly exist in the spectrum band which is used by many other wireless technologies. For patient monitoring in hospitals and nursing homes can also pose many challenges in spectrum management due to the co-existence of multiple networks of the same types.
There is a need for careful spectrum management features and cognition in body area networks. A reinforcement learning based algorithm for spectrum management is presented in this PhD work. The algorithm focuses on the channel conditions before making a decision whether to use it or not. This approach makes sure that best channel among available is selected in order to meet QoS requirements of the network. Furthermore, a transmission power assignment scheme is presented for cognitive body area networks. The proposed scheme is based on illumination problem from the convex optimization field. The proposed algorithm can minimize the interbody area network interference by reducing the transmission power by 4.5 dBm.
Clockwise: IEEE 802.15.6 Channel Models; (a) Average BER for the SCA algorithm; (b) Average BER for the RL-CAA algorithm; (a) Average throughput for the SCA algorithm; (b) Average throughput for the RL-CAA algorithm; (a) Comparison of individual dissatisfaction probabilities of 5–50% throughput margin values; (b) Comparison of total averaged dissatisfaction probabilities for 5–50% throughput margin values. Source: Ahmed, T.; Le Moullec, Y. (2017). A QoS Optimization Approach in Cognitive Body Area Networks for Healthcare Applications. Sensors, 17 (4), 1−23
Sensors and lab-on-chip
The design and implementation of small and smart wearable and implantable data acquisition systems in e.g. medicine and robotics require cost-efficient and energy-efficient, yet reliable, sensing and communication solutions. The overall research directions include microfluidic-based lab-on-a-chip (LoC) systems for point-of-care and need-of-care applications (e.g. for the detection of pathogens, mastitis), and sensors based on e.g. wideband technology, wired (galvanic electrode systems) and wireless (radio, radar, eddy current) excitation systems with the development of electrical circuits for high sensitivity differential electrical impedance spectroscopy (EIS) measurement and for amplitude limiting in the current-to-voltage converter part of the EIS measurement setup.
Moreover, these above topics are to be strategically combined with innovative techniques for sensor signal processing like generation of excitation signals, e.g. the analysis of the dependence of the electrical model fitting accuracy on the number of the excitation frequencies as well as enhanced iterative algorithm for the crest factor minimization of multisine, and IoT-based communication, specifically the smart and compressive signal processing as well as the cognitive communication topics mentioned in the two previous points. Time-frequency domain Fourier and wavelet/chirplet binary-ternary transforms.
Examples of PhD theses related to this research area:
Development of multiscale smart surfaces for neuromorphic applications
PhD student: Rauno Joemaa
Supervisors: Tamas Pardy, Toomas Rang
PhD thesis defence expected in 2022
Neuromorphic engineering is a novel interdisciplinary field, focusing on the design of biomimetic artificial neural systems that emulate the electrical activity of biological neural networks. These systems can greatly help in the development of neuroprosthetics and help better understand the human nervous system. Neuromorphic circuits of the past employed VLSI electronics, using purely electrical signals to communicate to live neurons in a hybrid system.
Microfluidic artificial neurons on the other hand are a novel approach that can mimic the internal electrochemical processes of live neurons, and therefore offer a great interface between silicon-based electronics and living cells or can replace the live cells entirely. The development, and potentially, networking of microfluidic artificial neurons brings together the knowledge from various domains: primarily electrical engineering and neurobiology, secondarily physics and chemistry.
An artificial microfluidic neuron must have an ion-selective membrane similar to the neural cell wall and must model the ion exchange processes characteristic of neurons, that produce an action potential. As the modelling of different microfluidic solution for producing electrochemical effects comparable to those of the natural action potential are explored using the Finite Element Method (FEM), the results are best proven in direct contact with the live cells.
Stable cellular incubation to test biomimetic concepts or conduct pharmacological studies requires an application which falls in the bounds of an Organ-on-Chip (OOC). Such a device can be advantageous in the pharmacological studies by creating an option to test neurological effects without animal test subjects or bypass the animal stage entirely and use human neurons instead, broadening into personalization.
In order to determine whether communication between a biological neuron and a microfluidic biomimetic analog can work in both directions or one which requires the use of multiple technological solutions, will be considered as different smart surfaces are evaluated. The thesis will take steps to include exploring an automated microfluidic incubation system utilizing immobilization methods which enable keeping live neurons active and in short reach of continuous external influence and monitoring.
One of the main approaches looked at is found in the largely underdeveloped area in microfluidic reactors - the patterning and deposition of electrospun nanofibers. Electrospinning is a method of nanofiber deposition, which is cheap, easily scalable and technologically simple to implement, but provides robust polymeric fiber meshes that intrinsically resemble the extracellular matrix (ECM).
Development of Bioimpedance Based Method and Low Cost Instrument for Non-invasive Tissue Characterisation with Eddy Currents
PhD student: Hip Koiv
Supervisors: Mart Min and Alvo Aabloo (University of Tartu)
PhD thesis defence expected in 2021
Check out Hip's video here: https://youtu.be/HdiOXbGlDn4
High blood pressure (hypertension) is a primary risk factor for different cardiovascular diseases and keeping it under control is vital to reduce the risk of many dangerous heart conditions. Brachial cuff sphygmomanometer is widely used to assess the pressure parameters, but we have known for over a half century that brachial pressure is a poor surrogate for central aortic pressure (CAP). CAP represents the true load imposed on the heart and large arteries, but there are many reasons why we are still so stuck in old methods (i.e., lack of proper guidelines and standards for alternative technologies).
Our impedance workgroup is developing a wearable device that measures bioimpedance from the wrist (on radial artery) and estimates blood pressure in the aorta using transfer functions (Figure 1).
Figure 1. Bioimpedance variations are registered from the radial artery with four-electrode configuration. From peripheral radial pressure wave it is possible to get central aortic pressure (CAP) using transfer function (TF).
For the measurement of small bioimpedance variations, we need reliable electrodes that enable low electrode-skin impedance. For a wearable device, similar to a wristwatch or a wristband, electrodes have to be dry or semi-dry and reusable. Unfortunately, it is quite a challenge because bioimpedance method requires a small excitation current flowing through the body and two electrode-skin interfaces as well. Hydrogel between the commonly used gel electrode (Ag/AgCl) and the skin can reduce this interface impedance but these commercial electrodes do not have reusable nature.
My research focuses on soft, flexible and dry (and semi-dry) electrodes that are based on silicone (PDMS), carbon nanofibers (CNF) and carbon fibers (CF). Longer carbon fiber strands that stick from the base material reduce the electrode-skin interface and make a better contact with the skin than the smooth texture (Figure 2).
Figure 2. A and B – side view of the CNF/CF-PDMS electrode and carbon fibers sticking from the base material. C – top view of the electrode.
In addition to dry electrodes I am working out semi-dry electrodes together with Tartu University IMS (Intelligent Materials and Systems) lab. These novel electrodes are based on ionic liquids (as electrolyte) and carbon textile or carbon/silicone composite. Ionic liquids have very low vapor pressure making the surface hold its moisture for a long time. This means that the electrolyte layer does not dry out as usual hydrogel does, which is very beneficial when making reusable electrodes. In addition, to test and validate the new electrode material, custom tissue phantoms are developed. Phantoms are expected to have tissue similar dielectric properties and precision across the band of interest and reasonable shelf-time to mimic the real testing situation.
PhD thesis defence expected in 2021
Methods for Agent Detection in Lab on Chip Applications
PhD student: Kaiser Parnamets
Supervisor: Tamas Pardy
PhD thesis defence expected in 2021
Under construction
Development of Sensors based on Wide Bandgap Heterostructures
PhD student: Udayan Sunil Patankar
Supervisors: Ants Koel, Tamas Pardy
PhD thesis defense expected in 2021
Heterostructures have become essential constituents of most advanced electronic devices. These structures are well suited for high frequency and fast switching digital electronic applications. Heterostructures are of great interest because; motion of charge carriers can be controlled by modifying energy band profiles of constituent materials. During the past few years, heterostructures based on silicon carbide (SiC) polytypes have come to prominence due to their promising physical and electrical properties. Very important efforts have been made in the last decade on the development of wide band gap materials. It is crucial for any industrial development to produce large-size materials with good quality at a reasonable cost.
Unfortunately, crystal growth of these refractory materials is difficult. The SiC has the potential to be obtained using very high temperature sublimation, PVD or CVD techniques. Silicon carbide (SiC) material possesses excellent physical robustness, chemical resistance and multiple options for smart devices through its electrical, chemical and optical properties It is also an ideal surface for developing another important material like graphene, with superior physical, chemical and electrical properties. Silicon carbide heterostructures fabricated from their most popular polytypes, 3C-SiC, 4H-SiC and 6H-SiC have high value of breakdown voltages and hole mobility. They are extensively used as power electronic devices, sensors and light emitting diodes.
Attractive properties and a wide range of applications of SiC heterostructures are attracting researchers to explore the use of SiC heterostructures in the field of sensorics. The objective research work is to understand the different characteristics of a wide band semiconductor heterostructures devices through thorough device simulations ie how they behave electrically in different conditions and problems associated with its fabrication. Investing in heterostructures can improve sensing capabilities of wide band SiC semiconductor like particle sensing, vibration, UV light sensing, emitting possibilities etc. under different conditions.
Figure 1. (a) (b) Device Structures and (c) Energy band diagram of SiC / Si nn Heterojunction
Figure 2. Device IV characteristics
Figure 3. CV Measurement of 6H-SiC 3.2mm 2 Substrate 80K to Room Temperature
Numerical Characterization of Interface Layer between Metal Film / Wide Bandgap Semiconductor and Development of Semiconductor Devices
PhD student: Mehadi Hsan Ziko
Supervisors: Ants Koel, Tamas Pardy
PhD defense expected in 2021
One of the research tasks in the semiconductor industry is inexpensive manufacturing of heterojunctions and reliable contacts. It is not very common technology to join the different material layers (like different SiC wafers and metal contacts to them) using direct bonding. Traditional sputtering and evaporation technologies have a number of drawbacks, naming for example cost, many processing steps, time-consumption, realization of metal layer with homogeneous thickness over a large area of contact, restrictions for the general thickness of the metal layer over the whole contact, etc.
This direct bonding will reduce the fabrication cost of the devices, improve their electrical properties, and solve so many processing problems. Direct bonding is possible with diffusion welding (DW) technology that has been proposed as the first topic in my research work. Generally, the DW technology gives the possibility to improve some quality features of contacts and improve the realization of metallization process. The drawbacks unfortunately also lie in quality - achieving defect less surfaces of two hard materials (like two SiC wafers). Shear micro deformation can take place during DW (at applied high pressure and temperature), in a subcontract surface layer of SiC.
During the development of metal contacts for p-SiC substrates, it has been observed that about 5 nm - 25 nm amorphous layer develops between the metal film semiconductor surfaces, which seems to influence the electrical characteristics of the Schottky structures. These traditional contact challenges in electronics and semiconductor structures are influencing the performance due to contact quality and reliability and finally influence on the device performance.
So, it is necessary to develop well-accepted understanding of the mechanism and hence the need to investigate the reasons missing behind this influence. Therefore, one of the main tasks (and method) is in the development of numerical models for such a multi-layer interface and try to develop the acceptable physical explanation of the behavior of for Metal-p-SiC interfaces.
Additionally, realization of large area contacts with experimental Schottky barrier diodes for power semiconductor structures is technologically enough tricky activity. The influence of such a layer improves on the base of our measurements specifically the forward characteristics of the structure.
Fig. Simulated and real Schottky barrier diode structure and its current- voltage characteristics.
Sources:
Simulations of Silicon Carbide (SiC) and Graphene based Novel Semiconductor Devices for Power Electronics and Gas Sensor Applications
(Formerly known as Characterization of Novel Photovoltaic Materials and Devices)
PhD student: Muhammad Haroon Rashid
Supervisor: Toomas Rang, Ants Koel
The goal of the project is to work in two different regimes to develop novel semiconductor devices for power electronics and gas sensing applications.
The first part of the thesis is the simulations of SiC based heterojunction devices (power diodes and LEDs) considering a novel technique called diffusion bonding. Heterostructures are of great interest because they give rise to interesting electrical and physical properties in the resultant semiconductor devices due to change of the semiconductor material at the heterojunction interface.
During the past few years heterostructures based on silicon carbide (SiC) have gained significant importance for power electronics application due to their high value of breakdown voltages and hole mobility. They are extensively used as power electronic devices, sensors and light emitting diodes. Currently, the most promising techniques with predicted results and stable yield for growing epilayers of SiC polytypes for the fabrication of electronic devices have been chemical vapour deposition (CVD), liquid-phase epitaxy (LPE) and molecular beam epitaxy (MBE).
These techniques are complex and require sophisticated apparatus and skills. Fabrication of SiC based heterostructure device with diffusion bonding will simplify the fabrication efforts and reduce the cost of the resultant devices. Micro- and nano-scale SiC based devices have been simulated with Silvaco TCAD Software and Quantumwise Atomistix Toolkit(ATK) respectively.
The second part of the project is to simulate graphene based gas sensors to detect a large variety of hazardous organic and inorganic gases in domestic and industrial environments to avoid accidents. The change in the electric current through the device in the presence of the target molecules is used as a molecule detection mechanism for the simulated sensors. A nano-scale semiconductor device simulator, Quantumwise Atomistix Toolkit has been used to simulate graphene based gas sensing devices.
Figure 1: Geometry of 3C/4H-SiC based NN-heterojunction diode simulated in Silvaco TCAD Software
Figure 2: 3D-view of 4H-6H/SiC nanoscale device with semiconductor electrodes simulated in ATK
Figure 3: ATK view of simulated structure of gas sensor
Microheating Solution for Molecular Diagnostics Device
PhD student: Tamas Pardy
Supervisors: Toomas Rang, Indrek Tulp, Ants Koel
PhD thesis defended in 2018. Link to fulltext.
Novel molecular diagnostics devices, primarily nucleic acid amplification tests (NAAT) mandate the use of precise temperature control. Labon-a-Chip (LoC) devices are self-contained molecular diagnostics devices, meaning they rely on no additional external instrumentation, thus the term noninstrumented NAAT (NINAAT). Integrating microheating into a noninstrumented molecular diagnostics device is challenging due mainly to the restrictions on cost, space and power. This limits commercialization efforts and therefore widespread use of rapid tests that would otherwise help decentralize clinical diagnostics, reduce waiting times for patients and healthcare costs in general.
This PhD work proposes a workflow and methodology for the development of temperature control for non-instrumented molecular diagnostics devices and demonstrates its application to a LoC NINAAT device, called the InTime NINAAT. In its first part the work details the evaluation process for temperature control options as well as describes the evaluation of four microheating options, namely chemical heating, self-regulating electrical heating, thermostat-regulated electrical heating and thermoelectric heating. For the evaluation, physical and simulated thermal models are constructed. These thermal models are simplified representations of the molecular diagnostics device in development and are used solely for thermal analysis. The work proposes self-regulating heating and thermostat-regulated heating for use as integrated microheating candidates and demonstrates that both are capable of temperature regulation in the specified target range with 0.5 °C steady-state error (SSE) in the LoC NINAAT system being developed.
The second part of the work describes the process by which the proposed microheating candidates are developed further and prepared for integration with the molecular diagnostics device. This part describes how to evaluate the candidates for multiple assay targets (demonstrated in the work through isothermal NAAT protocols) as well as in-detailed thermal analysis for a single assay target (the LAMP protocol in this work). Both microheating candidates developed for the NINAAT platform were demonstrated compliant with assay requirements and could maintain target temperatures for the LAMP protocol in 95% of the reaction volume in steady state as calculated from simulated thermal models. The self-regulating heating candidate was chosen as the final solution due to its simplicity and lower cost compared to the alternative. A microheating solution was proposed based on a polymer resin PTCR self-regulating heater for use with the functional prototype of the NINAAT microfluidic chip. The heating solution was demonstrated to be capable of maintaining reaction temperatures in the required 60-63 °C range for 20 minutes powered by 2xAAA alkaline batteries and to reach the target range in 10 minutes. From the simulated thermal model, it was calculated that about 85% of the reaction volume was in range in steady state. In a final experiment the functional prototype NAAT chip along with the microheating solution were demonstrated capable of executing the LAMP protocol and successfully detected the target DNA.
The final part details the integration of the microheating solution into the functional prototype of the developed molecular diagnostics device. In this part the proposed microheating solution is integrated into the InTime NINAAT functional prototype device, complete with all fluidic and user interface functions. The InTime NINAAT device is a self-contained non-instrumented NAAT platform that carries a DNA amplification workflow from sample input to result output. After integrating the heater and a final compliance check, the device is demonstrated to work, as it would be used by the end-user.
 |
 |
 |
 |
Clockwise: Developing microheating candidates into microheating solutions for use in the functional prototype molecular diagnostics device; Thermal modelling for the core chip functional prototype indicated about 85% of the reaction volume was in the defined target temperature range; Thermal model for the core chip functional prototype (a) including a frame, electrical connections (b) and a heater (c); LAMP testing in core chip functional prototype (left) resulted in successful DNA amplification (right). Source: Tamas Pardy, Microheating Solution for Molecular Diagnostics Device, PhD thesis, Tallinn University of Technology, 2018
Wearable Solutions for Monitoring Cardiorespiratory Activity
PhD student: Margus Metshein
Supervisors: Paul Annus; Mart Min; Alvo Aabloo
PhD thesis defended in 2018. Link to Fulltext.
The principle objective of this thesis is to propose the knowledge of the means for providing the best possible signal that contains the information of cardiorespiratory activity. Cardiorespiratory signal carries useful data of the condition and status of the subject: heart rate, respiratory rate, the condition of cardiovascular system, the presence of lung water etc. In order to gain access to cardiorespiratory activity, the goal of implementing the components of wearable experimental systems, based on capacitive, inductive and resistive coupling, was established and fulfilled.
The work is motivated by the need for wearable and unnoticeable solutions for monitoring the defined volumetric changes in an organism. The trend of continuously following the physiological parameters, either in everyday life for personal body condition monitoring, or as a medical observation of patients in the risk groups of some aggravated illness, is demanding the means for implementing the assigned tasks. Although the effect of motions is moderately researched, the techniques based on electrical methods for monitoring cardiorespiratory activity in remote are strongly affected, proposing an obstacle in the development of sensors. Currently, there is a lack of knowledge for the positioning of electrical sensors relative to body in order to monitor cardiorespiratory activity.
The principle of capacitive, inductive and resistive coupling is conceptually the same: by using some sort of electrical stimulation, the object is excited and the response in the form of an electrical signal is measured. The sensitivities and the relevant current distribution, affected by the volumetric changes, is expected to influence the measured response. The most vulnerable part is the interface between the sensors and the object i.e. coupling, which is capable of causing interferences that corrupt the interesting signal. This thesis deals with the following: the schemes of sensors for providing the signal of cardiorespiratory activity, the effect of positioning sensors on the surface of the body on relative organs, the effect of concurrent movements and the variations relative to cardiorespiratory activity.
The achieved results show the significance of the exact positioning and the need for aware choice of configurations and placements of electrodes. The solution of utilization of large area electrodes in capacitive connection for mapping the thoracic surface in order to measure the EBI for monitoring cardiorespiratory activity, constitutes a relatively new field and the results give grasp to the future of contactless monitoring.
The effect of two essentially different shapes of inductively coupled coils on the measured signal was experimentally determined, from which the Fo8 coil proposes a novel approach in focused field.
The most suitable positions and configurations of electrodes have been experimentally determined and proposed, showing the vantage of distal arrangement relative to radial artery in front of circular. The optimal pressure point for externally applied pressure on rigid electrodes, located on radial artery, was determined since which the modulation in the measured signal of EBI is not increasing significantly any more. As the resistive solution presumes the galvanic contact, the externally applied pressure contributes to further improve the quality of measured signal of pulse wave, designating a relevant outcome.
Clockwise: 4 Block diagram of the proposed idea of EBI measurement device; Attachment of the implemented prototype of EBI measurement device on ES1; Influence of the choice of EPC on the measured Sresp (parameter no. 1) and Scard (parameter no. 2); Influence of the choice of EPC on the measured Sresp:move (parameter no. 3) and Scard:move (parameter no. 4).. Source: Margus Metshein, Wearable Solutions for Monitoring Cardiorespiratory Activity, PhD Thesis, Tallinn University of Technology, 2018