Remote Powering and Wireless Communication of a Smart Knee Prosthesis

With the advance in microelectronics and microsystems, better instruments in every domain are designed and built. For many years, scientists have tried to understand nature, and engineers to mimic the perfect design. One example of this is the human knee, which is usually a very long-lasting joint. Due to accidents, infection and misuse, some patients need to have total knee replacement surgery. Therefore, knee prostheses are implanted. While some patients recover quickly and forget on which knee they have had the surgery, others cannot escape from the pain, and need additional operations. Understanding the reasons for this will ultimately help to design better prostheses, and speed up the healing process. In addition, due to many reasons that merit investigation, prostheses last at most for only 15 years. With the necessary sensors added to these prostheses, the necessary data for understanding the reasons can be collected. This leads to the design of an electronic knee prosthesis. However, creating an electronic knee opens up many challenges. The SImOS project was started to examine feasibility and to reach the goal of building an electronic knee prosthesis. It is going to hold some force sensors to measure the forces in two condyles to achieve a better balance in the knee prosthesis. In addition, it is going to have magnetometers to understand the kinematics of the knee. All these factors increase the power consumption of the implanted electronics. However while remote powering is already an issue, further increasing the power consumption becomes more challenging. On the other hand, with the development of better tools, such as complex field solvers, it is easier to perform more complex analyses. This helps to review the already-used methods and propose better ones. To design the electronic knee, the problems related to the remote powering and communica- tion should be also solved. Inductive powering is an attractive solution for this kind of case and is already being used in biomedical applications. The small distance requirements between the external base station and the implantable medical devices and the fact that they are easily achievable with fairly small-sized antennas makes this method widespread. However, in some cases, the decision of the design parameters could be more challenging. If the distance is increased further, and metallic parts are added, the remote powering efficiency decreases drastically. As the subject of this thesis, a better approach to designing the remote powering coils is presented to overcome this problem and to achieve sufficiently high efficiency. This is done by showing that the s-parameter simulation or measurement results can be used directly to calculate the remote powering efficiency of the coil pairs. This allows us to use the result of the available field solvers directly to define an objective function. Since the objective function is easy to compute, by using an optimizer, which is chosen as simulated annealing in this thesis, it is shown that better coils can be designed. The achievable remote powering efficiency of the coil pair increases by designing the coil independently of the load. In this case, the optimal load of the coil would be different from the actual load. Hence the achievable efficiency cannot be reached. On the other hand, the actual load could be converted to the optimal load by using a matching network. This method is usually avoided due to the losses of the matching network elements especially the inductors. However, it is shown that the efficiency can be maintained nearly as it is by using only capacitors as the matching network elements in this design. It is also shown that it is possible to decouple the link design from the load by using a matching network so that the global remote powering efficiency is increased. In this case the performance of the coil pair only depends on the available volume and the distance in between. In addition, instead of using load shift keying for wireless communication, which is generally used for RFID tags, it is shown that building a stand-alone transmitter gives better results in terms of power consumption for high power-demanding applications. This thesis shows that the direct power consumption of a stand-alone transmitter could be less than the indirect power consumption that would occur in the case of load shift keying. By designing this transmitter the efficiency is maintained nearly as it is. All in all, 30% inductive link efficiency is achieved by using inductive powering for a distance of 2.5 cm. The coil at the implant is designed as an air-core rectangular solenoid and has dimensions of 18.2 mm x 2.8 mm x 3.9 mm. The proposed simulated annealing method is used for dimensioning the coil pair. With the designed microelectronic circuits in 0.18 μm technol- ogy, and the external class E-type power amplifier, 14% overall remote powering efficiency is obtained. 2.8V in addition to 1.8V operating voltage is provided by decoupling the link from the microelectronic design. With the designed switched-capacitor voltage doubler, the 2.8V supply is generated with an remote powering efficiency of 12%. For wireless bi-directional communication, a stand-alone transmitter is designed instead of the load shift keying. The FSK/OOK transmitter consumes 350 μW while transmitting at a rate of 57 kbps. For sending the commands to the implant, the power amplifier is designed as an ASK modulator. The required ASK demodulator at the implant can demodulate up to 1 Mbps of data-rate, while consuming only 15 μW.

Electromagnetic Environment Associated with Lightning Strikes to Tall Strike Objects

Seyed Abbas Mosaddeghi

The study of the lightning interaction with tall strike objects has attracted considerable attention of lightning researchers lately. Many lightning measurements including current and associated electromagnetic fields were recently made all over the globe namely in Russia, South Africa, Germany, Brazil, Japan, and Austria. It is a novel area of studies, and the resolution of associated questions will have an impact upon many lightning-related applications such as lightning protection and the determination of lightning parameters from remote field measurements. The main objective of the thesis is to carry out further theoretical investigations and experimental measurements to understand and elucidate recently raised questions on the characteristics of lightning return-strokes to tall structures and their associated electromagnetic radiation. Chapter 2 presents a review on recent progress in the modeling of lightning strikes to tall towers and associated experimental data obtained during the last decade or so. Two types of return stroke models namely the Engineering Models, and the Electromagnetic or Antenna-Theory (AT) models, extended to take into account the presence of a tall strike object are discussed. The Chapter contains also a description of the computational methods for the evaluation of electromagnetic fields generated by a lightning strike to a tall structure, as well as an overview of available data on lightning current and associated electromagnetic fields. The chapter finally highlights some important questions raised by different research groups in the past few years which call for further investigations. These questions are as follows: No systematic theoretical analysis nor experimental data are available for electromagnetic fields in the immediate vicinity of a tall structure struck by lightning. The characterization of nearby electromagnetic fields is particularly important in the analysis of the interaction to nearby electrical and electronics systems. Why do lightning return stroke models not reproduce the far-field zero crossing associated with lightning to tall structures? How should these models be revised to be able to reproduce such an effect? How should the engineering models be revised in order to remove the associated current discontinuity at the return stroke wavefront? It is well-known that the measurements of electromagnetic fields from lightning are affected by the presence of nearby buildings and metallic structures. However, no systematic and quantitative analysis of such an effect is presently available in the literature. The work presented in this thesis addresses all of the above questions. The main original contributions of this thesis, consisting of both theoretical and experimental work, are presented in Chapters 3 through 6. Chapter 3 is devoted to a theoretical description of the signature of electric and magnetic fields at very close distance associated with lightning strikes to a tower. It is shown that the electric field generated by a lightning return stroke to a tall structure can change polarity at very close distance range. This change in the polarity seems to be a specific signature of the very close vertical electric field. A simple equation is derived which provides an estimate of the critical distance below which such an inversion of polarity might occur. It is also shown that the inversion of polarity depends on the value of the reflection coefficient at the base of the tower and disappears for reflection coefficients close to 1. On the other hand, other parameters such as the return stroke speed, the reflection coefficient at the top of the strike object, and the adopted return stroke model seem not to have an impact on the inversion of polarity. Simulation results also showed that the electric field peak at distances beyond the height of the tower or so exhibits the typical 1/r dependence. At closer distances, however, the E-field peak features a saturation, due to the so-called tower shadowing effect. This shadowing effect results in a substantial decrease of the nearby electric field. On the other hand, the magnetic field peak varies inversely proportional to the horizontal distance and does not depend significantly on the presence of an elevated strike object. Chapter 4 introduces an improved version of the engineering models for return-strokes to tall structures which accounts for (1) the presence of possible reflections at the return stroke wavefront, and, (2) a return stroke initiation above the structure due to an upward connecting leader. We also propose an elegant iterative solution that can be easily implemented into computer simulation programs to take into account in a straightforward way multiple reflections occurring at the discontinuities at the tower ends and at the return stroke wavefront. Simulation results for the magnetic fields are compared with experimental waveforms associated with lightning strikes to the CN Tower (553 m). It is shown that taking into account the reflections at the return-stroke wavefront results in better reproducing the fine structure of the magnetic field waveforms. Chapter 5 presents and discusses obtained measurements of electric (vertical and radial) and magnetic fields from leaders and return strokes associated with lightning strikes to the Gaisberg tower in Austria obtained in 2007 and 2008. The data include simultaneous records of vertical and radial electric fields, which were obtained for the first time at such close distances. It is found that the vertical and radial electric field waveforms appear as asymmetrical V-shaped pulses. For the vertical electric field, the initial, relatively slow, negative electric field change is due to the downward leader and the following fast positive field change is due to the upward return stroke phase of the lightning discharge. For the horizontal electric fields, however, the bottom of the V is not associated with the transition from the leader to the return stroke. The horizontal field change due to the return stroke is characterized by a short negative pulse of the order of one microsecond or so, starting with a fast negative excursion followed by a positive one. In addition, an analytical expression for the radial electric field, assuming a uniform charge distribution along the leader with constant speed is derived. It is also shown that the return-stroke vertical electric field changes appear to be significantly smaller than similar measurements obtained using triggered lightning. This finding confirms the shadowing effect of the tower predicted by the theoretical analysis of Chapter 3, which results in a significant decrease of the electric field at distances of about the height of the tower or less. Finally, the ability of two different models for the return stroke in reproducing measured vertical and horizontal electric fields is tested using the obtained measured data. The considered models are (1) the engineering MTLE (Modified Transmission Line with Exponential Decay) model, and (2) the electromagnetic model implemented using the Numerical Electromagnetics Code NEC-4. It is shown that both models predict electric field waveforms which are in reasonable agreement with measured waveforms. In general, the predicted fields by the electromagnetic model appear to be in better agreement with measured data, because of the direct use of the measured current waveform as an input and the more accurate representation of the tower. Chapter 6 reports on the effect of nearby buildings on electromagnetic fields from lightning. Indeed, sensors used for the measurement of lightning electric and magnetic fields are often placed close to or on top of buildings or other structures. Metallic beams and other conducting parts in those structures may cause enhancement or attenuation effects on the measured fields. Experimental waveforms radiated from distant natural lightning recorded during the summers of 2006 and 2007 are presented. Electric and magnetic field waveforms were measured simultaneously on the roof of a building and on the ground at different distances away from it. The results suggest that the measured electric field on the roof of the building could be enhanced by a factor of 1.7 to 1.9, whereas the electric fields on the ground experienced a significant reduction by a factor ranging from 5 to 20. Also, it is shown that for a sensor located on the ground close to a building, the magnetic field component perpendicular to the building can experience significant attenuation, presumably due to the effect of the induced currents in the building. The magnetic field on the roof of the building seems not to be significantly affected by the building. Simulations using the Numerical Electromagnetic Code (NEC-4) were also carried out in which the building was represented using a simple wire-grid model. The simulation results support in essence the findings of the experimental analysis, despite quantitative differences which are ascribed, at least in part, to the oversimplified model of the building.

EPFL2011

Localizing the Source of an Epidemic Using Few Observations

Brunella Marta Spinelli

Localizing the source of an epidemic is a crucial task in many contexts, including the detection of malicious users in social networks and the identification of patient zeros of disease outbreaks. The difficulty of this task lies in the strict limitations on the data available: In most cases, when an epidemic spreads, only few individuals, who we will call sensors, provide information about their state. Furthermore, as the spread of an epidemic usually depends on a large number of variables, accounting for all the possible spreading patterns that could explain the available data can easily result in prohibitive computational costs. Therefore, in the field of source localization, there are two central research directions: The design of practical and reliable algorithms for localizing the source despite the limited data, and the optimization of data collection, i.e., the identification of the most informative sensors. In this dissertation we contribute to both these directions. We consider network epidemics starting from an unknown source. The only information available is provided by a set of sensor nodes that reveal if and when they become infected. We study how many sensors are needed to guarantee the identification of the source. A set of sensors that guarantees the identification of the source is called a double resolving set (DRS); the minimum size of a DRS is called the double metric dimension (DMD). Computing the DMD is, in general, hard, hence estimating it with bounds is desirable. We focus on G(N,p) random networks for which we derive tight bounds for the DMD. We show that the DMD is a non-monotonic function of the parameter p, hence there are critical parameter ranges in which source localization is particularly difficult. Again building on the relationship between source localization and DRSs, we move to optimizing the choice of a fixed number K of sensors. First, we look at the case of trees where the uniqueness of paths makes the problem simpler. For this case, we design polynomial time algorithms for selecting K sensors that optimize certain metrics of interest. Next, turning to general networks, we show that the optimal sensor set depends on the distribution of the time it takes for an infected node u to infect a non-infected neighbor v, which we call the transmission delay from u to v. We consider both a low- and a high-variance regime for the transmission delays. We design algorithms for sensor placement in both cases, and we show that they yield an improvement of up to 50% over state-of-the-art methods. Finally, we propose a framework for source localization where some sensors (called dynamic sensors) can be added while the epidemic spreads and the localization progresses. We design an algorithm for joint source localization and dynamic sensor placement; This algorithm can handle two regimes: offline localization, where we localize the source after the epidemic spread, and online localization, where we localize the source while the epidemic is ongoing. We conduct an empirical study of offline and online localization and show that, by using dynamic sensors, the number of sensors we need to localize the source is up to 10 times less with respect to a strategy where all sensors are deployed a priori. We also study the resistance of our methods to high-variance transmission delays and show that, even in this setting, using dynamic sensors, the source can be localized with less than 5% of the nodes being sensors.

EPFL2018

All-printed low-power metal oxide gas sensors on polymeric substrates

Danick Briand, Saleem Khan

This paper presents a novel approach for the fabrication of miniaturized and fully printed metal-oxide (MOX) gas sensors on a polyimide (PI) substrate by using digital manufacturing of the functional layers, namely aerosol jet and inkjet printing technologies. We are reporting a stacking approach for all-printed MOX sensors for the first time. High resolution (similar to 20 mu m) printing of metallic patterns is enabled by aerosol jet of gold nanoparticles in solution, which leads to printed resistive thermal transducers with a reduced size of 500 x 500 mu m(2). The wide area printing feature of the aerosol jet is also applied for printing a thin (similar to 2 mu m) inter-dielectric PI layer in between the heater and gold electrodes. The gas sensing layer, i.e. Pd-doped tin oxide (SnO2) nanoparticles, is deposited through drop on demand inkjet printing. Proper operation of the printed sensors is evaluated in both dry and humid conditions under reducing and oxidizing gases i.e. CO and NO2, respectively. All the tests are performed at 250 degrees C, produced by the integrated micro-hotplate at a low power consumption of similar to 78 mW. The chemo-resistive responses of the sensors towards both the gases are found to be in the acceptable range as compared to conventional metal-oxide gas sensor responses. These results are very promising for future integration of MOX gas sensors notably in smart printed electronics, disposable systems and wearables.