Eric Slottke

Constrained maximum likelihood positioning for UWB based human motion tracking

In this paper, the problem of human motion tracking with ultra-wideband radio nodes is addressed. We provide a general maximum likelihood formulation of the positioning problem based on range measurements which can handle synchronous and asynchronous agents. Geometrical constraints on the node topology, which are imposed by the human body, are also taken into account. For a Gaussian ranging error model and the specific problem of arm motion tracking, we derive the maximum likelihood estimation rule and calculate an analytical expression for the unconstrained and constrained Cramér-Rao Lower Bound. With these results, we study analytically and via computer simulations under what circumstances the geometrical constraints lead to performance gains. It is found that the largest benefits are obtained in case of asynchronous agents and for certain arm positions. Intuitive reasons for this phenomenon are given. Finally, we verify these findings and evaluate the position location performance experimentally with range estimates obtained from measured ultra-wideband channel impulse responses including the impact of the human body.

Analysis of Channel Parameters for Different Antenna Configurations in Vehicular Environments

In order to be able to investigate future Vehicle-to-Vehicle (V2V) communication systems, the need for realistic simulation environments is increasing. Especially, the integration of realistic antenna models which account for several non-ideal effects such as e.g. mutual coupling or the influence of a finite ground plane size is important. In this paper, a comparison of two possible multiple-input and multiple-output (MIMO) antenna configurations consisting of three antennas mounted on the car at different positions is carried out using a complex simulation environment. In particular, the receive diversity and the MIMO capacity are considered since the link robustness as well as the increase of spectral efficiency are important issues.

A simulation study of traffic efficiency improvement based on Car-to-X communication

There are several promising applications for car-to-x communication that aim at the improvement of traffic efficiency. With the increasing maturity of car-to-x communication systems, insights into the impacts of these applications on vehicular traffic are required. In this paper, we present a simulation study of traffic efficiency improvements that can be obtained by the utilization of a merging assistance application ahead of a freeway lane drop. Our investigations are based on the traffic simulation tool AIMSUN. We connect the traffic simulator with a simulation model that implements the merging assistance application as well as the message exchange between vehicles and a roadside unit. Simulation results show significant traffic efficiency improvements indicated by travel time reductions of up to 30% in dense traffic under ideal conditions.

Accurate Localization of Passive Sensors Using Multiple Impedance Measurements

We consider the problem of locating a sensor in a network of nodes which use inductive coupling as means of interaction. Conventional approaches in this setting require the located node to have a power supply or complex circuitry. By analysis of the circuit model underlying inductive coupling, we develop a localization method which allows for passive nodes with a simple layout. The proposed method reconstructs the unknown position from measurements of the input impedance at a set of reference nodes via the Nelder-Mead simplex algorithm. We show that the accuracy of the localization depends highly on the initialization value used for optimization, and provide a heuristic grid search algorithm to increase the quality of the initial position and thus of the localization process. A lower bound on the performance is provided for a two-dimensional setup; it can be closely met by the proposed localization algorithms in a realistic setup. The results show that the discussed approach can provide very high localization accuracy with a low-complexity node configuration.

Circuit Based Near-Field Localization: Calibration Algorithms and Experimental Results

For a network consisting of inductively coupled nodes, we experimentally demonstrate circuit-based localization of a passive and chipless agent device. By using measurements of the input reflection at multiple anchor nodes, we reconstruct the unknown position of an agent node in a coplanar arrangement for both random and optimized anchor arrangements. A key step for applying this localization scheme in a practical setup is the introduction of a calibration phase which, in contrast to previous work, eliminates the need for explicit knowledge of the involved circuits. Furthermore, we improve localization performance by investigating imperfections in the measurement device and applying appropriate compensation methods. In the employed setup, we demonstrate a median localization RMSE of 1.1 mm, making the proposed scheme a promising method for precise localization in clouds of RFID-like wireless devices.

Single-anchor localization in inductively coupled sensor networks using passive relays and load switching

Wireless physical layers based on inductive coupling are receiving attention in the context of RFID and, more recently, sensor networks. In this work we present the first method for localization in inductively coupled networks which only requires a single anchor. The utilization of passive relays with known positions resolves the associated problem of localization ambiguity. We demonstrate that a single impedance measurement can be sufficient to locate a node in two dimensions. Localization performance can further be increased by deterministic load switching. Numerical results indicate that a median localization error in the millimeter range is possible under realistic conditions.

Magneto-inductive passive relaying in arbitrarily arranged networks

We consider a wireless sensor network that uses inductive near-field coupling for wireless powering or communication, or for both. The severely limited range of an inductively coupled source-destination pair can be improved using resonant relay devices, which are purely passive in nature. Utilization of such magneto-inductive relays has only been studied for regular network topologies, allowing simplified assumptions on the mutual antenna couplings. In this work we present an analysis of magneto-inductive passive relaying in arbitrarily arranged networks. We find that the resulting channel has characteristics similar to multipath fading: the channel power gain is governed by a non-coherent sum of phasors, resulting in increased frequency selectivity. We propose and study two strategies to increase the channel power gain of random relay networks: i) deactivation of individual relays by open-circuit switching and ii) frequency tuning. The presented results show that both methods improve the utilization of available passive relays, leading to reliable and significant performance gains.

Inductively Coupled Microsensor Networks: Relay Enabled Cooperative Communication and Localization

In this thesis, we study a novel paradigm for wireless sensor networks: we envision a dense microsensor network, consisting of hundreds or thousands of highly miniaturized wireless nodes with millimeter or sub-millimeter dimensions. Such a microsensor network has many interesting applications ranging from in vivo medical sensing to environmental monitoring. However, the design and operation of the envisioned type of network are challenging: the large number of nodes, in combination with the small form factor, imposes severe constraints on both node complexity as well as power consumption. We propose using inductive neareld coupling as advantageous physical layer choice, allowing an RFID-like operation of the network with wireless power supply from central reader devices and low-complexity tag design. The use of inductive neareld coupling has only rarely been studied in the context of microsensor networks. Our primary goals for inductively coupled microsensors are twofold: we want to enable reliable communication to and between sensor nodes, and perform accurate localization of individual sensors. Both tasks are a ected by the central limitation of a physical layer based on neareld coupling: the severely limited range of interaction. We will show throughout this thesis that the use of wireless relaying allows for overcoming this limitation. Based on a circuit-theoretic communication framework extended to inductively coupled microsensor networks, we rst explore fundamental properties and design limitations. We then investigate suitable approaches to relaying and study their use to increase both the communication range and link reliability. In addition to improving the communication performance of microsensor networks, we demonstrate the capability of inductively coupled relays to enable novel secondary uses of the wireless channel. To this end we show that relay-assisted microsensor networks can achieve distributed computation by introducing a scheme to implement wireless arti cial neural networks over multihop MIMO channels. We continue by showing the feasibility of accurate localization in an RFID-like setting. To this end, we propose a novel method that allows for the localization of a purely passive sensor only consisting of a simple loop antenna and a matching circuit. The unknown position of the sensor is hereby reconstructed from measuring

UWB Marine Engine Telemetry Sensor Networks: Enabling Reliable Low-Complexity Communication

We study wireless ultra wideband (UWB) sensor networks operating in the environment of marine engines for telemetry purposes. In the harsh propagation environment posed by the engine room, UWB enables robust communication without the need for complex sensor node implementations. For this work, we therefore present the first UWB channel measurements performed around, on, and inside a marine diesel engine, and use these measurements as basis for the design and reliability analysis of a low- complexity UWB communication system.

Universal computation with low-complexity wireless relay networks

We propose a method for enabling complex computations in a network of low-complexity wireless devices. By utilizing multihop relaying, such devices can form the wireless equivalent of an artificial neural network (ANN). We provide a method for programming the network functionality in a decentralized fashion and demonstrate the robustness of wireless ANNs against node failures and imperfections. Applications of this scheme exist in low-complexity sensor networks, where elaborate calculations can be carried out in a distributed fashion, or for creating powerful ANNs with very high degrees of interconnectivity realized by the wireless medium.