Thorsten Nowak

From radio telemetry to ultra-low-power sensor networks: tracking bats in the wild

Sensor networks have successfully been used for wildlife monitoring and tracking of different species. When it comes to small animals such as smaller birds, mammals, or even insects, the current approach is to use extremely lightweight RF tags located using radio telemetry. A new quantum leap in technology is needed to overcome this limitation and enable new ways to observe larger numbers of small animals. In an interdisciplinary team, we are working on the different aspects of such a new technology. In particular, we report on our findings on a sensor- network-based tracking solution for bats. Our system is based on integrated localization and wireless communication protocols for ultra-low-power systems. This requires coding techniques for improved reliability as well as ranging solutions for tracking hunting bats. We address the technological and methodical problems related to system design, software support, and protocol design. First field experiments have been conducted that showcase the capabilities of our system.

Protocol design for ultra-low power wake-up systems for tracking bats in the wild

We present a novel concept for a wake-up system based ultra-low power communication protocol for sensor networks. The main application field is monitoring contacts and even tracks of bats in the wild. Our sensor nodes can weigh at most 2 g out of which 1 g remains for the battery. We investigate a novel communication protocol design applicable to these systems and also showing great potentials for other ultra-low power sensor networks. In particular, we investigate the bat to ground communication by combining duty cycling with a multi-stage wake-up receiver. We employ Binary Offset Carrier (BOC) modulated signals that allow to accurately localize and track the bats while transmitting data in parallel. In a first step, we evaluated the conceptual design using a software-defined radio system to demonstrate the feasibility of the protocol design.

Combined localization and data transmission in energy-constrained wireless sensor networks

Many applications demand simultaneous localization and aggregation in wireless sensor networks. For such an application, e.g. wildlife monitoring, a signaling scheme for a combined localization and communication using a common set of subcarriers is proposed. The concept is based on binary offset carrier signals. But, in contrast to Global Navigation Satellite Systems the presented approach makes use of pure subcarrier localization, and thus enables data transmission in short burst signals. The ranging performance is assessed utilizing the Cramér-Rao Lower Bound depending on the amount of data transferred and considering bit errors.

Optimal network topology for a locating system using RSSI-based direction finding

Location-awareness is one of the major challenges in wireless sensor networks (WSNs) today. Received signal strength indicator (RSSI)-based localization techniques are a promising approach to provide location information in WSNs utilizing low-cost hardware. In this paper, the performance of RSSI-based direction-of-arrival (DOA) is assessed. Therefore, the Cramér-Rao Lower Bound (CRLB) for direction estimation with RSSI measurements is derived in an analytical form. Furthermore, a position CRLB RSSI-based DOA estimates is presented. Since RSSI-based DOA is highly dependent on the direction of the received signal, and hence position estimates are, the network topology has a major impact on the feasible precision of the locating system. A methodology is shown to find the optimal WSN topology evaluating the position CRLB for RSSI-based DOA measurements.

A Low-Cost RSSI-Based Localization System for Wildlife Tracking

This paper presents a low-cost localization system for the high-resolution tracking of bats. The system bases on a ground network consisting of multiple low-cost receiver stations, and ultra-lightweight transmitters mounted on the bats. A main challenge of the received signal strength based localization is the limited dynamic range of the employed low-cost receivers. This challenge is solved using an efficient 2-stage differential correlation concept. It significantly improves the dynamic wrt. conventional power detection methods. In addition, this concept requires low processing power and is robust wrt. frequency offsets. Finally, this paper presents a performance evaluation employing reference measurements recorded in the rain forest of Panama.

Fundamental limits in RSSI-based direction-of-arrival estimation

The use of wireless sensor networks is rapidly increasing. Also the demand of ubiquitous location sensors is swiftly expanding. Hence, energy and location-awareness come into focus of research today. A prospective approach for low-power locating sensor networks is received signal strength indicator (RSSI)-based direction finding. The presented approach is based on RSSI difference measurements retrieved by a array of directed antennas. In this paper, fundamental limits of RSSI-based direction finding are evaluated, beyond the Cramer-Rao Lower Bound (CRLB). That is not applicable for the design of a localization system topology due to the nature of the gain difference function that leads to an unbounded variance of the unbiased estimator. Thus, a maximum likelihood (ML) approach to the RSSI-based direction finding is presented. The ML estimator yields a limited variance for all signal directions. However, that benefit comes at the expense of being biased. Beyond treating direction estimates, mean square position errors are compared for both, the unbiased and the ML estimator.

A path loss and fading model for RSSI-based localization in forested areas

Location-awareness is a prerequisite for many applications, today. Demanding applications, e.g. wildlife monitoring, rely on precise location information. A prospective approach tracking tiny lightweight animals is to capture field-strength measurements utilizing a wireless sensor network. Apriori knowledge of propagation channel characteristics is essential for precise localization. Hence, in this paper a channel model for forested areas is proposed. Therefore, a reference set of field-strength measurements has been evaluated and a channel model has been derived from these measurements. Besides path loss and fading, also cross fading between two directional antennas has been characterized. Cross-fading between these antennas is crucial for RSSI-based direction finding by directional antennas. Furthermore, the impact of incorrect modeling measurement uncertainties on the position estimation error is assessed.

A low-cost RSSI based localization system design for wildlife tracking

We present a low cost localization system for bats tracking. The system is based on an angle-of-arrival estimation using field strength difference measurements of two directional antennas. The receivers are constructed of low-cost commercial broadcast receivers and an embedded processing platform. A concept for an efficient 2-stage differential correlation is shown to address the dynamic range limitations of the low-cost hardware, the processing power limitations and frequency offsets. We show that the dynamic range can be improved by >16dB compared to a power detection method. The full system has been validated in a field trial in the rain forest of Panama.

BATS: Adaptive Ultra Low Power Sensor Network for Animal Tracking

In this paper, the BATS project is presented, which aims to track the behavior of bats via an ultra-low power wireless sensor network. An overview about the whole project and its parts like sensor node design, tracking grid and software infrastructure is given and the evaluation of the project is shown. The BATS project includes a lightweight sensor node that is attached to bats and combines multiple features. Communication among sensor nodes allows tracking of bat encounters. Flight trajectories of individual tagged bats can be recorded at high spatial and temporal resolution by a ground node grid. To increase the communication range, the BATS project implemented a long-range telemetry system to still receive sensor data outside the standard ground node network. The whole system is designed with the common goal of ultra-low energy consumption while still maintaining optimal measurement results. To this end, the system is designed in a flexible way and is able to adapt its functionality according to the current situation. In this way, it uses the energy available on the sensor node as efficient as possible.

Impact of antenna pattern modeling errors on RSSI-based DOA estimation

In this paper, we elaborate a reference antenna in terms of direction of arrival (DOA) estimation accuracy to verify its reliability under realistic propagation conditions. Estimation accuracy decreases for the reference antenna due to environmental influences. Our main contribution is modeling such non-anechoic environment in the simulation by considering proximity to earth with surface roughness, antenna cover against climatic influences and height of the stacked-up system over earth in a real-world deployment. We developed a ring extension for the antenna, such environmental influences are minimized and the DOA estimation accuracy is improved again. The overall localization performance of a system with several DOA estimation sensor nodes for different antenna configurations is proven in a simulation environment. The results show an improvement of the localization accuracy for the optimized antenna.