Jan-Hinrich Hauer

Experimental Study of the Impact of WLAN Interference on IEEE 802.15.4 Body Area Networks

As the number of wireless devices sharing the unlicensed 2.4 GHz ISM band increases, interference is becoming a problem of paramount importance. We experimentally investigate the effects of controlled 802.11b interference as well as realistic urban RF interference on packet delivery performance in IEEE 802.15.4 body area networks. Our multi-channel measurements, conducted with Tmote Sky sensor nodes, show that in the low-power regime external interference is typically the major cause for substantial packet loss. We report on the empirical correlation between 802.15.4 packet delivery performance and urban WLAN activity and explore 802.15.4 cross-channel quality correlation. Lastly, we examine trends in the noise floor as a potential trigger for channel hopping to detect and mitigate the effects of interference.

Flexible hardware abstraction for wireless sensor networks

We present a flexible hardware abstraction architecture (HAA) that balances conflicting requirements of wireless sensor networks (WSNs) applications and the desire for increased portability and streamlined development of applications. Our three-layer design gradually adapts the capabilities of the underlying hardware platforms to the selected platform-independent hardware interface between the operating system core and the application code. At the same time, it allows the applications to utilize a platforms full capabilities-exported at the second layer, when the performance requirements outweigh the need for cross-platform compatibility. We demonstrate the practical value of our approach by presenting how it can be applied to the most important hardware modules that are found in a typical WSN platform. We support our claims using concrete examples from existing hardware abstractions in TinyOS and our implementation of the MSP430 platform that follows the architecture proposed in this paper.

Mitigating the effects of RF interference through RSSI-Based error recovery

On a common sensor node platform (Telos) we sample RSSI with high frequency during packet reception. We find that a packet collision (RF interference) often manifests as a measurable, temporal increase in RSSI. We investigate how the receiver can use this information to detect interference and, through temporal correlation, estimate the bit error positions in a corrupted packet. In an experimental study in two testbeds and several realistic BAN scenarios we show that a simple threshold-based algorithm often succeeds in estimating a large fraction of the bit error positions correctly. We develop an ARQ scheme that utilizes the error estimates to reduce the size of retransmitted packets. For this ARQ scheme we present an analytical model and verify it experimentally. Our results indicate that in comparison with a standard Send-and-Wait ARQ the expected number of bits per transmission can be reduced significantly (in our measurements by up to 14.7 %).

A component framework for content-based publish/subscribe in sensor networks

Component-based architectures are the traditional approach to reconcile application specific optimization with reusable abstractions in sensor networks. However, they frequently overwhelm the application designer with the range of choices in component selection and composition. We introduce a component framework that reduces this complexity. It provides a well-defined content-based publish/subscribe service, but allows the application designer to adapt the service by making orthogonal choices about: (1) the communication protocol components for subscription and notification delivery, (2) the supported data attributes and (3) a set of service extension components. We present TinyCOPS, our implementation of the framework in TinyOS 2.0, and demonstrate its advantages by showing experimental results for different application configurations on two sensor node platforms in a large-scale indoor testbed.

Opportunistic packet scheduling in body area networks

Significant research efforts are being devoted to Body Area Networks (BAN) due to their potential for revolutionizing healthcare practices. Energy-efficiency and communication reliability are critically important for these networks. In an experimental study with three different mote platforms, we show that changes in human body shadowing as well as those in the relative distance and orientation of nodes caused by the common human body movements can result in significant fluctuations in the received signal strength within a BAN. Furthermore, regular movements, such as walking, typically manifest in approximately periodic variations in signal strength. We present an algorithm that predicts the signal strength peaks and evaluate it on real-world data. We present the design of an opportunistic MAC protocol, named BANMAC, that takes advantage of the periodic fluctuations of the signal strength to achieve high reliability even with low transmission power.

Passive discovery of IEEE 802.15.4-based body sensor networks

In this paper we study passive discovery of IEEE 802.15.4 networks operating in the beacon-enabled mode. The task of discovery occurs in different scenarios. One example is a simple device that wishes to associate with a specific, pre-specified PAN coordinator (targeted discovery). Another example are opportunistic relaying applications, where arbitrary foreign coordinators should be discovered (untargeted discovery). We consider a simple class of listening strategies and provide different analytical models which allow to find optimal strategies for different listening scenarios. For the case of targeted discovery without constraints on the listening costs we give a dynamic programming formulation, for targeted discovery with bounded costs we present and validate a simple model and derive the desired performance measures. For untargeted discovery we present simulation results in a mobile scenario.

Flexible hardware abstraction of the TI MSP430 microcontroller in TinyOS

Wireless sensor networks (WSNs) promote energy-efficiency as the main design criterion. This introduces rather conflicting requirements for the hardware adaptation layer. Maximizing the efficiency requires that the presentation must closely mimic the underlying hardware model. On the other hand, increasing the level of abstraction simplifies the development, but at the cost of lowered efficiency because it obstructs the link between the application and the hardware. As new microcontrollers and radios are introduced for use in WSNs, applications must be able to effectively use new low power features and peripherals. The MSP430 family of microcontrollers by Texas Instruments is specifically designed for ultra-low-power applications. It incorporates a 16-Bit RISC CPU, peripherals and a clock system. The MSP430F149 is one of the most popular members of the family. As shown in Fig. 1, it has 60 KB Flash, 2 KB of RAM and a flexible clock system sourced by an internal digitally controlled oscillator (DCO) and/or two external oscillators. It also contains a 12-Bit A/D Converter, two independent timers and two USARTs.

Link quality ranking: Getting the best out of unreliable links

Link quality estimation has been an active area of research within the wireless sensor network community. It is now well known that the estimation of reliable links requires few sample packets — less than 10, while the estimation of unreliable links require many more — above 50. In scenarios where unreliable links are ubiquitous, and a rapid transfer of data is needed, traditional estimation techniques are not a viable option. In such scenarios, it is instead sufficient to identify the best link available at any given time. Within this context, we propose Link Quality Ranking (LQR), a mechanism that identifies the best link available when only unreliable links are present. Our testbed results indicate that with one sample packet, the delivery rate of LQR — with respect to the best link available — is above 93%. With 10 sample packets, the performance is above 96%.

IEEE 802.15.4 and ZigBee as Enabling Technologies for Low-Power Wireless Systems with Quality-of-Service Constraints

This book outlines the most important characteristics of IEEE 802.15.4 and ZigBee and how they can be used to engineer Wireless Sensor Network (WSN) systems and applications, with a particular focus on Quality-of-Service (QoS) aspects. It starts by providing a snapshot of the most relevant features of these two protocols, identifying some gaps in the standard specifications. Then it describes several state-of-the-art open-source implementations, models and tools that have been designed by the authors and have been widely used by the international community. The book also outlines the fundamental performance limits of IEEE 802.15.4/ZigBee networks, based on well-sustained analytical, simulation and experimental models, including how to dimension such networks to optimize delay/energy trade-offs.

The ANGEL IEEE 802.15.4 enhancement layer: Coupling priority queueing and service differentiation

The EU FP6 ANGEL project considers the usage of IEEE 802.15.4-based wireless sensor network technology in medical applications. Some key requirements found in this class of applications are not well supported by IEEE 802.15.4. In this paper we propose a wrapper layer on top of IEEE 802.15.4 that adds, amongst others, a mechanism combining priority queueing and per-packet parameter control to provide (stochastic) service differentiation. In this paper we describe the design and implementation of this mechanism and present measurement results, showing its effectiveness.