Usman Raza

Data prediction is proposed in wireless sensor networks (WSNs) to extend the system lifetime by enabling the sink to determine the data sampled, within some accuracy bounds, with only minimal communication from source nodes. Several theoretical studies clearly demonstrate the tremendous potential of this approach, able to suppress the vast majority of data reports at the source nodes. Nevertheless, the techniques employed are relatively complex,and their feasibility on resource-scarce WSN devices is often not ascertained. More generally, the literature lacks reports from real-world deployments, quantifying the overall system-wide lifetime improvements determined by the interplay of data prediction with the underlying network. These two aspects, feasibility and system-wide gains, are key in determining the practical usefulness of data prediction in real-world WSN applications.

In this paper, we describe Derivative-Based Prediction (DBP), a novel data prediction technique much simpler than those found in the literature. Evaluation with real data sets from diverse WSN deployments shows that DBP often performs better than the competition, with data suppression rates up to 99% and good prediction accuracy. However, experiments with a real WSN in a road tunnel show that, when the network stack is taken into consideration, DBP only triples lifetime---a remarkable result per se, but a far cry from the data suppression rates above. To fully achieve the energy savings enabled by data prediction, the data and network layers must be jointly optimized. In our testbed experiments, a simple tuning of the MAC and routing stack, taking into account the operation of \dbp, yields a remarkable seven-fold lifetime improvement w.r.t. the mainstream periodic reporting.

Energy autonomy and system lifetime are critical concerns in wireless sensor networks (WSNs), for which energy harvesting (EH) is emerging as a promising solution. Nevertheless, the tools supporting the design of EH-WSNs are limited to a few simulators that require developers to re-implement the application with programming languages different from WSN ones. Further, simulators notoriously provide only a rough approximation of the reality of low-power wireless communication.

In this paper we present SENSEH, a software framework that allows developers to move back and forth between the power and speed of a simulated approach and the reality and accuracy of in-field experiments. SENSEH relies on COOJA for emulating the actual, deployment-ready code, and provides two modes of operation that allow the reuse of exactly the same code in realworld WSN deployments. We describe the toolchain and software architecture of SENSEH, and demonstrate its practical use and benefits in the context of a case study where we investigate how the lifetime of a WSN used for adaptive lighting in road tunnels can be extended using harvesters based on photovoltaic panels.

One of the most challenging goals of many wireless sensor network (WSN) deployments is the reduction of energy consumption to extend system lifetime. This paper considers a novel combination of techniques that address energy savings at the hardware and application levels: wake-up receivers and node level power management, plus prediction-based data collection. Individually, each technique can achieve significant energy savings, but in combination, the results are impressive. This paper presents a case study of these techniques as applied in a road tunnel for light monitoring. Preliminary results show the potential for two orders of magnitude reduction in power consumption. This savings of 380 times allows the creation of an energetically sustainable system by considering integration with a simple, photovoltaic energy harvester.

In recent years, indoor localization has become a hot research topic with some sophisticated solutions reaching accuracy on the order of ten centimeters. While certain classes of applications can justify the corresponding costs that come with these solutions, a wealth of applications have requirements that can be met at much lower cost by accepting lower accuracy. This paper explores one specific application for monitoring patients in a nursing home, showing that sufficient accuracy can be achieved with a carefully designed deployment of low-cost wireless sensor network nodes in combination with a simple RSSI-based localization technique. Notably our solution uses a single radio sample per period, a number that is much lower than similar approaches. This greatly eases the power burden of the nodes, resulting in a significant lifetime increase. This paper evaluates a concrete deployment from summer 2012 composed of fixed anchor motes throughout one floor of a nursing home and mobile units carried by patients. We show how two localization algorithms perform and demonstrate a clear improvement by following a set of simple guidelines to tune the anchor node placement. We show both quantitatively and qualitatively that the results meet the functional and non-functional system requirements.

Model-driven data acquisition techniques aim at reducing the amount of data reported, and therefore the energy consumed, in wireless sensor networks (WSNs). At each node, a model predicts the sampled data; when the latter deviate from the current model, a new model is generated and sent to the data sink. However, experiences in real-world deployments have not been reported in the literature. Evaluation typically focuses solely on the quantity of data reports suppressed at source nodes: the interplay between data modeling and the underlying network protocols is not analyzed. In contrast, this paper investigates in practice whether i) model-driven data acquisition works in a real application; ii) the energy savings it enables in theory are still worthwhile once the network stack is taken into account. We do so in the concrete setting of a WSN-based system for adaptive lighting in road tunnels. Our novel modeling technique, Derivative-Based Prediction (DBP), suppresses up to 99% of the data reports, while meeting the error tolerance of our application. DBP is considerably simpler than competing techniques, yet performs better in our real setting. Experiments in both an indoor testbed and an operational road tunnel show also that, once the network stack is taken into consideration, DBP triples the WSN lifetime---a remarkable result per se, but a far cry from the aforementioned 99% data suppression. This suggests that, to fully exploit the energy savings enabled by data modeling techniques, a coordinated operation of the data and network layers is necessary.

Time series forecasting aims at improving energy efficiency in wireless sensor networks (WSNs) by reducing the amount of data traffic. One such technique has each node generate a model that predicts the sampled data. When the actual, sensed data deviates from the model, a new model is generated and transmitted to the sink. Reductions in application data traffic as high as two orders of magnitude can be achieved. However, our experience in applying such forecasting in a real world deployment shows that the actual lifetime improvement is significantly less due to networking overheads. The study reported here reveals that careful, coordinated network parameter tuning can leverage the reduced traffic of forecasting techniques to increase lifetime without compromising application performance.

Wireless Sensor Networks (WSNs) are distributed systems composed of battery-powered nodes that sense and collect information about the physical world. They enable applications in a wide variety of domains including but not limited to environmental monitoring, health care and disaster management. In such applications, nodes communicate the sensed information over multiple radio links until it reaches its destination referred to as the sink. As wireless communication is the most energy hungry operation, the data collection causes the biggest drain from battery. This motivates research on energy efficient mechanisms for data collection. Though there is a plethora of protocols proposed in research literature, they are not designed to collaborate with the applications. One opportunity from such collaboration is exploiting complete knowledge about application characteristics to make data collection more energy efficient. This enables underlying layers not to provision the resources more than the needs of the application and therefore save valuable battery power. The aim of this thesis is to explore a complex interplay between application characteristics and adaptive mechanisms across network stack using concrete real world deployments. It will propose a generic framework that integrates the adaptations to achieve near-optimal energy efficiency for heterogeneous applications.