We present a high performance logging system for embedded UNIX and GNU/Linux applications. Compared to the standard UNIX and GNU/Linux logging method, syslog, our method has two orders of magnitude lower latency and an order of magnitude higher message throughput. This speed-up is mainly due to the use of a memory-mapped file as the means of inter-process communication, fewer memory copies and the batching of output messages in the logging daemon. In addition, our logging system also accepts syslog messages, providing compatibility with existing applications. Our logging system is in production use in the Cisco UCS Virtual Interface Card.

In this paper we consider a sensor network deployed far away from the base station. Each sensor in the network is equipped with two radio interfaces: the low-power IEEE 802.15.4 radio and the high-bandwidth 3G radio. The low-power radios are used on all sensors to transmit data within the network, while the high-bandwidth radios are activated only on a subset of sensors, referred to as gateways, for sending data to the base station. We assume that not all sensors are required to transmit their sensed data to the base station, yet the base station does have a network throughput requirement. A high throughput requirement would cause more energy consumption on sensors and shorten the network lifetime. We investigate the problem of maximizing the network lifetime subject to the network throughput being guaranteed and the data delivery latency from the network to the base station being bounded. We first formulate a novel optimization problem, namely, the throughput guaranteed network lifetime maximization problem. We then devise a heuristic for it, with the key ingredients of gateway assignment technique and energy-efficient routing forest establishment strategy. We finally conduct extensive experiments through simulation to evaluate the performance of the proposed heuristic and show that it outperforms another two algorithms in terms of network lifetime.

Micro-solar power system design is challenging because it must address long-term system behavior under highly variable solar energy conditions and consider a large space of design options. Several micro-solar power systems and models have been made, validating particular points in the whole design space. We provide a general architecture of micro-solar power systems---comprising key components and interconnections among the components---and formalize each component in an analytical or empirical model of its behavior. To model the variability of solar energy, we provide three solar radiation models, depending on the degree of information available: an astronomical model for ideal conditions, an obstructed astronomical model for estimating solar radiation under the presence of shadows and obstructions, and a weather-effect model for estimating solar radiation under weather variation. Our solar radiation models are validated with a concrete design, the HydroWatch node, thus achieving small deviation from the long-term measurement. They can be used in combination with other micro-solar system models to improve the utility of the load and estimate the behavior of micro-solar power systems more accurately. Thus, our solar radiation models provide more accurate estimations of solar radiation and close the loop for micro-solar power system modeling.

Building a micro-solar power system is challenging because it must address long-term system behavior under highly variable solar energy and consider a large design space. We develop a practical theory of micro-solar power systems that is materialized in a simulation suite that models component and system behavior over a long time scale and in an external environment that depends on time, location, weather, and local variations. This simulation provides sufficient accuracy to guide specific design choices in a large design space. Unlike the many macro-solar calculators, this design tool models detailed behavior of milliwatt systems in the worst conditions, rather than typical behavior of kilowatt systems in the best conditions. Our simulation suite is validated with a concrete design of micro-solar power systems, the HydroWatch node. With our simulation suite, micro-solar power systems can be designed in a systematic fashion. Putting the model and empirical vehicle together, the design choices in each component of a micro-solar power system are studied to reach a deployable candidate. The deployment is evaluated by analyzing the effects of different solar profiles across the network. The analysis from the deployment can be used to refine the next system-design iteration.

This paper describes a systematic approach to building micro-solar power subsystems for wireless sensor network nodes. Our approach composes models of the basic pieces - solar panels, regulators, energy storage elements, and application loads - to appropriately select and size the components. We demonstrate our approach in the context of a microclimate monitoring project through the design of the node, micro-solar subsystem, and network, which is deployed in a challenging, deep forest setting. We evaluate our deployment by analyzing the effects of the range of solar profiles experienced across the network.

We present a building block approach to hardware platform design based on a decade of collective experience in this area, arriving at an architecture in which general-purpose modules that require expertise to de sign and incorporate commonly-used functionality are integrated with application-specific carriers that satisfy the unique sensing, power supply, and mechanical constraints of an application. Of course, modules are widespread, but our focus is far less on the performance of any individual module and far more on an overall architecture that supports the prototype, pilot, and production stages of design, and preserves the artifacts and learnings accumulated along the way.

We present the philosophy, design, and initial evaluation of the Trio Testbed, a new outdoor sensor network deployment that consists of 557 solar-powered motes, seven gateway nodes, and a root server. The testbed covers an area of approximately 50,000 square meters and was in continuous operation during the last four months of 2005. This new testbed in one of the largest solar-powered outdoor sensor networks ever constructed and it offers a unique platform on which both systems and application software can be tested safely at scale. The testbed is based on Trio, a new mote platform that provides sustainable operation, enables efficient in situ interaction, and supports fail-safe programming. The motivation behind this testbed was to evaluate robust multi-target tracking algorithms at scale. However, using the testbed has stressed the system software, networking protocols, and management tools in ways that have exposed subtle but serious weaknesses that were never discovered using indoor testbeds or smaller deployments. We have been iteratively improving our support software, with the eventual aim of creating a stable hardware-software platform for sustainable, scalable, and flexible testbed deployments.

A main challenge with developing applications for wireless embedded systems is the lack of visibility and control during execution of an application. In this paper, we present a tool suite called Marionette that provides the ability to call functions and to read or write variables on pre-compiled, embedded programs at run-time, without requiring the programmer to add any special code to the application. This rich interface facilitates interactive development and debugging at minimal cost to the node.

We present an incremental network programming mechanism which re programs wireless sensors quickly by transmitting the incremental changes for the new program version. Using the Rsync algorithm we generate the difference of the two program images, which allows us to distribute just the key changes of the program. Unlike previous approaches, our design does not assume any prior knowledge of the program code structure and can be applied to any hardware platform. To meet the resource constraints of wireless sensors we tuned the Rsync algorithm which was originally made for updating binary files among computationally powerful machines. In our design, the sensor node processes the delivery and the decoding of the difference script in separate steps. This makes it easy to extend for multi-hop network programming. We are able to achieve the speedup of 9.1 for changing a constant and 2.1 to 2.5 for changing a few lines in the source code over the non-incremental delivery.