We consider a canonical task in wireless sensor networks -- the extraction of information about environmental features -- and propose a multi-step solution that is fault-tolerant, self-organizing and energy-efficient. We explicitly take into account the possibility of sensor measurement faults and study a distributed algorithm for detecting and correcting such faults, showing through theoretical analysis and simulation results that 85-95% of faults can be corrected using this algorithm even when as many as 10% of the nodes are faulty. We present a self-organizing algorithm which combines shortest-path routing mechanisms with leader-election to permit nodes within each feature region to self-organize into routing clusters. These clusters are used in data aggregation schemes that we propose for feature extraction. We show that the best such aggregation scheme can result in an order-of-magnitude improvement in energy savings.

Key Words- immune nets; distributed evaluation; sensor network; process diagnosis; sensor faults; self-organization; probabilistic reasoning, distributed AI We deal with the problem of evaluating the system reliability where the probabilistic reasoning cannot apply due to the existence of cyclic structure of probabilistic dependence. The evaluation can be done by multiagents whose reliabilities are evaluated by other agents. The dynamics of the reliabilities will be expressed by a continuous dynamical equation, hence allowing the method applicable to the dynamical environment such as evaluating reliabilities of sensors of process plant, based on the dynamically changing sensor values. A sensor network for applying the model to the sensor self-diagnosis of process plants, is formalized. The sensor network introduces mutual testing among sensor values by the constraint between sensor values, thus allows locating faulty sensor. The simulation for the firing section of a cement plant is also presented as an example of industrial application. It is demonstrated that the model can diagnose multiple fault and that it can reflect ambiguous situation in the diagnosis. Several modifications on the model, which can be applied to sensor fault diagnosis of process plants,also is proposed. 1

Any sensory system can be viewed as a passive or dumb element which provides raw data. It can also be viewed as an intelligent element which returns "analyzed" information. Finally, it can be viewed as a commanding element which sends commands to the physical system. Each of these views is used in different situations and for different tasks. Commanding sensors are an extension to the logical sensor approach in which a mapping from events to actions is added to the sensor model. In a previous paper, we proposed a sensor-based distributed control scheme for mobile robots along with several simulation results [1]. In this paper, the application of this scheme to control a real mobile robot is presented and the results of several experiments are discussed. A server-client model is used to implement this scheme where the server is a process that carries out the commands to be executed, and each client is a process with a certain task. The logical sensor approach is used to model the sensor...

Advances in sensor technology and wireless communications have made networked microsensors possible, where each sensor individually senses the environment but collaboratively achieves complex information gathering and dissemination tasks. These networked sensors, however, possess several characteristics that have challenged many aspects of traditional computer network design, such as the scalability issue caused by the sheer amount of sensor nodes, the infrastructureless network, and the stringent resource onboard the sensors. These new features call for a re-design of overall structure of applications and services. It has been widely accepted that practical localized algorithms is probably the best solution to wireless sensor networks. In this article, we discuss recent research results on localized algorithms design in supporting services and applications in sensor networks.

We describe the use of the mobile agent paradigm to design an improved infrastructure for data integration in Distributed Sensor Network (DSN). We use the acronym MADSN to denote the proposed Mobile-Agent-based DSN. Instead of moving data to processing elements for data integration, as is typical of a client/server paradigm, MADSN moves the processing code to the data locations. This saves network bandwidth and provides an effective means for overcoming network latency, since large data transfers are avoided. We study two important problems related to MADSN design --- the distributed integration problem, and the optimum performance problem. Compared to DSNs, a mobile-agent implementation of multi-resolution data integration saves up to 90% of the data transfer time. For a given set of network parameters, we analyze the conditions under which MADSN performs better than DSN and determine the condition under which MADSN reaches its optimum performance level.

Sensor systems are becoming ubiquitous throughout society, yet their design, construction and operation are still more of an art than a science. In this paper, we define, develop, and apply a formal semantics for sensor systems that provides a theoretical framework for an integrated software architecture for modeling sensor-based control systems. Our goal is to develop a design framework which allows the user to model, analyze and experiment with different versions of a sensor system. This includes the ability to build and modify multisensor systems and to monitor and debug both the output of the system and the affect of any modification in terms of robustness, efficiency, and error measures. The notion of Instrumented Logical Sensor Systems (ILSS) that are derived from this modeling and design methodology is introduced. The instrumented sensor approach is based on a sensori-computational model which defines the components of the sensor system in terms of their functionality, accuracy, robustness and efficiency. This approach provides a uniform specification language to define sensor systems as a composition of smaller, predefined components. From a software engineering standpoint, this addresses the issues of modularity, reusability, and reliability for building complex systems. An example is given which compares vision and sonar techniques for the recovery of wall pose.

Recent advances in wireless communications along with developments in low-power circuit design and micro-electro mechanical systems (MEMS) have heralded the advent of compact and inexpensive wireless micro-sensor devices. A large network of such sensor nodes capable of communicating with each other provides significant new capabilities for automatically collecting and analyzing data from physical environments.

The self-organizing diagnosis has been studied by applying the idea of autonomous and decentralized systems extracted from the concept of immune network. The model implements network-level recognition by connecting information from local recognition units by dynamical evaluation chain. The model has been further elaborated for engineering concerns of identifying not only sensor faults but process faults. The sensor faults will be identified by evaluating reliability of data from sensor, while the process faults will be identified by evaluating that of constraints that must be satisfied among these data. We have demonstrated that the extended sensor network will work against both sensor faults and process faults by an illustrative example. 1

The use of inexpensive devices in large-scale wireless sensor networks is likely to result in higher rates of temporary or lasting failures for individual nodes. It is important for real-time information routing algorithms in this space to provide tolerance to such failures in an energy-efficient manner. The conventional

In previous work, we introduced the notion of Instrumented Logical Sensor Systems (ILSS) that are derived from a modeling and design methodology [2, 4]. The instrumented sensor approach is based on a sensori-computational model which defines the components of the sensor system in terms of their functionality, accuracy, robustness and efficiency. This approach provides a uniform specification language to define sensor systems as a composition of smaller, predefined components. From a software engineering standpoint, this addresses the issues of modularity, reusability, and reliability for building complex multisensor systems. In this report