J. Canada-Bago

Knowledge-based duty cycle estimation in wireless sensor networks: Application for sound pressure monitoring

Wireless sensor networks comprise an important research area and a near future for industry and communications. Wireless sensor networks contain resource-constrained sensor nodes that are powered by small batteries, limited process and memory and wireless communication. These features give sensors their versatility and drawbacks, such as their limited operating lifetimes. To feasibly deploy wireless sensor networks with isolated motes, several approaches and solutions have been developed; the most common, apart from using alternative power sources such as solar panels, are those that put sensors to sleep for time periods established by the application. We thus propose a fuzzy rule-based system that estimates the next duty cycle, taking the magnitude being tested and battery charge as input. To show how it works, we compare an analytical delta system to our contribution. As an application to test both systems, a sound pressure monitoring application is presented. The results have shown that the fuzzy rule-based system better predicts the evolution of the magnitude by which errors committed by idle periods decrease. This work also shows that application-oriented duty cycle control can be an alternative for measuring systems, thus saving battery and improving sensor node lifetime, with a reasonable loss of precision.

Optimisation of control parameters for genetic algorithms to test computer networks under realistic traffic loads

Although many studies have focused on testing computer networks under realistic traffic loads by means of genetic algorithms (GAs), little attention has been paid to optimising the parameters of the GAs in this problem. The objective of this work is to design and validate a system that, given some constraints on traffic bandwidth, generates the worst-case traffic for a given computer network and finds the traffic configuration (critical background traffic) that minimises throughput in that computer network. The proposed system is based on a meta-GA, which is combined with an adaptation strategy that finds the optimum values for the GA control parameters and adjusts them to improve the GAs performance. To validate the approach, different comparisons are performed with the goal of assessing the acceptable optimisation power of the proposed system. Moreover, a statistical analysis was conducted to ascertain whether differences between the proposed system and other algorithms are significant. The results demonstrate the effectiveness of the system and prove that, when the background traffic is driven by a GA that uses the parameters obtained from the system, the computer networks performance is much lower than when the traffic is generated by Poisson statistical processes or by other algorithms. This system has identified the worst traffic pattern for the protocol under analysis.

Wireless Intelligent Sensors Management Application Protocol-WISMAP.

Although many recent studies have focused on the development of new applications for wireless sensor networks, less attention has been paid to knowledge-based sensor nodes. The objective of this work is the development in a real network of a new distributed system in which every sensor node can execute a set of applications, such as fuzzy ruled-base systems, measures, and actions. The sensor software is based on a multi-agent structure that is composed of three components: management, application control, and communication agents; a service interface, which provides applications the abstraction of sensor hardware and other components; and an application layer protocol. The results show the effectiveness of the communication protocol and that the proposed system is suitable for a wide range of applications. As real world applications, this work presents an example of a fuzzy rule-based system and a noise pollution monitoring application that obtains a fuzzy noise indicator.

A New Collaborative Knowledge-Based Approach for Wireless Sensor Networks

This work presents a new approach for collaboration among sensors in Wireless Sensor Networks. These networks are composed of a large number of sensor nodes with constrained resources: limited computational capability, memory, power sources, etc. Nowadays, there is a growing interest in the integration of Soft Computing technologies into Wireless Sensor Networks. However, little attention has been paid to integrating Fuzzy Rule-Based Systems into collaborative Wireless Sensor Networks. The objective of this work is to design a collaborative knowledge-based network, in which each sensor executes an adapted Fuzzy Rule-Based System, which presents significant advantages such as: experts can define interpretable knowledge with uncertainty and imprecision, collaborative knowledge can be separated from control or modeling knowledge and the collaborative approach may support neighbor sensor failures and communication errors. As a real-world application of this approach, we demonstrate a collaborative modeling system for pests, in which an alarm about the development of olive tree fly is inferred. The results show that knowledge-based sensors are suitable for a wide range of applications and that the behavior of a knowledge-based sensor may be modified by inferences and knowledge of neighbor sensors in order to obtain a more accurate and reliable output.

Improving hybrid ad hoc networks

Intelligent selection of a gateway in ad hoc networks based in performance metrics.Selection method based on a fuzzy system optimized by a genetic algorithm.The fitness function of the genetic algorithm is designed with another fuzzy system in order to accomplish it optimization task.This system over-performs other similar systems in the Internet gateway election problem. The selection of an appropriate and stable route that enables suitable load balancing of Internet gateways is an important issue in hybrid mobile ad hoc networks. The variables employed to perform routing must ensure that no harm is caused that might degrade other network performance metrics such as delay and packet loss. Moreover, the effect of such routing must remain affordable, such as low losses or extra signaling messages. This paper proposes a new method, Steady Load Balancing Gateway Election, based on a fuzzy logic system to achieve this objective. The fuzzy system infers a new routing metric named cost that considers several networks performance variables to select the best gateway. To solve the problem of defining the fuzzy sets, they are optimized by a genetic algorithm whose fitness function also employs fuzzy logic and is designed with four network performance metrics. The promising results confirm that ad hoc networks are characterized by great uncertainty, so that the use of Computational Intelligence methods such as fuzzy logic or genetic algorithms is highly recommended.

Knowledge-based wireless sensors using sound pressure level for noise pollution monitoring

Over the last few years, there is a growing interest in monitoring noise pollution in urban areas and some recent studies have proposed the deployment of Wireless Sensor Networks for this task. Although the noise indicators defined by European Union directive 2002/49/EC can be calculated by sensor nodes, the noise perception is affected by subjective factors and there is not a direct correlation between the indicators and the subjective perception of noise. In this work, we present a mathematic algorithm for calculating the sound pressure level in a sensor node and a Fuzzy Noise Indicator that allows sensor nodes to infer the degree of subjective noise annoyance. Each sensor node executes an adapted Fuzzy Rule-Based System which has two inputs: a) the A-weighting equivalent noise level value and b) its persistence in time. The results show that the use of this Fuzzy Indicator helps to distinguish between situations with noise annoyance and other situations less annoying.

An Architecture for Performance Optimization in a Collaborative Knowledge-Based Approach for Wireless Sensor Networks

Over the past few years, Intelligent Spaces (ISs) have received the attention of many Wireless Sensor Network researchers. Recently, several studies have been devoted to identify their common capacities and to set up ISs over these networks. However, little attention has been paid to integrating Fuzzy Rule-Based Systems into collaborative Wireless Sensor Networks for the purpose of implementing ISs. This work presents a distributed architecture proposal for collaborative Fuzzy Rule-Based Systems embedded in Wireless Sensor Networks, which has been designed to optimize the implementation of ISs. This architecture includes the following: (a) an optimized design for the inference engine; (b) a visual interface; (c) a module to reduce the redundancy and complexity of the knowledge bases; (d) a module to evaluate the accuracy of the new knowledge base; (e) a module to adapt the format of the rules to the structure used by the inference engine; and (f) a communications protocol. As a real-world application of this architecture and the proposed methodologies, we show an application to the problem of modeling two plagues of the olive tree: prays (olive moth, Prays oleae Bern.) and repilo (caused by the fungus Spilocaea oleagina). The results show that the architecture presented in this paper significantly decreases the consumption of resources (memory, CPU and battery) without a substantial decrease in the accuracy of the inferred values.

A fuzzy balancing load system to improve hybrid ad hoc networks

In this paper, we propose the use of an expert system in order to improve a load balancing mechanism in a hybrid multihop network. The use of Fuzzy Logic is particularly appropriate in this type of networks because of the uncertainty associated with the movement of mobile nodes. In our proposal, mobile nodes select routes to the Internet based on a knowledge-based system. The expert system uses a set of parameters to choose the best path to the Internet so that the links are not saturated, allowing a load balancing between them. Preliminary results show that the proposed algorithm improves the performance of other load balancing mechanisms in terms of network latency and packet loss.

Propagation of Agent Performance Parameters in Wireless Sensor Networks

Wireless sensor networks are composed of resource-constrained sensor nodes, powered by batteries, with limited CPU and memory, and wireless communication. In spite of the fact that sensor nodes are resource-constrained devices, several soft computing technologies have been adapted to them. In order to save battery, sensor nodes work in cycles based on awake and sleep modes. In this work we propose a method, based on a differential decision system, to calculate dynamic parameters that control the awake-sleep cycle in a multi-agent sensor structure and their propagation to other sensor nodes in a network. As an application of the proposed system, a sound pressure monitoring application is presented. Results have shown that the proposed method utilizes less work cycles than continuous measuring systems, saving battery and improving the lifetime of sensor nodes, with a reasonable lost of precision.

From a Genetic Fuzzy Rule-Based System to a Intelligent Sensor Network

Nowadays, intelligent systems, e.g. fuzzy systems, are being incorporated into sensor networks. In this way, this paper presents an intelligent sensor network which has been developed as a genetic fuzzy rule-based system. The objectives of the present work include: first, the design of the fuzzy rule-based sensor which incorporates a new inference engine specially designed for the intelligent sensor; and, second, the design of an evolutionary algorithm, which is adapted to the sensor and based on genetic algorithm, in order to evolve the knowledge of the system. The sensor network is composed of a computer and a set of sensors. Two possible implementations of the sensor are presented: the first one includes a fuzzy ruled-based sensor; the second implementation is based on a genetic fuzzy rule-based sensor. The sensor network can incorporate expert knowledge and evolve the knowledge bases. This intelligent sensor has been tested using a sensor which is based on an 8051 microcontroller, and an inference engine which has been designed for this sensor. The evolutionary algorithm has been tested using a simulated system. In conclusion, sensor networks can incorporate fuzzy rule-based system and evolutionary algorithms. The former group allows controlling a system by the knowledge base; the latter allows evolving knowledge bases in order to obtain new knowledge.