Wireless sensor networks (WSNs) have recently been under the focus of the research community. The applications for this technology, as well as the problems it poses, are many. In this work, we consider using such technology to implement a detection system. Detection systems are based on statistical signal processing and provide the capability of deciding whether something has or has not happened. They found application in surveillance, monitoring and control systems, among others. This work is divided into two parts. First, we carry out the design, development and implementation of a WSN testbed where algorithms of all sorts can be tested in a real environment. Second, we concentrate on the design and development of a detection system that we implement on the testbed. We put into practice various approaches and architectures using a design approach outlined here. We analyze the algorithms in terms of detection performance. We address the issue of packet losses through the inclusion of heuristics. We analyze, from a theoretical point of view, the communication cost of the various schemes considered. In addition, we discuss other trade-offs that stem from the use of WSN. Our results show closeness of the designed and measured detection performance for most of the cases, which leads us to the conclusion that we have a valid design procedure. The heuristics to counteract packet losses provide the algorithms with robustness against that issue. Variations on the detection physical scenario are also well stood. We conclude that using WSNs for implementing detection systems brings advantages but also many questions. It is relatively easy to conceive a detection system, using a WSN, that is robust, has low deployment and running costs and achieves good performance. Nevertheless, considerations on features such as energy, communication bandwidth and more have to be done so that we can obtain an optimized system.

Taking resource limitations into account in the design of wireless sensor networks are important in many emerging applications. The need for minimizing the communication and power consumption of individual nodes and other units pose interesting challenges for estimation and control strategies.This document describes the design, implementation, obtained results and conclusions of a cooperative localization and tracking system based on two types of sensors. Practical investigations are made to reach the optimal sensor scheduling based on offline and covariance-based scheduler approaches. One sensor gives low quality measurements and is based on an ultrasound sensor and the other is a web-camera with high precision but with delayed results. The ultrasound sensor is connected to wireless sensor nodes which are part of the KTH Wireless Sensor Network Testbed and the web-camera is connected to a data processing unit and placed in the same area. An overview about localization techniques and solutions are presented. The design, development, and implementation of the KTH Wireless Sensor Network Testbed is also discussed. The software implemented on the system is fully detailed as well as the necessary hardware. A presentation of the filtering methods used to perform localization and tracking is put forward. Analysis and conclusions of all the different approaches used are discussed. Guidelines for future work are also proposed.

Today the armed forces of many countries need to strike accurately potential enemies, wherever they might be, from a safe place. Since naval units can be deployed almost everywhere in the open sea, the idea of a naval cruise missile emerged in the 70’s. These missiles are designed to be launched from various naval vehicles such as frigates or submarines and strike deeply in the enemy territory. A program called Missile de Croisière Naval (MdCN Naval Cruise Missile) was therefore launched in 2006 by the DGA, the French procurement agency. MBDA is the industrial company appointed by the DGA to design and build the missile. Control aspects on a cruise missile are of primary interest since they impact the reliability, performance and availability of the weapon. In the aeronautics and weapon industry, gain scheduled controllers are used in most cases. However, many non-linear techniques have  been developed in the literature and might improve the behaviour of the missile. The main objective of the present thesis is to apply non-linear techniques on the control and guidance loops of the MdCN too see whether of not they can improve such a system. Based on this report it should be easy for the engineers of the DGA to compare the controllers of the thesis and the classical gain scheduled controllers used in the industry. To achieve this task some basic knowledge of flight dynamics are recalled and a model of the MdCN is computed and divided into the control loop and the guidance loop. Then a non-linear controller for the launch phase using a Lyapunov based technique called back-stepping is designed and tested through a statistic analysis. During the cruise phase different anti-windup strategies are applied on the propulsion control loop of the missile and compared. Finally a software interface with a navigation-dedicated tool is coded and implemented in Simulink to analyse the complete Guidance-Navigation-Control loop and to see how navigation errors impact the control algorithms. The main contributions of this thesis are the controllers designed for the launch phase and the propulsion loop that will be compared with the controller that MBDA is going to deliver next year to see whether or not the non-linear techniques used in the thesis should be used on the missile. Furthermore, all the tools and procedure set up to interface the control and guidance laws with the navigation models and filters will give the possibility to the DGA to have a deeper understanding of the algorithms used by MBDA and to make sure that navigation and estimation issues are properly taken in account when designing the control and guidance laws.

Ad-hoc wireless networks use multi-hop transmissions to communicate, without exploiting any infrastructure. Logical links create unreliable connections between nodes: the capacity of the channel can unpredictably change due to the presence of obstructions, interferences between nodes and stochastic phenomena such as fading. Moreover the medium is multi-access and resources are contended by different users. It is possible to notice that adjusting the power at the physical layer affects the interference perceived, which in turn modifies the resources availability, alters the queue length at the network layer and eventually influences the source rates at the transport layer. To this end, a lot of works have shown that a better understanding of inherent coupling between different layers in the networking stack is worth. Specifically, network performances can be increased if the traditionally separated network layers are jointly optimized. Network utility maximization has emerged as a powerful framework for studying such cross-layer issues and optimizing performances overall the network. In particular, we focus on distributed cross-layer algorithms that achieve a global optimum recurring to local information only. Although the literature is vast in this field, most of works remain as theory. We aim at clarifying the practical feasibility of such theoretical dissertations and what considerations are needed in order to establish a bridge between theory and practice. After analyzing different crosslayer methods, we focus on the work of Papandriopoulos et al.. We first discuss its theoretical benefits in an interference limited system such as CDMA, without accounting physical constraints of the network. In order to validate the performances in a more realistic scenario, we implement the algorithm of Papandriopoulos et al. in the NS-2 network simulator without breaking up the hierarchy of the standard ISO/OSI stack. We focus on modeling physical, data link, network and transport layer, underlining issues and possible solutions from a practical perspective. For instance, we propose a novel approach to calculate the congestion prices of the network. After discussing benefits and drawbacks of the underlying theory, we propose several results under many simulation scenarios and eventually a comparison with the standard protocol for wireless networks IEEE/802.11.

This thesis is part of the SOCRADES project, a European research and advanced development project with the primary objective to develop a design, execution and management platform for next-generation industrial automation systems. In the speci¯c case of work it concerns the networked control systems. These are becoming more important in the industrial automation ¯eld thanks to the many advantages introduced by the networks. In fact, the use of a network to connect the devices permits to eliminate unnecessary wirings, reducing the complexity and the overall cost in designing and implementing the control systems. In the last years the fast spread of the wireless technologies has opened new scenarios for the communication in the automation ¯eld. The bene¯ts introduced by the use of wireless communication in the networked control system are many. First of all the simplicity and the convenience of the sensors placement. The price to be paid is a lower reliability due to the interference that can easily a®ect the medium (radio frequencies) with the consequent possible loss of communication. This work is focused on the study of the problem of losing packets (the information in a wireless network is formed by packets of bits) in a WirelessHART networked control system and on the possible solutions to avoid the problem. WirelessHART is a wireless protocol that provides a low cost, relatively low speed (e.g., compared to IEEE 802.11g) wireless connection. The aims of the thesis are multiple, ¯rst of all the implementation of a tool that permits to simulate a wirelessHART network. In fact, since it is a quite new protocol, the most used network simulators do not give the possibility to simulate WirelessHART networks. This simulator has been implemented modifying the original version of the TrueTime network simulator, adding this new protocol and some other functions to make the simulator as close as possible to the reality. Another objective of the thesis is to study the e®ects, and the possible solutions, of the loss of communication in a wirelessHART network. The last part of the thesis deals with a level control problem in a mineral flotation plant and with the possibility to use a wirelessHART network for that plant.

No description available.

Designing efficient and reliable communication protocols for wireless sensor networks in indoor monitoring applications is a challenging task, due to the uncertainty and dynamics of the environment. We consider SERAN, a two-layer semi-random protocol that specifies a routing algorithm and a MAC layer for clustered wireless sensor networks. It combines a randomized and a deterministic approach: the former provides robustness over unreliable channels, the latter reduces the packet collisions. We provide a mathematical model for the protocol that allows us to analyze its behavior and optimize performance. We define an optimization problem, considering the energy consumption as objective function and constraints in terms of error rate and end-to-end delay. A TinyOS implementation of the protocol on a WSN test bed composed by Moteiv’s Tmote Sky wireless sensors is presented. Experimental results validate the model and show excellent performance for low data rate transmissions, with low average node duty cycle, which yields a long network lifetime.

We consider a peer-to-peer approach to wireless sensor networks using the IEEE 802.15.4 standard, with sensors not synchronized, and without any routing protocol. Only communications between neighbors are allowed. In this scenario we do a comparison between four distributed algorithms that solve a special class of optimization problems, which are of great interest in networking applications. We want to retrieve, without a central node, the average of a scalar measurement from all sensors in the network. In the nal state, each sensor should have the global average of the considered measurement. To evaluate performances of the algorithms, we build an application for the network simulator ns2, and we do several simulations to evaluate convergence delay, and nal error, respect to time and to number of packets sent. In this thesis we present the algorithms from theoretical and practical point of views, we describe our application for ns2, and we show the results obtained, which show that this types of algorithms, if tuned properly, work well and are ready to be practically implemented in a real sensor network.

Within the Third Generation Partnership Project (3GPP), the new project entitled Long Term Evolution (LTE) is going to be a new standard. LTE is characterized by a broadband optimized radio access network, which focuses on supporting a high throughput with low latency by an ip based transport network. During the handover, which is the procedure to disconnect a mobile user from a base station and connect it to another, LTE supports the inter-base station packet forwarding to achieve a seamless transition. However, given the mobility of the end users and the high bandwidth required, the handover may cause sudden degradation of the throughput of the tcp connection if the process is not correctly controlled. Moreover, during the handover, congestions in the transport network could lead to a poor utilization of the transport and radio resources available, and degrade significantly the user’s throughput. The aim of this thesis project is to study the impact on the user throughput and system performance when users with high bit rates tcp services are moving through cells of the network, and to propose enhancing solutions to counteract to possible throughput degradations. First, we have implemented an accurate LTE simulator in the ns-2 . The simulator implements the tunneling mechanism of LTE for the end users ip packets. It allows us to switch the ip tunnel between the gateway and radio base station associated with the moving mobile in the gateway towards the target base station when the terminal arrives there. The simulator has then been used to investigate the performance of the handover event during a tcp connection with focus on the data forwarding and the router buffer dimension, and to propose two solutions to achieve a better end user throughput. The first solution is based on a prediction technique that avoids data forwarding, whereas the second solution acts in the transport network with an active queue management. The simulation results show that the handover prediction could increase the tcp performance and, second, that the router buffer dimension does matter for the end user throughput in case of a congested transport network. The project has been performed at the Automatic Control Lab at KTH in Stockholm and in collaboration with Ericsson Research Lab. The work is part of a joint effort done with Matteo Pacifico. Further results are available in the master thesis “Modeling and Performance Improvement of tcp over LTE Handover” [1].

The Long Term Evolution (LTE) of the UMTS Terrestrial Radio Access and Radio Access Network is a new communication standard aimed for commercial deployment in 2010. Goals for LTE include support for improved system capacity and coverage, high peak data rates, low latency, reduced operating costs, multiantenna support, flexible bandwidth operations and seamless integration with existing systems. The aim of this thesis project is to study the impacts on the end-user and system performance when users with high bit rates tcp services are moving through the network. These impacts affect the reduced end-user or system throughput, e.g., due to congestion in the transport network, leading to poor utilization of the transport and radio resources available. To reach such an aim, it has been necessary (1) to create a new simulator with the ns-2 and perform simulation in different network settings and (2) formulate a mathematical model able to capture the principal dynamics of the real system. Possible solutions to mitigate the impacts are investigated by comparing the simulations results of tcp performance in the radio and transport network, with the mathematical model of it. The project has been performed at the Automatic Control Lab at KTH in Stockholm and in collaboration with Ericsson Research Lab. The work is part of a joint effort with Davide Pacifico. Further results are available in the master thesis “ Analysis and Performance Improvement of tcp during handover of LTE”[1].