In most range-based localization methods, inferring distance from radio signal strength using mathematical modeling becomes increasingly unreliable and complicated in indoor and extreme environments, due to effects such as multipath propagation and signal interference. We propose FuzLoc, a range-based, anchor-based, fuzzy logic enabled system system for localization. Quantities like RSS and distance are transformed into linguistic variables such as Low, Medium, High etc. by binning. The location of the node is then solved for using a nonlinear system in the fuzzy domain itself, which outputs the location of the node as a pair of fuzzy numbers. An included destination prediction system activates when only one anchor is heard; it localizes the node to an area. It accomplishes this using the theoretical construct of virtual anchors, which are calculated when a single anchor is in the node’s vicinity. The fuzzy logic system is trained during deployment itself so that it learns to associate an RSS with a distance, and a set of distances to a probability vector. We implement the method in a simulator and compare it against other methods like MCL, Centroid and Amorphous. Extensive evaluation is done based on a variety of metrics like anchor density, node density etc.

Wireless sensor networks (WSNs) have received huge attention during the recent years due to their applications in a large number of areas such as environmental monitoring, health and traffic monitoring, surveillance and tracking, and monitoring and control of factories and home appliances. Also, the rapid developments in the micro electro-mechanical systems (MEMS) technology and circuit design lead to a faster spread and adoption of WSNs. Wireless sensor networks consist of a number of nodes featured in general with energy-limited sensors capable of collecting, processing and transmitting information across short distances. Clock synchronization plays an important role in designing, implementing, and operating wireless sensor networks, and it is essential in ensuring a meaningful information processing order for the data collected by the nodes. Because the timing message exchanges between different nodes are affected by unknown possibly time-varying network delay distributions, the estimation of clock offset parameters represents a challenge. This dissertation presents several robust estimation approaches of the clock offset parameters necessary for time synchronization of WSNs via the two-way message exchange mechanism. In this dissertation the main emphasis will be put on building clock phase offset estimators robust with respect to the unknown network delay distributions. Under the assumption that the delay characteristics of the uplink and the downlink are asymmetric, the clock offset estimation method using the bootstrap bias correction approach is derived. Also, the clock offset estimator using the robust Mestimation technique is presented assuming that one underlying delay distribution is mixed with another delay distribution. Next, although computationally complex, several novel, efficient, and robust estimators of clock offset based on the particle filtering technique are proposed to cope with the Gaussian or non-Gaussian delay characteristics of the underlying networks. One is the Gaussian mixture Kalman particle filter (GMKPF) method. Another is the composite particle filter (CPF) approach viewed as a composition between the Gaussian sum particle filter and the KF. Additionally, the CPF using bootstrap sampling is also presented. Finally, the iterative Gaussian mixture Kalman particle filter (IGMKPF) scheme, combining the GMKPF with a procedure for noise density estimation via an iterative mechanism, is proposed.

Network coding and duty-cycling are two popular techniques for saving energy in wireless sensor networks. To the best of our knowledge, the idea to combine these two techniques, for even more aggressive energy savings, has not been explored. One explanation is that these two techniques achieve energy efficiency through conflicting means, e.g., network coding saves energy by exploiting overhearing, whereas dutycycling saves energy by cutting idle listening and, thus, overhearing. In this thesis, we thoroughly evaluate the use of network coding in duty-cycled sensor networks. We propose a scheme called DutyCode, in which a MAC protocol implements packet streaming and allows the application to decide when a node can sleep. Additionally, a novel, efficient coding scheme decision algorithm, ECSDT, assists DutyCode to reduce further energy consumption by minimizing redundant packet transmissions, while an adaptive mode switching algorithm allows smooth and timely transition between DutyCode and the default MAC protocol, without any packet loss. We investigate our solution analytically, implement it on mote hardware, and evaluate it in a 42-node indoor testbed. Performance evaluation results show that our scheme saves 30-46% more energy than solutions that use network coding, without using duty-cycling.

A local reconfiguration algorithm (INP) for reliable routing in wireless sensor networks that consist of many static (fixed) energy-constrained nodes is introduced in the dissertation. For routing around crash fault nodes, a communication tree structure connecting sensor nodes to the base station (sink or root) is dynamically reconfigured during information dissemination. Unlike other location based routing approaches, INP does not take any support from a high costing system that gives position information such as GPS. For reconfigurations, INP uses only local relational information in the tree structure among nearby nodes by collaboration between the nodes that does not need global maintenance, so that INP is energy efficient and it scales to large sensor networks. The performance of the algorithm is compared to the single path with repair routing scheme (SWR) that uses a global metric and the modified GRAdient broadcast scheme (GRAB-F) that uses interleaving multiple paths by computation and by simulations. The comparisons demonstrate that using local relative information is mostly enough for reconfigurations, and it consumes less energy and mostly better delivery rates than other algorithms especially in dense environments. For the control observer to know the network health status, two new diagnosis algorithms (Repre and Local) that deal with crash faults for wireless sensor networks are also introduced in the dissertation. The control observer knows not only the static faults found by periodic testing but also the dynamic faults found by a path reconfiguration algorithm like INP that is invoked from evidence during information dissemination. With based on this information, the control observer properly treats the network without lateness. Local algorithm is introduced for providing scalability to reduce communication energy consumption when the network size grows. The performance of these algorithms is computationally compared with other crash faults identification algorithm (WSNDiag). The comparisons demonstrate that maintaining the communication tree with local reconfigurations in Repre and Local needs less energy than making a tree per each diagnosis procedure in WSNDiag. They also demonstrate that providing scalability in Local needs less energy than other approaches.

Much research has been done on wireless sensor networks. However, most protocols and algorithms for such networks are based on the ideal model Unit Disk Graph (UDG) model or do not assume any model. Furthermore, many results assume the knowledge of location information of the network. In practice, sensor networks often deviate from the UDG model significantly. It is not uncommon to observe stable long links that are more than five times longer than unstable short links in real wireless networks. A more general network model, the quasi unit-disk graph (quasi-UDG) model, captures much better the characteristics of wireless networks. However, the understanding of the properties of general quasi-UDGs has been very limited, which is impeding the design of key network protocols and algorithms. In this dissertation we study the properties for general wireless sensor networks and develop new topological/geometrical techniques for wireless sensor networking. We assume neither the ideal UDG model nor the location information of the nodes. Instead we work on the more general quasi-UDG model and focus on figuring out the relationship between the geometrical properties and the topological properties of wireless sensor networks. Based on such relationships we develop algorithms that can compute useful substructures (planar subnetworks, boundaries, etc.). We also present direct applications of the properties and substructures we constructed including routing, data storage, topology discovery, etc. We prove that wireless networks based on quasi-UDG model exhibit nice properties like separabilities, existences of constant stretch backbones, etc. We develop efficient algorithms that can obtain relatively dense planar subnetworks for wireless sensor networks. We also present efficient routing protocols and balanced data storage scheme that supports ranged queries. We present algorithmic results that can also be applied to other fields (e.g., information management). Based on divide and conquer and improved color coding technique, we develop algorithms for path, matching and packing problem that significantly improve previous best algorithms. We prove that it is unlikely for certain problems in operation science and information management to have any relatively effective algorithm or approximation algorithm for them.

In this thesis, we consider decentralized multiclass classification problem in wireless sensor networks. In literature, the decentralized detection using error correcting code has been shown to have good fault-tolerance capability. In this thesis, we provide fault-tolerance capability by employing the code with a particular structure so that the decoding at the fusion center can be efficient. Specifically, the convolution code is employed to decode the local decision vector sent from all the local sensors. In addition, we proposed an efficient convolution code design algorithm by using simulated annealing. The simulation result shows that the proposed approach has good performance.

This thesis considers the problem of distributed sequential detection in wireless sensor networks (WSNs), where the number of operating sensors is unknown to the fusion center. Since the energy and bandwidth of communication channel are limited in WSNs, we employ the censoring scheme in the sequential detection to achieve energy-efficiency and low communication rate. Specifically, we show by simulations that employing censoring scheme can reduce the number of local decisions that required for the fusion center to make a final decision. The results implies that the energy conservation does not necessary degrade the performance of sequential detection in terms of the expected local decisions required for making a final decisions.

In this thesis, we consider the distributed classification problem in wireless sensor networks (WSNs). Sensor nodes in WSNs detect environmental variations and make their decisions individually, after which their decisions, possibly in the presence of faults, are transmitted to a fusion center. In literature, the distributed classification fusion using error correcting codes has been shown to have good sensor fault-tolerance capability. In this thesis, we extend the fault-tolerant classification system using error correcting code by using two-dimensional channel coding. We also extend the binary coding in literature to the M-ary code. This thesis then suggests a code construction method with low computational complexity. Based on the suggest code construction method, this thesis then conducts a series experiment to investigate the performance of the suggested method.

Thesis (M. Phil.)--University of Hong Kong, 2010. / Includes bibliographical references (leaves 93-103). Also available in print.

Thesis (Ph. D.)--Rutgers University, 2009. / "Graduate Program in Computer Science." Includes bibliographical references (p. 103-106).