Multi-modal person identity authentication is gaining more and more attention in the biometrics area. Combining different modalities increases the performance and robustness of identity authentication systems. The authentication problem is a binary classification problem. The fusion of different modalities can be therefore performed by binary classifiers. We propose to evaluate different binary classification schemes (SVM, MLP, C4.5, Fisher's linear discriminant, Bayesian classifier) on a large database (295 subjects) containing audio and video data. The identity authentication is based on two modalities: face and speech.

This paper presents a brief overview of the DDM algorithms, systems, applications, and the emerging research directions. The structure of the paper is organized as follows. We first present the related research of DDM and illustrate data distribution scenarios. Then DDM algorithms are reviewed. Subsequently, the architectural issues in DDM systems and future directions are discussed

Collaboration in sensor networks must be fault tolerant due to the harsh environmental conditions in which such networks can be deployed. This paper focuses on finding algorithms for collaborative target detection that are efficient in terms of communication cost, precision, accuracy, and number of faulty sensors tolerable in the network. Two algorithms, namely value fusion and decision fusion are identified first. When comparing their performance and communication overhead, decision fusion is found to become superior to value fusion as the ratio of faulty sensors increases.

We investigate task-decomposition and collaboration in a two-tiered sensor network for habitat monitoring. The system recognizes and localizes a specified type of birdcalls. The system has a few powerful macro nodes in the first tier, and many less-powerful micro nodes in the second tier. Each macro node combines data collected by multiple micro nodes for target classification and localization. We describe two types of lightweight preprocessing, which significantly reduce data transmission from micro nodes to macro nodes. Micro nodes classify events according to their cross-zero rates and discard irrelevant events. Data about events of interest are reduced and compressed before being transmitted to macro nodes for target localization. Preliminary experiments illustrate the effectiveness of event filtering and data reduction at micro nodes.

In wireless sensor networks, sensor nodes are spread randomly over the coverage area to collect information of interest. Data fusion is used to process these collected information before they are sent to the base station, the observer of the sensor network. We study the security of the data fusion process in this work. In particular, we propose a witness-based solution to assure the validation of the data sent from data fusion nodes to the base station. We also present the theoretical analysis for the overhead associated with the mechanism, which indicates that even in an extremely harsh environment the overhead is low for the proposed mechanism.

Signal estimation from a sequential encoding in the form of quantized noisy measurements is considered. As an example context, this problem arises in a number of remote sensing applications, where a central site estimates an information-bearing signal from low-bandwidth digitized information received from remote sensors, and may or may not broadcast feedback information to the sensors. We demonstrate that the use of an appropriately designed and often easily implemented additive control input before signal quantization at the sensor can significantly enhance overall system performance. In particular, we develop efficient estimators in conjunction with optimized random, deterministic, and feedback-based control inputs, resulting in a hierarchy of systems that trade performance for complexity.

Abstract – Collaborative signal processing algorithms in sensor networks must be robust to device failures because one expects a large number of failures due to the harsh conditions in which they are usually deployed. In this paper, we study two distinct approaches, value-fusion and decisionfusion, for achieving fault-tolerance in collaborative target detection algorithms. In value-fusion, sensor devices first exchange their measured values to arrive at a fault-tolerant consensus on the measurement. Then each device makes an independent decision as to whether or not a target is present based on the consensus measurement. In contrast, in decision-fusion, each device first makes an independent decision as to whether or not a target is present and then the devices exchange their decisions to arrive at a faulttolerant consensus decision. In this paper, we compare the performance of value and decision fusion using two measures: probability of correct detection and probability of false alarm. The results show that if fault-tolerance is not required, then value-fusion is better than decision-fusion and whereas if fault-tolerance is essential, then decisionfusion is better than value-fusion.

Abstract — We develop coding strategies for estimation under communication constraints in tree-structured sensor networks. The strategies have a modular and decentralized architecture. This promotes the flexibility, robustness, and scalability that wireless sensor networks need to operate in uncertain, changing, and resource-constrained environments. The strategies are based on a generalization of Wyner-Ziv source coding with decoder side information. We develop solutions for general trees, and illustrate our results in serial (pipeline) and parallel (hub-and-spoke) networks. Additionally, the strategies can be applied to other network information theory problems. They have a successive coding structure that gives an inherently less complex way to attain a number of prior results, as well as some novel results, for the CEO problem, multiterminal source coding, and certain classes of relay channels. Index Terms — sensor networks, distributed estimation, data fusion, side information, Wyner-Ziv coding, rate distortion theory, CEO problems, multiterminal source coding, distributed detection, relay channels. I.

Abstract—In this paper, we consider two important problems for distributed fault-tolerant detection in wireless sensor networks: 1) how to address both the noise-related measurement error and sensor fault simultaneously in fault-tolerant detection and 2) how to choose a proper neighborhood size n for a sensor node in fault correction such that the energy could be conserved. We propose a fault-tolerant detection scheme that explicitly introduces the sensor fault probability into the optimal event detection process. We mathematically show that the optimal detection error decreases exponentially with the increase of the neighborhood size. Experiments with both Bayesian and Neyman-Pearson approaches in simulated sensor networks demonstrate that the proposed algorithm is able to achieve better detection and better balance between detection accuracy and energy usage. Our work makes it possible to perform energyefficient fault-tolerant detection in a wireless sensor network. Index Terms—Distributed event detection, fault tolerance, sensor fusion, energy-efficiency, wireless sensor networks. 1

This study examines the problem of target detection by a wireless sensor network. Sensors acquire measurements emitted from the target that are corrupted by noise and initially make individual decisions about the presence/absence of the target. We propose the Local Vote Decision Fusion algorithm, in which sensors first correct their decisions using decisions of neighboring sensors, and then make a collective decision as a network. An explicit formula that approximates the system’s decision threshold for a given false alarm rate is derived using limit theorems for random fields, which provides a theoretical performance guarantee for the algorithm. We examine both distance- and nearest neighbor-based versions of the local vote algorithm for grid and random sensor deployments and show that, for a fixed system false alarm, the local vote correction achieves significantly higher target detection rate than decision fusion based on uncorrected decisions. The local vote decision fusion framework is extended to the sequential case, where information becomes available over time.