Intelligent signal processing and data mining underlie numerous important applications in science and engineering. This book aims to present the most recent developments in these areas with a focus on big data processing and reasoning in real time. It consists of 12 standalone chapters grouped into two parts.

Following an introductory chapter, the first part (chapters 2 to 6) is devoted to novel methods for intelligent signal processing, with a focus on sequential Monte Carlo methods. Chapter 2 presents several sequential Monte Carlo and Markov chain Monte Carlo methods for tracking objects moving in groups. The proposed methods categorize objects into groups and can scale well to a large number of objects. The authors also propose a causality inference framework for identifying dominant agents based on their trajectories.

Chapter 3 focuses on the problem of multi-target tracking of an unknown number of targets. The authors propose an improved sequential Monte Carlo implementation of the intensity filter that achieves good performance in real-world problems.

Chapter 4 begins with a survey of the state of the art in localization techniques for wireless networks, and a discussion of sequential Monte Carlo methods for the localization of mobile nodes in wireless networks based on received signal strength indicators. The performance of the proposed methods is evaluated in urban and suburban environments, using both simulated and real-world data.

Chapter 5 studies the problem of electroencephalography (EEG) source localization. It proposes a solution that combines sequential Monte Carlo methods and spatial filtering by beamforming. The proposed solution is validated for brain source localization using generated EEG data from dipoles with known locations and moments, which are artificially corrupted with noise.

Chapter 6, the last chapter in this part, is devoted to the application of sequential Monte Carlo methods to the automotive industry for lane detection and tracking. It discusses model-based lane detection algorithms and focuses on a recent lane feature extraction approach that offers improved tracking performance.

The second part (chapters 7 to 12) is devoted to recent advances in data mining and machine learning approaches and their application. Chapter 7 explores anomaly detection methods for sensor signals. The chapter explains how patterns for the detection of anomalies can be expressed with the help of relational algebra, and presents an application of an anomaly detection system for monitoring and analyzing air traffic.

Chapter 8 presents a tutorial overview of hierarchical clustering methods. Together with the presented methods, the authors discuss related data visualization techniques, such as the cartoon cluster plot, the dendrogram, and the bicluster plot. The authors discuss scalability issues on the runtime; memory requirements of the hierarchical clustering methods are also presented. They then introduce a hybrid clustering approach to improve the runtime of the average link hierarchical clustering algorithm, and evaluate it on 16 benchmark datasets.

Chapter 9 studies the problem of automatic object recognition in complex scenes, and proposes a method that uses shape context representation, many-to-one matching, and subtractive clustering to solve it. The proposed method is experimentally shown to be robust to image transformations and ambient conditions.

Chapter 10 presents an overview of historical consistent neural networks (HCNNs), a new type of time-delay recurrent neural networks that can be useful in several applications, including financial market forecasting and risk analysis. HCNNs make no distinction between input and output variables, and can be used to model high-complexity systems made up of a number of interacting subdynamics.

In chapter 11, the authors provide a general overview of reinforcement learning, and discuss the implementation of reinforcement learning using neural networks. The chapter includes two example board game applications of reinforcement learning to sequential decision-making tasks, one focusing on tic-tac-toe and the other on Chung Toi.

Chapter 12 concludes the book, covering the data analysis and modeling of biomedical nonlinear and nonstationary time series. The authors present an online variant of empirical mode decomposition and its application to biomedical time series.

In summary, this book is well structured and provides good coverage of several state-of-the-art approaches to intelligent signal processing and data mining. Most of the chapters contain examples and algorithms and are accompanied by useful illustrations (several of which are in color) that help to explain the described approaches. The audience for the book is researchers and practitioners in signal processing and data mining who are interested in the latest developments in these areas.