Information Theory and Reliable Communication

This paper presents control and coordination algorithms for groups of vehicles. The focus is on autonomous vehicle networks performing distributed sensing tasks where each vehicle plays the role of a mobile tunable sensor. The paper proposes gradient descent algorithms for a class of utility functions which encode optimal coverage and sensing policies. The resulting closed-loop behavior is adaptive, distributed, asynchronous, and verifiably correct.

Coverage control for mobile sensing networks

Gigantically comprehensive and carefully researched, Security Engineering makes it clear just how difficult it is to protect information systems from corruption, eavesdropping, unauthorized use, and general malice. Better, Ross Anderson offers a lot of thoughts on how information can be made more secure (though probably not absolutely secure, at least not forever) with the help of both technologies and management strategies. His work makes fascinating reading and will no doubt inspire considerable doubt--fear is probably a better choice of words--in anyone with information to gather, protect, or make decisions about. Be aware: This is absolutely not a book solely about computers, with yet another explanation of Alice and Bob and how they exchange public keys in order to exchange messages in secret. Anderson explores, for example, the ingenious ways in which European truck drivers defeat their vehicles' speed-logging equipment. In another section, he shows how the end of the cold war brought on a decline in defenses against radio-frequency monitoring (radio frequencies can be used to determine, at a distance, what's going on in systems--bank teller machines, say), and how similar technology can be used to reverse-engineer the calculations that go on inside smart cards. In almost 600 pages of riveting detail, Anderson warns us not to be seduced by the latest defensive technologies, never to underestimate human ingenuity, and always use common sense in defending valuables. A terrific read for security professionals and general readers alike. --David Wall Topics covered: How some people go about protecting valuable things (particularly, but not exclusively, information) and how other people go about getting it anyway. Mostly, this takes the form of essays (about, for example, how the U.S. Air Force keeps its nukes out of the wrong hands) and stories (one of which tells of an art thief who defeated the latest technology by hiding in a closet). Sections deal with technologies, policies, psychology, and legal matters.

Security Engineering: A Guide to Building Dependable Distributed Systems

Neural networks for control systems: a survey

This chapter introduces the subject of statistical pattern recognition (SPR). It starts by considering how features are defined and emphasizes that the nearest neighbor algorithm achieves error rates comparable with those of an ideal Bayes’ classifier. The concepts of an optimal number of features, representativeness of the training data, and the need to avoid overfitting to the training data are stressed. The chapter shows that methods such as the support vector machine and artificial neural networks are subject to these same training limitations, although each has its advantages. For neural networks, the multilayer perceptron architecture and back-propagation algorithm are described. The chapter distinguishes between supervised and unsupervised learning, demonstrating the advantages of the latter and showing how methods such as clustering and principal components analysis fit into the SPR framework. The chapter also defines the receiver operating characteristic, which allows an optimum balance between false positives and false negatives to be achieved.

Statistical Pattern Recognition

We review techniques for sensor fusion in robot navigation, emphasizing algorithms for self-location. These find use when the sensor suite of a mobile robot comprises several different sensors, some complementary and some redundant. Integrating the sensor readings, the robot seeks to accomplish tasks such as constructing a map of its environment, locating itself in that map, and recognizing objects that should be avoided or sought. The review describes integration techniques in two categories: low-level fusion is used for direct integration of sensory data, resulting in parameter and state estimates; high-level fusion is used for indirect integration of sensory data in hierarchical architectures, through command arbitration and integration of control signals suggested by different modules. The review provides an arsenal of tools for addressing this (rather ill-posed) problem in machine intelligence, including Kalman filtering, rule-based techniques, behavior based algorithms, and approaches that borrow from information theory, Dempster-Shafer reasoning, fuzzy logic and neural networks.

Sensor fusion for mobile robot navigation

Z. Chair and P.R. Varshney (1986) solved the data fusion problem for fixed binary local detectors with statistically independent decisions. Their solution is generalized by using the Bahadur-Lazarsfeld expansion of probability density functions. The optimal data fusion rule is developed for correlation local binary decisions, in terms of the conditional correlation coefficients of all orders. It is shown that when all these coefficients are zero, the rule coincides with the original Chair-Varshney design. >

Optimal data fusion of correlated local decisions in multiple sensor detection systems

Active authentication is the problem of continuously verifying the identity of a person based on behavioral aspects of their interaction with a computing device. In this paper, we collect and analyze behavioral biometrics data from 200 subjects, each using their personal Android mobile device for a period of at least 30 days. This data set is novel in the context of active authentication due to its size, duration, number of modalities, and absence of restrictions on tracked activity. The geographical colocation of the subjects in the study is representative of a large closed-world environment such as an organization where the unauthorized user of a device is likely to be an insider threat: coming from within the organization. We consider four biometric modalities: 1) text entered via soft keyboard, 2) applications used, 3) websites visited, and 4) physical location of the device as determined from GPS (when outdoors) or WiFi (when indoors). We implement and test a classifier for each modality and organize the classifiers as a parallel binary decision fusion architecture. We are able to characterize the performance of the system with respect to intruder detection time and to quantify the contribution of each modality to the overall performance.

Active Authentication on Mobile Devices via Stylometry, Application Usage, Web Browsing, and GPS Location

Two important computational features of neural networks are associative storage and retrieval of knowledge, and uniform rate of convergence of network dynamics independent of network dimension. It is indicated how these properties can be used for adaptive control through the use of neural network computation algorithms, and resulting computational advantages are outlined. The neuromorphic control approach is compared to model reference adaptive control on a specific example. It is shown that the utilization of neural networks for adaptive control offers definite speed advantages over traditional approaches for very-large-scale systems. >

Neural network architecture for control

The complex computing systems employed by governments, corporations, and other institutions are frequently targeted by cyber-attacks designed for espionage and sabotage. The malicious software used in such attacks are typically custom-designed or obfuscated to avoid detection by traditional antivirus software. Our goal is to create a malware detection system that can quickly and accurately detect such otherwise difficult-to-detect malware. We pose the problem of malware detection as a multi-channel change-point detection problem, wherein the goal is to identify the point in time when a system changes from a known clean state to an infected state. We present a host-based malware detection system designed to run at the hypervisor level, monitoring hypervisor and guest operating system sensors and sequentially determining whether the host is infected. We present a case study wherein the detection system is used to detect various types of malware on an active web server under heavy computational load.

/pdf/multi-channel-change-point-malware-detection-2p05w4jdbj.pdf

Moshe Kam

Papers

Sensor fusion for mobile robot navigation

Optimal data fusion of correlated local decisions in multiple sensor detection systems

Active Authentication on Mobile Devices via Stylometry, Application Usage, Web Browsing, and GPS Location

Neural network architecture for control

Multi-channel Change-Point Malware Detection