There is, I think, something ethereal about i —the square root of minus one. I remember first hearing about it at school. It seemed an odd beast at that time—an intruder hovering on the edge of reality.

Usually familiarity dulls this sense of the bizarre, but in the case of i it was the reverse: over the years the sense of its surreal nature intensified. It seemed that it was impossible to write mathematics that described the real world in …

I and i

Machine Learning is the study of methods for programming computers to learn. Computers are applied to a wide range of tasks, and for most of these it is relatively easy for programmers to design and implement the necessary software. However, there are many tasks for which this is difficult or impossible. These can be divided into four general categories. First, there are problems for which there exist no human experts. For example, in modern automated manufacturing facilities, there is a need to predict machine failures before they occur by analyzing sensor readings. Because the machines are new, there are no human experts who can be interviewed by a programmer to provide the knowledge necessary to build a computer system. A machine learning system can study recorded data and subsequent machine failures and learn prediction rules. Second, there are problems where human experts exist, but where they are unable to explain their expertise. This is the case in many perceptual tasks, such as speech recognition, hand-writing recognition, and natural language understanding. Virtually all humans exhibit expert-level abilities on these tasks, but none of them can describe the detailed steps that they follow as they perform them. Fortunately, humans can provide machines with examples of the inputs and correct outputs for these tasks, so machine learning algorithms can learn to map the inputs to the outputs. Third, there are problems where phenomena are changing rapidly. In finance, for example, people would like to predict the future behavior of the stock market, of consumer purchases, or of exchange rates. These behaviors change frequently, so that even if a programmer could construct a good predictive computer program, it would need to be rewritten frequently. A learning program can relieve the programmer of this burden by constantly modifying and tuning a set of learned prediction rules. Fourth, there are applications that need to be customized for each computer user separately. Consider, for example, a program to filter unwanted electronic mail messages. Different users will need different filters. It is unreasonable to expect each user to program his or her own rules, and it is infeasible to provide every user with a software engineer to keep the rules up-to-date. A machine learning system can learn which mail messages the user rejects and maintain the filtering rules automatically. Machine learning addresses many of the same research questions as the fields of statistics, data mining, and psychology, but with differences of emphasis. Statistics focuses on understanding the phenomena that have generated the data, often with the goal of testing different hypotheses about those phenomena. Data mining seeks to find patterns in the data that are understandable by people. Psychological studies of human learning aspire to understand the mechanisms underlying the various learning behaviors exhibited by people (concept learning, skill acquisition, strategy change, etc.).

Machine learning

Probability Distributions.- Linear Models for Regression.- Linear Models for Classification.- Neural Networks.- Kernel Methods.- Sparse Kernel Machines.- Graphical Models.- Mixture Models and EM.- Approximate Inference.- Sampling Methods.- Continuous Latent Variables.- Sequential Data.- Combining Models.

Pattern Recognition and Machine Learning

We will review some of the major results in random graphs and some of the more challenging open problems. We will cover algorithmic and structural questions. We will touch on newer models, including those related to the WWW.

Random graphs

Statistics for Spatial Data.

Recent emphasis on green communications has generated great interest in the investigations of energy harvesting communications and networking. Energy harvesting from ambient energy sources can potentially reduce the dependence on the supply of grid or battery energy, providing many attractive benefits to the environment and deployment. However, unlike the conventional stable energy, the intermittent and random nature of the renewable energy makes it challenging in the realization of energy harvesting transmission schemes. Extensive research studies have been carried out in recent years to address this inherent challenge from several aspects: energy sources and models, energy harvesting and usage protocols, energy scheduling and optimization, implementation of energy harvesting in cooperative, cognitive radio, multiuser and cellular networks, etc. However, there has not been a comprehensive survey to lay out the complete picture of recent advances and future directions. To fill such a gap, in this paper, we present an overview of the past and recent developments in these areas and highlight a number of possible future research avenues.

/pdf/advances-in-energy-harvesting-communications-past-present-3oubbkeycy.pdf

Advances in Energy Harvesting Communications: Past, Present, and Future Challenges

http://sig.umd.edu/publications/Wang_ComNet_201010.pdf

Game theory for cognitive radio networks: An overview

In recent decades, we have witnessed the evolution of information technologies from the development of VLSI technologies, to communication and networking infrastructure, to the standardization of multimedia compression and coding schemes, to effective multimedia content search and retrieval. As a result, multimedia devices and digital content have become ubiquitous. This path of technological evolution has naturally led to a critical issue that must be addressed next, namely, to ensure that content, devices, and intellectual property are being used by authorized users for legitimate purposes, and to be able to forensically prove with high confidence when otherwise. When security is compromised, intellectual rights are violated, or authenticity is forged, forensic methodologies and tools are employed to reconstruct what has happened to digital content in order to answer who has done what, when, where, and how. The goal of this paper is to provide an overview on what has been done over the last decade in the new and emerging field of information forensics regarding theories, methodologies, state-of-the-art techniques, major applications, and to provide an outlook of the future.

http://sig.umd.edu/publications/Stamm_IEEEAccess_201305.pdf

Information Forensics: An Overview of the First Decade

In this paper, symbol-error-rate (SER) performance analysis and optimum power allocation are provided for uncoded cooperative communications in wireless networks with either decode-and-forward (DF) or amplify-and-forward (AF) cooperation protocol, in which source and relay send information to destination through orthogonal channels. In case of the DF cooperation systems, closed-form SER formulation is provided for uncoded cooperation systems with PSK and QAM signals. Moreover, an SER upper bound as well as an approximation are established to show the asymptotic performance of the DF cooperation systems, where the SER approximation is asymptotically tight at high signal-to-noise ratio (SNR). Based on the asymptotically tight SER approximation, an optimum power allocation is determined for the DF cooperation systems. In case of the AF cooperation systems, we obtain at first a simple closed-form moment generating function (MGF) expression for the harmonic mean to avoid the hypergeometric functions as commonly used in the literature. By taking advantage of the simple MGF expression, we obtain a closed-form SER performance analysis for the AF cooperation systems with PSK and QAM signals. Moreover, an SER approximation is also established which is asymptotically tight at high SNR. Based on the asymptotically tight SER approximation, an optimum power allocation is determined for the AF cooperation systems. In both the DF and AF cooperation systems, it turns out that an equal power strategy is good, but in general not optimum in cooperative communications. The optimum power allocation depends on the channel link quality. An interesting result is that in case that all channel links are available, the optimum power allocation does not depend on the direct link between source and destination, it depends only on the channel links related to the relay. Finally, we compare the performance of the cooperation systems with either DF or AF protocol. It is shown that the performance of a systems with the DF cooperation protocol is better than that with the AF protocol. However, the performance gain varies with different modulation types and channel conditions, and the gain is limited. For example, in case of BPSK modulation, the performance gain cannot be larger than 2.4 dB; and for QPSK modulation, it cannot be larger than 1.2 dB. Extensive simulation results are provided to validate the theoretical analysis.

/pdf/cooperative-communication-protocols-in-wireless-networks-45v4kxk5p5.pdf

Cooperative Communication Protocols in Wireless Networks: Performance Analysis and Optimum Power Allocation

Merging the fundamental principles of resource allocation with the state-of-the-art in research and application examples, Han and Liu present a novel and comprehensive perspective for improving wireless systems performance. Cross-layer multiuser optimization in wireless networks is described systematically. Starting from the basic principles, such as power control and multiple access, coverage moves to the optimization techniques for resource allocation, including formulation and analysis, and game theory. Advanced topics such as dynamic resource allocation and resource allocation in antenna array processing, and in cooperative, sensor, personal area, and ultrawideband networks, are then discussed. Unique in its scope, timeliness, and innovative author insights, this invaluable work will help graduate students and researchers to understand the basics of wireless resource allocation whilst highlighting modern research topics, and will help industrial engineers to improve system optimization.

/pdf/resource-allocation-for-wireless-networks-basics-techniques-1epprmuvka.pdf

K. J. Ray Liu

Papers

Advances in Energy Harvesting Communications: Past, Present, and Future Challenges

Game theory for cognitive radio networks: An overview

Information Forensics: An Overview of the First Decade

Cooperative Communication Protocols in Wireless Networks: Performance Analysis and Optimum Power Allocation

Resource Allocation for Wireless Networks: Basics, Techniques, and Applications