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

Wireless indoor positioning systems have become very popular in recent years. These systems have been successfully used in many applications such as asset tracking and inventory management. This paper provides an overview of the existing wireless indoor positioning solutions and attempts to classify different techniques and systems. Three typical location estimation schemes of triangulation, scene analysis, and proximity are analyzed. We also discuss location fingerprinting in detail since it is used in most current system or solutions. We then examine a set of properties by which location systems are evaluated, and apply this evaluation method to survey a number of existing systems. Comprehensive performance comparisons including accuracy, precision, complexity, scalability, robustness, and cost are presented.

Survey of Wireless Indoor Positioning Techniques and Systems

The PASCAL Visual Object Classes Challenge

An overview of the self-organizing map algorithm, on which the papers in this issue are based, is presented in this article.

The Self-Organizing Map

An uncertain optimal control model with Hurwicz criterion

This paper presents the design and implementation of a unique multi-radio Wireless Sensor Network (WSN) platform compared to current WSN nodes that have only one radio interface. Four independent, low energy radio transceivers allow simultaneous reception and transmission to either another multi-radio WSN node or up to four different WSN nodes. This also enables high interference tolerance, low latency, and high mesh-networking performance. The platform is based on synthesizable multi-processor System-on-Chip implementation on FPGA. The radios are compatible with ultra-low energy microcontroller based WSN nodes, which can be freely mixed in the network. The platform applicability is demonstrated by an ultra low latency WSN router and high data rate file transfer applications. Theoretical hop delay is as low as 106 µs, while 3.3 Mbps network throughput is achievable. Performance measurements for up to four parallel Nios II softcore processors are also presented.

High-performance multi-radio WSN platform

Specification and Description Language (SDL) is a high abstraction level system design language with a clear graphical notation. Because of formal presentation, an SDL model can be automatically converted into source C code for implementation. However, the high abstraction level creates a conceptual gap between a general SDL model and its implementation in a final operational platform. The paper studies the SDL development of an embedded Medium Access Control (MAC) protocol for a wireless local area network (WLAN) demonstrator. The SDL design flow for the protocol is first started by architectural design without target platform dependencies. Functionality is added to the model using the top-down design approach. Functional simulations are used for verifying the operation of the protocol. Next, the performance is estimated using performance simulations in a workstation environment. Performance improvements can be achieved by optimising the SDL model.

Using SDL for implementing a wireless medium access control protocol

This paper presents an intrusion detection system architecture based on the artificial immune system concept. In this architecture, an innate immune mechanism through unsupervised machine learning methods is proposed to primarily categorize network traffic to “self” and “non-self” as normal and suspicious profiles respectively. Unsupervised machine learning techniques formulate the invisible structure of unlabeled data without any prior knowledge. The novelty of this work is utilization of these methods in order to provide online and real-time training for the adaptive immune system within the artificial immune system. Different methods for unsupervised machine learning are investigated and DBSCAN (density-based spatial clustering of applications with noise) is selected to be utilized in this architecture. The adaptive immune system in our proposed architecture also takes advantage of the distributed structure, which has shown better self-improvement rate compare to centralized mode and provides primary and secondary immune response for unknown anomalies and zero-day attacks. The experimental results of proposed architecture is presented and discussed.

Artificial Immune System Based Intrusion Detection: Innate Immunity using an Unsupervised Learning Approach

Finite state machine (FSM) optimisation has usually been studied through state assignment, state vector encoding, and combinational logic optimisation. Such details should not be consequential in behavioural descriptions. On the other hand, describing correct and efficient hardware structures in VHDL (VHSIC hardware description language), or generally in any high-level description language, is more a question of description style than correct language statements. Therefore, more or less conscious choices are made in the design description itself that guide the synthesis software toward a specific implementation. The best implementation is also dependent on the target technology and, therefore, there is no single best description style for all FSMs. The paper is a study of the kind of performance trade-offs that can be made by changing the description style. A program is shown to be able to generate these different descriptions from an intermediate format (kiss2) describing the FSM. Therefore, this process for finding a better description could be automated and performed by the synthesis software itself. Descriptions are tested on a set of 13 FSMs most from a benchmark suite LGSynth93. The results show at least two times better performance of speed or area in the best description compared with the worst. In performance critical applications this difference can be of a crucial importance.

Timo Hämäläinen

Papers

An uncertain optimal control model with Hurwicz criterion

High-performance multi-radio WSN platform

Using SDL for implementing a wireless medium access control protocol

Artificial Immune System Based Intrusion Detection: Innate Immunity using an Unsupervised Learning Approach

Finite state machine encoding for VHDL synthesis