Timo Ala-Kleemola

GAIS: A Method for Detecting Interleaved Sequential Patterns from Imperfect Data

This paper introduces a novel method, GAIS, for detecting interleaved sequential patterns from databases. A case, where data is of low quality and has errors is considered. Pattern detection from erroneous data, which contains multiple interleaved patterns is an important problem in a field of sensor network applications. We approach the problem by grouping data rows with the help of a model database and comparing groups with the models. In evaluation GAIS clearly outperforms the greedy algorithm. Using GAIS desired sequential patterns can be detected from low quality data.

New aspects to knowledge-aided clutter analysis

Digital signal processing allows improvements in site-specific clutter prediction. With digital terrain maps and a flight obstacle register, land clutter origin can be solved. An efficient, knowledge-aided approach to extracting homogeneous clutter from radar signal is presented. Once homogeneous clutters statistic has been recognized, also mixture models can be constructed. The suggested aspects are illustrated through an air surveillance radar simulation. The enhancement attained in clutter analysis and thus in clutter models is the novelty of the presented aspects.

Case based reasoning approach to automatic comparison of models

In recent years much work has been done to find regularities in data streams. Normally streams are complete and well ordered. Major defects, such as incomplete or misleading events, occur in practice at least after some finite time. In this paper we describe a method developed for recognizing previously known patterns from a data stream. The solution is based on case based reasoning (CBR) with an extension to handle imperfect information. We also introduce a new idea to use two different case-bases instead of only one. The performance has been evaluated using simulated data and the obtained results are considered encouraging.

Classification of items in a walk-through metal detector using time series of eigenvalues of the polarizability tensor

During the last decade, the safety regulations of the airports have been set to a new level. As the number of passengers is constantly increasing, yet effective but quick security control at checkpoints sets great requirements to the 21st century security systems. In this paper, we shall introduce a novel metal detector concept that enables not only to detect but also to classify hidden items, though their orientation and accurate location are unknown. Our new prototype walk-through metal detector generates mutually orthogonal homogeneous magnetic fields so that the measured dipole moments allow classification of even the smallest of the items with high degree of accuracy in real-time. Invariant to different rotations of an object, the classification is based on eigenvalues of the polarizability tensor that incorporate information about the item (size, shape, orientation etc.); as a further novelty, we treat the eigenvalues as time series. In our laboratory settings, no assumptions concerning the typical place, where an item is likely situated, are made. In that case, 90 % of the dangerous and harmless items, including knives, guns, gun parts, belts etc. according to a security organisation, are correctly classified. Made misclassifications are explained by too similar electromagnetic properties of the items in question. The theoretical treatment and simulations are verified via empirical tests conducted using a robotic arm and our prototype system. In the future, the state-of-the-art system is likely to speed-up the security controls significantly with improved safety.

On Sequential On-Line Outlier Detection and a Linescan Application

Industrial quality monitoring is increasing rapidly, and challenging signal environments with requirement of steady performance pose conflicting demands to on-line tests. The sequential probability ratio test (SPRT) and the Kalman filter (KF) are proposed as two tools for detection and recognition-oriented signal processing. A modified sequential test is suggested and applied to a linescan problem

Metadata in Sequential Real-Time 2-D Detection

Extraction of patterns is an important low-level operation in several vision applications. In particular, this work is motivated by the problem of detecting unknown vague structures in images with a minimum sample number. Tracing of an unknown structure is allowed by the spatial 2-D placement of extremum. Often, this permits near-optimum testing of hypothesis, which also accentuates the descriptive value of extrema. The proposed sequential approach guarantees predictable detection delay, which is a significant advantage in real-time use. The performance is demonstrated with simulations and real experiments, where transients have unknown starting time

New detection algorithm for poor signal-to-noise conditions

A new detection algorithm for radar signal processing is presented. Especially in fighter applications, well-performing detection algorithms play a crucial role for enhancing situational awareness. By utilizing the local characteristics of the received signal more efficiently, better performance can be achieved. As phenomena-considered-disturbances in the received target signal are often discarded, valuable information is lost. However, this associated information can benefit the system, although even the latest statistical inference and modeling, are applied in practice with synthetic array radar (SAR) image processing. The usefulness of this new approach is demonstrated.

Incorporating weather conditions and various scatterers into volumetric radar clutter simulation

This paper presents a method for generating volumetric clutter for air surveillance radar simulation. Complex valued radar signal consists of magnitude and phase. In the presented simulation, radar clutter signal is created from magnitude and phase distribution and then filtered imitating the radar signal formation. Radar geometry can be integrated to the simulation by manipulating magnitude, phase, and phase difference distributions. Magnitude is affected by range bin size and distance from radar. Also weather condition and polarization effect on the signal. These can be controlled with adjustments to the distribution that the matrix is created from. This solution offers a simple way to create background to realistic radar simulation. Different distributions are used for signal magnitude and phase of various clutter sources. Typically, volumetric clutter source consists of many evenly sized scatterers. Preliminary phase, originating from randomly distributed particles, can be considered evenly distributed. Phase difference in long time, on the other hand, shows the radial movement of particles. Therefore, phase difference can be modeled, for example, with Gaussian distribution and magnitude with Weibull distribution, of course, depending on true environment. As an example, chaff is simulated with differing radial wind.

On joint phase-envelope use in radar CFAR processing

A method assuming linear phase drift is presented to improve radar detection performance. Its use is based on the assumption that the target illumination time comprises multiple coherent pulses or coherent processing intervals (CPI). For example in a conventional scanning radar, this often inaccurate information can be used for statistical data mapping to point out possible target presence. If coherent integration is desired in a beam-agile system, the method should allow sequential detection. Discussion involves a pragmatic example on the echo phase progress utilization in the constant false alarm rate (CFAR) processing of a moving target indication (MTI) system. The detection performance is evaluated with scanning radar simulations. The method has been tested using real-world recordings and some observations are briefly outlined.