Knowledge-based Signal Research Articles

A data-driven bi-level program for knowledge-based signal control system under uncertainty

A data-driven knowledge based system (DDKS) is considered for urban signal control with hazardous material (hazmat) transportation. A data-driven bi-level program (DDBP) is presented to determine generalized travel cost for hazmat carriers and regular traffic flows. A risk-averse (RA) signal control is developed for DDKS with uncertain risk in the presence of hazmat transportation. Since DDBP is generally non-convex, a stochastic program using two-stage approach is proposed to find local optimal solutions. Numerical computations using a real-data city network are made and good results are obtained. As compared with conventional signal controls such as delay-minimizing (DM) and risk-neutral (RN) signal control, the proposed RA exhibits considerable advantage on mitigation of public risk exposure whilst incurred less cost loss as compared to other data-driven alternatives in all cases.

Design and application of dynamic environmental knowledge base

The authors propose a two-stage approach to design radar dynamic environmental knowledge base (DEKB), which provides a priori information for knowledge-based (KB) signal processing and radar system design. Specifically, the first stage is to use historical knowledge such as maps to establish the static environmental knowledge base (SEKB), which is the initial state of the DEKB. The second stage is to employ radar returns and other sensors information to update the knowledge base dynamically. Moreover, an exemplar establishment of the SEKB based on real scene and a specific update method based on two Anderson–Darling tests are introduced. Finally, as an application of the DEKB, the authors present a new KB constant false alarm rate (CFAR) detector utilising multidimensional knowledge from the DEKB. The performance of the new detector is analysed by real radar data, collected by a linear frequency-modulated continues wave radar, and compared with the classical cell averaging CFAR detector and the KB CFAR detector which uses only one kind of knowledge.

Knowledge-based radar signal and data processing: a tutorial review

Radar systems are an important component in military operations. In response to increasingly severe threats from military targets with reduced radar cross sections (RCSs), slow-moving and low-flying aircraft hidden in foliage, and in environments with large numbers of targets, knowledge-based (KB) signal and data processing techniques offer the promise of significantly improved performance of all radar systems. Radars under KB control can be deployed to utilize valuable resources such as airspace or runways more effectively and to aid human operators in carrying out their missions. As battlefield scenarios become more complex with increasing numbers of sensors and weapon systems, the challenge will be to use already available information effectively to enhance radar performance, including positioning, waveform selection, and modes of operation. KB processing fills this need and helps meet the challenge.

New signal processing techniques for the decomposition of EMG signals.

This paper relates to the use of knowledge-based signal processing techniques in the decomposition of EMG signals. The aim of the research is to automatically decompose EMG signals recorded at force levels up to 20 per cent maximum voluntary contraction (MVC) into their constituent motor unit action potentials (MUAPS), and to display the MUAP shapes and firing times for the clinician. This requires the classification of nonoverlapping MUAPs and superimposed waveforms formed from overlapping MUAPs in the signal. Nonoverlapping MUAPs are classified using a statistical pattern-recognition method. The decomposition of superimposed waveforms uses a combination of procedural and knowledge-based methods. The decomposition method was tested on real and simulated EMG data recorded at force levels up to 20 per cent MVC. The different EMG signals contained up to six motor units (MUs). The new decomposition program classifies the total number of MUAP firings in an EMG signal with an accuracy always greater than 95 per cent. The decomposition program takes about 15s to classify all nonoverlapping MUAPs in EMG signal of length 1.0s and, on average, an extra 9s to classify each superimposed waveform.

A Knowledge-based Signal-symbol Processing Approach for Change Detection

Recently, computerized fault diagnostic systems have been applied to catch up and localize incipient failures in complex plants. The early indication of abmpt changes in the dynamic properties of aignals is a critical issue in the operation of such systems. To date, systems performing change detection functions effectively, have generally required the continuous, constructive interaction with the skilled analyst for intelligent assistance in commissioning signal processing activities. Quite recently, remarkable efforts have been made to automate the nontrivial decision-making activities of the human operator in this field. In this paper we describe a knowledge-based approach to solve the autonomous, change detection problem. The focus of the paper is on the expert skill-based decision process because this is the critical but nevertheless neglected, aspect in the realization of knowledge-based signal processing systems. In the framework of a combined expert-decision support system (ES/DSS) we aim to provide a methodology to embody expert trategies as control knowledge and signal-symbol processing skills as domain knowledge applied to the change detection task. The overall scheme and behavior of the system is briefly reviewed.

Neural network signal understanding for instrumentation

A report is presented on the use of neural signal interpretation theory and techniques for the purpose of classifying the shapes of a set of instrumentation signals, in order to calibrate devices, diagnose anomalies, generate tuning/settings, and interpret the measurement results. Neural signal understanding research is surveyed, and the selected implementation and its performance in terms of correct classification rates and robustness to noise are described. Formal results on neural net training time and sensitivity to weights are given. A theory for neural control using functional link nets is given, and an explanation facility designed to help neural signal understanding is described. The results are compared to those obtained with a knowledge-based signal interpretation system using the same instrument and data.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Knowledge-based signal processing for radar ESM systems

Radar electronic support measures (ESM) systems perform the functions of threat detection and area surveillance to determine the identity and bearing of surrounding radar emitters. Automatic ESM systems incorporate a passive receiver to measure the parameters of detected radar pulses and an automatic processor to rapidly sort pulses and identify the emitters. Current processors use algorithmic processing methods which are inflexible and do not fully utilise available sources of a priori information. The paper discusses the role of knowledge-based processing methods and how they may be applied to the key ESM signal-processing functions of deinterleaving, merge and emitter identification. ESM processors are required to sort input pulse data streams exceeding one million pulses per second and minimise the reporting latency of new emitters. The paper further discusses the requirements to achieve real-time operation of knowledge-based ESM processing techniques.

Rosette: An object-oriented concurrent systems architecture

A number of concurrent computers have been designed, such as [Ametek 1988] and [NCUBE 1986]. Taking advantage of concurrency on these computers will require new languages, operating systems, and en...

Knowledge-based signal processing in the decomposition of myoelectric signals

The mixture of quantitative and qualitative information utilized in decomposition, which suggests that a purely numerical approach to the task would be inefficient, is discussed. The incorporation of symbolic processing methods into the decomposition system to complement the numerical aspects of the task is discussed. The overall philosophy used in the decomposition system is examined, focusing on the principles of detection used. The organization of the software used by the system is described.