José M. N. Vieira

Artificial neural networks and neuro-fuzzy systems for modelling and controlling real systems: a comparative study

Abstract This article presents a comparison of artificial neural networks and neuro-fuzzy systems applied for modelling and controlling a real system. The main objective is to model and control the temperature inside of a kiln for the ceramic industry. The details of all system components are described. The steps taken to arrive at the direct and inverse models using the two architectures: adaptive neuro fuzzy inference system and feedforward neural networks are described and compared. Finally, real-time control results using internal model control strategy are presented. Using available Matlab software for both algorithms, the objective is to show the implementation steps for modelling and controlling a real system. Finally, the performances of the two solutions were compared through different parameters for a specific real didactic case.

Interpolation, spectrum analysis, error-control coding, and fault-tolerant computing

This paper uncovers relations between the topics mentioned in the title, relations that we believe to have gone nearly unnoticed so far. More precisely, we show that four often studied problems in signal processing, spectrum analysis, information theory, and computing are closely related or even equivalent in a certain sense (if one of them can be solved, so can any of the others, and using essentially the same algorithms). The problems are (i) a nonlinear band-limited finite-dimensional interpolation problem (ii) the problem of estimating a signal that is the superposition of a finite number of harmonics (iii) an error-control coding problem in the real field, and (iv) certain techniques that occur in algorithm-based fault tolerant computing. The advantages of recognizing these problems as equivalent are obvious: the techniques commonly used in one field can be imported to the others, the duplication of research efforts is prevented, and the overall degree of understanding of the four problems increases. New algorithms are suggested as a result of these investigations.

Stable DFT codes and frames

Discrete Fourier transform codes (DFT codes) or real number codes have been studied and recognized as useful (as joint source-channel codes, for example) but are not stable under bursty losses. This paper introduces a two-channel DFT code with an interleaver and shows that its numerical stability far exceeds that of the corresponding single-channel DFT code (the ratio of the frame bounds for the two-channel system can be smaller by many orders of magnitude). This leads to a stable way of dealing with bursts of errors using DFT codes.

A sliding window solution for the on-line implementation of the Levenberg-Marquardt algorithm

The Levenberg-Marquardt algorithm is considered as the most effective one for training artificial neural networks but its computational complexity and the difficulty to compute the trust region have made it very difficult to develop a true iterative version to use in on-line training. The algorithm is frequently used for off-line training in batch versions although some attempts have been made to implement iterative versions. To overcome the difficulties in implementing the iterative version, a batch-sliding window with Early Stopping, which uses a hybrid Direct/Specialized evaluation procedure, is proposed and tested with a real system. m.

Gigasample Time-Interleaved Delta-Sigma Modulator for FPGA-Based All-Digital Transmitters

This paper presents the architecture and implementation of an FPGA-based all digital transmitter with a baseband sampling rate of 1 GHz. Hardware complexity and operation frequency analysis are performed to find the best hardware topology to improve the transmitter performance. The used polyphase decomposition techniques for delta sigma modulators considerably improve the bandwidth and SNR figures of merit of state-of-the-art FPGA integrated RF transmitters. The experimental results obtained show the feasibility of this approach and the improvement in the signal modulator oversampling ratio, enabling higher bandwidth in all-digital transmitters.

Implementing the Levenberg-Marquardt Algorithm on-line: a Sliding Window Approach with Early Stopping

Abstract The Levenberg-Marquardt algorithm is considered as the most effective one for training Artificial Neural Networks but its computational complexity and the difficulty to compute the trust region have made it very difficult to develop a true iterative version to use in on-line training. The algorithm is frequently used for off-line training in batch versions although some attempts have been made to implement iterative versions. To overcome the difficulties in implementing the iterative version, a batch sliding window with Early Stopping version, which uses a hybrid Direct/Specialized evaluation procedure is proposed and tested with a real system.

Analog Filter Bank for Cochlear Radio

In this paper, we propose a design for the analog part of an Hybrid Filter Bank (HFB) system targeted to radio-frequency (RF) signals, inspired on the mammals cochlea, with the basilar membrane mimicked by microstrip lines. A prototype is presented, operating from 840 MHz to 1470 MHz using ADCs working at 210 Msps. Simulation results revealed that it is possible to design a HFB with low values of distortion and aliasing. With this cochlear analog front-end it is possible to increase the bandwidth of the signal while keeping a high dynamic range. Further developments of this technology could fulfill the promise of Software-defined Radio to build a radio able to deal with wideband signals.

All-Digital Transmitter With a Mixed-Domain Combination Filter

In this brief, we present a new reconfigurable filtering technique for all-digital transmitters using signal interference and cancellation. The proposed method allows the change of some filtering characteristics digitally. This technique is used to reduce the quantization noise from the all-digital transmitters output, reducing the analog output filter requirements. Without a narrow band filter, a greater degree of flexibility for the transmitters carrier frequency and bandwidth is attained, improving its suitability for multiband and multistandard operation. The results show the quantization noise reduction without an output analog filter, contributing to an increase of the transmitters coding efficiency up to more than 95%.

High Accuracy 3D Indoor Positioning Using Broadband Ultrasonic Signals

In this paper we propose a high accuracy indoor positioning system based on broadband ultrasonic signals and time of arrival (ToA) measurements. Using low cost transducers we were able to use acoustic chirps of between 20 and 45KHz as pulse signals. This overcomes most of the problems faced by the narrow band signals usually used with common piezo-ultrasonic transducers, which include poor resolution, low environment noise immunity, short range and low robustness to the Doppler effect. Using synchronized ultrasonic anchor nodes and time division multiplexing to share the medium, we build a GPS-like system for indoor pervasive applications. A set of experiments were performed to evaluate the proposed system. Very stable 3D position estimates were obtained (absolute standard deviation less than 2.3cm) and a position refresh rate of 350ms was achieved.

The Rank of Random Binary Matrices and Distributed Storage Applications

Random binary matrices appear in a variety of signal processing and encoding problems. They play an important role in rateless codes and in distributed storage applications. This paper focuses on block angular matrices, a class of random rectangular binary matrices that are particularly suited to distributed storage applications. We address one of the key issues regarding binary random matrices in general, and block angular matrices in particular: the probability of obtaining a full rank matrix, when drawing uniformly at random from the set of binary matrices with compatible structure. This paper gives a closed-form expression for this probability, as well as some bounds and approximations.