Vehicle classification in Sensor Networks using time-domain signal processing and Neural Networks

AbstractCellular micromotion—a tiny movement of cell membranes on the nm-µm scale—has been proposed as a pathway for inter-cellular signal transduction and as a label-free proxy signal to neural activity. Here we harness several recent approaches of signal processing to detect such micromotion in video recordings of unlabeled cells. Our survey includes spectral filtering of the video signal, matched filtering, as well as 1D and 3D convolutional neural networks acting on pixel-wise time-domain data and a whole recording respectively.

Download Full-text

Statistical View on Phase and Magnitude Information in Signal Processing

Artificial Satellites ◽

10.2478/v10018-012-0018-6 ◽

2012 ◽

Vol 47 (3) ◽

pp. 127-136

Author(s):

Waldemar Popiński

Keyword(s):

Signal Processing ◽

Time Domain ◽

Signal Reconstruction ◽

Time Domain Signal ◽

Reconstructed Signal ◽

Band Pass ◽

Expected Values ◽

Random Fluctuations ◽

Phase Distortions ◽

Complex Valued

Statistical View on Phase and Magnitude Information in Signal ProcessingIn this work the problem of reconstruction of an original complex-valued signalot,t= 0, 1, …,n- 1, from its Discrete Fourier Transform (DFT) spectrum corrupted by random fluctuations of magnitude and/or phase is investigated. It is assumed that the magnitude and/or phase of discrete spectrum values are distorted by realizations of uncorrelated random variables. The obtained results of analysis of signal reconstruction from such distorted DFT spectra concern derivation of the expected values and bounds on variances of the reconstructed signal at the observation moments. It is shown that the considered random distortions in general entail change in magnitude and/or phase of the reconstructed signal expected values, which together with imposed random deviations with finite variances can blur the similarity to the original signal. The effect of analogous random amplitude and/or phase distortions of a complex valued time domain signal on band pass filtration of distorted signal is also investigated.

Download Full-text

Time-Domain Signal Processing Via Four-Wave Mixing In Nonlinear Delay Lines

Optical Engineering ◽

10.1117/12.7972888 ◽

1982 ◽

Vol 21 (2) ◽

Cited By ~ 5

Author(s):

T. R. O'Meara ◽

Amnon Yariv

Keyword(s):

Signal Processing ◽

Time Domain ◽

Four Wave Mixing ◽

Wave Mixing ◽

Delay Lines ◽

Time Domain Signal ◽

Nonlinear Delay

Download Full-text

A new spectral estimation-based feature extraction method for vehicle classification in distributed sensor networks

TURKISH JOURNAL OF ELECTRICAL ENGINEERING & COMPUTER SCIENCES ◽

10.3906/elk-1807-49 ◽

2019 ◽

pp. 1120-1131 ◽

Cited By ~ 2

Author(s):

ERDEM KÖSE ◽

ALİ KÖKSAL HOCAOĞLU

Keyword(s):

Sensor Networks ◽

Frequency Domain ◽

Time Domain ◽

False Positive Rate ◽

Spectral Characteristics ◽

Classification Performance ◽

Vehicle Classification ◽

Distributed Sensor Networks ◽

Efficient System ◽

Distributed Sensor

Ground vehicle detection and classification with distributed sensor networks is of growing interest for border security. Different sensing modalities including electro-optical, seismic, and acoustic were evaluated individually and in combination to develop a more efficient system. Despite previous works that mostly studied frequency-domain features and acoustic sensors, in this work we analyzed the classification performance for both frequency and time-domain features and seismic and acoustic modalities. Despite their infrequent use, we show that when fused with frequency-domain features, time-domain features improve the classification performance and reduce the false positive rate, especially for seismic signals. We investigated the performance of seismic sensors and showed that the classification performance varies with the type of road due to the distinct spectral characteristics of the medium. Our proposed classifier fuses time and frequency-domain features and acoustic and seismic modalities to achieve the highest classification rate of 98.6% using a relatively small number of features.

Download Full-text