This dataset is proposed for human activity recognition tasks. The static activities including sitting, standing, and laying, as well as walking, running, cycling, and walking upstairs/downstairs. Each activity lasts for 2 minutes, 23 subjects were involved in the experiments.


Twenty-three subjects were involved in 8 activities, including 3 static ones and 5 periodic activities of walking, running, cycling, and walking upstairs/downstairs. Each activity lasts around 2 minutes.

For each column, the sensor signals are organized as " 'wri_Acc_X', 'wri_Acc_Y', 'wri_Acc_Z', 'wri_Gyr_X', 'wri_Gyr_Y', 'wri_Gyr_Z', 'wri_Mag_X', 'wri_Mag_Y', 'wri_Mag_Z', 'ank_Acc_X', 'ank_Acc_Y', 'ank_Acc_Z', 'ank_Gyr_X', 'ank_Gyr_Y', 'ank_Gyr_Z', 'ank_Mag_X', 'ank_Mag_Y','ank_Mag_Z', 'bac_Acc_X', 'bac_Acc_Y', 'bac_Acc_Z', 'bac_Gyr_X', 'bac_Gyr_Y', 'bac_Gyr_Z', 'bac_Mag_X', 'bac_Mag_Y', 'bac_Mag_Z' ".


A qualitative and quantitative extension of the chaotic models used to generate self-similar traffic with long-range dependence (LRD) is presented by means of the formulation of a model that considers the use of piecewise affine onedimensional maps. Based on the disaggregation of the temporal series generated, a valid explanation of the behavior of the values of Hurst exponent is proposed and the feasibility of their control from the parameters of the proposed model is shown.


fGn series used for simulations in the article "Sobre la Generación de Tráfico Autosimilar con Dependencia de Largo Alcance Empleando Mapas Caóticos Unidimensionales Afines por Tramos (Versión Extendida)", "On the Generation of Self-similar with Long-range Dependent Traffic Using Piecewise Affine Chaotic One-dimensional Maps (Extended Version)". Available at:

They should be used in MATLAB R2009a.


This Matlab model and the included results are submitted as reference for the paper ''. 

Presenting a comparative study of the Sequential Unscented Kalman Filter (SUKF), Least-squares (LS) Multilateration and standard Unscented Kalman Filter (UKF) for localisation that relies on sequentially received datasets. 

The KEWLS and KKF approach presents a novel solution using Linear Kalman Filters (LKF) to extrapolate individual sensor measurements to a synchronous point in time for use in LS Multilateration. 



This article explores the required amount of time series points from a high-speed computer network to accurately estimate the Hurst exponent. The methodology consists in designing an experiment using estimators that are applied to time series addresses resulting from the capture of high-speed network traffic, followed by addressing the minimum amount of point required to obtain in accurate estimates of the Hurst exponent.


fGn series used for simulations in the article "Preliminaries on the Accurate Estimation of the Hurst Exponent Using Time Series".  Available at:

They should be used in Selfis01b.


The data set contains electrical and mechanical signals from experiments on three-phase induction motors. The experimental tests were carried out for different mechanical loads on the induction motor axis and different severities of broken bar defects in the motor rotor, including data regarding the rotor without defects. Ten repetitions were performed for each experimental condition.


Experimental Setup:

The experimental workbench consists of a three-phase induction motor coupled with a direct-current machine, which works as a generator simulating the load torque, connected by a shaft containing a rotary torque wrench.

- Induction motor: 1hp, 220V/380V, 3.02A/1.75A, 4 poles, 60 Hz, with a nominal torque of 4.1 Nm and a rated speed of 1715 rpm. The rotor is of the squirrel cage type composed of 34 bars.

- Load torque: is adjusted by varying the field winding voltage of direct current generator. A single-phase voltage variator with a filtered full-bridge rectifier is used for the purpose. An induction motor was tested under 12.5, 25, 37.5, 50, 62.5, 75, 87.5 and 100% of full load.

- Broken rotor bar: to simulate the failure on the three-phase induction motor's rotor, it was necessary to drill the rotor. The rupture rotor bars are generally adjacent to the first rotor bar, 4 rotors have been tested, the first with a break bar, the second with two adjacent broken bars, and so on rotor containing four bars adjacent broken.

Monitoring condition:

All signals were sampled at the same time for 18 seconds for each loading condition and ten repetitions were performed from transient to steady state of the induction motor.

- mechanical signals: five axial accelerometers were used simultaneously, with a sensitivity of 10 mV/mm/s, frequency range from 5 to 2000Hz and stainless steel housing, allowing vibration measurements in both drive end (DE) and non-drive end (NDE) sides of the motor, axially or radially, in the horizontal or vertical directions.

- electrical signals: the currents were measured by alternating current probes, which correspond to precision meters, with a capacity of up to 50ARMS, with an output voltage of 10 mV/A, corresponding to the Yokogawa 96033 model. The voltages were measured directly at the induction terminals using voltage points of the oscilloscope and the manufacturer Yokogawa.

Data Set Overview:

-          Three-phase Voltage

-          Three-phase Current

-          Five Vibration Signals



            The database was acquired in the Laboratory of Intelligent Automation of Processes and Systems and Laboratory of Intelligent Control of Electrical Machines, School of Engineering of São Carlos of the University of São Paulo (USP), Brazil.


The dataset consists of two populations of fetuses: 160 healthy and 102 Late Intra Uterine Growth Restricted (IUGR). Late IUGR is an adverse pathological condition encompassing chronic hypoxia as a consequence of placental insufficiency, resulting in an abnormal rate of fetal growth. In standard clinical practice, Late IUGR diagnosis can only be suspected in the third trimester and ultimately confirmed at birth. This data collection comprises of a set of 31 Fetal Heart Rate (FHR) indices computed at different time scales and domains accompanied by the clinical diagnosis.


The data for healthy and Late IUGR populations are included in a single .xlsx file.

Participants are listed by rows and features by columns. In the following we report an exhaustive list of features contained in the dataset accompanied by their units, time interval employed for the computation, and scientific literature references: 

 Fetal and Maternal Domains

  • Clinical Diagnosis [HEALTHY/LATE IUGR]: binary variable to report the clinical diagnosis of the participant
  • Gestational Age [days]: gestational age at the time of CTG examination
  • Maternal Age [years]: maternal age at the time of CTG examination
  • Sex [Male (1)/Female (2)]: fetal sex


Morphological and Time Domains

  • Mean FHR [bpm] – 1-min epoch: the mean of FHR excluding accelerations and decelerations 
  • Std FHR [bpm] – 1-min epoch: the standard deviation of FHR excluding accelerations and decelerations 
  • DELTA [ms] – 1-min epoch: defined in accordance with [1], [2] excluding accelerations and decelerations 
  • II [] – 1-min epoch: defined in accordance with [1], [2] excluding accelerations and decelerations 
  • STV [ms] – 1-min epoch: defined in accordance with [1], [2] excluding accelerations and decelerations 
  • LTI [ms] – 3-min epoch: defined in accordance with [1], [2] excluding accelerations and decelerations 
  • ACC_L [#] – entire recording: the count of large accelerations defined in accordance with [3], [4] 
  • ACC_S [#] – entire recording: the count of small accelerations defined in accordance with [3], [4] 
  • CONTR [#]– entire recording: the count of contractions defined in accordance with [3], [4] 


Frequency Domain 

  • LF [ms²/Hz] – 3-min epoch: defined in accordance with [2], LF band is defined in the range [0.03 - 0.15] Hz 
  • MF [ms²/Hz] – 3-min epoch: defined in accordance with [2], MF band is defined in the range [0.15 - 0.5] Hz 
  • HF [ms²/Hz] – 3-min epoch: defined in accordance with [2], HF band is defined in the range HF [0.5 - 1 Hz] 


Complexity Domain 

  • ApEn [bits] – 3-min epoch: defined in accordance with [5], m = 1, r = 0.1*standard deviation of the considered epoch 
  • SampEn [bits] – 3-min epoch: defined in accordance with [6], m = 1, r = 0.1*standard deviation of the considered epoch 
  • LCZ_BIN_0 [bits] – 3-min epoch: defined in accordance with [7], binary coding and p = 0 
  • LCZ_TER_0 [bits] – 3-min epoch: defined in accordance with [7], tertiary coding and p = 0 
  • AC/DC/DR [bpm] – entire recording: defined in accordance with [8]–[10], considering different combinations of parameters T and s, L is constant and equal 100 samples; e.g, AC_T1_s2 is defined as the acceleration capacity computed setting the parameters T = 1 and s = 2



[1]       D. Arduini, G. Rizzo, A. Piana, P. Bonalumi, P. Brambilla, and C. Romanini, “Computerized analysis of fetal heart rate—Part I: description of the sys- tem (2CTG),” J Matern Fetal Invest, vol. 3, pp. 159–164, 1993.

[2]       M. G. Signorini, G. Magenes, S. Cerutti, and D. Arduini, “Linear and nonlinear parameters for the analysis of fetal heart rate signal from cardiotocographic recordings,” IEEE Trans. Biomed. Eng., vol. 50, no. 3, pp. 365–374, 2003.

[3]       FIGO, “Guidelines for the Use of Fetal Monitoring,” Int. J. Gynecol. Obstet., vol. 25, pp. 159–167, 1986.

[4]       R. Rabinowitz, E. Persitz, and E. Sadovsky, “The relation between fetal heart rate accelerations and fetal movements.,” Obstet. Gynecol., vol. 61, no. 1, pp. 16–18, 1983.

[5]       S. M. Pincus and R. R. Viscarello, “Approximate entropy: a regularity measure for fetal heart rate analysis.,” Obstet. Gynecol., vol. 79, no. 2, pp. 249–55, 1992.

[6]       D. E. Lake, J. S. Richman, M. P. Griffin, and J. R. Moorman, “Sample entropy analysis of neonatal heart rate variability,” Am. J. Physiol. - Regul. Integr. Comp. Physiol., vol. 283, no. 3, pp. R789–R797, 2002.

[7]       A. Lempel and J. Ziv, “On the complexity of finite sequences,” IEEE Trans. Inf. Theory, vol. 22, no. 1, pp. 75–81, 1976.

[8]       A. Bauer et al., “Phase-rectified signal averaging detects quasi-periodicities in non-stationary data,” Phys. A Stat. Mech. its Appl., vol. 364, pp. 423–434, 2006.

[9]       A. Fanelli, G. Magenes, M. Campanile, and M. G. Signorini, “Quantitative assessment of fetal well-being through ctg recordings: A new parameter based on phase-rectified signal average,” IEEE J. Biomed. Heal. Informatics, vol. 17, no. 5, pp. 959–966, 2013.

[10]     M. W. Rivolta, T. Stampalija, M. G. Frasch, and R. Sassi, “Theoretical Value of Deceleration Capacity Points to Deceleration Reserve of Fetal Heart Rate,” IEEE Trans. Biomed. Eng., pp. 1–10, 2019.


This dataset aims at providing a toy demo for the comparison between the doubly orthogonal matching pursuit (DOMP) and its predecessor, the OMP.


File descriptions:

- toydemo.m: Main file. It provides a comparison of the algorithms by loading the measurement  and performing a behavioral model of the PA.

- demo_meas.mat: example of the input and output measurements of a power amplifier (PA) working under a 15-MHz LTE signal sampled at a sampling rate of 92.16MHz.

- model_gmp_domp_omp.m: generates the model structure and calls the pruning techniques.

- omp_domp.m: executes the OMP and DOMP techniques.



Reinforcement Learning (RL) agents can learn to control a nonlinear system without using a model of the system. However, having a model brings benefits, mainly in terms of a reduced number of unsuccessful trials before achieving acceptable control performance. Several modelling approaches have been used in the RL domain, such as neural networks, local linear regression, or Gaussian processes. In this article, we focus on a technique that has not been used much so far:\ symbolic regression, based on genetic programming.


The proposed signals are used  for electromagnetic-based stroke classification.  Six realistic head phantom computed from MRI scans, is surrounded by an antenna array of 16 dipole antennas distributed uniformly around the head. These antennas are deployed in a fixed circular array around the head, at a distance of approximately 2-3 mm from the head. A Gaussian pulse covering the bandwidth from 0:7 to 2 GHz is emitted from each of the antennas, sequentially, while all of the antennas capture the scattered signals. Since 16 antennas were used, there are a total of 256 channel signals (i.e.