A nonparametric data mapping technique for active initialization of the multilayer perceptron

This research aims to develop a gamelan music genre classifier based on the musical mode system determined based on the dominant notes in a certain order. Only experts can discriminate the musical mode system of compositions. The Feed Forward Neural Networks method was used to classify gamelan compositions into three musical mode systems. The challenge is to recognize the musical mode system of compositions between the initial melody without having to analyze the entire melody using a small amount of data for the dataset. Instead of conducting a melodic extraction from audio signal data, the text-based skeletal melody data, which is a form of extracted melodic features, are used for the dataset. Unique corpuses are controlled based on the cardinality of the one-to-many relationship, and a data mapping technique based on the bars is used to increase the number of corpuses. The results show that the proposed method is suitable to solve the specified problems, where the accuracy in recognizing the class of unseen compositions between the initial melody achieves at 86.7%.

Download Full-text

Design and Simulation of parallel CDMA System Based on 3D-Hadamard Transform

Diyala Journal of Engineering Sciences ◽

10.24237/djes.2013.06404 ◽

2013 ◽

Vol 6 (4) ◽

pp. 45-69

Author(s):

Ali T. Shaheen

Keyword(s):

Fading Channels ◽

Spectral Efficiency ◽

Hadamard Matrix ◽

Additive White Gaussian Noise ◽

Mapping Technique ◽

Cdma System ◽

Data Mapping ◽

Two Dimensional ◽

And Performance ◽

Code Division Multiple

Future wireless systems aim to provide higher transmission data rates, improved spectral efficiency and greater capacity. In this paper a spectral efficient two dimensional (2-D) parallel code division multiple access (CDMA) system is proposed for generating and transmitting (2-D CDMA) symbols through 2-D Inter-Symbol Interference (ISI) channel to increase the transmission speed. The 3D-Hadamard matrix is used to generate the 2-D spreading codes required to spread the two-dimensional data for each user row wise and column wise. The quadrature amplitude modulation (QAM) is used as a data mapping technique due to the increased spectral efficiency offered. The new structure simulated using MATLAB and a comparison of performance for serial one-dimensional (1-D) CDMA and parallel (2-D) CDMA is made under Additive White Gaussian Noise (AWGN), flat fading and multi-path selective fading channels conditions. It is found that 2-D CDMA has better speed and performance than serial 1-D CDMA.

Download Full-text

Landslide susceptibility mapping of Phewa catchment using multilayer perceptron artificial neural network

Nepal Journal of Environmental Science ◽

10.3126/njes.v4i0.22718 ◽

2016 ◽

Vol 4 ◽

pp. 1-9 ◽

Cited By ~ 1

Author(s):

Ananta Man Singh Pradhan ◽

Yun-Tae Kim

Keyword(s):

Neural Network ◽

Artificial Neural Network ◽

Multilayer Perceptron ◽

Landslide Susceptibility ◽

Susceptibility Mapping ◽

Geographical Information ◽

Landslide Susceptibility Mapping ◽

Mapping Technique ◽

Inventory Map ◽

Artificial Neural

The aim of this study was to prepare landslide susceptibility mapping technique using multilayer perceptron artificial neural network (MLP-ANN) and then to apply this method to Phewa catchment in western Nepal. To determine the effect of causative factors on landslides, data layers of aspect, elevation, slope, internal relief, slope shape, drainage proximity, drainage density, stream power index, topographic wetness index, sediment transport index, land cover and geology were analysed in R-statistical package and final map was produced using geographical information system environment. A GIS-based landslide inventory map of 88 landslide locations was prepared using data from previous reports and satellite image interpretation. A MLP-ANN model was generated from a training set consisting of ~70% randomly selected landslide in the inventory map, with the remaining ~30% landslides used for validation of the susceptibility map. According to analysis, the model had a success rate of 82.1% and the prediction accuracy of 91.4%, indicating a good performance.

Download Full-text

Mapping Nonsquare and Unevenly Spaced 2-D SLDV Data of an Aircraft Fuselage by Using Spatial DFT-IDFT Techniques

Shock and Vibration ◽

10.1155/1996/195675 ◽

1996 ◽

Vol 3 (2) ◽

pp. 135-140 ◽

Cited By ~ 2

Author(s):

William Xinzuo Li ◽

Larry D. Mitchell ◽

Min-Fu Lu ◽

Michael L. Neumann

Keyword(s):

Laser Scanning ◽

Mapping Technique ◽

Data Mapping ◽

Point Distribution ◽

Aircraft Fuselage ◽

Sense Data ◽

Data Points ◽

Rectangular Grids ◽

Data Point ◽

Minimum Distortion

The scanning laser Doppler vibrometry (SLDV) technique provides velocities of a structure at 2-dimensional (2-D) angularly evenly spaced (in the laser scanning sense) data points. This causes an unevenly spaced data point distribution on the surface of the test structure. In many cases evenly spaced data point distribution with square or rectangular grids is highly desirable. In this study the SLDV velocity data of a partial surface area of an aircraft fuselage were mapped to truly spatial evenly spaced coordinates by using the spatial DFT-IDFT technique with minimum distortion. This 2-D data mapping technique certainly is not limited to the fuselage, hut can he very useful for many other 3-D structures.

Download Full-text