Classification of Sonar Data Set Using Neural Network Trained by Gray Wolf Optimization

Mosavi, Mohammad Reza; Khishe, Mohammad; Ghamgosar, Abolfazl

doi:10.14311/nnw.2016.26.023

Cited by 70 publications

(41 citation statements)

References 43 publications

(57 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [24], a standard neural network trained using the Grey Wolf algorithm was used for categorizing a sonar dataset. The research stated that the GWO had a tremendous ability for resolving higher dimension issues.…”

Section: The State Of the Art Of The Grey Wolf Optimizermentioning

confidence: 99%

See 1 more Smart Citation

A multi hidden recurrent neural network with a modified grey wolf optimizer

2019

View full text Add to dashboard Cite

Identifying university students’ weaknesses results in better learning and can function as an early warning system to enable students to improve. However, the satisfaction level of existing systems is not promising. New and dynamic hybrid systems are needed to imitate this mechanism. A hybrid system (a modified Recurrent Neural Network with an adapted Grey Wolf Optimizer) is used to forecast students’ outcomes. This proposed system would improve instruction by the faculty and enhance the students’ learning experiences. The results show that a modified recurrent neural network with an adapted Grey Wolf Optimizer has the best accuracy when compared with other models.

show abstract

Section: The State Of the Art Of The Grey Wolf Optimizermentioning

confidence: 99%

“…O k (t) = f ∑ w gk h g 2 (t) H2 g (24) where w gk represents the weight connection between the output layer O k (t) and the second hidden layer h g 2 (t).…”

Section: L 1 (T) = H J 1 (T − 1)mentioning

confidence: 99%

A multi hidden recurrent neural network with a modified grey wolf optimizer

2019

View full text Add to dashboard Cite

show abstract

“…Bearing-time and range-time terms hint to Range Doppler, and as all target energy stand in constant range direction then range curvature will not depend upon target bearing position. [7], [10] and [12].…”

Section:  Chirp Scalingmentioning

confidence: 99%

“…It can be exerted to all range frequencies which are dependent to bearing frequencies. Adapted filter corrects bulky range cell migration and secondary range compression with removing the second and the forth exponential expressions from the equation (12) which leads to S 3 (f t , f η ) and can be obtained by below equation:…”

Section:  Chirp Scalingmentioning

confidence: 99%

The Implementation of Chirp Scaling Algorithm Via Fast Fourier Transform

2019

International Journal of Research Studies in Computer Science a

View full text Add to dashboard Cite

It is necessary to use bulky aperture to produce high resolution images in optical systems. Since frequency in synthetic aperture radar is lower than that of optical systems then even obtaining mediocre resolution in synthetic aperture radar need a bulky antenna. For doing this, it needs antenna with hundreds meters' length [1]. An aerial radar can collect data while making advance of distance to few hundred meters and process data in such way that it seems to be obtained from a physically bulky antenna. Synthetic aperture is defined as distance through which a flying object goes to make synthetic antenna. As a synthetic aperture radar moves it sends a pulse to each target. Reflecting echoes are taken by the receiver and kept in an echo-accumulating space. Since relating to the ground the radar is moving, reflecting echoes take Doppler displacement [2]. Comparing Doppler displacement frequency to a referent frequency makes it possible that large number of reflecting echo concentrate on one point; as a result, the length of antenna capturing a particular point will be increased effectively [3]. To produce image from raw data four algorithms have been developed so far. Among these algorithms Chirp scaling with excellent power of the focusing on the target (higher resolution), optimized processing [4], and simple implementation has been cared for. In spite of the Range Migration and Range Doppler Algorithms this algorithm can be implemented without the need of inter-acquire easily, it can also be used to point and band capturing. In point capturing Range Migration Algorithm, Range Doppler Algorithm, and Chirp Scaling Algorithm are used but in point capturing Polar Algorithm, Range Doppler Algorithm, and Chirp Scaling Algorithm are used. Fourier Transform is used to both implement and phase-keeping effectively. Another advantage of using Chirp Scaling Algorithm in terms of Secondary Range Compression is to provide range for bilateral processing of adapting filter. Secondary Range Compression is an effective method for joining Range-Bearing, it increasingly can be applied to pass large radar beam, large beam width, and high Squint inclination. The structure of paper is as follow: section II present the chirp scaling algorithm. Section III demonstrate the results and conclusion is represented in section IV.

show abstract

“…There are several other examples of research that are applying the inspiration by brain structures into the field of artificial intelligence, such as grey wolf optimization [4], genetic feature selector [5], associative memory [6] or fuzzy neural network [7]. In many applications, the biological motivation for the paradigm of neural networks is an advantage because the modeled processes are also of biological origin, such as automated analysis of medical or physiological data [8,9] or mental processes modeling [10][11][12].…”

Section: Introductionmentioning

confidence: 99%

Homeostatic Learning Rule for Artificial Neural Networks

Ruzek¹

2018

NNW

View full text Add to dashboard Cite

This article presents an improvement of learning algorithm for an artificial neural network that makes the learning process more similar to a biological neuron, but still simple enough to be easily programmed. This idea is based on autonomous artificial neurons that are working together and at same time competing for resources; every neuron is trying to be better than the others, but also needs the feed back from other neurons. The proposed artificial neuron has similar forward signal processing as the standard perceptron; the main difference is the learning phase. The learning process is based on observing the weights of other neurons, but only in biologically plausible way, no back propagation of error or 'teacher' is allowed. The neuron is sending the signal in a forward direction into the higher layer, while the information about its function is being propagated in the opposite direction. This information does not have the form of energy, it is the observation of how the neuron's output is accepted by the others. The neurons are trying to find such setting of their internal parameters that are optimal for the whole network. For this algorithm, it is necessary that the neurons are organized in layers. The tests proved the viability of this concept-the learning process is slower; but has other advantages, such as resistance against catastrophic interference or higher generalization.

show abstract

Classification of Sonar Data Set Using Neural Network Trained by Gray Wolf Optimization

Cited by 70 publications

References 43 publications

A multi hidden recurrent neural network with a modified grey wolf optimizer

A multi hidden recurrent neural network with a modified grey wolf optimizer

The Implementation of Chirp Scaling Algorithm Via Fast Fourier Transform

Homeostatic Learning Rule for Artificial Neural Networks

Contact Info

Product

Resources

About