Evolution Performance of Symbolic Radial Basis Function Neural Network by Using Evolutionary Algorithms

Radial Basis Function Neural Network(RBFNN)ensembles have long suffered from non-efficient training,where incorrect parameter settings can be computationally disastrous.This paper examines different evolutionary algorithms for training the Symbolic Radial Basis Function Neural Network(SRBFNN)through the behavior’s integration of satisfiability programming.Inspired by evolutionary algorithms,which can iteratively find the nearoptimal solution,different Evolutionary Algorithms(EAs)were designed to optimize the producer output weight of the SRBFNN that corresponds to the embedded logic programming 2Satisfiability representation(SRBFNN-2SAT).The SRBFNN’s objective function that corresponds to Satisfiability logic programming can be minimized by different algorithms,including Genetic Algorithm(GA),Evolution Strategy Algorithm(ES),Differential Evolution Algorithm(DE),and Evolutionary Programming Algorithm(EP).Each of these methods is presented in the steps in the flowchart form which can be used for its straightforward implementation in any programming language.With the use of SRBFNN-2SAT,a training method based on these algorithms has been presented,then training has been compared among algorithms,which were applied in Microsoft Visual C++software using multiple metrics of performance,including Mean Absolute Relative Error(MARE),Root Mean Square Error(RMSE),Mean Absolute Percentage Error(MAPE),Mean Bias Error(MBE),Systematic Error(SD),Schwarz Bayesian Criterion(SBC),and Central Process Unit time(CPU time).Based on the results,the EP algorithm achieved a higher training rate and simple structure compared with the rest of the algorithms.It has been confirmed that the EP algorithm is quite effective in training and obtaining the best output weight,accompanied by the slightest iteration error,which minimizes the objective function of SRBFNN-2SAT.

NEW PERIODIC SOLUTIONS OF ITO'S 5th-ORDER mKdV EQUATION AND ITO'S 7th-ORDER mKdV EQUATION

Based on the modified Jocobi elliptic function expansion method and the modified extended tanh function method,a new algebraic method is presented to obtain mu ltiple travelling wave solutions for nonlinear wave equations.By using the metho d,Ito's 5th order and 7th order mKdV equations are studied in detail and more new exact Jocobi elliptic function periodic solutions are found.With modulus m→1 or m→0,these solutions degenerate into corresponding solitary wave s olutions,shock wave solutions and trigonometric function solutions.

Performance of M-QAM, M-DPSK and M-PSK with MRC diversity in a Nakagami-m fading channel

The nature of a wireless communication channel is very unpredictable. To design a good communication link, it is required to know the statistical model of the channel accurately. The average symbol error probability(ASER) was analyzed for different modulation schemes. A unified analytical framework was presented to obtain closed-form solutions for calculating the ASER of M-ary differential phase-shift keying(M-DPSK), coherent M-ary phase-shift keying(M-PSK), and quadrature amplitude modulation(QAM) over single or multiple Nakagami-m fading channels. Moreover, the ASER was estimated and evaluated by using the maximal ratio-combining(MRC) diversity technique. Simulation results show that an error rate of the fading channel typically depends on Nakagami parameters(m), space diversity(N), and symbol rate(M). A comparison between M-PSK, M-DPSK, and M-QAM modulation schemes was shown, and the results prove that M-ary QAM(M-QAM) demonstrates better performance compared to M-DPSK and M-PSK under all fading and non-fading conditions.

Accuracy of Multivariable Refinable Functions and Two-Scale Similarity Transform

In this paper,we prove that if two multiresolutions satisfy a relation,then they have the same accuracy in the multivariate FSI case.

A CHARACTERIZATION OF N-DIMENSIONAL DAUBECHIES TYPE TENSOR PRODUCT WAVELET

In this paper, we consider the problem of the existence of general non-separable variate orthonormal compactly supported wavelet basis when the symbol function has a special form. We prove that the general non-separable variate orthonormal wavelet basis doesn't exist if the symbol function possesses a certain form. This helps us to explicate the difficulty of constructing the non-separable variate wavlet basis and to hint how to construct non-separable variate wavlet basis.

The Symbolic Extension Theory in Topological Dynamics

In this survey we will present the symbolic extension theory in topological dynamics,which was built over the past twenty years.