z-logo
Premium
Special issue on 2017 International Conference of Intelligence Computation and Evolutionary Computation (ICEC2017)
Author(s) -
Du Zhenyu
Publication year - 2018
Publication title -
concurrency and computation: practice and experience
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.309
H-Index - 67
eISSN - 1532-0634
pISSN - 1532-0626
DOI - 10.1002/cpe.4672
Subject(s) - computer science , cloud computing , computation , curse of dimensionality , big data , dimension (graph theory) , evolutionary computation , adaptation (eye) , process (computing) , computational intelligence , artificial intelligence , data mining , algorithm , mathematics , physics , pure mathematics , optics , operating system
This editorial describes a special issue of papers from the 2017 International Conference of Intelligence Computation and Evolutionary Computation.1 There are 12 papers in this special issue. Cloud Parallel Spatial-Temporal Data Model with Intelligent Parameter Adaptation for Spatial-Temporal Big Data proposes a cloud parallel spatial-temporal data model with intelligent parameter adaptation for spatial-temporal big data that is able to divide a spatial-temporal problem into a lot of subdivided spatial-temporal problems and to map the subdivided problems onto different cloud parallel computing nodes to process. This paper includes the concept, division methods, and mathematical formulas of cloud parallel spatial-temporal data model, and provides the method to intelligently find the best parameter of cloud parallel spatial-temporal data model for solving the problem with highest parallel speed-up or highest parallel efficiency in cloud parallel computing environment.2 Self-Balancing Robot Bionic Intelligence Multidimensional Decision-Making Evaluation Algorithm proposes a new bionic intelligent algorithm, which adopts an overall updating evaluation strategy for the solution. For solving the multidimensional function optimization problem, due to the mutual interference between the various dimensions, the convergence rate and the quality of the solution will be deteriorated by using the overall update evaluation strategy. In order to compensate for this defect, a multidimensional decision-making algorithm for self-balancing robot based on bionic intelligence is proposed. In the iterative process of the improved algorithm, updating evaluation strategy by the dimensionality is adopted for the solution. It combines the updated value of each dimension with the value of other dimensions to form a new solution and accept the update value using greedy way; this can improve the solution quality. The results show that the improved strategy can effectively improve the convergence rate of the SRMDE algorithm, and the quality of the solution is improved. Compared with the relative improved self-balancing robot multidimensional decision-making algorithm and other evolutionary algorithms, the improved algorithm is competitive in solving the continuous function optimization problem.3 Adaptive Predistortion Algorithm Based on the Multilinear Multiplier Fusion proposes an adaptive predistortion method for dual loop learning structure based on the fusion of multilinear multiplier model recognition to solve the problem of the insufficiency in the predistortion structure and low accuracy of the memory nonlinear power amplifier in the wireless communication system. On the basis of the real number delay multilinear multiplier model, an improved quantum behaved optimization (QBO) algorithm is adopted in this method to carry out indirect learning structure off-line training of the multilinear multiplier, so as to determine the model parameters as an initial value of the predistorter. Then, the least square method (LSM) algorithm is used to carry out direct learning structure on-line micro adjustment of the parameters of the predistorter and to conduct fitting on the nonlinearity and memory effect of the power amplifier. The method has the advantages of simple structure, fast convergence, and high precision, which has avoided the local optimum. The experimental results show that the adjacent channel power in this scheme is improved by approximately 7 dB compared with that of the classical dual loop structure predistortion method, and the linearization performance of the amplifier has been significantly improved; thus, its feasibility has been verified.4 A 3D Face Registration Algorithm Based on Conformal Mapping proposes a novel conformal mapping algorithm to deal with the 3D face registration. Besides, the calculation is about harmonic energy makes the method applicable to low-quality meshes. They begin with a harmonic mapping, and then minimize the harmonic energy by a specific boundary condition on surfaces and to obtain the conformal mapping; finally, they use a landmark-constrained surface registration algorithm to register faces. Numerical experiments on various surfaces demonstrate the efficiency and robustness of the method.5 Opto-electric Target Tracking Algorithm Based on Local Feature Selection and Particle Filter Optimization studies the binocular stereo vision camera for opto-electric target tracking. The RGB and LBP features of the electronic images are used. It calculates strategy through center-surround difference, and further obtained the small scale map of electronic image as a significant area. It is fused in the particle filter tracking algorithm, and the mean shift algorithm is introduced in the stage prediction stage. Based on this, it proposes target tracking algorithm based on local weighted combining with particle filter optimization (TFFLE). In the process of the experiment, it takes qualitative analysis, selects the Bhattacharyya similarity and tracking error as the performance indexes, and tests it to demonstrate that the algorithm is effective.6 Multi-objective Optimization Algorithm of Adaptive Entropy Wireless Network Based on Embedded DSP proposes a new algorithm based on adaptive entropy and multi-objective optimization is proposed. In adaptive entropy, the objective function and constraint condition of the maximum entropy and entropy increasing principle are used to select the objective, which satisfies the maximization of the system, and the optimal scheduling is carried out, and then the multi-objective optimization is carried out on the selected target to migrate, release the target of heavy optimization and long-term unavailability, which can reduce the energy consumption, realize the optimization balance, and improve the system utilization rate. The simulation

This content is not available in your region!

Continue researching here.

Having issues? You can contact us here