Yuanyuan Fan

Self-adapting control parameters with multi-parent crossover in differential evolution algorithm

The performance of differential evolution DE algorithm is influenced by the setting of control parameters, which is quite dependent on the problem and difficult to be determined. Therefore, the studies on parameter adaptation mechanisms have gradually become more popular. In this paper, we present a self-adaptive DE algorithm GaDE, in which the adaptation of amplification factor and crossover rate is executed with a multi-parent crossover, while the adaptation timing is decided by the comparative result between the target vector and its offspring. The performance of GaDE algorithm is evaluated on a suite of bound-constrained numerical optimisation problems. The results show that our algorithm is better than, or at least comparable to, the canonical DE, and the two other adaptive DE algorithms.

MR-COF: A Genetic MapReduce Configuration Optimization Framework

Hadoop/MapReduce has emerged as a de facto programming framework to explore cloud-computing resources. Hadoop has many configuration parameters, some of which are crucial to the performance of MapReduce jobs. In practice, these parameters are usually set to default or inappropriate values. This severely limits system performance (e.g., execution time). Therefore, it is essential but also challenging to investigate how to automatically tune these parameters to optimize MapReduce job performance. In this paper, we propose an automatic MapReduce configuration optimization framework named as MR-COF. By monitoring and analyzing the runtime behavior, the framework adopts a cost-based performance prediction model that predicts the MapReduce job performance. In addition, we design a genetic search algorithm which iteratively tunes parameters in order to find out the best one. Testbed-based experimental results show that the average MapReduce job performance is increased by 35 % with MR-COF compared to the default configuration.

A Software-Defined Intelligent Method for Antenna Design

The personalized and diverse demands of modern communication present new challenges to antenna design. While the emergence of Software-Defined Everything provides an innovative hardware design idea that hardware structure is modeled in a software way and designed with intelligence optimization algorithms. Inspired by the design idea, in this paper we propose a software-defined intelligent method for antenna design. The optimal variables are described with the software-defined antenna structure method, and the optimal target is determined with the software-defined antenna performance method. Based on the above the software-defined antenna model will be abstracted, which convert the work of antenna design to the optimal problem on antenna structure. Meanwhile considering the features of the long simulation and the intensive computation, an intelligent algorithm is proposed suitable for the intelligent design method. This software-defined intelligent method is applied to an antenna design example and several optimal antennas satisfying the requirements have been obtained. It demonstrates that automation and intelligentization in antenna design can be achieved with this innovative method.

SAPSN: A Sensor Network for Signal Acquisition and Processing

Software defined wireless sensor network can be adapted to different application needs through dynamic programming. In this paper, we propose a signal acquisition and processing wireless sensor network (SAPSN). SAPSN consists of sampling nodes, processing nodes and remote controllers. At first, the sampling node completes the local signal sampling by analog-digital conversion. Next, according to the different demand from the remote controller, processing node completes time domain or frequency domain analysis of signal processing, and transfers the results back to the remote controller. Finally, the application in remote controller will display the results according to different users needs. SAPSN is capable of time domain or frequency domain signal analysis and processing, depending on different application requirements. In this paper, we present the concept underlying SAPSN, its architecture. We also present preliminary experimental results.

Multi-label Classification based on Particle Swarm Algorithm

Multi-label classification is a generalization of single-label classification, and its samples belong to multiple labels. The K-nearest neighbor algorithm can solve this problem as an optimization problem. It finds the optimum solution by caculating the distance between each sample in general. But in fact, the distance of K-nearest neighbor algorithm may be miscalculated due to the caused by the redundant or irrelevant characteristic value. In order to solve this problem, in this paper, we propose a novel method that uses the particle swarm algorithm to optimize the feature weights to improve the accuracy of distance calculation. As a result, it can improve classification accuracy further. The experimental results show that applying particle swarm algorithms optimization technique to improving K-nearest neighbor algorithm for multi-label classification problem, can improve the accuracy of classification effectively.

A Complete On-chip Evolvable Hardware Technique Based on Pareto Dominance

To increase the speed of evolvable hardware, a complete on-chip evolvable hardware technique is adopted, where both hardware evaluation and evolutionary algorithm itself are configured on chip. At the same time, a multi-objective evolutionary algorithm based on Pareto dominance is proposed to satisfy and conciliate multiple objectives in many combinational circuits design. This method is applied to the design of a 1-bit full adder and its feasibility is validated by the result of the experiment. The data of result also shows that the speed of evolvable hardware is dramatically increased.

A Study on modeling evolutionary antenna based on ST-5 antenna and NEC2

Plentiful applications of evolutionary algorithm (EA) to antenna design have formed a new hot research topic - evolutionary antenna. A novel antenna has been designed through EA in NASA Ames Research Center, and has been successfully applied in ST-5 mission [1]. It has represented the first evolved hardware in space, and the first deployed evolved antenna [1, 2]. We present a mathematical model and an algorithmic model for the evolutionary antenna, according to the requirements for the ST-5 antenna and the characteristics of NEC2. And also we show the experiment results and make a conclusion.

An Optimal Sink Placement for High Coverage and Low Deployment Cost in Mobile Wireless Sensor Networks

Reliable communication quality and reasonable cost control are two of important goals of Sink node location problem in modern communication. How to achieve an optimization solution by balancing both the two goals is a difficult tradeoff, especially in mobile wireless sensor networks. Therefore, with the maximum signal coverage as the communication quality goal, a constrained multi-objective optimization model is proposed by optimizing cost and communication quality at the same time. In order to deal with the constraints, we improve the classic multi-objective algorithm NSGA_II and adopt the strategy of binary tournament based on CV violation value to select the parent, so that the superior individuals have a greater probability to participate in the genetic operation. Finally, the problem model and algorithm proposed in this paper are applied to a set of Sink node location problems with different cost and network parameters. Compared with the NSGA_II algorithm, experimental results show that the new algorithm can improve the signal coverage better without much increase in cost. At the same time, the variances of the optimization results obtained in new algorithm are smaller, which means its optimization is more stable.

A Fast Vision-Based Localization Algorithm for Spacecraft in Deep Space

Star light navigation can provide the current attitude and position of the spacecraft in deep space. However, the accuracy of stellar-inertial attitude determination is degraded because of star image smearing under high dynamic condition. To solve this problem, two key work, including accuracy star extraction and fast star identification, should be done. In this paper, we bring interpolation algorithm into contiguous area pixel searching for star extraction, and get sub-pixel coordinate information of the star points. In addition, a divisional method is proposed to improve star identification algorithm speed based on Hausdorff distance. The simulation results show that work not only has accuracy identification rate but also has better recognition speed. It was used successfully in the actual projects.

Executing Time and Cost-Aware Task Scheduling in Hybrid Cloud Using a Modified DE Algorithm

Task scheduling is one of the basic problem on cloud computing. In hybrid cloud, tasks scheduling faces new challenges. In order to better deal the multi-objective task scheduling optimization in hybrid clouds, on the basis of the GaDE and Pareto optimum of quick sorting method, we present a multi-objective algorithm, named NSjDE. This algorithm also makes considerations to reduce the frequency of evaluation Comparing with experiment of Min-Min algorithm, GaDE algorithm and NSjDE algorithm, results show that for the single object task scheduling, GaDE and NsjDE algorithms perform better in getting the approximate optimal solution. The optimization speed of multi-objective NSjDE algorithm is faster than the single-objective jDE algorithm, and NSjDE can produce more than one non-dominated solution meeting the requirements, in order to provide more options to the user.