Search results

The purpose of this paper is to propose distributed learning-based three different metaheuristic algorithms for the identification of nonlinear systems. The proposed algorithms are experimented in this study to address problems for which input data are available at different geographic locations. In addition, the models are tested for nonlinear systems with different noise conditions. In a nutshell, the suggested model aims to handle voluminous data with low communication overhead compared to traditional centralized processing methodologies.

Population-based evolutionary algorithms such as genetic algorithm (GA), particle swarm optimization (PSO) and cat swarm optimization (CSO) are implemented in a distributed form to address the system identification problem having distributed input data. Out of different distributed approaches mentioned in the literature, the study has considered incremental and diffusion strategies.

Performances of the proposed distributed learning-based algorithms are compared for different noise conditions. The experimental results indicate that CSO performs better compared to GA and PSO at all noise strengths with respect to accuracy and error convergence rate, but incremental CSO is slightly superior to diffusion CSO.

This paper employs evolutionary algorithms using distributed learning strategies and applies these algorithms for the identification of unknown systems. Very few existing studies have been reported in which these distributed learning strategies are experimented for the parameter estimation task.

Localization is a fundamental problem in wireless sensor networks. In many applications, sensor location information is critical for data processing and meaning. While the global positioning system (GPS) can be used to determine mote locations with meter precision, the high hardware cost and energy requirements of GPS receivers often prohibit the ubiquitous use of GPS for location estimates. This high cost (in terms of hardware price and energy consumption) of GPS has motivated researchers to develop localization protocols that determine mote locations based on cheap hardware and localization algorithms. The purpose of this paper is to present a comprehensive review of wireless sensor network localization techniques, and provide a detailed overview for several distance‐based localization algorithms.

To provide a detailed summary of wireless sensor network localization algorithms, the authors outline a tiered classification system in which they first classify algorithms as distributed, distributed‐centralized, or centralized. From this broad classification, the paper then further categorizes localization algorithms using their protocol techniques. By utilizing this classification system, the authors are able to provide a survey of several wireless sensor network localization algorithms and summarize relative algorithm performance based on the algorithms' classification.

There are numerous localization algorithms available and the performance of these algorithms is dependent on network configuration, environmental variables, and the ranging method implemented. When selecting a localization algorithm, it is important to understand basic algorithm operation and expected performance. This tier‐based algorithm classification system can be used to gain a high‐level understanding of algorithm performance and energy consumption based on known algorithm characteristics.

Localization is a widely researched field and given the quantity of localization algorithms that currently exist, it is impossible to present a complete review of every published algorithm. Instead, the paper presents a holistic view of the current state of localization research and a detailed review of ten representative distance‐based algorithms that have diverse characteristics and methods. This review presents a new classification structure that may help researchers understand, at a high‐level, the expected performance and energy consumption of algorithms not explicitly addressed by our work.

This paper gives a bibliographical review of the finite element and boundary element parallel processing techniques from the theoretical and application points of view. Topics include: theory – domain decomposition/partitioning, load balancing, parallel solvers/algorithms, parallel mesh generation, adaptive methods, and visualization/graphics; applications – structural mechanics problems, dynamic problems, material/geometrical non‐linear problems, contact problems, fracture mechanics, field problems, coupled problems, sensitivity and optimization, and other problems; hardware and software environments – hardware environments, programming techniques, and software development and presentations. The bibliography at the end of this paper contains 850 references to papers, conference proceedings and theses/dissertations dealing with presented subjects that were published between 1996 and 2002.

Heretofore, it has been extremely expensive to install and use distributed databases. With the advent of Java, JDBC and other Internet technologies, it has become easy and inexpensive to connect multiple databases and form distributed databases, even where the various host computers run on different platforms. These types of databases can be used in many peer‐to‐peer applications which are now receiving much attention from researchers. Although it is easy to form a distributed database via Internet/intranet, effective sharing of information continues to be problematic. We need to pay more attention to the enabling algorithms, as dedicated links between computers are usually not available in peer‐to‐peer systems. The lack of dedicated links can cause poor performance, especially if the databases are connected via Internet. Discusses the problems of distributed database operation with reference to an example. Presents two statistical selection algorithms which are designed to select the jth smallest key from a very large file distributed over many computers. The objective of these algorithms is to minimise the number of communication messages necessary to the selection operation. One algorithm is for the intranet with broadcast/multicast facilities while the other is for Internet without broadcast/multicast facilities.

Presents a review on implementing finite element methods on supercomputers, workstations and PCs and gives main trends in hardware and software developments. An appendix included at the end of the paper presents a bibliography on the subjects retrospectively to 1985 and approximately 1,100 references are listed.

Pervasive computing envisions seamless support for user tasks through cooperating devices that are present in an environment. Fluctuating availability of devices, induced by mobility and failures, requires mechanisms and algorithms that allow applications to adapt to their ever‐changing execution environments without user intervention. To ease the development of adaptive applications, Becker et al. (3) have proposed the peer‐based component system PCOM. This system provides fundamental mechanisms to support the automated composition of applications at runtime. In this article, we discuss the requirements on algorithms that enable automatic configuration of pervasive applications. Furthermore, we show how finding a configuration can be interpreted as Distributed Constraint Satisfaction Problem. Based on this, we present an algorithm that is capable of finding an application configuration in the presence of strictly limited resources. To show the feasibility of this algorithm, we present an evaluation based on simulations and real‐world measurements and we compare the results with a simple greedy approximation.

This paper aims to present a distributed Bayesian approach with connectivity maintenance to manage a multi-agent network search for a target on a two-dimensional plane.

The Bayesian framework is used to compute the local probability density functions (PDFs) of the target and obtain the global PDF with the consensus algorithm. An inverse power iteration algorithm is introduced to estimate the algebraic connectivity λ₂ of the network. Based on the estimated λ₂, the authors design a potential field for the connectivity maintenance. Then, based on the detection probability function, the authors design a potential field for the search target. The authors combine the two potential fields and design a distributed gradient-based control for the agents.

The inverse power iteration algorithm can distributed estimate the algebraic connectivity by the agents. The agents can efficient search the target with connectivity maintenance with the designed distributed gradient-based search algorithm.

Previous study has paid little attention to the multi-agent search problem with connectivity maintenance. Our algorithm guarantees that the strongly connected graph of the multi-agent communication topology is always established while performing the distributed target search problem.

This paper aims to evaluate different approaches for the parallelization of compute-intensive tasks. The study compares a Java multi-threaded algorithm, distributed computing solutions with MapReduce (Apache Hadoop) and resilient distributed data set (RDD) (Apache Spark) paradigms and a graphics processing unit (GPU) approach with Numba for compute unified device architecture (CUDA).

The paper uses a simple but computationally intensive puzzle as a case study for experiments. To find all solutions using brute force search, 15! permutations had to be computed and tested against the solution rules. The experimental application comprises a Java multi-threaded algorithm, distributed computing solutions with MapReduce (Apache Hadoop) and RDD (Apache Spark) paradigms and a GPU approach with Numba for CUDA. The implementations were benchmarked on Amazon-EC2 instances for performance and scalability measurements.

The comparison of the solutions with Apache Hadoop and Apache Spark under Amazon EMR showed that the processing time measured in CPU minutes with Spark was up to 30% lower, while the performance of Spark especially benefits from an increasing number of tasks. With the CUDA implementation, more than 16 times faster execution is achievable for the same price compared to the Spark solution. Apart from the multi-threaded implementation, the processing times of all solutions scale approximately linearly. Finally, several application suggestions for the different parallelization approaches are derived from the insights of this study.

There are numerous studies that have examined the performance of parallelization approaches. Most of these studies deal with processing large amounts of data or mathematical problems. This work, in contrast, compares these technologies on their ability to implement computationally intensive distributed algorithms.

Wireless local area networks (WLANs) are the predominant option for broadband wireless access network, and multiple access points (APs) will be much more available for wireless stations (WSTAs). Call admission control (CAC) on AP selection problem over 802.11 WLAN is a critical issue. In the existing architecture, strongest‐signal‐first is the default AP selection mechanism in 802.11 WLAN which uses the single criterion, received signal strength indicator, to select AP. However, this method suffers from bandwidth deficiency and unbalanced load among APs due to the uneven distribution of user load, thus degrading the system throughput. Instead, the purpose of this paper is to propose a multi‐criteria CAC on AP selection algorithms.

The distributed multi‐criteria considered in order are RSSI, minimum required bandwidth of WSTA, estimated effective bandwidth (EEB) and AP‐WSTA distance. A semiMarkov model considering both packet retransmission limit, packet error rate and collision effect is proposed to predict the system throughput and validated through simulation results. Two multi‐criteria AP selection algorithms after EEB is evaluated are proposed and compared based on this analytical model.

The proposed algorithms outperform the traditional SSF algorithm in terms of the balance index for AP and the average system throughput.

The paper presents performance analysis for multi‐criteria CAC for distributed access point selection in WLANs.

The multi-robot task allocation (MRTA) problem is a challenging issue in the robotics area with plentiful practical applications. Expanding the number of tasks and robots increases the size of the state space significantly and influences the performance of the MRTA. As this process requires high computational time, this paper aims to describe a technique that minimizes the size of the explored state space, by partitioning the tasks into clusters. In this paper, the authors address the problem of MRTA by putting forward a new automatic clustering algorithm of the robots' tasks based on a dynamic-distributed double-guided particle swarm optimization, namely, ACD³GPSO.

This approach is made out of two phases: phase I groups the tasks into clusters using the ACD³GPSO algorithm and phase II allocates the robots to the clusters. Four factors are introduced in ACD³GPSO for better results. First, ACD³GPSO uses the k-means algorithm as a means to improve the initial generation of particles. The second factor is the distribution using the multi-agent approach to reduce the run time. The third one is the diversification introduced by two local optimum detectors LODpBest and LODgBest. The last one is based on the concept of templates and guidance probability Pguid.

Computational experiments were carried out to prove the effectiveness of this approach. It is compared against two state-of-the-art solutions of the MRTA and against two evolutionary methods under five different numerical simulations. The simulation results confirm that the proposed method is highly competitive in terms of the clustering time, clustering cost and MRTA time.

The proposed algorithm is quite useful for real-world applications, especially the scenarios involving a high number of robots and tasks.

In this methodology, owing to the ACD³GPSO algorithm, task allocation's run time has diminished. Therefore, the proposed method can be considered as a vital alternative in the field of MRTA with growing numbers of both robots and tasks. In PSO, stagnation and local optima issues are avoided by adding assorted variety to the population, without losing its fast convergence.