Smart Clustering Research Articles

A Q-learning-based smart clustering routing method in flying Ad Hoc networks

Flying ad hoc networks (FANETs) have particular importance in various military and civilian applications due to their specific features, including frequent topological changes, the movement of drones in a three-dimensional space, and their restricted energy. These features have created challenges for designing cluster-based routing protocols. In this paper, a Q-learning-based smart clustering routing method (QSCR) is suggested in FANETs. In QSCR, each node discovers its neighbors through the periodic exchange of hello messages. The hello time interval is different in each cluster, and cluster leaders determine this interval based on the average speed similarity. Next, an adaptive clustering process is presented for categorizing drones in the clusters. In this step, the cluster leader is selected based on a new parameter called merit value, which includes residual energy, centrality, neighbor degree, speed similarity, and link validity time. Then, a centralized Q-learning model is presented to tune weight coefficients related to merit parameters dynamically. In the last step, the routing process is done using a greedy forwarding technique. Finally, QSCR is run on NS2, and the simulation results of QSCR are compared with those of ICRA, WCA, and DCA. These results show that QSCR carries out the clustering process rapidly but has less cluster stability than ICRA. QSCR gets energy efficiency and improves network lifetime. In the routing process, QSCR has a high packet delivery rate compared to other schemes. Also, the number of isolated clusters created in QSCR is less than other clustering methods. However, the proposed scheme has a higher end-to-end delay than ICRA. Also, this scheme experiences more communication overhead than ICRA slightly.

Robust and fast algorithm design for efficient Wi-Fi fingerprinting based indoor positioning systems

Indoor positioning systems (IPS) based on Wi-Fi fingerprinting have gained significant attention due to their potential for providing location-based services. Large scale IPS deployments require implementation of robust, accurate, and fast algorithms. Data analytics assisted algorithms provide positioning accuracy improvements, however their integration within real-time and scalable solutions significantly depends on the computational complexity. Therefore, we propose robust and computationally efficient algorithms for performance enhancements in Wi-Fi fingerprinting-based IPS. A robust radio map algorithm based on enhanced statistical cluster initialization was designed for efficient indoor environment characterization. The proposed data filtering algorithm leveraged smart clustering to mitigate real-time data variations. The designed area classification algorithm was based on smart dual-band data aggregation. We evaluated the performance of proposed algorithms based on accuracy and computation time. The performance evaluations signified accuracy and computational enhancement, in comparison to related benchmark techniques. The data filtering and area classification algorithms required 40% less computation time. Simultaneously, 14.5% and 36% accuracy improvements were recorded for the area classification and radio map algorithms respectively. The proposed algorithms have the potential to significantly enhance IPS performance in a variety of real-time applications, including indoor navigation and asset tracking.

ІТ-кластери як інструмент забезпечення смарт-спеціалізації регіонів України

Smart specialization strategies are actively implemented as part of a large-scale action plan for the development of the EU economy. Ukraine also comprehensively implements approaches to smart specialization in the development and implementation of regional development strategies. At the same time, IT clusters are developing both in the EU and in Ukraine. But the vast majority of studies do not pay attention to the specific characteristics of IT clusters in the context of the formation and implementation of smart specialization strategies. Therefore, the purpose of the study is to determine the role of IT clusters as a tool for ensuring smart specialization strategies in Ukraine. The concepts of smart specialization and regional clusters are similar in their emphasis on (1) productivity and innovation as key factors of competitiveness; (2) regional integration of local advantages. The differences between the concepts lie in the final result. For smart specialization, the search and dissemination of knowledge for the emergence of new opportunities on the market are important. At the same time, the increase in productivity of cluster members and the development of related industries are important for regional clusters. Regional clusters can be considered as an element of the smart specialization strategy. We can talk about smart clustering in the process of smart specialization. A comparative analysis of the functioning of foreign and domestic IT clusters showed that the action of foreign clusters is aimed at cooperation with local/regional authorities, educational institutions and business. The clusters presented in the work provide support to enterprises (including industrial ones) in the transition to sustainable digital technologies. The majority of domestic IT clusters are formed in regions with limited industrial potential and are involved in educational and social projects, projects on the interaction of IT clusters with the local authorities of the regions. Based on the nature and modern role of IT clusters, they can be considered as a resource and accelerator for all areas of smart specialization. Ukraine has two main ways of using the potential of domestic IT clusters to ensure smart specialization. The first way involves the maximally accelerated integration into the S3 Platform of all regional initiatives regarding smart specialization. The second way involves the creation of the national S3 Platform (for Ukraine) with the aim of consolidating all initiatives within the framework of regional smart specialization strategies, their coordination at the national level with further integration into the pan-European S3 platform. There is also a third intermediate path, which involves the simultaneous integration of individual initiatives into the European S3 platform, and the creation of its own nationwide S3 Platform (for Ukraine).

Robust integrated optimization of well placement and control under field production constraints

Field development and control optimization aim to maximize the economic profit of oil and gas production while respecting various constraints such as production limits imposed by the available fluid processing capacity and/or reservoir management strategies. The limitations of the existing optimization workflows are 1) well locations or production/injection controls are optimized independently despite the fact that one affects another, and 2) forthcoming field production limits are ignored during at least one of these optimization stages. This paper presents a robust, multi-level framework for field development and control optimization under fluid processing capacity constraints while considering reservoir description uncertainty. The developed framework is based on sequential iterative optimization of control variables at different levels, where the loop of well placement followed by control optimization continues until no significant improvement is observed in the expected objective value. Simultaneous perturbation stochastic approximation (SPSA) algorithm is employed as the optimizer at all optimization levels. Field production constraints are imposed on the reservoir model using a simplified production network. Smart clustering techniques are applied to systematically select an ensemble of reservoir model realizations as the representative of all available realizations at each optimization level. The developed framework is tested on the Brugge benchmark field case study with a maximum field liquid production limit imposed via the production network. A comparative analysis is performed for each case to investigate the impact of field liquid production constraints on optimal well placement and control strategy. Results demonstrate that ignoring fluid processing capacity constraints, in one or multiple levels of optimization, results in a sub-optimal scenario, highlighting the significance of the proposed optimization framework in robust field development and management. • A robust, integrated framework is presented for field development and control under field production constraints while considering reservoir uncertainty. • Clustering techniques are developed to systematically select representative reservoir model realizations at each optimization level. • Ignoring field production constraints in one or multiple levels of optimization results in sub-optimal scenarios, showing the significance of developed framework in robust field development and control.

Diagnostics of Parkinson’s disease by positron-emission tomography with 18F-DOPA: new approach based on smart clustering

PET investigation of human brain for the purposes of the effective diagnostics of some neurodegenerative diseases requires the definition of the normal indices. Here we present the figures to be used as the reference values of the necessary indices to be used in medical investigations.

Smart-Cluster-Based Distributed Caching for Fog-IoT Networks

The idea of co-operative caching in a cache-enabled wireless network has gained much interest due to its services in terms of short service delay and improved transmission rate at the user end. In this article, we consider a co-operative caching mechanism for a fog-enabled Internet of Things (IoT) network. We propose a delay-minimizing policy for fog nodes (FNs), where the goal is to reduce the service delay for the IoT nodes, also known as terminal nodes (TNs). To this end, a novel smart clustering mechanism is proposed, aiming to efficiently assign FNs to the TNs while improving the network benefit by finding a tradeoff between the delay and the network’s energy consumption. We perform mathematical analysis and extensive simulations to highlight the potential gain and the proposed policy.

Deep neural network-based clustering technique for secure IIoT

The advent of Industrial Internet of Things (IIoT) has determined the proliferation of smart devices connected to the Internet and injected a vast amount of data into it, which may undergo many computational stages at several clusters. On the one hand, the benefits brought by these technologies are well known; however, in the envisaged scenario, the exposure of data, services and infrastructures to malicious attacks has definitely grown. Even a single breach on any of the links of the data–service–infrastructure chain may seriously compromise the security of the end-user application. Therefore, the logical and smart clustering while satisfying security and reliability is a key issue for IIoT networks. A novel clustering method proposed based on power demand assures security of data information in IIoT-based applications. First, security capacity of the system is calculated from mutual information of primary channel and eavesdropping channel. Then, under the maximum transmit power constraint, an optimal transmit power is found based on deep learning technique, which maximizes security capacity of the system. Finally, the network is clustered according to the calculated power demand. Experimental results accredit the proposed method has higher security and reliability, as well as lower network time overhead and power consumption.

Re-centring peripheries along ring roads under a smart growth agenda: case studies of Espoo Innovation Garden and Vantaa Aviapolis in Finland

ABSTRACTThe study expands on the recent cluster-oriented smart growth vision in Greater Helsinki, Finland, and examines with two case studies the pertinent locational factor and locality-specific issues. The contemporary regional development tendencies of smart clustering have reframed place-making schemes with an aim of placing large concentrations of business activities on the periphery. The case studies of Espoo Innovation Garden and Vantaa Aviapolis draw on descriptive before-and-after analysis over a defined timeframe. The study addresses the formation and transformation of the spatial structure and its relative and relational qualities on an aggregated scale. It demonstrates the potential, challenges, tools, and strategies of the operating smart growth agenda generating spatial-economic competitiveness and attractiveness factors in these particular projects. The findings explain how the locational factor along ring roads parallel with the land-use management and former model of spatial and economic concentration and centralization are hampering the restructuring of new urban centralities.

Smart Clustering LEACH (SC-LEACH) Protocol for Improved Network Survivability in Wireless Sensor Networks

Wireless Sensor Networks (WSN) are becoming increasingly popular by the virtue of its diversified applications. Nevertheless WSN faces various research challenges despite its ease of implementation. One of the major issues of concern is the energy efficiency which is closely related to the lifetime of the network. Low-Energy Adaptive Clustering Hierarchy (LEACH) is proven to be an efficient way to route packets in WSNs. However the performance of the LEACH can be further enhanced by lessening the frequency of cluster reformation. In this paper we propose a novel Smart Clustering LEACH (SC-LEACH) protocol which is an advancement over the traditional LEACH protocol. The results of the proposed algorithm were validated through simulations.

Prototype of Power Optimization Based on Converter Topologies Reconfiguration Using PV String Smart Clustering Method for Static Miniature Photovoltaic Farm Under Partially Shaded Condition

Photovoltaic is a technology that converts sunlight radiation into electrical energy and has a great chance to develop in tropical countries such as Indonesia. The effect of the partially shaded condition reduces the energy generated and the efficiency of the photovoltaic farm. This paper proposes a new method for power optimization by reconfiguring the converter topologies using the Smart Clustering Method. The reconfiguration of converter topologies is based on the voltages magnitude of each string on the PV Farm clustered using the smart clustering method. Each converter has MPPT algorithm, so the total power generated by the PV farm is equal to the sum of power generated by each converter topology. The modified P&O Algorithm is used in the uniform clustering, and the firefly Algorithm is used in the non-uniform clustering. For the validation of the proposed method, this paper compared the performance of the proposed method with the performance of MPPT using single converter topologies as a conventional method.

A Smart and Balanced Energy-Efficient Multihop Clustering Algorithm (Smart-BEEM) for MIMO IoT Systems in Future Networks.

Wireless Sensor Networks (WSNs) are typically composed of thousands of sensors powered by limited energy resources. Clustering techniques were introduced to prolong network longevity offering the promise of green computing. However, most existing work fails to consider the network coverage when evaluating the lifetime of a network. We believe that balancing the energy consumption in per unit area rather than on each single sensor can provide better-balanced power usage throughout the network. Our former work—Balanced Energy-Efficiency (BEE) and its Multihop version BEEM can not only extend the network longevity, but also maintain the network coverage. Following WSNs, Internet of Things (IoT) technology has been proposed with higher degree of diversities in terms of communication abilities and user scenarios, supporting a large range of real world applications. The IoT devices are embedded with multiple communication interfaces, normally referred as Multiple-In and Multiple-Out (MIMO) in 5G networks. The applications running on those devices can generate various types of data. Every interface has its own characteristics, which may be preferred and beneficial in some specific user scenarios. With MIMO becoming more available on the IoT devices, an advanced clustering solution for highly dynamic IoT systems is missing and also pressingly demanded in order to cater for differing user applications. In this paper, we present a smart clustering algorithm (Smart-BEEM) based on our former work BEE(M) to accomplish energy efficient and Quality of user Experience (QoE) supported communication in cluster based IoT networks. It is a user behaviour and context aware approach, aiming to facilitate IoT devices to choose beneficial communication interfaces and cluster headers for data transmission. Experimental results have proved that Smart-BEEM can further improve the performance of BEE and BEEM for coverage sensitive longevity.

Design clustering of offshore wind turbines using probabilistic fatigue load estimation

In large offshore wind farms fatigue loads on support structures can vary significantly due to differences and uncertainties in site conditions, making it necessary to optimize design clustering. An efficient probabilistic fatigue load estimation method for monopile foundations was implemented using Monte-Carlo simulations. Verification of frequency domain analysis for wave loads and scaling approaches for wind loads with time domain aero-elastic simulations lead to 95% accuracy on equivalent bending moments at mudline and tower bottom. The computational speed is in the order of 100 times faster than typical time domain tools. The model is applied to calculate location specific fatigue loads that can be used in deterministic and probabilistic design clustering. Results for an example wind farm with 150 turbines in 30–40 m water depth show a maximum load difference of 25%. Smart clustering using discrete optimization algorithms leads to a design load reduction of up to 13% compared to designs based on only the highest loaded turbine position. The proposed tool improves industry-standard clustering and provides a basis for design optimization and uncertainty analysis in large wind farms.

A Smart Clustering Based Approach To Dynamic Bandwidth Allocation In Wireless Networks

A Smart Clustering Based Approach To Dynamic Bandwidth Allocation In Wireless Networks

The Effect of Fault Tolerance in a WSN Cluster on the QoS

The principle issue with existing convention lies in the arbitrary choice of cluster heads. There exists a likelihood that the cluster heads structured are unequal and may stay in one part of the system making another part of the system inaccessible. In this work an attempt has been made to enhance the system existence with smart clustering component by dissecting the Connectivity, Distance and Residual proficiencies. The work will additionally decrease the rate of getting a particular node dead, rapidly. The proposed scheme will similarly control the amount of clusters in the system by indicating as far as possible for number of nodes in the cluster. This work will give an attractive balanced circulation of cluster and also sensor nodes over the system.

Modeling of the seismotectonic provinces of Iran using the self-organizing map algorithm

Various studies have been conducted thus far to classify the seismicity regions of Iran into zones that share similar seismic characteristics. In these studies, seismotectonic models have been presented that implement various information layers based mostly on human expert judgment. The purpose of the present research is to develop a seismotectonic model using all of the available information about the layers along with an automated smart clustering algorithm. In this paper, based on a review of past research studies, a combined model is presented, which involves the combination of the four most-referenced seismotectonic models.To build upon past studies, a smart algorithm called the self organizing map (SOM) was selected for clustering the seismotectonic data of provinces. Input layers, such as earthquake coordinates, seismicity, faults distribution, topography, gravity, and magnetism, were used to execute the SOM smart algorithm. Based upon previous studies, the number of clusters was selected in the range of 5–25. Three of the SOM algorithm outputs, which were more compatible with the combined model, were selected for use in this study. According to the clustering validation methods and expert judgments, the highest compatibility belongs to the models that were produced either by all of the information or by the combination of seismicity, faults, and topography layers. Implementation of the clustering validation methods indicated that the optimum number of seismotectonic zones is between 9 and 13.This study emphasizes that the application of smart methods along with various geological, structural, subsurface, and aerial information layers will produce informative and reliable results.

Clustering‐based power‐controlled routing for mobile wireless sensor networks

SUMMARYThis paper presents two new routing protocols for mobile sensor networks, viz. power‐controlled routing (PCR) and its enhanced version, i.e. Enhanced Power‐Controlled Routing (EPCR). In both the protocols, fixed transmission power is employed in the clustering phase but when ordinary nodes are about to send their data to their respective cluster‐heads, they change their transmission power according to their distance from their cluster‐head. While in PCR, the nodes are associated with the cluster‐head on the basis of weight, in EPCR it is done on the basis of distance. In addition to the protocols, we are suggesting a packet loss recovery mechanism for the PCR and EPCR. Both protocols work well for both mobile and static networks and are designed to achieve high network lifetime, high packet delivery ratio, and high network throughput. These protocols are extensively simulated using mass mobility model, with different speeds and different number of nodes to evaluate their performance. Simulation results show that both PCR and EPCR are successful in achieving their objectives by using variable transmission powers and smart clustering. Copyright © 2011 John Wiley & Sons, Ltd.

InGeno – an integrated genome and ortholog viewer for improved genome to genome comparisons

BackgroundSystematic genome comparisons are an important tool to reveal gene functions, pathogenic features, metabolic pathways and genome evolution in the era of post-genomics. Furthermore, such comparisons provide important clues for vaccines and drug development. Existing genome comparison software often lacks accurate information on orthologs, the function of similar genes identified and genome-wide reports and lists on specific functions. All these features and further analyses are provided here in the context of a modular software tool "inGeno" written in Java with Biojava subroutines.ResultsInGeno provides a user-friendly interactive visualization platform for sequence comparisons (comprehensive reciprocal protein – protein comparisons) between complete genome sequences and all associated annotations and features. The comparison data can be acquired from several different sequence analysis programs in flexible formats. Automatic dot-plot analysis includes output reduction, filtering, ortholog testing and linear regression, followed by smart clustering (local collinear blocks; LCBs) to reveal similar genome regions. Further, the system provides genome alignment and visualization editor, collinear relationships and strain-specific islands. Specific annotations and functions are parsed, recognized, clustered, logically concatenated and visualized and summarized in reports.ConclusionAs shown in this study, inGeno can be applied to study and compare in particular prokaryotic genomes against each other (gram positive and negative as well as close and more distantly related species) and has been proven to be sensitive and accurate. This modular software is user-friendly and easily accommodates new routines to meet specific user-defined requirements.