This paper proposes a system able to collect information about pedestrian congestion levels in Bucharest, by offering an Android application for monitoring its users' location anonymously. The application sends th...
详细信息
ISBN:
(纸本)9781728123318
This paper proposes a system able to collect information about pedestrian congestion levels in Bucharest, by offering an Android application for monitoring its users' location anonymously. The application sends the data to a server which calculates the most crowded areas of the city using the DBSCAN clustering algorithm. The application displays congested areas, as well as their level of congestion by using different coloured circles on the map presented to the user. The history of busy areas of Bucharest can also be visited at the request of the user.
Personalized recommender systems have become increasingly popular in recent years, as they have the ability to make appropriate choices for each active user. Collaborative filtering (CT) is the most successful and wid...
详细信息
ISBN:
(纸本)9781538619810
Personalized recommender systems have become increasingly popular in recent years, as they have the ability to make appropriate choices for each active user. Collaborative filtering (CT) is the most successful and widely used technique in recommender systems, which aims at discovering similar users or items based on the history user rating records, i.e., user-item matrix. However, CF may not generate good recommendations when user-item matrix is very sparse. To address this problem, we explore the property category and semantic content to reduce the amount of items, which lead to more accurate performance when estimating user similarity. In addition, since the amount of users is quite huge, we first profile similar users with the aid of clustering algorithm before recommendation. Then, for each active user, the CF recommender system returns top recommendations from the narrow-down cluster the same as the active user by calculating user similarity with the help of item semantic information. The experiments have been performed on the benchmark dataset in NITCC 2017 to recommend point-of interest (POI) for each active user. The comparative results demonstrate that our proposed model outperforms the two baselines (i.e., a user-based CF system and an item-based CF system).
Cell tower locations are not publicly available due to business interests of wireless providers. Very often wireless providers provide exaggerated coverage maps that may mislead the public. In addition to providing a ...
详细信息
ISBN:
(纸本)9780738113302
Cell tower locations are not publicly available due to business interests of wireless providers. Very often wireless providers provide exaggerated coverage maps that may mislead the public. In addition to providing a neutral check on the coverage maps, prediction of cell tower locations hosting multiple operators' access nodes could also be helpful in disaster communications and public safety in general. The localization of the disaster-affected towers can be very conducive to respond and reach to the victims. Further, victims' devices could utilize this knowledge to initiate device-to-device (D2D) or unmanned aerial vehicular (UAV) communications as alternatives to the damaged cellular infrastructure. Publicly available crowdsourced cell (base station) locations and FCC's sites can be used to predict the cell tower/site locations in the United States. In this work, we utilized a weighted k-means algorithm to predict cell tower locations from OpenCellid crowdsourced dataset and implemented a mapping algorithm to locate nearest physical towers. We map the predicted towers to two different sources of physical towers. Our comparison shows a significant accuracy in predicting tower locations regardless of sources of physical towers. The technique can be used to predict the tower locations in other countries as well.
Event-driven wireless sensor networks are special networks that just need the sensors which detected the events gathering and sending data through a reasonable routing to the single information processing center calle...
详细信息
ISBN:
(纸本)9781467378222
Event-driven wireless sensor networks are special networks that just need the sensors which detected the events gathering and sending data through a reasonable routing to the single information processing center called sink node, usually used in monitoring emergency events in many large-scale severe environments. However, the battery energy of sensors are limited and difficult to be charged or replaced, it requires effective network routing protocols in communication and message passing process to extend their lifetime. In this paper, we propose a real-time, weighted clustering routing algorithm to organize the event detected sensors in the event regions into clusters. The algorithm has a full consideration of the sensors' residual energy, distance, event excited intensity and received signal strength during the three parts: choosing cluster-heads, clustering and establishing routing path. Simulation results have demonstrated the effectively routing performance of our algorithm, and its network lifetime is 16% longer than the TEEN algorithm.
Classification technique has become the necessary tool to detect products which has texture patterns in machine vision. The design used for the process of texture discrimination is critical to the performance of the o...
详细信息
ISBN:
(纸本)9781424448692
Classification technique has become the necessary tool to detect products which has texture patterns in machine vision. The design used for the process of texture discrimination is critical to the performance of the outcome. In this study we propose a design process model which can systematically maintain and improve on the performance of texture segmentation in terms of speed and segmentation accuracy. The overall accuracy and time tested for the synthetic textures segmentation has shown significant improvement throughout the parameters tuning of the chosen methods.
Inspired by clustering and energy harvesting techniques, we study multiple-cluster wireless sensor networks with energy harvesting (EH) sensors serving as relay for cluster heads. In this paper, we derive the model fo...
详细信息
ISBN:
(纸本)9781479980550
Inspired by clustering and energy harvesting techniques, we study multiple-cluster wireless sensor networks with energy harvesting (EH) sensors serving as relay for cluster heads. In this paper, we derive the model for realistic energy harvesting rate. Then we propose distributed matching algorithm for EHs to serve as relay for CHs. The proposed algorithm could find optimal/near-optimal CH-EH matching in short time and still achieve good performance. We evaluate the performance of our method through theoretical analysis as well as simulation.
Optimizing the transmission ranges between nodes and balancing the number of nodes among clusters are two feasible approaches to conserve the limited battery power in exchanging messages when designing power-efficient...
详细信息
ISBN:
(纸本)9781457702556
Optimizing the transmission ranges between nodes and balancing the number of nodes among clusters are two feasible approaches to conserve the limited battery power in exchanging messages when designing power-efficient (or green) clustered wireless ad hoc networks. To achieve this objective, we first employ the concept of Relative Neighborhood Graph (RNG) to obtain a power-efficient logical network topology in which the transmission ranges between nodes are adjusted to the optimal. Then, based on the obtained RNG-based logical network topology, we present a green clustering algorithm to organize the wireless ad hoc network into a clustered architecture in which the number of nodes among clusters is balanced. Simulation results confirm that the presented approaches not only optimize the transmission ranges between nodes but also balance the number of nodes among clusters. Thus, the organized clustered wireless ad hoc network is regarded as power-efficient (or green).
This paper focuses on structure of dialog graphical system for pattern recognition with help of distance function. In this paper is evaluate the performance of different criterion functions and algoritms for the probl...
详细信息
ISBN:
(纸本)9789531841306
This paper focuses on structure of dialog graphical system for pattern recognition with help of distance function. In this paper is evaluate the performance of different criterion functions and algoritms for the problem of clustering large datasets.
clustering of data is a fundamental data analysis step that has been widely studied across in data mining. Adaptive resonance theory network (ART) is an important algorithm in clustering. ART is also very popular in t...
详细信息
ISBN:
(纸本)3540343857
clustering of data is a fundamental data analysis step that has been widely studied across in data mining. Adaptive resonance theory network (ART) is an important algorithm in clustering. ART is also very popular in the unsupervised neural network. Type I adaptive resonance theory network (ART1) deals with the binary numerical data, whereas type II adaptive resonance theory network (ART2) deals with the general numerical data. Several information systems collect the mixing type attitudes, which included numeric attributes and categorical attributes. However, ART1 and ART2 do not deal with mixed data. If the categorical data attributes are transferred to the binary data format, the binary data do not reflect the similar degree. It influences the clustering quality. Therefore, this paper proposes a modified adaptive resonance theory network (M-ART) and the conceptual hierarchy tree to solve similar degrees of mixed data. This paper utilizes artificial simulation materials and collects a piece of actual data about the family income to do experiments. The results show that the M-ART algorithm can process the mixed data and has a great effect on clustering.
one of the major restrictions for the application areas in wireless sensor network is the threat of limited energy resources. The sensor nodes are equipped with ir-rechargable batteries as the source of power. To maxi...
详细信息
ISBN:
(纸本)9781479923526
one of the major restrictions for the application areas in wireless sensor network is the threat of limited energy resources. The sensor nodes are equipped with ir-rechargable batteries as the source of power. To maximize the entire network life time it is required to optimize the energy usage for the sensor nodes in WSNs. It is already established that clustering is an efficient routing technique requires less energy consumption to transmit data to the base station with compare to the direct communication techniques. Efficiency of clustering technique is directly dependent on formation of the cluster and cluster head selection techniques. The uneven distribution of the sensors among the clusters will cause energy leakage problem for some cluster heads and may cause premature deaths to them. In this paper, we present a novel cluster based routing algorithm that ensures to form clusters with even load as well as select cluster heads in a better way using fuzzy based environment so that optimal data transmission on single or multi hop environment is maintained. The simulation results more than 40% prolonged life time for the sensor nodes.
暂无评论