Traffic monitoring and managing in urban intelligent transportation systems (ITS) can be carried out based on vehicular sensor networks. In a vehicular sensor network, vehicles equipped with sensors such as GPS, can a...
详细信息
Traffic monitoring and managing in urban intelligent transportation systems (ITS) can be carried out based on vehicular sensor networks. In a vehicular sensor network, vehicles equipped with sensors such as GPS, can act as mobile sensors for sensing the urban traffic and sending the reports to a traffic monitoring center (TMC) for traffic estimation. The energy consumption by the sensor nodes is a main problem in the wireless sensor networks (WSNs);moreover, it is the most important feature in designing these networks. Clustering the sensor nodes is considered as an effective solution to reduce the energy consumption of WSNs. Each cluster should have a Cluster Head (CH), and a number of nodes located within its supervision area. The cluster heads are responsible for gathering and aggregating the information of clusters. Then, it transmits the information to the data collection center. Hence, the use of clustering decreases the volume of transmitting information, and, consequently, reduces the energy consumption of network. In this paper, fuzzy C-Means (FCM) and fuzzy subtractive algorithms are employed to cluster sensors and investigate their performance on the energy consumption of sensors. It can be seen that the FCM algorithm and fuzzysubtractive have been reduced energy consumption of vehicle sensors up to 90.68% and 92.18%, respectively. Comparing the performance of the algorithms implies the 1.5 percent improvement in fuzzy subtractive algorithm in comparison.
This study aims to produce a diagnosis system for breast masses related to breast cancer. The dataset consisting of 60 digital mammograms is acquired from Istanbul University Faculty of Medicine Hospital. 78 masses in...
详细信息
This study aims to produce a diagnosis system for breast masses related to breast cancer. The dataset consisting of 60 digital mammograms is acquired from Istanbul University Faculty of Medicine Hospital. 78 masses in the mammograms are extracted manually for this study by the experts. It is a fuzzy based comperative study of malignant-benign classification for breast masses which has the accuracy of 74.36% with k-means and 93.75% with ANFIS based fuzzy c-means and subtractive clustering.
暂无评论