End-To-End Throughput for Vanet With and Without Cloud Effect

Abstract

Vehicle Ad-hoc network (VANET) have sensing nodes moving with different speeds. The vehicular cloud is a group of vehicles communicating with each other to share the resources provided from the central cloud. The end-to-end throughput for multi-hop wireless network is analysed in this paper using a set of non-linear equations. The interference due to the hidden nodes and the neighbouring nodes is considered into computing the end-to-end throughput. The contention graph is used to clarify the connection among the nodes into the network to find the capacity of the individual links to facilitate estimating the end-to-end throughput. Effect of the cloud on the end-to-end throughput is also illustrated in this paper. The results show that the throughput with employing cloud is lower than that without using the cloud.

Predicting Perceptual Quality in Internet Television Based on Unsupervised Learning

“…All the tested topologies with their classification rates are shown in Table 4. Since there are no target data (learning without a teacher), a Kohonen map can easily avoid overfitting, which improves the credibility of the delivered output [26,27]. Another benefit of using a Kohonen map is that the results can be easily presented in graphical form.…”

“…False positive and false negative are Type I and Type II errors. This is extremely useful for the measuring of recall, precision, and F1 score, which are briefly described below [27].…”