No abstract

The latest trend in the research filed implies the importance of data security in wireless sensor networks. There are various approaches identified for securing the data by using trust wide security such as cryptographic systems and routing protocols. However, these approaches are very critical to identify the optimal path in the network and attacks by unauthorized node cannot be prevented. In this paper, a new algorithm for combining the AODV (Ad Hoc On-Demand Distance Vector) routing protocol and particle swarm optimization (PSO) is implemented to produce trustable routing in every location through block chain. The possible routing procedure will enhance the routing nodes to acquire routing information among all the nodes but will never allow the node to capture the information. The routing protocol on the blockchain is used to utilize the path efficiently without deviation caused by other anchor nodes. It also identifies the congestion in the entire path of the particular network and avoids tampering of information between the nodes. The blockchain enabled with PSO algorithm and AODV routing protocol provides the simulation results about the efficient packet delivery system. The Security has been performed in every node used to identify the best route for producing the efficient throughput and quality of services.

The deep learning concept for performing object detection plans to minimize the labeling cost by identifying the samples which increase the detection into the unlabeled pool. According to the object detection than the classification process as the designing and selection of procedure is very essential. The related works have been implemented the aggregation data process of several outputs and batch box process for improving the performance evaluation. The mean Average Precision 𝒎𝑨𝑷 is the main performance metric for identifying the accuracy of the object detection. The class imbalance problem has been solved using the background class in every group of sample images. The loss related weight algorithm for training group is proposed in this paper utilizing the batch boxes, aggregating data and also the 𝒎𝑨𝑷 enhancements are addressed to solve the class imbalance problem. Additionally, a sampling process is used for identifying the uncertainty and enhancing the object detection process. The performance results illustrate that the proposed framework generates good performance than the relevant technique and it will be used for realtime applications in an efficient manner.

No abstract

No abstract