Decision fusion is used to fuse classification results and improve the classification accuracy in order to reduce the consumption of energy and bandwidth demand for data transmission. Decentralized classification fusion problem was the reason to use belief function based decision fusion approach in Wireless Sensor Networks (WSNs). With the consideration of improving the belief function fusion approach, we have proposed four classification techniques namely Enhanced K-Nearest Neighbor (EKNN), Enhanced Extreme Learning Machine (EELM), Enhanced Support Vector Machine (ESVM), and Enhanced Recurrent Extreme Learning Machine (ERELM). In addition, WSNs are fallible to errors and faults because of their different software, hardware failures, and their deployment in diverse fields. These challenges require efficient fault detection methods to be used to detect faults in WSNs in a timely manner. We induced four type of faults: offset fault, gain fault, stuck-at fault, and out of bounds fault and used enhanced classification methods to solve the sensor failure issues. Experimental results show that ERELM has given the first best result for the improvement of belief function fusion approach. The other three proposed techniques ESVM, EELM, and EKNN have provided the second, third, and fourth best results, respectively. Proposed enhanced classifiers are used for fault detection and are evaluated using three performance metrics ,i.e., Detection Accuracy (DA), True Positive Rate (TPR), and Error Rate (ER).
In this thesis, the owner of the (Internet of Thing) IoT device can generate revenueby selling IoT device’s data to interested users.
However, on the other hand, users do not trust the owner of IoT device for data trading and are not confident about the quality of data. Traditional data trading systems have many limitations, as they involve third party and lack: decentralization, security and reputation mechanisms. Therefore, in this thesis, we have leveraged the IoTs with blockchain technology to provide trustful data trading through automatic review system for monetizing IoT’s data. We have developed blockchain based review system for IoT data monetization using Ethereum smart contracts. Review system encourages the owner to provide authenticated data and solve the issues regarding data integrity, fake reviews and conflict between entities. Data quality is ensured to users through reviews and ratings about the data, stored in blockchain. To maintain the data integrity, we have used Advanced Encryption Standard (AES)-256 encryption technique to encrypt data. All transactions are secure and payments are automated without any human intervention. Arbitrator entity is responsible to resolve problems between data owner and users. Incentive is provided to users and arbitrator in order to maintain the user participation and honesty. Additionally, Ethereum blockchain system requires gas for every transaction. Simulations are performed for the validation of our system. We have examined our model using three parameters: gas consumption, mining time and encryption time. Simulations show that the proposed methods outperform the existing techniques and give better results for belief function and fault detection in datascience WSNs. Additionally, blockchain based data trading in IoT system requires gas for every transaction. We have examined our model using three parameters: gas consumption, mining time and encryption time.