Wireless sensor network is recent trend development in remote technologies for ubiquitous computing in various applications, like monitoring, sensing, geo location-based applications. By the sense the transmission nodes are most probably affected by close attacks leads devastating communication defects. From the communication, ensuring security is the important aspect for communication nodes to protecting the data transmission without any malicious attacks to resolve this problem, we propose an advanced transfer learning model to identifying the clone attacks based on Neural Fuzzy intensive-Sub spectral scaling feature selection (NFI-SSFS) to secure using Cooperative Secure Optimal Link Stability Routing Allocation (CS-OLSR). The communication logs is collected to consume the variance feature level of packet difference rate under memory and transmission defect fact with sport of False injection impact rate (FIIR) and Time stamp communication behaviour rate (TSCBR). Then the intensive feature factor is obtained using NFI-SSFS to marginalize the clone attack rate. Then CS-OLSR is applied to ensure the secure routing based on the identified clone attack region. The proposed system effectively identifies the data replacement node effetely to find the clone attacks. This system makes effective clone attack route finding approach to ensure the security to transfer another route to make data safely.

Attack detection, Wireless mobile network, Machine learning, ANN, node behaviour analysis, feature selection and classification