• Dr. Lata B T
• Srikantha S P

Wireless Internet of Sensor Networks (WISNs) are currently extensively used in applications from smart cities to environmental monitoring. Their open channels of communication do, however, expose them to both active and passive threats that endanger the data's integrity, confidentiality, and availability. Ongoing vulnerability of WISNs to sophisticated cyberattacks. Conventional security measures often cannot identify and mitigate these risks dynamically, hence an enhanced approach that can adapt and learn with time is needed. Inception-ResNet-v2 model is included into a novel security architecture we provide for threat detection and mitigation in real time. The method is to train the model using a sizable dataset including several attack fingerprints and normal traffic patterns. This deep learning model is used to predict potential network security breaches by using its well-known high accuracy and quick processing. Results of our studies show improvements in the detection and mitigation of both active and passive attacks. Less than 1% of false positives and 98.5% of active attack detection accuracy and 97.8% of passive attack detection accuracy were achieved by the proposed model. Less than 200 millisecond response time of the system ensured timely threat mitigation as well.

Cyber-attacks, Deep Learning, Inception-ResNet-v2, Security, Wireless Internet of Sensor Networks

"Enhancing Security in Wireless Sensor Networks using Inception-Resnet-V2", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.9, Issue 12, page no.h1-h10, December-2022, Available :http://www.jetir.org/papers/JETIR2212701.pdf

"Enhancing Security in Wireless Sensor Networks using Inception-Resnet-V2", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.9, Issue 12, page no. pph1-h10, December-2022, Available at : http://www.jetir.org/papers/JETIR2212701.pdf