Detecting selective forwarding attacks on cloud

—The Internet of Things (IoT) technology is now ubiquitous in both daily life and industrial production. Wireless Sensor Networks (WSNs) are a crucial part of the IoT infrastructure, providing the underlying network. However, WSNs are vulnerable to attacks due to challenging environmental conditions and their open communication nature. One particularly elusive threat is the selective forwarding attack, where malicious nodes intentionally drop parts or all of the received data packets, causing information loss and delays. This paper addresses the transition from the limited resources of central processing units (CPUs) to the expansive capabilities of cloud platforms within the IoT framework. By harnessing the robust computing power of cloud platforms, we investigate the effectiveness of the Time Convolution Network (TCN) time series prediction algorithm in detecting malicious nodes. Comparative results demonstrate that our proposed scheme achieves a Miss Detection Rate (MDR) of below 1% and a False Detection Rate (FDR) of less than 5%.

Authors

• Haozhen Wang
• Yuhang Wang School of Computer Science and Engineering, Sun Yat-sen University, Guangzhou, China Guangdong Key Laboratory of Big Data Analysis and Processing, Guangzhou, Guangdong, China
• Yuanming Wu

Keywords

• Internet of Things cloud platform
• Time convolution network
• Selective forwarding attack
• Wireless sensor network