Remote Industrial Sensor Network Monitoring Using M2M Based Ethical Sniffers

Diagnosing the deployed network efficiency and anomaly detection, which is an important research issue in traditional networking systems, has not been carefully addressed in industrial wireless sensor networks. Although recent wireless systems for industrial automation such as ISA100.11a employ device management protocols, these protocols generate and report a large amount of status information from individual sensor nodes. Also, these protocols do not capture influences on network performance from external sources such as malicious nodes or interference from other networks. We propose a latent network diagnosis system (LaNDS) for industrial sensor networks. LaNDS employs a packet sniffing method for efficiently evaluating network performance and instantly identifying degradation causes of networking performance. LaNDS adopts an efficient network evaluation approach for detecting abnormalities from both internal and external causes. In our proposed monitoring scenario, special sniffer devices having M2M capability (WiMAX interface) are used to monitor the industrial sensor network by employing ethical sniffing. Our approach does not incur additional traffic overhead for collecting desired information. For evaluation, we have tested LaNDS locally on an ISA100.11a based sensor network in a lab environment and have validated the efficiency of the system based on the possible erroneous cases of industrial sensor network.