Search

The Internet of Things (IoT) has seen expeditious growth in recent times with 7 billion connected devices in 2020, thus leading to the vital importance of real-time monitoring of IoT devices. Through this paper, we demonstrate the idea of building a cloud-native application to monitor smart home devices. The application intends to provide valuable performance metrics from the perspective of end-users and react to anomalies in real-time. In this demo paper, we conduct the demonstration using Autoencoder (an unsupervised technique) based Deep Neural Networks (DNNs) to learn the normal operating conditions of power consumption of smart devices. When an anomaly is detected, the DNNs take proactive action and send appropriate commands back to the device. In addition, the users are provided with a real-time graphical representation of power consumption. This will help to save electricity on a domestic as well as industrial level. Finally, we discuss the future prospects of monitoring IoT devices.

Cyber-security is concerned with protecting information, a vital asset in today’s world. The volume of data that is generated can be usefully analyzed when cyber-security systems are effectively implemented with the aid of software support. Our approach is to determine normal behavior of a system based on sequences of system call traces made by the kernel processes in the system. This paper describes a robust and computationally efficient anomaly based host based intrusion detection system using an Encoder-Decoder mechanism. Using CuDNNLSTM networks, it is possible to obtain a set of comparable results with reduced training times. The Bidirectional Encoder and a unidirectional Decoder is trained on normal call sequences in the ADFA-LD dataset. Intrusion Detection is evaluated based on determining the probability of a sequence being reconstructed by the model yes