The exponential growth of big data and deep learning has increased the data exchange traffic in society. Machine Learning as a Service, (MLaaS) which leverages deep learning techniques for predictive analytics to enhance decision-making, has become a hot commodity. However, the adoption of MLaaS introduces data privacy challenges for data owners and security challenges for deep learning model owners. Data owners are concerned about the safety and privacy of their data on MLaaS platforms, while MLaaS platform owners worry that their models could be stolen by adversaries who pose as clients. Consequently, Privacy-Preserving Deep Learning (PPDL) arises as a possible solution to this problem. Recently, several papers about PPDL for MLaaS have been published. However, to the best of our knowledge, no previous paper has summarized the existing literature on PPDL and its specific applicability to the MLaaS environment. In this paper, we present a comprehensive survey of privacy-preserving techniques, starting from classical privacy-preserving techniques to well-known deep learning techniques. Additionally, we present a detailed description of PPDL and address the issue of using PPDL for MLaaS. Furthermore, we undertake detailed comparisons between state-of-the-art PPDL methods. Subsequently, we classify an adversarial model on PPDL by highlighting possible PPDL attacks and their potential solutions. Ultimately, our paper serves as a single point of reference for detailed knowledge on PPDL and its applicability to MLaaS environments for both new and experienced researchers.

Currently, it is rapidly increasing convergence services based on various mobile devices with sensors like Smart Home. Specifically the mobility of the sensors in Smart Home merged with wireless sensor networks (WSN) brings security issues such as re-authentication and tracing the node movement. We extend our novel and efficient node authentication and key exchange protocol that support Irregular distribution. Compared with previous protocols, our protocol has only a third of communication and computational overhead. We expect our protocol to be the efficient solution that increases the lifetime of sensor network.

The rise of interconnected devices through wireless networks provides two sides consequences. On one side, it helps many human tasks; on the other hand, the prone wireless medium opens the vulnerable system to be exploited by adversaries. An Intrusion Detection System (IDS) is one method to inspect the network traffic by leveraging state-of-the-art anomaly detection techniques. Deep learning models have been utilized to distinguish the benign and malicious traffic. However, projecting the tabular data into images before the image classification has been the main challenge of leveraging deep learning for IDS purposes. We propose the novel projection of tabular data into 2-coded color mapping for IDS purposes. The proposed method employs a feature selection method to ensure optimal dimensionality. We examined the different number of attribute subsets to obtain the relationship between the attributes. Furthermore, it takes advantage of the Convolutional Neural Network (CNN) model to classify the Wi-Fi attacks. We evaluate the proposed model using the most common Wi-Fi attacks dataset, Aegean Wi-Fi Intrusion Dataset (AWID2). The proposed method achieved an F1 score of 99.73% and a false positive rate of 0.24%. This study highlights the importance of addressing the mapping procedures from tabular data into grid-based data before deep learning training and validates the effectiveness of CNN to detect multiple types of wireless network attacks.