Cyber-Threat Detection System Using a Hybrid Approach of Transfer Learning and Multi-Model Image Representation.

Currently, Android apps are easily targeted by malicious network traffic because of their constant network access. These threats have the potential to steal vital information and disrupt the commerce, social system, and banking markets. In this paper, we present a malware detection system based on word2vec-based transfer learning and multi-model image representation. The proposed method combines the textual and texture features of network traffic to leverage the advantages of both types. Initially, the transfer learning method is used to extract trained vocab from network traffic. Then, the malware-to-image algorithm visualizes network bytes for visual analysis of data traffic. Next, the texture features are extracted from malware images using a combination of scale-invariant feature transforms (SIFTs) and oriented fast and rotated brief transforms (ORBs). Moreover, a convolutional neural network (CNN) is designed to extract deep features from a set of trained vocab and texture features. Finally, an ensemble model is designed to classify and detect malware based on the combination of textual and texture features. The proposed method is tested using two standard datasets, CIC-AAGM2017 and CICMalDroid 2020, which comprise a total of 10.2K malware and 3.2K benign samples. Furthermore, an explainable AI experiment is performed to interpret the proposed approach.

Fast File Transfers from IoT Devices by Using Multiple Interfaces.

The Internet-of-Things (IoT) is a modern technological revolution that enables communication amongst a plethora of different devices. To date, about 30 billion devices have been connected to the internet and more than 75 billion devices are probably to be connected worldwide by 2025. These can range from small sensors and actuators to larger devices such as smartphones, drones or even buildings and interconnected cars. Devices are often mobile and battery powered thus their communication requires fast and energy efficient solutions. To this end, this paper studies the use of multi-interface communication for fast and energy efficient communication. In particular, we consider the basic operation of data transfer between smartphones in the form of files. This task can be performed for backup purposes, and hence it represents a useful and frequent operation that users perform. Our aim is to provide a new and easy means that optimises file transfers with respect to time and energy consumption. In particular, as smartphones are endowed with various connecting interfaces like Bluetooth, WiFi and 4G, we conduct experimental studies by varying different parameters in order to understand the best setting, including which interface is more appropriate to accomplish file transfer. To this respect, we also implemented an innovative and light app that allows the use of two or more interfaces concurrently. The experimental results show how the coupling of some interfaces might be effective in terms of time, while consuming a negligible amount of energy. Actually, such results become more and more interesting as the size of the file to be transferred grows. The best combination experienced is by making use of WiFi at 5 GHz concurrently with 4G, whereas WiFi at 2.4 GHz caused interference complications.

A Hash-Based RFID Authentication Mechanism for Context-Aware Management in IoT-Based Multimedia Systems.

With the technological advances in the areas of Machine-To-Machine (M2M) and Device-To-Device (D2D) communication, various smart computing devices now integrate a set of multimedia sensors such as accelerometers, barometers, cameras, fingerprint sensors, gestures, iris scanners, etc., to infer the environmental status. These devices are generally identified using radio-frequency identification (RFID) to transfer the collected data to other local or remote objects over a geographical location. To enable automatic data collection and transition, a valid RFID embedded object is highly recommended. It is used to authorize the devices at various communication phases. In smart application devices, RFID-based authentication is enabled to provide short-range operation. On the other hand, it does not require the communication device to be in line-of-sight to gain server access like bar-code systems. However, in existing authentication schemes, an adversary may capture private user data to create a forgery problem. Also, another issue is the high computation cost. Thus, several studies have addressed the usage of context-aware authentication schemes for multimedia device management systems. The security objective is to determine the user authenticity in order to withhold the eavesdropping and tracing. Lately, RFID has played a significant for the context-aware sensor management systems (CASMS) as it can reduce the complexity of the sensor systems, it can be available in access control, sensor monitoring, real time inventory and security-aware management systems. Lately, this technology has opened up its wings for CASMS, where the challenging issues are tag-anonymity, mutual authentication and untraceability. Thus, this paper proposes a secure hash-based RFID mechanism for CASMS. This proposed protocol is based on the hash operation with the synchronized secret session-key to withstand any attacks, such as desynchronization, replay and man-in-the-middle. Importantly, the security and performance analysis proves that the proposed hash-based protocol achieves better security and performance efficiencies than other related schemes. From the simulation results, it is observed that the proposed scheme is secure, robust and less expensive while achieving better communication metrics such as packet delivery ratio, end-to-end delay and throughput rate.

Intelligent decision making for energy efficient fog nodes selection and smart switching in the IOT: a machine learning approach.

With the emergence of Internet of Things (IoT) technology, a huge amount of data is generated, which is costly to transfer to the cloud data centers in terms of security, bandwidth, and latency. Fog computing is an efficient paradigm for locally processing and manipulating IoT-generated data. It is difficult to configure the fog nodes to provide all of the services required by the end devices because of the static configuration, poor processing, and storage capacities. To enhance fog nodes' capabilities, it is essential to reconfigure them to accommodate a broader range and variety of hosted services. In this study, we focus on the placement of fog services and their dynamic reconfiguration in response to the end-device requests. Due to its growing successes and popularity in the IoT era, the Decision Tree (DT) machine learning model is implemented to predict the occurrence of requests and events in advance. The DT model enables the fog nodes to predict requests for a specific service in advance and reconfigure the fog node accordingly. The performance of the proposed model is evaluated in terms of high throughput, minimized energy consumption, and dynamic fog node smart switching. The simulation results demonstrate a notable increase in the fog node hit ratios, scaling up to 99% for the majority of services concurrently with a substantial reduction in miss ratios. Furthermore, the energy consumption is greatly reduced by over 50% as compared to a static node.