Bambang Susilo

The Internet of Things (IoT) has emerged as a crucial element in everyday life. The IoT environment is currently facing significant security concerns due to the numerous problems related to its architecture and supporting technology. In order to guarantee the complete security of the IoT, it is important to deal with these challenges. This study centers on employing deep learning methodologies to detect attacks. In general, this research aims to improve the performance of existing deep learning models. To mitigate data imbalances and enhance learning outcomes, the synthetic minority over-sampling technique (SMOTE) is employed. Our approach contributes to a multistage feature extraction process where autoencoders (AEs) are used initially to extract robust features from unstructured data on the model architecture’s left side. Following this, long short-term memory (LSTM) networks on the right analyze these features to recognize temporal patterns indicative of abnormal behavior. The extracted and temporally refined features are inputted into convolutional neural networks (CNNs) for final classification. This structured arrangement harnesses the distinct capabilities of each model to process and classify IoT security data effectively. Our framework is specifically designed to address various attacks, including denial of service (DoS) and Mirai attacks, which are particularly harmful to IoT systems. Unlike conventional intrusion detection systems (IDSs) that may employ a singular model or simple feature extraction methods, our multistage approach provides more comprehensive analysis and utilization of data, enhancing detection capabilities and accuracy in identifying complex cyber threats in IoT environments. This research highlights the potential benefits that can be gained by applying deep learning methods to improve the effectiveness of IDSs in IoT security. The results obtained indicate a potential improvement for enhancing security measures and mitigating emerging threats.

The autonomic Internet of Things is the creation of self-management capability in the Internet of Things system by embedding some autonomic properties, with the goal of freeing humans from all detail of the operation and management of the system. At same time, this provides a system to always operate on the best performance. This paper presents a review of the recent studies related to the design of network communication protocol, which can support autonomic Internet of Things. Many of the studies come from the research and development in Wireless Sensor Network protocols, as it becomes one of the key technologies for the Internet of Things. The identified autonomic properties are self-organization, self-optimization, and self-protection. We review some protocols with the objective of energy consumption reduction and energy harvesting awareness, as it can support the self-energy-awareness property. As the result, the protocol designs are mapped according to each autonomic property supported, including protocols for MAC layer, protocols for clustering, protocols for routing, and protocols for security. This can be used to map the advances of communication protocol research for the autonomic Internet of Things and to identify the opportunities for future research.

The internet has become an inseparable part of human life, and the number of devices connected to the internet is increasing sharply. In particular, Internet of Things (IoT) devices have become a part of everyday human life. However, some challenges are increasing, and their solutions are not well defined. More and more challenges related to technology security concerning the IoT are arising. Many methods have been developed to secure IoT networks, but many more can still be developed. One proposed way to improve IoT security is to use machine learning. This research discusses several machine-learning and deep-learning strategies, as well as standard datasets for improving the security performance of the IoT. We developed an algorithm for detecting denial-of-service (DoS) attacks using a deep-learning algorithm. This research used the Python programming language with packages such as scikit-learn, Tensorflow, and Seaborn. We found that a deep-learning model could increase accuracy so that the mitigation of attacks that occur on an IoT network is as effective as possible.

The Internet has become a major need of the world community leading to a new era of sophisticated communication strategies. This is to provide services to develop the architecture and improve the content widely. The main obstacle to the TCP / IP protocol for retrieving data or using a service is the consumer who wants to know the endpoint (IP address) that provides the desired object or service. Another problem is the efficient use of network power available. The last problem that motivates the paradigm shift in the network is the inherent problem in IP. Neither the integrity of the IP packet nor its authenticity can be accessed by the recipient or by the network forwarding entity.Named Data Networking (NDN) offers an alternative solution as the next generation internet architecture. Named Data Networking (NDN) is a theory that has been used based on progress in data-driven processes. In this thesis, we have studied the strategy of Forwarding Replica Routing (NRR) to some forwarding strategies in NDN by analyzing their performance against triple-play services (data, voice and video). Ideal Nearest Replica Routing (NRR) is a strategy implemented on the CCnSIM simulator where this forwarding strategy looks for the nearest cache and selects the shortest path accordingly. Experiments were performed using NS3 and ndnSIM 2.0 on Ubuntu 14.04.5 with different forwarding strategies and increasing the number of consumers.The experimental results show that the Nearest Replica Routing (NRR) strategy provides the total data sent / total overhead cost more efficiently by 20% to 58% compared to other forwarding strategies in terms of data transfer. And the total cost per kilobyte is better by 18% to 45% compared to other forwarding strategies.

Internet has become the primary need of the world community that led to a new era of sophisticated communication strategy. This enforced service providers to develop a new architecture and boosted wide range distribution of content. Named Data Networking (NDN) offers an alternative solution as the successor for the architecture of the Internet. In this paper, we evaluated the ideal Nearest Replica Routing (NRR) forwarding strategy against multiple forwarding strategies that exist in NDN by evaluating their performance against triple-play service with specific context. In our experiment we used NS3 and ndnSIM 2.0 on Ubuntu 14.04.5. The experiment results show that the ideal Nearest Replica Routing (NRR) strategy has a good performance in various circumstances.