Edge intelligence is rapidly emerging as a pivotal platform for supporting future IoT networks. The integration of artificial intelligence and machine learning (AI/ML) with edge computing furnishes a new era in which edge systems can learn environment dynamics and optimize resource autoscaling policies. However, the heterogeneity of IoT networks, characterized by diverse applications and requirements, necessitates edge systems with advanced intelligence to tailor resource autoscaling policies to specific environments. Despite ongoing research in edge intelligence, most studies have focused on a single environment or service type, thereby limiting their applicability to real-world scenarios with varied IoT services. To address this limitation, we propose the intelligent and autonomous edge slicing (IAES) system, a novel approach designed to recognize diverse IoT environments and implement per-slice resource allocation policies. IAES leverages deep reinforcement learning (DRL), namely, dueling double deep Q-networks (D3QN), to optimize resource autoscaling across distinct IoT environments, such as smart cities, eHealth, and smart factories. Additionally, IAES incorporates an intelligent environment classification component that utilizes joint traffic prediction and classification models. Several AI algorithms such as long short-term memory (LSTM), convolutional neural networks (CNN), and multilayer perceptron networks (MLP), are evaluated for their efficacy in predicting IoT environments. Simulation experiments demonstrate that the IAES system achieves a 50-60% reduction in system costs compared to both rule-based commercial autoscaling employed in Kubernetes systems and an intelligent prediction-based autoscaling algorithm.

The proliferation of Internet of Things (IoT) services across diverse sectors such as healthcare, industrial IoT, and smart cities has introduced unprecedented complexity in network Management and Orchestration (MO). Contemporary MO systems are challenged to support the coexistence of IoT services with varying Quality of Service (QoS) requirements while ensuring end-to-end (E2E) performance across heterogeneous technologies and multi-administrator domain networks. To address these requirements, emerging technologies such as Intelligent Intent-Based Network Slicing (I-IBNS) systems are increasingly employed, leveraging advancements in network automation, Artificial Intelligence (AI), and Network Slicing (NS). This paper focuses on three critical challenges in developing I-IBNS systems for IoT: E2E resource allocation, privacy preservation in multi-administrator systems, and E2E QoS assurance across heterogeneous networks and time-varying traffic. We propose a privacy-preserving I-IBNS framework, named Harmony Slice Master (H-SliceMaster), which integrates a knowledge management framework for privacy-aware data aggregation, an intent propagation mechanism for translating high-level intents into network configurations, and a novel Promise and Price Network Operation (PPNO) principle for optimizing E2E resource allocation while maintaining intra-domain privacy. A proof-of-concept design of the H-SliceMaster is presented, utilizing Deep Q-Networks (DQN) for intra-domain resource allocation and a centralized algorithm for optimizing E2E network slice deployment. Simulation results demonstrate that the H-SliceMaster efficiently satisfies various IoT applications and delay requirements. Moreover, the proposed system achieved an 20% and 50% reduction in system costs compared to a Branching Dueling Q-Network (BDQ) and greedy algorithm, respectively.

The rapid integration of intelligence and automation into future Internet of Things (IoT) systems, empowered by Intent-based Networking (IBN) and Network Slicing (NS) technologies, is transforming the way novel services are envisioned and delivered. The automation capabilities of IBN depend significantly on key facilitators, including data management and resource management. A robust data management methodology is essential for leveraging large-scale data, encompassing service-specific and network-specific data, enabling IBN systems to extract insights and facilitate real-time decision-making. Another critical enabler involves deploying intent-based mechanisms within an NS system that translate and ensure user intents by mapping them to precise Management and Orchestration (MO) commands. Nevertheless, data management in IoT systems faces significant security and operational challenges due to the diverse range of services and technologies involved. Furthermore, intent-based resource management demands intelligent proactive, and adaptive MO mechanisms that can fulfill a wide range of intent requirements. Existing surveys within the field have focused on technology-specific advancements, often overlooking these challenges. In response, this paper defines Intelligent Intent-Based Network Slicing (I-IBNS) systems exemplifying the integration of intelligent IBN and NS for the MO of IoT systems. Furthermore, the paper surveys I-IBNS systems, focusing on two critical domains: resource management and data management. The resource management segment examines recent developments in IBN mechanisms within an NS system. Meanwhile, the second segment explores data management complexities within IoT networks. Moreover, the paper envisions the roles of intent, NS, and the IoT ecosystem, thereby laying the foundation for future research directions.

Traffic modeling is a valuable tool for simulating traffic characteristics and assessing the effectiveness of new network mechanisms and protocol designs. The emergence of the Internet of Things (IoT) has led to a growing interest in IoT traffic modeling due to the unique characteristics of IoT traffic, such as sudden data bursts and application-dependent traffic characteristics. The focus of the literature has been on modeling the arrival distribution of IoT traffic. However, this approach fails to capture important characteristics of time-series traffic, such as IoT traffic behaviors and seasonality patterns. Such characteristics provide crucial insights for the effective management and optimization of IoT networks. By exploiting time-series characteristics, dynamic resource allocation mechanisms can be designed instead of resource provisioning for peak usage. Additionally, comprehending the traffic generation behavior of IoT sensors can provide insight into the energy consumption of the sensor layer, which has a multitude of uses. In this article, we present a novel IoT traffic modeling framework called the tiered Markov-modulated stochastic process (TMMSP). The TMMSP framework can produce application-specific IoT time-series traffic traces that mimic the behaviors, e.g., the temporal dynamics, of real IoT traffic. Our results illustrate the flexibility and capability of the TMMSP framework in modeling the traffic behaviors of three IoT applications, specifically, telehealth, asset monitoring, and building security applications. Finally, we illustrate how the TMMSP framework can be used to evaluate the performance of an autonomous edge slicing (AES) mechanism.

Future wireless networks will exhibit heterogeneity of traffic generating sources originated by numerous Internet of Things (IoT) nodes as well as traditional mobile phones. Moreover, the space of novel IoT services is expanding the simple monitoring tasks of IoT nodes to more complex services in which a node can be in a monitoring state and transition autonomously to an alarm state when predefined conditions are detected. The complexity of the envisioned future wireless networks is indeed new to the community with challenges affecting many aspects such as protocol design and network operation mechanisms. Traffic modeling lies at the core of these issues. As the advancement of technologies continues, faithful performance evaluation measures are dependent on the underlying traffic model. In this scope, we propose a Tiered Markov Modulated Poisson Process (TMMPP) that is capable of capturing IoT traffic characteristics, e.g. patterns and seasonality, which occur in long time spans, e.g days, with the flexibility of modeling different IoT service behaviors. Moreover, we study an autonomous edge scaling mechanism as a use case illustrating the benefits of the proposed TMMPP traffic model.

The Internet of Things (IoT) technologies are transforming traditional businesses into digital-based platforms allowing for more service innovation, performance efficiency and customer satisfaction. Novel services enable users to utilize their personal devices (eg. mobile phones or laptops) to access the IoT platform, process data, and control the IoT infrastructure. However, these services impose critical user authentication and access control requirements. In this paper, we propose a decentralized user authentication and access control system for the IoT platforms via a permissioned blockchain network. We define an authorization sensitivity factor to provide clients with specific access control privileges and we consider an ehealth system as a use case example to demonstrate our solution. The proposed system is implemented using Ethereum platform. Besides, we investigate a threat model that considers an insider Distributed Denial of Service (DDoS) attack. The proposed defense mechanism utilizes a modifier function in the smart contract and keeps a real-time record of legitimate users to restrict function calls. The results illustrate the benefits of the defense mechanism in terms of the system response time.