Network operators are faced with a strategic puzzle on how to balance limited resource availability, dynamic IoT traffic requirements, and dynamic IoT device behavior in an end-to-end communication paradigm. This work formally defines the IoT-aware Virtual Network Function (VNF) placement problem. We consider realistic IoT traffic statistics from a smart-city PoC to evaluate an indicative set of placement algorithms with different objective functions under static and dynamic traffic scenarios, to study their impact on the overall performance.

Y. Rafique, A. Leivadeas, and M. Ibnkahla, “An IoT-Aware VNF Placement Proof of Concept in a Hybrid Edge-Cloud Smart City Environment,” in 2022 IEEE Wireless Communications and Networking Conference (WCNC), Apr. 2022, pp. 1395–1400. doi: 10.1109/WCNC51071.2022.9772004.

For more details: 

Although the market valuation and adoption of IoT in various sectors is in an uptrend, the actual deployment is lagging when compared to the industrial forecasted data. The main reason behind this drawback is the lifespan of IoT devices due to their limited battery capacities. A solution to the problem is to deploy energy harvesting units (e.g., solar to replenish the batteries). However, due to the time varying availability of both irradiance and temperature and their effect on the power output, it is essential to predict both variables. To this end, we propose in this paper a prediction system that does not consume a lot of energy and that can be deployed on low computational nodes. This model consists of a Seasonal Auto Regressive Integrated Moving Average (SARIMA) with a Kalman filtering (KF) component. We build this model using an actual dataset for Ottawa, Ontario, Canada. We then demonstrate its effectiveness by presenting the results for randomly selected days in the Winter Season. In this context, we show that the SARIMA-KF outperforms the SARIMA in all scenarios with an average error reduction of 59.3%.

M. Azzam, Z. Bouida, and M. Ibnkahla, “Irradiance and Temperature Forecasting for Energy Harvesting Units in IoT Sensors using SARIMA-KF,” in 2022 IEEE Wireless Communications and Networking Conference (WCNC), Apr. 2022, pp. 1701–1706. doi: 10.1109/WCNC51071.2022.9771763.

For more details: 

The wide adoption of large-scale Internet of Things (IoT) systems has led to an unprecedented increase in backhaul traffic congestion, making it critical to optimize traffic management at the network edge. In IoT systems, the backhaul network is supported by various backhauling technologies that have different characteristics. Also, the characteristics of the backhaul links can be sometimes time-varying and have an unknown state, due to external factors such as having the resources shared with other systems. It is the responsibility of the edge devices to be able to forward IoT traffic through the unknown-state backhaul network by selecting the suitable backhaul link for each collected data flow. To the best of our knowledge, this type of backhaul selection problem is not addressed in the literature. Therefore, there is a crucial need to develop intelligent approaches enabling edge devices to learn how to deal with unknown-state (partially observable) components of the backhaul network, which is the primary goal of this paper. We propose an edge-based backhaul selection technique for improving traffic delivery by exploiting multi-objective feedback on delivery performance. The proposed approach relies on the Advantage-Actor-Critic Deep Reinforcement Learning (DRL) methods. Moreover, to improve the DRL training performance in large-scale deployments of distributed IoT systems, Federated Learning (FL) is applied enabling multiple edge devices to collaborate in training a shared backhaul selection policy. The proposed Federated DRL (F-DRL) approach is able to solve the backhaul selection problem as verified and demonstrated through extensive simulations.

A. Jarwan and M. Ibnkahla, “Edge-Based Federated Deep Reinforcement Learning for IoT Traffic Management,” IEEE Internet of Things Journal, pp. 1–1, 2022, doi: 10.1109/JIOT.2022.3174469.

For more details: 

In cluster-based Information-Centric Wireless Sensor Networks (ICWSNs), mobile sensor nodes are grouped into clusters in rounds. In each cluster, a Cluster Head (CH) is selected, which collects, aggregates, and forwards locally sensed data to a sink node. CHs further store a copy of data for the round period to act as cache nodes and deliver data to mobile users upon requests. Nevertheless, clustering and securing mobile ICWSNs is challenging. This is because, in addition to sensor nodes’ and users’ mobility, sensor nodes are often resource-constrained. Therefore, clustering and security resource allocation in mobile ICWSNs should be carefully re-designed to ensure efficient ICWSN operation, data security, and timely data access to mobile users. This paper proposes a Node Embedding with Security Resource Allocation (NESRA) clustering algorithm for mobile ICWSNs in rounds. NESRA allocates security resources to sensor nodes based on the location, mobility, and energy resources available in the first step. An optimization problem is formulated to select CHs that maximizes network coverage and minimizes data delivery delay to mobile users in the second step. In the third step, NESRA utilizes network representation learning that embeds sensor nodes’ location, mobility, and expected energy expenditure features into a 2D space to form well-separated clusters of sensing nodes. Compared to existing works, NESRA achieves lower energy consumption, nodes’ death rate, and latency, and allows higher throughput and cache nodes’ utilization with stable data security. Still, NESRA has some challenges to overcome in high-mobility networks.

A. Gharib and M. Ibnkahla, “Node Embedding for Security-Aware Clustering of Mobile Information-Centric Sensor Networks,” in IEEE Internet of Things Journal, doi: 10.1109/JIOT.2022.3152183.

For more details: