The wide-spread Internet of Things (IoT) utilization in almost every scope of our life made it possible to automate daily life tasks with no human intervention. This promising technology has immense potential for making life much easier and open new opportunities for newly developed applications to emerge. However, meeting the diverse Quality of Service (QoS) demands of different applications remains a formidable topic due to diverse traffic patterns, unpredictable network traffic, and resource-limited nature of IoT devices. In this context, application-tailored QoS provisioning mechanisms have been the primary focus of academic research. This paper presents a literature review on QoS techniques developed in academia for IoT applications and investigates current research trends. Background knowledge on IoT, QoS metrics, and critical enabling technologies will be given beforehand, delving into the literature review. According to the comparison presented in this work, the commonly considered QoS metrics are Latency, Reliability, Throughput, and Network Usage. The reviewed studies considered the metrics that fit their provisioning solutions.

Internet of Things (IoT) systems usually produce massive amounts of data, while the number of devices connected to the internet might reach billions by now. Sending all this data over the internet will overhead the cloud and consume bandwidth. Fog computing's (FC) promising technology can solve the issue of computing and networking bottlenecks in large-scale IoT applications. This technology complements the cloud computing by providing processing power and storage to the edge of the network. However, it still suffers from performance and security issues. Thus, machine learning (ML) attracts attention for enabling FC to settle its issues. Lately, there has been a growing trend in utilizing ML to improve FC applications, like resource management, security, lessen latency and power usage. Also, intelligent FC was studied to address issues in industry 4.0, bioinformatics, blockchain and vehicular communication system. Due to the ML vital role in the FC paradigm, this work will shed light on recent studies utilized ML in a FC environment. Background knowledge about ML and FC also presented. This paper categorized the surveyed studies into three groups according to the aim of ML implementation. These studies were thoroughly reviewed and compared using sum-up tables. The results showed that not all studies used the same performance metric except those worked on security issues. In conclusion, the simulations of proposed ML models are not sufficient due to the heterogeneous nature of the FC paradigm.

Clustering schemes in vehicular networks organize vehicles into logical groups. They are vital for improving network performance, accessing the medium, and enabling efficient data dissemination. Most schemes rely on periodically broadcast hello messages to provide up-to-date information about the vehicles. However, the periodic exchange of messages overwhelms the system and reduces efficiency. This paper proposes the Flag-based Vehicular Clustering (FVC) scheme. The scheme leverages a combination of Fitness Score (FS), Link Expiration Time (LET), and clustering status flags to enable efficient cluster formation in a hybrid manner. The FVC relies on the periodic broadcast of the basic safety message in the Dedicated Short-Range Communications (DSRC) standard for exchanging the vehicle’s status, FS, and joining request. Piggybacking extra information onto the existing periodic beacon reduces the overhead of exchanging additional control messages, which is the main contribution of this work. The scheme is implemented in a hybrid manner by utilizing a Road Side Unit (RSU) to implement a clustering algorithm. This work considered the FastPAM algorithm, a fast version of the Partitioning Around Medoids (PAM) clustering algorithm, to generate a list of potential cluster heads. The FVC scheme uses the LET as the clustering metric with the FastPAM algorithm. Moreover, the Lightweight FastPAM Vehicular Clustering (LFPVC) algorithm is considered by selecting the initial cluster heads based on the FS instead of the greedy FastPAM’s build stage. In the absence of the RSU, the vehicles utilize the FS with proper back-off time to self-elect the cluster head. The hybrid FVC scheme increased the cluster lifetime by 32% and reduced the control-message overhead by 63% compared to the related work. Moreover, the LFPVC algorithm achieved similar results to the FastPAM algorithm.

Implementing machine learning algorithms in the non-conducive environment of the vehicular network requires some adaptations due to the high computational complexity of these algorithms. K-clustering algorithms are simplistic, with fast performance and relative accuracy. However, their implementation depends on the initial selection of clusters number (K), the initial clusters’ centers, and the clustering metric. This paper investigated using Scott’s histogram formula to estimate the K number and the Link Expiration Time (LET) as a clustering metric. Realistic traffic flows were considered for three maps, namely Highway, Traffic Light junction, and Roundabout junction, to study the effect of road layout on estimating the K number. A fast version of the PAM algorithm was used for clustering with a modification to reduce time complexity. The Affinity propagation algorithm sets the baseline for the estimated K number, and the Medoid Silhouette method is used to quantify the clustering. OMNET++, Veins, and SUMO were used to simulate the traffic, while the related algorithms were implemented in Python. The Scott’s formula estimation of the K number only matched the baseline when the road layout was simple. Moreover, the clustering algorithm required one iteration on average to converge when used with LET.