Deep learning (DL) has revolutionized wireless communication systems by introducing data-driven end-to-end (E2E) learning, where the physical layer (PHY) is transformed into DL architectures to achieve peak optimization. Leveraging DL for E2E optimization in PHY significantly enhances its adaptability and performance in complex wireless environments, meeting the demands of advanced network systems such as 5G and beyond. Furthermore, this evolution of data-driven PHY optimization has also enabled advanced semantic applications across various modalities, including text, image, audio, video, and multimodal transmissions. These applications elevate communication from bit-level to semantic-level intelligence, making it capable of discerning context and intent. Although the PHY, as a DL architecture, plays a crucial role in enabling semantic communication (SemCom) systems, comprehensive studies that integrate both E2E communication and SemCom systems remain significantly underexplored. This highlights the novelty and potential of these integrative fields, marking them as a promising research domain. Therefore, this article provides a comprehensive review of the emerging field of data-driven PHY for E2E communication systems, emphasizing their role in enabling semantic applications across various modalities. It also identifies key challenges and potential research directions, serving as a crucial guide for future advancements in DL for E2E communication and SemCom systems.

Underwater wireless sensor networks (UWSNs) are a prominent research topic in academia and industry, with many applications such as ocean, seismic, environmental, and seabed explorations. The main challenges in deploying UWSN are high ocean interference and noise, which results in longer propagation time, low bandwidth, and changes in network topology. To mitigate these problems, routing protocols have been identified as an efficient solution. Over the years, several protocols have been proposed in this direction and among them, the most popular are the ones that use multi-path propagation. However, there is a lack of compilation of studies that highlight the advancement of multi-path routing protocols of UWSN through the years. Hence, getting a heuristic idea of the existing protocols is crucial. In this study, we present a comprehensive survey of UWSNs multi-path routing protocols and categorize them into three main categories; energy-based routing protocols, geographic information-based routing protocols, and data-based routing protocols. Furthermore, we sub-classify them into several categories and identify their advantages and disadvantages. In addition, we identify the application of UWSN, open challenges and compare the protocols. The findings of our study will allow researchers to better understand different categories of UWSN multi-path routing protocols in terms of their scope, advantages, and limitations.

Various advanced and mission-critical applications are enabled by the emerging technologies in fifth-generation (5G) mobile communication systems. To ensure improved quality of experience (QoE) of users, 5G and beyond networks require ultra-reliable low-latency communications (URLLC). The successful realization of the URLLC entails the advent of new technological concepts. Therefore, this article presents an overview of the enabling techniques for the URLLC. Classification of the enabling techniques is done and an extensive review of the literature is presented to identify the state-of-the-art techniques, limitations, and the potential approaches for alleviating the limitations. It is observed that artificial intelligence (AI)-enabled edge computing and caching solutions are widely explored as promising techniques to effectively guarantee low latency and reliable content acquisition while reducing redundant network traffic and improving the QoE. Therefore, we present a classification of the AI-enabled edge caching solutions and discuss various mechanisms of the caching agents. In particular, we investigate the use of deep learning (DL), deep reinforcement learning (DRL), and federated learning (FL) algorithms. Subsequently, we analyze the performance of the state-of-the-art edge caching schemes and demonstrate the performance gains of FL frameworks over conventional centralized and decentralized DL and DRL frameworks. We confirm that FL edge caching is a viable mechanism in 5G and beyond networks. On the other hand, it is shown that the IEEE 802.1 time sensitive networking and the emerging IETF deterministic networking standards present effective mechanisms when deterministic networks with bounded ultra-low latency are considered. Finally, we present the open issues and opportunities for further research.

Achieving high source location privacy is critical when Wireless Sensor Networks (WSNs) are used in sensitive applications such as in asset or battlefield monitoring. Due to the sensitivity of information in these applications, it is important to ensure the flow of data between sensor nodes is secure and it does not expose any information about the monitored assets to an adversary. This paper proposes a routing scheme with stronger source location privacy than the privacy of traditional routing schemes. The paper addresses some limitations of four existing schemes by providing highly random routing paths between the source nodes and sink node. The scheme randomly sends packet to the sink node through tactically positioned proxy nodes to guarantee the routes are highly befuddling to the adversary. To achieve high privacy, the proposed scheme uses a randomizing factor to generate a new random route for every successive packet. Simulation results demonstrate that the proposed scheme provides longer safety period and stronger privacy to outperform other schemes. Furthermore the scheme provides stronger privacy against both, patient and cautious adversary models.

Wireless sensor networks (WSNs) are deployed in sensitive applications, such as in military and asset monitoring. In these applications, it is important to ensure good source location privacy. This is owing to the open nature of WSNs and the easiness of an adversary to eavesdrop on sensor communication and back trace the location of the source node. This paper proposes a scheme to preserve the source location privacy based on random routing techniques. To achieve high privacy, packets are randomly routed from the source to the sink node through strategically positioned mediate or diversion nodes. The random selection of mediate or diversion nodes is location-based. Depending on the location of the source node, packets are forwarded through different regions of the network. The proposed scheme guarantees that successive packets are routed through very different routing paths and adversaries find it confusing to back trace them to the source node location. Simulation results demonstrate that the proposed scheme effectively confuses the adversary and provides higher source location privacy to outperform other routing-based source location privacy schemes.

In wireless sensor networks, clustering routing algorithms have been widely used owing to their high energy-efficiency and scalability. In clustering schemes, the nodes are organized in the form of clusters, and each cluster is governed by a cluster head. Once the cluster heads are selected, they form a backbone network to periodically collect, aggregate, and forward data to the base station using minimum energy (cost) routing. This approach significantly improves the network lifetime. Therefore, a new cluster head selection method that uses a weighted sum method to calculate the weight of each node in the cluster and compare it with the standard weight of that particular cluster is proposed in this paper. The node with a weight closest to the standard cluster weight becomes the cluster head. This technique balances the load distribution and selects the nodes with highest residual energy in the network. Additionally, a data routing scheme is proposed to determine an energy-efficient path from the source to the destination node. This algorithm assigns a weight function to each link on the basis of a fuzzy membership function and intra-cluster communication cost within a cluster. As a result, a minimum weight path is selected using Dijkstra’s algorithm that improves the energy efficiency of the overall system. The experimental results show that the proposed algorithm shows better performance than some existing representative methods in the aspects of energy consumption, network lifetime, and system throughput.