Recently, broadband maritime communication has attracted much attention due to the rapid development of blue economy. In addition to the conventional MF/HF/VHF bands, there has been increasing interests in the utilization of higher frequency bands to provide broadband data service to the sea area. To design efficient maritime communication systems, the first and a fundamental requirement is to develop a framework to understand the wireless channels. In an integrated air-ground-sea communications network, there are two major type of channels to be investigated, namely the air-to-sea channel (e.g., for communication links from the aircraft-based base stations or relays) and the near-sea-surface channel (for land-to-ship/ship-to-land or ship-to-ship communications). Due to the unique features of the maritime propagation environment such as sparse scattering, sea wave movement, and the ducting effect over the sea surface, the modeling of these maritime channel links differs from conventional terrestrial wireless channels in many aspects and, consequently, will result in significant impact on the transceiver design. In this survey, we highlight the most notable differences from the modeling perspective as well as the channel characteristics for the air-to-sea and near-sea-surface channel links, with more focus on the latter. After a thorough review of existing modeling approaches and measurement campaigns, we conclude that the sparse and the location-dependent properties constitute the most important and distinctive characteristics of the maritime wireless channels. As such, we further remark on the challenges and research topics for future development of maritime communications.

Mobile edge computing (MEC) is considered a novel paradigm for computation-intensive and delay-sensitive tasks in fifth generation (5G) networks and beyond. However, its uncertainty, referred to as dynamic and randomness, from the mobile device, wireless channel, and edge network sides, results in high-dimensional, nonconvex, nonlinear, and NP-hard optimization problems. Thanks to the evolved reinforcement learning (RL), upon iteratively interacting with the dynamic and random environment, its trained agent can intelligently obtain the optimal policy in MEC. Furthermore, its evolved versions, such as deep reinforcement learning (DRL), can achieve higher convergence speed efficiency and learning accuracy based on the parametric approximation for the large-scale state-action space. This paper provides a comprehensive research review on RL-enabled MEC and offers insight for development in this area. More importantly, associated with free mobility, dynamic channels, and distributed services, the MEC challenges that can be solved by different kinds of RL algorithms are identified, followed by how they can be solved by RL solutions in diverse mobile applications. Finally, the open challenges are discussed to provide helpful guidance for future research in RL training and learning MEC.

This paper proposes a new artificial noise (AN) scheme that combines with the relay selection techniques to improve the physical layer security for a wiretap full-duplex decode-and-forward cooperative network. The communication consists of two phases. In the first phase, the source transmits a message to the selected full-duplex relay, which simultaneously injects AN to mitigate eavesdropping attacks. In the second phase, the selected relay decodes the source message, XORs the source message and AN, and then broadcasts the mixed-signal to the destination. Different form existing schemes, the proposed AN scheme makes the eavesdropper fail in obtaining any information about the source message directly in the second phase. We take into account the correlation of the channel coefficients for the same link during different phases in the performance analysis of AN-aided max-min (MM) and partial (Par) relay selection schemes, and characterize the security-reliability trade-off (SRT) of the proposed schemes by the closed-form expressions of outage probability and intercept probability. Numerical results show that the proposed AN-aided MM/Par relay selection achieves a better SRT performance than the corresponding full-duplex mode assisted AN-aided relay selection scheme in existing literature. Furthermore, the MM scheme outperforms the Par scheme in terms of the SRT, and the SRT of the MM scheme can be substantially improved by increasing the number of relays, while the improvement of the Par scheme is limited.

The demand for wideband communication in the coastal area (i.e., ≤ 100 km from the coastline) has been rapidly increasing in recent years. Compared to the terrestrial scenario, the coastal environment has long-distance and highly dynamic channels, and the communication devices are more strictly constrained by energy supplies. While the RLNC has the fountain erasure-correction property and is suitable for transmissions over the long-distance dynamic channels, it suffers from high coding coefficient delivery cost and decoding complexity. In this article, we look into the application of sparse network coding in coastal communication systems. We identify two typical multicast scenarios that may appear in coastal communications, namely the relay-aided multicast and multicast from a shore-based base station with D2D communication enabled among the subscribers. We provide a detailed comparison of existing sparse network coding schemes. Based on that, we demonstrate through simulations that an appropriate choice of sparse codes is critical to meet the unique requirements in coastal communication systems. We show that batched sparse code is suitable for relay-aided multicast, and subset-based sparse codes are preferable for D2D-enabled multicast.

Extra-large massive multiple-input multiple-output (XL-MIMO) is envisioned to be a promising technology for 6G wireless, where the number of antennas is greatly increased to a new extent, resulting in extra large aperture arrays. In this case, different users may see different array parts (termed as visible region, VR) due to the spatial non-stationarity. Exploiting the VR information can facilitate low-complexity transmission design, but how to acquire this information for a large amount of users is still challenging. To this end, we first establish a VR model for XL-MIMO systems and show that a user's VR is location-dependent. Assuming that the VRs of some beacon users can be known a priori, we propose three location-based VR recognition schemes, including Voronoi cell partition, weighted Voronoi cell partition, and a neural network (NN) approach termed VR-Net, which takes the location of a user as input and returns its VR index as output. Simulation results show that all schemes can achieve high VR recognition accuracy if the number of beacon users is sufficiently large. Notably, in the more practical scenario that the available location-VR dataset is limited, the proposed VR-Net is able to achieve much better recognition performance.