In response to the increasing number of devices expected in next-generation networks, a shift to over-the-air (OTA) computing has been proposed. By leveraging the superposition of multiple access channels, OTA computing enables efficient resource management by supporting simultaneous uncoded transmission in the time and frequency domains. To advance the integration of OTA computing, our study presents a theoretical analysis that addresses practical issues encountered in current digital communication transceivers, such as transmitter synchronization (sync) errors and intersymbol interference (ISI). To this end, we investigate the theoretical mean squared error (MSE) for OTA transmission under sync errors and ISI, while also exploring methods for minimizing the MSE in OTA transmission. Using alternating optimization, we also derive optimal power policies for both the devices and the base station. In addition, we propose a novel deep neural network (DNN)-based approach to design waveforms that improve OTA transmission performance under sync errors and ISI. To ensure a fair comparison with existing waveforms such as raised cosine (RC) and better-than-raised-cosine (BTRC), we incorporate a custom loss function that integrates energy and bandwidth constraints along with practical design considerations such as waveform symmetry. Simulation results validate our theoretical analysis and demonstrate performance gains of the designed pulse over RC and BTRC waveforms. To facilitate testing of our results without the need to rebuild the DNN structure, we also provide curve-fitting parameters for the selected DNN-based waveforms.

The optimal solution to an optimization problem depends on the problem’s objective function, constraints, and size. While deep neural networks (DNNs) have proven effective in solving optimization problems, changes in the problem’s size, objectives, or constraints often require adjustments to the DNN architecture to maintain effectiveness, or even retraining a new DNN from scratch. Given the nature of wireless networks, which involves multiple and diverse objectives that can have conflicting requirements and constraints, we propose a multi-task learning (MTL) framework to enable a single DNN to jointly solve a range of diverse optimization problems. In this framework, optimization problems with varying dimensionality values, objectives, and constraints are treated as distinct tasks. To jointly address these tasks, we propose a conditional computation-based MTL approach with routing. The multi-task DNN consists of two components, the base DNN (bDNN), which is the single DNN used to extract the solutions for all considered optimization problems, and the routing DNN (rDNN), which manages which nodes and layers of the bDNN to be used during the forward propagation of each task. The output of the rDNN is multiplied with all bDNN’s weights during the forward propagation, creating a unique computational path through the bDNN for each task. This setup allows the tasks to either share parameters or use independent ones, with the decision controlled by the rDNN. The proposed framework supports both supervised and unsupervised learning scenarios. Based on this framework, a soft modularization and a hard parameter sharing approach of lower complexity are proposed. The numerical results demonstrate the efficiency of the proposed MTL approaches compared to several multi-task benchmarks and the single-task DNN approach.

One of the main applications expected to be enabled by next-generation networks is computing. The goal-oriented nature of computing allows the use of different implementation techniques, with over-the-air (OTA) computation being one of the main proposed schemes due to its effective resource management and computational efficiency. In this work, we aim at optimizing the waveform of the system in the presence of intersymbol interference (ISI) and sampling error. To this end, we propose a deep neural network (DNN) framework that generates an optimal waveform that minimizes the mean square error (MSE) of the OTA computation system. To ensure that the generated waveform exhibits the same behavior as other common waveforms, weighted energy and spectrum constraints are included in the loss function of the training phase. To better mitigate ISI, the spectrum constraint integrates the roll-off factor of the waveform, allowing for the generation of different waveforms. Simulation results verify that the desired constraints are met and show a significant performance gain over state-of-the-art waveforms.

In this paper, we propose a novel approach to minimize the inference delay in semantic segmentation using split learning (SL), tailored to the needs of real-time computer vision (CV) applications for resource-constrained devices. Semantic segmentation is essential for applications such as autonomous vehicles and smart city infrastructure, but faces significant latency challenges due to high computational and communication loads. Traditional centralized processing methods are inefficient in such scenarios, often resulting in unacceptable inference delays. SL offers a promising alternative by partitioning deep neural networks (DNNs) between edge devices and a central server, enabling localized data processing and reducing the amount of data required for transmission. Our contribution includes the joint optimization of bandwidth allocation, cut layer selection of the edge devices’ DNN, and the central server’s processing resource allocation. We investigate both parallel and serial data processing scenarios and propose low-complexity heuristic solutions that maintain near-optimal performance while reducing computational requirements. Numerical results show that our approach effectively reduces inference delay, demonstrating the potential of SL to improve real-time CV applications in dynamic, resource-constrained environments.

Deep neural networks (DNNs) have demonstrated their efficacy in delivering accurate solutions to a range of optimization problems. However, in the context of wireless communications, the size of these problems may vary across adjacent time slots, due to fast changes in the networks’ architecture, e.g., the number of users. It is essential to note that this time-varying dimensionality of optimization problems in wireless networks necessitates adjustments in the DNN architecture, resulting in different numbers of input and output nodes. To address this challenge, in our paper, optimization problems of varying size are treated as distinct tasks. To tackle these tasks, a multi-task learning (MTL) approach based on modular sharing is proposed. The multi-task approach consists of a DNN, which is used to extract the solutions for all the optimization problems, and a router which manages which nodes and layers of the input and output layer of the DNN to be used during the forward propagation of each task. Consequently, all tasks share common parameters of the DNN, while the DNN dynamically adjusts to the number of nodes of its output and input layers. Numerical results demonstrate the superiority of the suggested approach over zero-padding, which is the current solution for handling resource allocation problems of varying size.

Unmanned aerial vehicles (UAVs) equipped with full-duplex relays (FDRs) are pivotal in overcoming connectivity challenges by dynamically establishing effective communication channels. However, despite their potential in network performance via trajectory optimization, integrating energy consumption models for UAV-mounted FDRs remains unexplored, crucial for trajectory design adhering to existing energy constraints. To this end, we introduce an energy-aware trajectory optimization framework to maximize network performance and user fairness within the UAV’s energy constraints. Specifically, we present a detailed energy consumption model describing the operational needs of UAV-mounted FDRs and formulate a joint time-division multiple access (TDMA) user scheduling-UAV trajectory optimization problem considering the power dynamics of UAV-mounted FDRs. Finally, our simulation results highlight the role of energy awareness in achieving optimal trajectory and scheduling, contributing to UAV-mounted FDRs’ performance in future networks.

Semantic communications offer a promising approach to decrease network congestion and improve reliability, leading to more sustainable and energy-efficient wireless networks. However, the design of semantic transceivers constrain their effectiveness. This paper introduces a novel multi-carrier system that combines both semantic and Shannon communications, with a focus on text transmission. We formulate an optimization problem that jointly selects the transmission method and allocates power to reduce the transmission delay. Despite the challenges of solving this non-convex problem, we employ alternating optimization techniques to address it and the closed-form solution of the power allocation is extracted. The simulation results verify that jointly selecting semantic and Shannon communications decreases the transmission delay compared to using only one of the schemes.

In the evolving landscape of sixth-generation (6G) wireless networks, unmanned aerial vehicles (UAVs) have emerged as transformative tools for dynamic and adaptive connectivity. However, dynamically adjusting their position to offer favorable communication channels introduces operational challenges in terms of energy consumption, especially when integrating advanced communication technologies like reconfigurable intelligent surfaces (RISs) and full-duplex relays (FDRs). To this end, by recognizing the pivotal role of UAV mobility, the paper introduces an energy-aware trajectory design for UAV-mounted RISs and UAV-mounted FDRs using the decode-and-forward (DF) protocol, aiming to maximize the network’s minimum rate and enhance user fairness, while taking into consideration the available on-board energy. Specifically, this work highlights their distinct energy consumption characteristics and their associated integration challenges by developing appropriate energy consumption models for both UAV-mounted RISs and FDRs that capture the intricate relationship between key factors such as weight, and their operational characteristics. Furthermore, a joint time-division multiple access (TDMA) user scheduling-UAV trajectory optimization problem is formulated, considering the power dynamics of both systems, while assuring that the UAV energy is not depleted mid-air. Finally, simulation results underscore the importance of energy considerations in determining the optimal trajectory and scheduling and provide insights into the performance comparison of UAV-mounted RISs and FDRs in UAV-assisted wireless networks.

This article focuses on the latest research and innovations in fundamental next-generation multiple access (NGMA) techniques and the coexistence with other key technologies for the sixth generation (6G) of wireless networks. In more detail, we first examine multiaccess edge computing (MEC), which is critical to meeting the growing demand for data processing and computational capacity at the edge of the network, as well as network slicing. We then explore over-the-air (OTA) computing, which is considered to be an approach that provides fast and efficient computation of various functions. We also explore semantic communications, identified as an effective way to improve communication systems by focusing on the exchange of meaningful information, thus minimizing unnecessary data and increasing efficiency. The interrelationship between machine learning (ML) and multiple access technologies is also reviewed, with an emphasis on federated learning (FL), federated distillation (FD), split learning (SL), reinforcement learning (RL), and the development of ML-based multiple access protocols. Finally, the concept of digital twinning and its role in network management is discussed, highlighting how virtual replication of physical networks can lead to improvements in network efficiency and reliability.

Semantic communications are considered a promising beyond-Shannon paradigm to reduce network traffic and increase reliability, thus making wireless networks more energy efficient, robust, and sustainable. However, the performance is limited by the efficiency of the semantic transceivers, i.e., the achievable “similarity” between the transmitted and received signals. Under strict similarity conditions, semantic transmission may not be applicable and Shannon communication is mandatory. In this paper, for the first time in the literature, we propose a multi-carrierHybrid Semantic-Shannoncommunication system where, without loss of generality, the case of text transmission is investigated. To this end, a joint semantic-Shannon transmission selection and power allocation optimization problem is formulated, aiming to minimize two transmission delay metrics widely used in the literature, subject to strict similarity thresholds. Despite their non-convexity, both problems are decomposed into a convex and a mixed linear integer programming problem by using alternating optimization, both of which can be solved optimally. Furthermore, to improve the performance of the proposed hybrid schemes, a novel association of text sentences to subcarriers is proposed based on the data size of the sentences and the channel gains of the subcarriers. We show that the proposed association is optimal in terms of transmission delay. Numerical simulations verify the effectiveness of the proposed hybrid semantic-Shannon communication scheme and the derived sentence-to-subcarrier association, and provide useful insights into the design parameters of such systems.

Distributed optimization is ubiquitous in emerging applications, such as robust sensor network control, smart grid management, machine learning, resource slicing, and localization. However, the extensive data exchange among local and central nodes may cause a severe communication bottleneck. To overcome this challenge, over-the-air computing (AirComp) is a promising medium access technology, which exploits the superposition property of the wireless multiple access channel (MAC) and offers significant bandwidth savings. In this work, we propose an AirComp framework for general distributed convex optimization problems. Specifically, a distributed primal-dual (DPD) subgradient method is utilized for the optimization procedure. Under general assumptions, we prove that DPD-AirComp can asymptotically achieve zero expected constraint violation. Therefore, DPD-AirComp ensures the feasibility of the original problem, despite the presence of channel fading and additive noise. Moreover, with proper power control of the users’ signals, the expected non-zero optimality gap can also be mitigated. Two practical applications of the proposed framework are presented, namely, smart grid management and wireless resource allocation. Finally, numerical results confirm DPD-AirComp’s excellent performance, while it is also shown that DPD-AirComp converges an order of magnitude faster compared to two digital orthogonal multiple access schemes, specifically, time-division multiple access (TDMA), and orthogonal frequency-division multiple access (OFDMA).

We introduce a set of multiple access protocols, calledhierarchical multiple access (HiMA), which are based on non-orthogonal multiple access (NOMA) and time-division multiple access (TDMA), optimized for the hierarchical network scenario. The proposed protocols can be efficiently utilized in various network configurations with an hierarchical form, such as relay networks, cloud-radio access networks (C-RANs), and fog-radio access networks (F-RANs). In particular, C-RANs and, more recently, F-RANs are regarded as promising paradigms to fully utilize the edge of the networks. Therefore, the implementation of novel multiple access protocols to properly exploit these configurations is critical for the fifth generation and beyond of wireless access. Furthermore, the resource allocation problem is formulated for each protocol with respect to the timeslot duration and power. As a result two fairness metrics are optimized, namely max-min rate fairness and proportional fairness. Finally, numerical results reveal the effectiveness of the joint design in the hierarchical network and an interesting trade-off is identified between fairness and achievable rate. Interestingly, despite NOMA being a very promising alternative to conventional multiple access schemes, the protocol that is solely based on NOMA does not always outperform the rest.