Distributed machine learning (DML) techniques, such as federated learning, partitioned learning, and distributed reinforcement learning, have been increasingly applied to wireless communications. This is due to improved capabilities of terminal devices, explosively growing data volume, congestion in the radio interfaces, and increasing concern of data privacy. The unique features of wireless systems, such as large scale, geographically dispersed deployment, user mobility, and the massive amount of data, give rise to new challenges in the design of DML techniques. There is a clear gap in the existing literature that the DML techniques are yet to be systematically reviewed for their applicability to wireless systems. This survey bridges the gap by providing a contemporary and comprehensive survey of DML techniques with a focus on wireless networks. Specifically, we review the latest applications of DML in power control, spectrum management, user association, and edge cloud computing. The optimality, accuracy, convergence rate, computation cost, and communication overhead of DML are analyzed. We also discuss the potential adversarial attacks faced by DML applications, and describe state-of-the-art countermeasures to preserve privacy and security. Last but not least, we point out a number of key issues yet to be addressed, and collate potentially interesting and challenging topics for future research.

The advancements in smart power grid and the advocation of “green communications” have inspired the wireless communication networks to harness energy from ambient environments and operate in an energy-efficient manner for economic and ecological benefits. This article presents a contemporary review of recent breakthroughs on the utilization, redistribution, trading and planning of energy harvested in future wireless networks interoperating with smart grids. This article starts with classical models of renewable energy harvesting technologies. We embark on constrained operation and optimization of different energy harvesting wireless systems, such as point-to-point, multipoint-to-point, multipoint-to-multipoint, multi-hop, and multi-cell systems. We also review wireless power and information transfer technologies which provide a special implementation of energy harvesting wireless communications. A significant part of the article is devoted to the redistribution of redundant (unused) energy harvested within cellular networks, the energy planning under dynamic pricing when smart grids are in place, and two-way energy trading between cellular networks and smart grids. Applications of different optimization tools, such as convex optimization, Lagrangian dual-based method, subgradient method, and Lyapunov-based online optimization, are compared. This article also collates the potential applications of energy harvesting techniques in emerging (or upcoming) 5G/B5G communication systems. It is revealed that an effective redistribution and two-way trading of energy can significantly reduce the electricity bills of wireless service providers and decrease the consumption of brown energy. A list of interesting research directions are provided, requiring further investigation.

The high computational complexity and energy consumption of artificial intelligence (AI) algorithms hinder their application in augmented reality (AR) systems. However, mobile edge computing (MEC) makes it possible to solve this problem. This paper considers the scene of completing video-based AI inference tasks in the MEC system. We formulate a mixed-integer nonlinear programming problem (MINLP) to reduce inference delays, energy consumption and to improve recognition accuracy. We give a simplified expression of the inference complexity model and accuracy model through derivation and experimentation. The problem is then solved iteratively by using alternating optimization. Specifically, by assuming that the offloading decision is given, the problem is decoupled into two sub-problems, i.e., the resource allocation problem for the devices set that completes the inference tasks locally, and that for the devices set that offloads tasks. For the problem of offloading decision optimization, we propose a Channel-Aware heuristic algorithm. To further reduce the complexity, we propose an alternating direction method of multipliers (ADMM) based distributed algorithm. The ADMM-based algorithm has a low computational complexity that grows linearly with the number of devices. Numerical experiments show the effectiveness of proposed algorithms. The trade-off relationship between delay, energy consumption, and accuracy is also analyzed.

Federated learning (FL) is a viable technique to train a shared machine learning model without sharing data. Hierarchical FL (HFL) system has yet to be studied regrading its multiple levels of energy, computation, communication, and client scheduling, especially when it comes to clients relying on energy harvesting to power their operations. This paper presents a new two-phase deep deterministic policy gradient (DDPG) framework, referred to as “TP-DDPG”, to balance online the learning delay and model accuracy of an FL process in an energy harvesting-powered HFL system. The key idea is that we divide optimization decisions into two groups, and employ DDPG to learn one group in the first phase, while interpreting the other group as part of the environment to provide rewards for training the DDPG in the second phase. Specifically, the DDPG learns the selection of participating clients, and their CPU configurations and the transmission powers. A new straggler-aware client association and bandwidth allocation (SCABA) algorithm efficiently optimizes the other decisions and evaluates the reward for the DDPG. Experiments demonstrate that with substantially reduced number of learnable parameters, the TP-DDPG can quickly converge to effective polices that can shorten the training time of HFL by 39.4% compared to its benchmarks, when the required test accuracy of HFL is 0.9.

This paper proposes a new fully decentralized approach to online placement and optimization of virtual machines (VMs) for network functions virtualization (NFV). The approach is of practical value, as network services comprising a chain of virtual network functions (VNFs) are proposed to be queued on the basis of leading unexecuted VNFs at every server, rather than on the typical basis of services, reducing queues per server and facilitating queue management and signaling. It is also non-trivial because the VNFs of network services must be executed correctly in order at different VMs, coupling the optimal decisions of VMs on processing or offloading. Exploiting Lyapunov optimization techniques, we decouple the optimal decisions by deriving and minimizing the instantaneous upper bound of the NFV cost in a distributed fashion, and achieve the asymptotically minimum time-average cost. We also reduce the queue length by allowing individual VMs to (un)install VNFs based on local knowledge, achieving stable redeployment of VNFs, adapting to the network topology and the temporal and spatial variations of services. Simulations show that the proposed approach is able to reduce the time-average cost of NFV by 71% and reduce the queue length (or delay) by 74%, as compared to existing approaches.

Mobile edge computing (MEC) offers an opportunity for devices relying on wireless power transfer (WPT), to accomplish computationally demanding tasks. Such WPT-powered MEC systems have yet to be optimized for long-term efficiency, due to random and changing task demands and wireless channel states of the devices. This paper presents an augmented two-staged deep Q-network (DQN), referred to as “TS-DQN”, for online optimization of WPT-powered MEC systems, where the WPT, offloading schedule, channel allocation, and the CPU configurations of the edge server and devices are jointly optimized to minimize the long-term average energy requirement of the systems. The key idea is to design a DQN for learning the channel allocation and task admission, while the WPT, offloading time and CPU configurations are efficiently optimized to precisely evaluate the reward of the DQN and substantially reduce its action space. Another important aspect is that a new action generation method is developed to expand and diversify the actions of the DQN, further accelerating its convergence. As validated by simulations, the proposed TS-DQN is much more energy efficient and converges much faster, than its potential alternative directly using the state-of-the-art Deep Deterministic Policy Gradient algorithm to learn all decision variables.