Federated Learning for 6G:Applications,Challenges,and Opportunities

Standard machine-learning approaches involve the centralization of training data in a data center,where centralized machine-learning algorithms can be applied for data analysis and inference.However,due to privacy restrictions and limited communication resources in wireless networks,it is often undesirable or impractical for the devices to transmit data to parameter sever.One approach to mitigate these problems is federated learning(FL),which enables the devices to train a common machine learning model without data sharing and transmission.This paper provides a comprehensive overview of FL applications for envisioned sixth generation(6G)wireless networks.In particular,the essential requirements for applying FL to wireless communications are first described.Then potential FL applications in wireless communications are detailed.The main problems and challenges associated with such applications are discussed.Finally,a comprehensive FL implementation for wireless communications is described.

Recent Advances in Data-Driven Wireless Communication Using Gaussian Processes: A Comprehensive Survey

Data-driven paradigms are well-known and salient demands of future wireless communication. Empowered by big data and machine learning techniques,next-generation data-driven communication systems will be intelligent with unique characteristics of expressiveness, scalability, interpretability, and uncertainty awareness, which can confidently involve diversified latent demands and personalized services in the foreseeable future. In this paper, we review a promising family of nonparametric Bayesian machine learning models,i.e., Gaussian processes(GPs), and their applications in wireless communication. Since GP models demonstrate outstanding expressive and interpretable learning ability with uncertainty, they are particularly suitable for wireless communication. Moreover, they provide a natural framework for collaborating data and empirical models(DEM). Specifically, we first envision three-level motivations of data-driven wireless communication using GP models. Then, we present the background of the GPs in terms of covariance structure and model inference. The expressiveness of the GP model using various interpretable kernels, including stationary, non-stationary, deep and multi-task kernels,is showcased. Furthermore, we review the distributed GP models with promising scalability, which is suitable for applications in wireless networks with a large number of distributed edge devices. Finally, we list representative solutions and promising techniques that adopt GP models in various wireless communication applications.

Transmission Energy Allocation for Over-the-Air Computation with Energy Harvesting

Over-the-air computation(AirComp)has recently emerged as a promising multiple-access technique for fast wireless data aggregation(WDA)from distributed wireless devices(WDs). This paper investigates an energy harvesting (EH) AirComp system, in which multiple EH-powered single-antenna WDs simultaneously send wireless signals to a single-antenna access point (AP) with conventional energy supply for WDA via AirComp. Under this setup, we minimize the average computation mean square error(MSE)over a particular time period, by jointly optimizing the transmit energy allocation at the WDs and the AirComp denoising factors at the AP over time, subject to the energy causality constraints at individual WDs. First, we consider the offline scenario by assuming that the energy state information(ESI)and channel state information (CSI) are non-causally known at the beginning of the period, in which the formulated average MSE minimization corresponds to a non-convex optimization problem. We present a high-quality converged solution by using the techniques of alternating optimization and convex optimization. It is shown that for each WD,if the EH rate is sufficiently high,then the channel inversion power allocation is adopted;while if the EH rate is low, then all the harvested energy should be used up for transmission with proper energy allocation over time. Next, we consider the online scenario with causal ESI and CSI,in which the MSE minimization becomes a stochastic optimization problem.In this scenario, we present an offline-inspired online algorithm to obtain efficient online energy allocation designs by utilizing the obtained offline solutions. Finally,numerical results show that the proposed designs significantly outperform two benchmark schemes with power-halving and full-power transmission,respectively.

Numerical evaluation on sub-Nyquist spectrum reconstruction methods

1 Introduction As wireless technology continues to expand,there is a growing concern about the efficient use of spectrum resources.Even though a significant portion of the spectrum is allocated to licensed primary users(PUs),studies indicate that their actual utilization is often limited to between 5%to 10%[1].The underutilization of spectrum has given rise to cognitive radio(CR)technology,which allows secondary users(SUs)to opportunistically access these underused resources[2].However,wideband spectrum sensing,the key of CR,is limited by the need for high-speed analog-to-digital converters(ADCs),which are costly and power-hungry.Compressed spectrum sensing(CSS)addresses this challenge by employing sub-Nyquist rate sampling.The efficiency of active transmission detection heavily depends on the quality of spectrum reconstruction.

Theory and techniques for“intellicise”wireless networks

With the acceleration of a new round of global scientific,technological,and industrial revolution,the next generation of information and communication technology,i.e.,6G,will inject new momentum into industry transformation and upgrading,as well as into economic innovation and development.This will subsequently promote a global industrial integration.Wireless communication will be ubiquitous in all areas of future society,supporting novel applications with various performance requirements.

Sub-Nyquist spectrum sensing and learning challenge

1 Introduction The fact that the spectrum resource is underutilised in certain bands has motivated the dynamic spectrum access(DSA)approach,which enables unlicensed secondary users(SUs)equipped with cognitive radio(CR)devices to access the spectrum without causing significant interference to primary users(PUs).Nowadays,the increasing bandwidth for wireless communication in millimetre-wave and Terahertz frequency bands puts higher requirements on the performance of spectrum sensing technique,the primary enabler of DSA.

TV White Space Spectrum Analysis Based on Machine Learning

Exploration of TV white space(TVWS)is a promising solution to mitigate the spectrum shortage and provide opportunities for new applications.In this paper,we present a detailed analysis of spectrum utilisation over TVWS at different locations in London.Both short-term and long-term outdoor measurement campaigns are conducted over large scales to better understand the spectrum features and variations across multiple locations and time periods.Different from most fixed-location-only measurements,we also drive along the main streets of London with a portable moving node to measure the on-route spectrum density along with the corresponding geographical information,which allows us to study the features and variations of spectrum use through a continuous space.To better analyse the dynamic spectrum utilisation,a machine learning based analysis algorithm is developed over the real-world measurements.This approach allows us to characterise the similarity and variability in spectrum usage within and among different channels,locations,and time instances,which is critical for the secondary system deployment to efficiently exploit the white space.