Non-independent and identically distributed (non-IID) data distribution among clients is considered as the key factor that degrades the performance of federated learning (FL). Several approaches to handle non-IID data, such as personalized FL and federated multitask learning (FMTL), are of great interest to research communities. In this work, first, we formulate the FMTL problem using Laplacian regularization to explicitly leverage the relationships among the models of clients for multitask learning. Then, we introduce a new view of the FMTL problem, which, for the first time, shows that the formulated FMTL problem can be used for conventional FL and personalized FL. We also propose two algorithms FedU and decentralized FedU (dFedU) to solve the formulated FMTL problem in communication-centralized and decentralized schemes, respectively. Theoretically, we prove that the convergence rates of both algorithms achieve linear speedup for strongly convex and sublinear speedup of order 1/2 for nonconvex objectives. Experimentally, we show that our algorithms outperform the conventional algorithm FedAvg, FedProx, SCAFFOLD, and AFL in FL settings, MOCHA in FMTL settings, as well as pFedMe and Per-FedAvg in personalized FL settings.

We consider network-assisted full-duplex (NAFD) cell-free massive multiple-input multiple-output (CF-mMIMO) systems, where full-duplex (FD) transmission is virtually realized via half-duplex (HD) hardware devices. The HD access points (APs) operating in uplink (UL) mode and those operating in downlink (DL) mode simultaneously serve DL and UL user equipments (UEs) in the same frequency bands. We comprehensively analyze the performance of NAFD CF-mMIMO from both a spectral efficiency (SE) and energy efficiency (EE) perspectives. Specifically, we propose a joint optimization approach that designs the AP mode assignment, power control, and large-scale fading (LSFD) weights to improve the sum SE and EE of NAFD CF-mMIMO systems. We formulate two mixed-integer nonconvex optimization problems of maximizing the sum SE and EE, under realistic power consumption models, and the constraints on minimum individual SE requirements, maximum transmit power at each DL AP and UL UE. The challenging formulated problems are transformed into tractable forms and two novel algorithms are proposed to solve them using successive convex approximation techniques. More importantly, our approach can be applied to jointly optimize power control and LSFD weights for maximizing the sum SE and EE of HD and FD CF-mMIMO systems, which, to date, has not been studied. Numerical results show that: (a) our joint optimization approach significantly outperforms the heuristic approaches in terms of both sum SE and EE; (b) in CF-mMIMO systems, the NAFD scheme can provide approximately 30% SE gains, while achieving a remarkable EE gain of up to 200% compared with the HD and FD schemes.

This work proposes novel synchronous, asynchronous, and session-based designs for energy-efficient massive multiple-input multiple-output networks to support federated learning (FL). The synchronous design relies on strict synchronization among users when executing each FL communication round, while the asynchronous design allows more flexibility for users to save energy by using lower computing frequencies. The session-based design splits the downlink and uplink phases in each FL communication round into separate sessions. In this design, we assign users such that one of the participating users in each session finishes its transmission and does not join the next session. As such, more power and degrees of freedom will be allocated to unfinished users, resulting in higher rates, lower transmission times, and hence, higher energy efficiency. In all three designs, we use zero-forcing processing for both uplink and downlink, and develop algorithms that optimize user assignment, time allocation, power, and computing frequencies to minimize the energy consumption at the base station and users, while guaranteeing a predefined maximum execution time of each FL communication round.