We consider a cell-free massive multiple-input multiple-output (MIMO) system with multiple antennas on the users and access points (APs). In previous works, the downlink spectral efficiency (SE) has been evaluated using the hardening bound that requires no downlink pilots. This approach works well for single-antenna users. In this paper, we show that much higher SEs can be achieved if downlink pilots are sent when having multi-antenna users. The reason is that the effective channel matrix does not harden. We propose a pilot-based downlink estimation scheme, derive a new SE expression, and show numerically that it yields substantially higher performance when having correlated Rayleigh fading channels. In cases with multi-antenna users, the APs can either coherently transmit the same data streams, or alternatively they transmit separate data streams non-coherently. The latter approach reduces the fronthaul signaling overhead, albeit with a potential SE penalty. For both strategies, we propose novel precoding and combining schemes. Specifically, we develop precoders based on the minimum mean square error (MMSE) criterion and investigate receive combining methods, including an MMSE combiner. Furthermore, we consider sharing different levels of channel knowledge between the APs. Finally, we present a comprehensive numerical analysis to validate our findings, evaluating the performance trade-offs associated with the number of users, APs, and antennas, as well as the choice of transmission strategy and the level of CSI sharing among APs.

This work investigates a collaborative sensing and data collection system in which multiple unmanned aerial vehicles (UAVs) sense an area of interest and transmit images to a cloud server (CS) for processing. To accelerate the completion of sensing missions, including data transmission, the sensing task is divided into individual private sensing tasks for each UAV and a common sensing task that is executed by all UAVs to enable cooperative transmission. Unlike existing studies, we explore the use of an advanced cell-free multiple-input multiple-output (MIMO) network, which effectively manages inter-UAV interference. To further optimize wireless channel utilization, we propose a hybrid transmission strategy that combines time-division multiple access (TDMA), non-orthogonal multiple access (NOMA), and cooperative transmission. The problem of jointly optimizing task splitting ratios and the hybrid TDMA-NOMA-cooperative transmission strategy is formulated with the objective of minimizing mission completion time. Extensive numerical results demonstrate the effectiveness of the proposed task allocation and hybrid transmission scheme in accelerating the completion of sensing missions.

We consider a cell-free massive MIMO system with multiple antennas on the users and access points. In previous works, the downlink spectral efficiency (SE) has been evaluated using the hardening bound that requires no downlink pilots. This approach works well when having single-antenna users. In this paper, we show that much higher SEs can be achieved if downlink pilots are sent since the effective channel matrix does not harden when having multi-antenna users. We propose a pilot-based downlink estimation scheme and derive a new SE expression that utilizes zero-forcing combining. We show numerically how the number of users and user antennas affects the SE.

This work proposes a novel deep learning approach to tackle multitask optimization problems in multi-user multi-antenna downlink systems. In practice, there is a tradeoff between maximizing the weighted sum spectral efficiency (WSSE) and weighted sum energy efficiency (WSEE) in wireless systems. Traditional beamforming algorithms face limitations in jointly addressing multiple optimization tasks, as they heavily rely on task-specific processes aimed at maximizing specific metrics. As a result, the multiple computations to deal with the multitask problems lead to poor computation and memory efficiency at the base station (BS), which is a challenging aspect to overcome. To address these issues, we present a novel multitask learning approach that effectively achieves the desired tradeoff while reducing the memory burden. We demonstrate the advantages of the proposed scheme that utilizes a single neural network over both existing model-based and data-driven algorithms.

Deep learning (DL) techniques have been intensively studied for the optimization of multi-user multiple-input single-output (MU-MISO) downlink systems owing to the capability of handling nonconvex formulations. However, the fixed computation structure of existing deep neural networks (DNNs) lacks flexibility with respect to the system size, i.e., the number of antennas or users. This paper develops a bipartite graph neural network (BGNN) framework, a scalable DL solution designed for multi-antenna beamforming optimization. The MU-MISO system is first characterized by a bipartite graph where two disjoint vertex sets, each of which consists of transmit antennas and users, are connected via pairwise edges. These vertex interconnection states are modeled by channel fading coefficients. Thus, a generic beamforming optimization process is interpreted as a computation task over a weighted bipartite graph. This approach partitions the beamforming optimization procedure into multiple suboperations dedicated to individual antenna vertices and user vertices. Separated vertex operations lead to scalable beamforming calculations that are invariant to the system size. The vertex operations are realized by a group of DNN modules that collectively form the BGNN architecture. Identical DNNs are reused at all antennas and users so that the resultant learning structure becomes flexible to the network size. Component DNNs of the BGNN are trained jointly over numerous MU-MISO configurations with randomly varying network sizes. As a result, the trained BGNN can be universally applied to arbitrary MU-MISO systems. Numerical results validate the advantages of the BGNN framework over conventional methods.

We consider the conductivity problem with a simply connected or multi-coated inclusion in two dimensions. The potential perturbation due to an inclusion admits a classical multipole expansion whose coefficients are the so-called generalized polarization tensors (GPTs). The GPTs have been fundamental building blocks in conductivity inclusion problems. In this paper, we present a new concept of geometric multipole expansion and its expansion coefficients, named the Faber polynomial polarization tensors (FPTs), using the conformal mapping and the Faber polynomials associated with the inclusion. The proposed expansion leads us to a series solution method for a simply connected or multi-coated inclusion of general shape, while the classical expansion leads us to a series solution only for a single- or multilayer circular inclusion. We also provide matrix expressions for the FPTs using the Grunsky matrix of the inclusion. In particular, for the simply connected inclusion with extreme conductivity, the FPTs admit simple formulas in terms of the conformal mapping associated with the inclusion. As an application of the concept of the FPTs, we construct semi-neutral inclusions of general shape that show relatively negligible field perturbations for low-order polynomial loadings. These inclusions are of the multilayer structure whose material parameters are determined such that some coefficients of geometric multipole expansion vanish.

Autoencoder (AE) techniques have been intensively studied for the optimization of wireless transceivers. However, fixed computational structures of existing AE models lack the flexibility to the lengths of message bits and codewords. This work proposes a versatile AE framework, termed by autoencoding graph neural network (AEGNN), where both encoder and decoder are realized by GNNs. The viability of the proposed AEGNN is demonstrated in various application scenarios.

Fog radio access networks (F-RANs), which consist of a cloud and multiple edge nodes (ENs) connected via fronthaul links, have been regarded as promising network architectures. The F-RAN entails a joint optimization of cloud and edge computing as well as fronthaul interactions, which is challenging for traditional optimization techniques. This paper proposes a Cloud-Enabled Cooperation-Inspired Learning (CECIL) framework, a structural deep learning mechanism for handling a generic F-RAN optimization problem. The proposed solution mimics cloud-aided cooperative optimization policies by including centralized computing at the cloud, distributed decision at the ENs, and their uplink-downlink fronthaul interactions. A group of deep neural networks (DNNs) are employed for characterizing computations of the cloud and ENs. The forwardpass of the DNNs is carefully designed such that the impacts of the practical fronthaul links, such as channel noise and signling overheads, can be included in a training step. As a result, operations of the cloud and ENs can be jointly trained in an end-to-end manner, whereas their real-time inferences are carried out in a decentralized manner by means of the fronthaul coordination. To facilitate fronthaul cooperation among multiple ENs, the optimal fronthaul multiple access schemes are designed. Training algorithms robust to practical fronthaul impairments are also presented. Numerical results validate the effectiveness of the proposed approaches.

This letter investigates a learning solution for robust beamforming optimization in downlink multi-user systems. A base station (BS) identifies efficient multi-antenna transmission strategies only with imperfect channel state information (CSI) and its stochastic features. To this end, we propose a robust training algorithm where a deep neural network (DNN), which only accepts estimates and statistical knowledge of the perfect CSI, is optimized to fit to real-world propagation environment. Consequently, the trained DNN can provide efficient robust beamforming solutions based only on imperfect observations of the actual CSI. Numerical results validate the advantages of the proposed learning approach compared to conventional schemes.

This paper designs wireless fronthauling techniques for deep learning (DL) enabled fog radio access networks (F-RANs) where computation and communication processes at a cloud and edge nodes (ENs) are carried out by deep neural networks (DNNs). Coordination among ENs and the cloud is realized by wireless fronthaul links, which incurs undesired randomness in forwardpass calculations of DNNs. To address this issue, we propose a robust training strategy whereby a group of DNNs can mitigate the impairment from fronthaul fading and additive noise. Numerical results demonstrate the superiority of the proposed robust wireless fronthauling scheme.

Non-orthogonal multiple access (NOMA) is an effective technique to improve spectral efficiency and connectivity level of multi-user systems by means of successive interference cancellation (SIC) decoding. However, the optimization of NOMA scheme for uplink systems has not been well addressed in the literature. This work tackles the joint design of the users' transmit power levels and the SIC decoding order at the BS for a multi-user uplink system. Since it is difficult to find the optimal solution, an iterative algorithm is proposed that alternately optimizes transmit powers and SIC decoding order until convergence is reached. The optimization of transmit powers is addressed by adopting the majorization minimization (MM) approach. The SIC decoding order is designed by evaluating per-user mean squared error (MSE) at each SIC decoding step. Numerical results are presented to identify the configurations in which the proposed design achieves relevant gains.

This paper studies deep learning-based beamforming design schemes for multi-user downlink systems. Two distinct objectives are considered: sum-rate maximization and min-rate maximization. Each of formulations is first tackled by classical majorization-minimization (MM) algorithms that find a locally optimum point iteratively. To reduce computational overheads of the MM algorithms, deep neural networks (DNNs) are introduced which yield optimized beamforming solutions from channel vector inputs. Performance of trained DNNs is evaluated in terms of bit-error rate (BER) measure. Numerical results show that deep learning approaches achieve the BER performance very close to MM algorithms with much reduced complexity. Also, it is desirable to adopt the minimum-rate criterion to achieve low BER performance rather than sum-rate.

This work studies the uplink of a multi-tenant cloud radio access network (C-RAN) system with spectrum pooling. In the system, each operator has a cloud processor (CP) connected to a set of proprietary radio units (RUs) through finite-capacity fronthaul links. The uplink spectrum is divided into private and shared subbands, and all the user equipments (UEs) of the participating operators can simultaneously transmit signals on the shared subband. To mitigate inter-operator interference on the shared subband, the CPs of the participating operators can exchange compressed uplink baseband signals on finite-capacity backhaul links. This work tackles the problem of jointly optimizing bandwidth allocation, transmit power control and fronthaul compression strategies. In the optimization, we impose that the inter-operator privacy loss be limited by a given threshold value. An iterative algorithm is proposed to find a suboptimal solution based on the matrix fractional programming approach. Numerical results validate the advantages of the proposed optimized spectrum pooling scheme.

This letter studies deep learning (DL) approaches to optimize beamforming vectors in downlink multi-user multi-antenna systems that can be universally applied to arbitrarily given transmit power limitation at a base station. We exploit the sum power budget as side information so that deep neural networks (DNNs) can effectively learn the impact of the power constraint in the beamforming optimization. Consequently, a single training process is sufficient for the proposed universal DL approach, whereas conventional methods need to train multiple DNNs for all possible power budget levels. Numerical results demonstrate the effectiveness of the proposed DL methods over existing schemes.

A multi-user downlink system is considered, in which a base station (BS) serves multiple user equipments (UEs). It is assumed that, to improve the downlink performance, the UEs can exchange compressed versions of the downlink received signals on out-of-band device-to-device (D2D) communication links. Since the D2D links from a UE to the other UEs may have different channel qualities across the receiving UEs, this letter proposes a layered cooperation strategy which combines broadcast coding and successive refinement quantization techniques. With this approach, a UE can have a better reconstruction of the downlink signal of the transmitting UE than another UE with worse D2D channel quality. The problem of jointly optimizing the downlink precoding and layered UE cooperation strategies is tackled, and numerical results show the performance gains of the proposed layered scheme as compared to baseline schemes with no or single-layer cooperation.

The field perturbation induced by an elastic or electrical inclusion admits a multipole expansion. In the classical expansion, basis functions are independent of the inclusion. This paper presents a new concept of geometric multipole expansion for the conductivity or anti-plane elasticity problem in two-dimensions. The coefficients of the expansion are denoted by the Faber polynomial polarization tensor (FPT). In the derivation, we use the series expansion for the complex logarithm by the Faber polynomials that are associated with the exterior conformal mapping of the inclusion. As an application, we construct multi-coated inclusions of general smooth shape that have negligible perturbation for low-order polynomial loadings. These inclusions are a layered structure composed of level sets of the exterior conformal mapping; material parameters of each layer are determined such that the second FPTs of low-order vanish. We provide numerical examples to validate the results.