This paper studies the performance of a user positioning system using Channel State Information (CSI) of a Massive MIMO (MaMIMO) system. To infer the position of the user from the CSI, a Convolutional Neural Network is designed and evaluated through a novel dataset. This dataset contains indoor MaMIMO CSI measurements using three different antenna topologies, covering a 2.5 m by 2.5 m indoor area. We show that we can train a Convolutional Neural Network (CNN) model to estimate the position of a user inside this area with a mean error of less than half a wavelength. Moreover, once the model is trained on a given scenario and antenna topology, Transfer Learning is used to repurpose the acquired knowledge towards another scenario with significantly different antenna topology and configuration. Our results show that it is possible to further train the CNN using only a small amount of extra labelled samples for the new topology. This transfer learning approach is able to reach accurate results, paving the road to a practical CSI-based positioning system powered by CNNs.

A massive MIMO network can serve ten’s of users simultaneously. However, in dense scenarios the users are potentially closely-spaced, potentially resulting in substantial inter-user interference. Scheduling can overcome this by selecting the users that lead to the highest combined spectral efficiency. As scheduling comes with a significant pilot overhead, an alternative strategy could minimize user correlation by distributing the antenna elements in space. In this paper, we propose a comprehensive system study including antenna topology and distribution, user scheduling and pilot overhead reduction. Our user scheduling and pilot reduction algorithms are evaluated using system level simulations relying on indoor line-of-sight channel measurements from a 64 antenna base station at 2.61GHz. To have a thorough evaluation of the proposed algorithm, we consider four different antenna topologies, including co-located and distributed placement of the base station arrays. Our evaluation shows that in a conference room with 64 densely deployed users, our proposed low complexity algorithm can improve the spectral efficiency by at least 14% compared to random user selection with the best antenna distribution strategy. Finally, our results show that by relying on channel hardening, we reduce the pilot overhead by 3.2x.

Assistive Technology (AT) maintains and improves the individual’s functioning and independence, thereby promoting their well-being.

Massive MIMO (MaMIMO) is a technology of primary interest for sub-6 GHz operation in the next generation cellular systems. While MaMIMO is most often linked to macrocell scenarios, where a single cell serves many users distributed over a large area, network densification will also result in scenarios where many users are served by a MaMIMO base station (BS) that is nearby. A key question is how to scale up MaMIMO: should we add more antennas to a given cell, or create multiple smaller and distributed cells that can cooperate? This paper documents the measured performance of a very dense MaMIMO system for an indoor-to-outdoor propagation environment. The impact of the number of antennas, and the distribution of the antenna elements is experimentally verified for a simplified linear deployment of the BSs. Concretely, we serve 12 closely located users with 16, 32 or 64 antennas. We compare a centrally positioned collocated array and two distributed arrays with their uplink throughput in a licensed 2.6 GHz band. The experimental results show that 12 users can be served with only 32 antennas for the distributed topology, which is effectively only 16 antennas per MaMIMO BS. For the specific case analyzed in our measurement campaign, with the centralized deployment, 64 antennas are needed to obtain good performance, while distributing the antenna elements in two sub-arrays improves total performance and fairness between the users.

Localization is crucial for nowadays’ communication systems, especially for beamforming techniques in massive MIMO systems. Large-scale MIMO systems have exhibited their advantages in communications. In the meantime, they also have the potential to provide accurate localization with their high angular resolution. In this paper, we study indoor localization performance of a Massive MIMO system with a 64-antenna Uniform Linear Array (ULA). Based on the sparse reconstruction method, we propose a Mixed field Sparse Bayesian Learning (MSBL) algorithm to localize devices for both near-field and far-field scenarios. Using the measurement results from our massive MIMO testbed, we show that our proposed MSBL algorithm can improve the localization accuracy by 49% with only a few snapshots. The performance of our algorithm is also robust to low Signal-to-Noise Ratio (SNR) conditions.

Cell-Free Massive MIMO is a promising solution for next-generation radio access networks. Especially the user-centric variant, where users are served by a subset of access points, a cluster, has garnered significant interest. However, managing these clusters in case of user mobility remains a significant issue. Additionally, the practical implementation of such a network remains an open problem. This work introduces a realistic temporal channel model for Cell-Free Massive MIMO. Subsequently, we propose a framework to facilitate handover procedures in a distributed topology considering limited channel knowledge for cluster formation. Next, we develop two clustering and handover strategies: 1) a fixed clustering strategy where the central processing unit computes the ideal cluster when a cluster handover threshold is exceeded and 2) an opportunistic strategy where access points make autonomous decisions based on the local channel. Finally, we establish a significant connection between our findings and O-RAN, an emerging network architecture that offers open interfaces and network-wide control capabilities. Simulation results show that even low-complexity handover mechanisms can significantly improve spectral efficiency under user mobility compared to cellular networks.