Search

While the problem of jointly controlling the pilot-and-data power in cell-based systems has been extensively studied, this problem is difficult to solve in cell-free systems due to two reasons. First, both the large- and small-scale fading are markedly different between a served user and the multiple serving access points. Second, due to the user-centric architecture, there is a need for decentralized algorithms that scale well in the cell-free environment. In this work, we study the impact of joint pilot-and-data power control and receive filter design in the uplink of cell-free systems. The problem is formulated as optimization tasks considering two different objectives: 1) maximization of the minimum spectral efficiency (SE) and 2) maximization of the total SE. Since these problems are non-convex, we resort to successive convex approximation and geometric programming to obtain a local optimal centralized solution for benchmarking purposes. We also propose a decentralized solution based on actor-critic deep reinforcement learning, in which each user acts as an agent to locally obtain the best policy relying on minimum information exchange. Practical signaling aspects are provided for such a decentralized solution. Finally, numerical results indicate that the decentralized solution performs very close to the centralized one and outperforms state-of-the-art algorithms in terms of minimum SE and total system SE., QC 20230227

Joint pilot and data power control (JPDPC) is known to have a large impact on both the overall spectral/energy efficiency and fairness of cell-based systems. However, the impact of JPDPC on the inherent spectral/energy efficiency and fairness trade-off in cell-free (CF) systems is much less understood. In this letter, considering pilot contamination, user-centric clustering and multi-antenna access points, we formulate novel JPDPC problems in CF systems as distinct optimization tasks, whose objectives are maximizing the minimum spectral efficiency (SE), maximizing the total SE and maximizing the product of the individual signal-to-interference-plus-noise ratios. Since these problems are non-convex, we solve them by combining successive convex approximation and geometric programming. To the best of our knowledge, this is the first letter analyzing and optimizing JPDPC in user-centric CF systems. Our results indicate that JPDPC allows users to save more energy than the disjoint optimization of pilot and data powers when maximizing the minimum SE, while showing that JPDPC plays a crucial role in balancing between SE and fairness also in CF systems., QC 20220307

Existing works concluded that coherent transmission outperforms non-coherent transmission in the downlink of cell-free systems when the fronthaul links have unlimited capacity. Since the capacity of the fronthaul links of cell-free networks is typically limited, in this paper we ask the question whether this conclusion holds under more realistic assumptions on the fronthaul capacity. To answer this question, we study and compare the performance of these transmission strategies by formulating novel energy efficiency (EE) maximization problems for both strategies, where we explicitly consider realistic fronthaul capacity and power consumption constraints. Despite the non-convexity of these problems, we derive closed-form equations to find suboptimal solutions of both problems using a unified framework that combines successive convex approximation and the Dinkelbach algorithm. Numerical results show that the performance of coherent transmission is severely impacted by limited fronthaul capacities, power consumption on the fronthaul links, user-centric cluster size and the number of antennas at the access points, such that in many cases non-coherent transmission achieves higher EE than coherent transmission. Based on these results, we provide deployment guidelines on when to use coherent or non-coherent transmission to maximize the EE of cell-free systems with limited fronthauls., QC 20221101

Several previous works have proposed game-theoretic approaches to controlling the pilot and data power levels in the uplink of both single and multi-cell multi-user multiple input multiple output (MU-MIMO) systems. Unfortunately, the vast majority of existing works design these power control schemes under the assumption that the wireless channels between the mobile terminals and the serving base station are block fading. Meanwhile, several letters have shown that modeling fast fading channels as autoregressive (AR) processes with known or estimated state transition matrices give much more accurate results than those suggested by block fading models. Thus, this letter proposes a game-theoretic approach to controlling the uplink pilot and data power levels in a MU-MIMO system, in which the wireless channels are AR processes with mobile terminal-specific state transition matrices. We find that the proposed approach outperforms a classical cellular path-loss compensating fractional power control scheme and a game-theoretic power control scheme designed for block fading channels., QC 20220907

Maximizing the rate in multiple input single output (MISO) systems using distributed algorithms is an important task that typically incurs high computational cost. In this work, we propose two deep Q-learning-based user scheduling schemes to solve the beamforming problem of sum-rate maximization with per base station power constraints in multicell MISO scenarios. The two key features of the proposed algorithms are that they are executed in a distributed fashion and are robust with respect to channel state information (CSI) errors. Simulation results show that in the presence of CSI errors the proposed schemes outperform state-of-the-art algorithms both in terms of average spectral efficiency and execution time., QC 20210113

Hybrid analog-digital (HAD) beamforming architectures have been proposed to facilitate the practical implementation of massive multiple-input multiple-output (MIMO) systems by reducing the number of employed radio frequency chains. While most prior studies have aimed to maximize spectral efficiency (SE), the present paper proposes a two-stage HAD beamforming design for multi-user MIMO systems that can be used to maximize either the system's overall energy efficiency (EE) or SE. This problem is nonconvex and NP-hard due to the joint optimization between the analog and digital domains and the constant modulus constraints required by the analog domain. To address this problem, we propose a decoupled two-stage design wherein the first stage, the analog beamforming parts are updated, which are then taken into account in the second stage to design the digital beamforming parts to maximize the system's EE or SE. We consider two widely-used HAD beamforming techniques that utilize either fully-connected (FC) or partially-connected (PC) architectures employing variable phase-shifters. Using the most recently available data for the circuitry power consumption of the components, we compare the performance of these two HAD architectures with that of the fully-digital (FD) architecture in terms of the total circuitry power consumption, and achieved SE and EE. We find that there is a certain number of users above which the FC architecture has higher circuitry power consumption than the FD counterpart, in contrast to the PC architecture that always has lower circuitry power consumption. More importantly, our results reveal, contrary to the common opinion, that depending on the circuitry parameters the FD architecture may achieve not only higher SE, but also higher EE than the HAD architectures., QC 20200309

Employing dynamic time division duplexing can increase the system-wide spectral efficiency of applications with varying and unbalanced uplink and downlink data traffic requirements. However, in order to achieve this efficiency gain, it is necessary to manage the effects of cross-link interference, which are generated among cells transmitting in opposite link directions. This paper considers bidirectional sum-power minimization beamforming as a means to deal with this cross-link interference, by forcing a minimum signal-to-interference-plus-noise ratio constraint for both uplink and downlink. We propose two iterative approaches to solve this beamforming problem. The first approach assumes centralized processing and requires the availability of global channel state information. The second approach is performed in a decentralized manner, based on the alternating direction method of multipliers and requires only local channel state information and reduced signaling load. Both approaches are shown to converge to a minimum network power expenditure, whereas close-to-optimum performance can be obtained when limiting the number of iterations., QC 20200110

Hybrid beamfiorming (BF) architectures employing phase shifters or switches reduce the number of required radio frequency chains and the power consumption of base stations that employ a large number of antennas. Due to the inherent tradeoff between the number of radio frequency chains, the complexity of the employed analog and digital BF algorithms and the achieved spectral and energy efficiency, designing hybrid BF architectures is a complex task. To deal with this ormplexity, we propose a unifying design that is applicable to architectures employing either phase shifters or switches. In our design, the analog part (!if the hybrid BF architecture maximizes the capacity of the equivalent channel, while the digital part is updated using the well-known block diagonalizat' approach. We then employ the proposed joint analog-digital beamforming algorithm on lour recently proposed hybrid architectures and compare their performance in terms of spectral and energy efficiency, and find that the proposed analog-digital BF algorithm outperforms previously proposed schemes. We also find that phase shifterbased architectures achieve high spectral efficiency, whereas switching-based architectures can boost energy efficiency with increasing number of base station antennas., QC 20190107

In dynamic time division duplexing (DTDD) systems, the uplink (UL) and downlink (DL) resources can be configured to adapt to changing traffic conditions. Therefiwe, DTDD systems are advantageously deployed in scenarios in which the UL and DL traffic demands are asymmetric and timevarying. Unfortunately, multicell DTDD systems give rise to base station-to-base station and user equipment-to-user equipment interference, that can severely degrade the system performance. Previous works on DTDD either assumed that the UL/DL configurations are given, or they did not take into account the negative impact of multicell DTDD interference. In this letter, we propose a novel cell reconfiguration technique that considers both the prevailing traffic conditions and multicell interference levels. The proposed technique is based on an efficient solution of a mixed integer linear program, whose objective is to maximize the overall system throughput taking into account users' traffic preferences. Realistic system level simulations indicate that the proposed scheme outperforms not only the static TDD system but also other reference schemes, that disregard the DTDD specific interference effects., QC 20181218

This letter proposes distributed beamforming as a means of reducing interference in dynamic time division duplexing multiple input multiple output networks. Specifically, we formulate an optimization task, whose objective is to minimize the base station (BS) transmit power, while satisfying a predefined signal-to-interference-plus-noise ratio threshold for each downlink user equipment and keeping the BS to BS interference power below a tolerable level. An iterative decentralized solution requiring reduced signaling load based on primal decomposition is proposed. This decentralized algorithm is shown to iterate towards the centralized solution, while feasible but suboptimal solutions can be obtained at any iteration., QC 20181108