Cooperative Distributed Beamforming Design for Multi-RIS Aided Cell-Free Systems

Cell-free systems significantly improve network capacity by enabling joint user service without cell boundaries,eliminating intercell interference.However,to satisfy further capacity demands,it leads to high-cost problems of both hardware and power consumption.In this paper,we investigate multiple reconfigurable intelligent surfaces(RISs)aided cell-free systems where RISs are introduced to improve spectrum efficiency in an energy-efficient way.To overcome the centralized high complexity and avoid frequent information exchanges,a cooperative distributed beamforming design is proposed to maximize the weighted sum-rate performance.In particular,the alternating optimization method is utilized with the distributed closed-form solution of active beamforming being derived locally at access points,and phase shifts are obtained centrally based on the Riemannian conjugate gradient(RCG)manifold method.Simulation results verify the effectiveness of the proposed design whose performance is comparable to the centralized scheme and show great superiority of the RISs-aided system over the conventional cellular and cell-free system.

Distributed Beamforming Based Wireless Power Transfer:Analysis and Realization

This paper presents a new Wireless Power Transfer(WPT)approach by aligning the phases of a group of spatially distributed Radio Frequency(RF)transmitters(TX)at the target receiver(RX)device.Our approach can transfer energy over tens of meters and even to targets blocked by obstacles.Compared to popular beamforming based WPTs,our approach leads to a drastically different energy density distribution:the energy density at the target receiver is much higher than the energy density at other locations.Due to this unique energy distribution pattern,our approach offers a safer WPT solution,which can be potentially scaled up to ship a higher level of energy over longer distances.Specifically,we model the energy density distribution and prove that our proposed system can create a high energy peak exactly at the target receiver.Then we conduct detailed simulation studies to investigate how the actual energy distribution is impacted by various important system parameters,including number/topology of transmitters,transmitter antenna directionality,the distance between receiver and transmitters,and environmental multipath.Finally,we build an actual prototype with 17 N210 and 4 B210 Universal Software Radio Peripheral(USRP)nodes,through which we validate the salient features and performance promises of the proposed system.

Distributed Beamforming for Energy-Harvesting Relaying in Vehicular Networks

Recently,battery capacity and charging speed have been the bottlenecks of mobile communication networks.Energy harvesting(EH)relaying has become a promising solution for green 5th generation mobile communication with the advancement of wireless power transfer technology.In this paper,we investigate EH relaying in vehicular networks and adopt distributed beamforming(DB)to enhance the reliability and capacity of EH relaying.To be more specific,we propose a DB solution based on the joint optimization of power-splitting(PS)factors.For amplify-and-forward relaying,to transform the optimization problem into a quasi-convex one,we apply the semidefinite relaxation(SDR)method so that we can effectively attain the global optimal solution,while the suboptimal DB solution with distributed optimal PS factor which only requires local channel state information is also proposed.For decode-and-forward relaying,to get the optimal PS factors,we set a signal-to-noise ratio threshold at the relays,which can reduce the system error rate caused by the poor transmission link.Simulation results demonstrate the efficiency of our proposed DB-based EH relaying scheme in vehicular networks.

Local Observations-Based Energy-Efficient Multi-Cell Beamforming via Multi-Agent Reinforcement Learning

With affordable overhead on information exchange,energy-efficient beamforming has potential to achieve both low power consumption and high spectral efficiency.This paper formulates the problem of joint beamforming and power allocation for a multiple-input single-output(MISO)multi-cell network with local observations by taking the energy efficiency into account.To reduce the complexity of joint processing of received signals in presence of a large number of base station(BS),a new distributed framework is proposed for beamforming with multi-cell cooperation or competition.The optimization problem is modeled as a partially observable Markov decision process(POMDP)and is solved by a distributed multi-agent self-decision beamforming(DMAB)algorithm based on the distributed deep recurrent Q-network(D2RQN).Furthermore,limited-information exchange scheme is designed for the inter-cell cooperation to boost the global performance.The proposed learning architecture,with considerably less information exchange,is effective and scalable for a high-dimensional problem with increasing BSs.Also,the proposed DMAB algorithms outperform distributed deep Q-network(DQN)based methods and non-learning based methods with significant performance improvement.