D2D/蜂窝通信模式切换与联合功率控制方案

设备直连（Device-to-Device, D2D）通信技术通过复用蜂窝系统的频谱资源提高频谱利用率,但D2D的引入会给蜂窝系统带来干扰。如何合理地选择D2D/蜂窝通信模式并进行功率优化控制,成为减小D2D和蜂窝系统间干扰、提升网络性能的关键。本文考虑D2D用户复用蜂窝上行链路场景,提出了一种基于距离和联合功率控制的通信模式选择方案。在该方案中,D2D用户和蜂窝用户与基站距离的比值决定了D2D用户是否采用Underlay模式进行通信,进而在约束蜂窝用户和D2D用户发射功率的条件下实现D2D链路和蜂窝链路的联合功率控制,最终推导出能够最大化系统总吞吐量的最优用户功率分配方案。根据仿真结果,本文提出的联合功率控制方案能够在降低系统间干扰的同时有效提高D2D和蜂窝系统的总吞吐量,进而提高了系统的性能。      

基于大规模MIMO和D2D技术的混合网络能量效率研究

文章研究了基于大规模多输入多输出（multiple-input multiple-output,MIMO）和端到端（device-to-device,D2D）技术的双层混合网络及其能量效率,面向D2D用户（D2D user,DU）提出了无线能量采集方案。在该混合网络模型中,宏蜂窝基站由大规模MIMO天线阵列构成,D2D网络覆盖在宏蜂窝之上,与蜂窝网共享频谱。为了保证D2D通信,系统配置了由小规模MIMO天线构成的专用能量发射基站（power beacon,PB）,其通过迫零预编码向能量约束的D2D发射机传送能量;同时,为了提高网络的能量效率,DU还可以从附近蜂窝网用户（cellular user,CU）的射频（radio frequency,RF）干扰信号中采集能量。对此混合网模型和能量采集方案,文章通过假设网络中CU、DU以及PB的位置服从独立泊松点过程和利用随机几何分析方法,基于反功率控制方案得到了网络中任一D2D发射机通过采集足够能量来建立通信链路的概率,其以简单的积分形式给出。分析和数值结果表明,文章提出的无线能量采集方案由于PB的配置保证了D2D通信的可靠性;同时又因为充分利用了附近CU的RF干扰信号,有效提高了D2D用户采集足够能量的概率,节约了PB的能量,改善了系统的能量效率。      

过时CSI下全双工中继辅助D2D网络的物理层安全

中继辅助终端直通（devicetodevice,D2D）网络通过与蜂窝网络共享频谱提高D2D用户的频谱效率和蜂窝用户（cellular user,CU）的物理层安全性。为进一步改善其性能,可以在基站和D2D链路的中继节点采用天线选择以及在中继节点采用全双工技术。然而,由于存在反馈时延和移动性,用于蜂窝链路和所有D2D链路天线选择的信道状态信息（channelstateinformation,CSI）均可能是过时的,针对该场景下的物理层安全性和可靠性问题,提出一种主动窃听和过时CSI场景下基站和中继节点均采用发射天线选择的全双工中继辅助D2D网络安全模型,推导CU的中断概率、遍历容量、非零安全容量概率、安全中断概率、渐近安全中断概率的解析表达式。数值计算与仿真结果均表明,基站发射天线数、中继干扰天线数越多,CU的安全性能越好;过时的CSI会降低CU的中断性能和安全性能。     

D2D辅助的协作中继NOMA系统中TSRS策略及其中断性能分析

该文研究了D2D辅助的协作中继NOMA（DC-NOMA）网络,在中继用户（RU）转发信息的第二时隙,利用RU到蜂窝中心用户（CCU）的D2D通信链路传输新的信号。为了提高传输可靠性,设计了新的信号检测策略,通过完全利用边缘信息消除了用户间的干扰。基于新的信号检测策略,提出了一种两阶段中继选择策略（TSRS）。在满足蜂窝边缘用户（CEU）可靠接收的中继集合中,选择使CCU成功解码概率最大的中继。为了评估所提出的DC-NOMA方案,推导了每个用户的确切中断概率且通过仿真得到验证。仿真结果表明,在高信噪比区域,该文所提出的基于新的信号检测策略的DC-NOMA在TSRS方案下的中断性能优于部分中继选择方案（PRSS）和传统的DC-NOMA。特别地,增大中继数量可有效提高DC-NOMA系统的中断性能。      

面向蜂窝网络的D2D多播通信的分簇和中继选择方法

传统蜂窝网络中,信道衰减的随机性和不确定性导致小区边缘用户的接收性能很差,尤其是面向视频传输等速率要求较高时其弊端更加凸显。D2D通信因其配置灵活性可作为传统蜂窝网络架构的有利补充,能有效改善边缘用户的性能。该文针对D2D通信的多播传输,分析了系统最小时延成本下的中继数量和分簇算法,提出一种基于分簇和中继选择的低时延D2D多播方案。该方案可以自适应选择多播重传中的中继的数量和中继节点到基站的距离,同时给出最优的带宽资源分配机制。仿真结果表明,与其他方案相比,所提方法能有效减少系统时延,提高边缘用户体验和系统性能。     

3D MIMO在无线携能通信系统中的应用和优化设计

为了提高无线携能通信(SWIPT)的效率,在基站采用3维(3D)定向天线,通过动态调整天线下倾角,来开发垂直维度,从而增强SWIPT中能量和信息的传输效率。该文研究单小区多输入多输出(MIMO)SWIPT系统,基站和用户分别使用迫零(ZF)预编码和功率划分(PS)技术。构建以最小化发送功率为目标的优化问题,在信噪比(SNR)和获取功率的约束下,联合优化了天线下倾角,PS比和分配的功率。最佳PS比和最优功率分配以闭合表达式给出。仿真表明,所得到的优化解的性能优于具有可调下倾角的常规MIMO系统和不考虑垂直维度的2维(2D)方案。     

D2D网络中信道选择与功率控制策略研究

针对D2D通信的资源分配问题,该文研究了D2D信道选择与功率控制策略。在保证蜂窝用户服务质量(QoS)的前提下,提出一种基于启发式的D2D信道选择算法,为系统内的D2D用户找到合适的信道复用资源。同时,利用拉格朗日对偶方法求解得到D2D用户最优传输功率。仿真结果表明当蜂窝用户与多对D2D用户共享信道资源时能够大幅度提升系统平均吞吐量。在相同条件下,该算法的性能要明显优于现有算法。     

基于多智能体深度强化学习的D2D通信资源联合分配方法

设备对设备(D2D)通信作为一种短距离通信技术,能够极大地减轻蜂窝基站的负载压力和提高频谱利用率。然而将D2D直接部署在授权频段或者免授权频段必然导致与现有用户的严重干扰。当前联合部署在授权和免授权频段的D2D通信的资源分配通常被建模为混合整数非线性约束的组合优化问题,传统优化方法难以解决。针对这个挑战性问题,该文提出一种基于多智能体深度强化学习的D2D通信资源联合分配方法。在该算法中,将蜂窝网络中的每个D2D发射端作为智能体,智能体能够通过深度强化学习方法智能地选择接入免授权信道或者最优的授权信道并发射功率。通过选择使用免授权信道的D2D对(基于“先听后说”机制)向蜂窝基站的信息反馈,蜂窝基站能够在非协作的情况下获得WiFi网络吞吐量信息,使得算法能够在异构环境中执行并能够确保WiFi用户的QoS。与多智能体深度Q网络(MADQN)、多智能体Q学习(MAQL)和随机算法相比,所提算法在保证WiFi用户和蜂窝用户的QoS的情况下能够获得最大的吞吐量。     

时延约束下采用SVC编码的D2D协作视频多播传输方案

为缓解基站的视频流量过载，本文针对时延敏感的实时视频业务，设计一种D2D协作视频多播传输方案。该方案采用可伸缩视频编码（Scalable Video Coding, SVC）对视频流进行编码处理，利用SVC流的分层结构特征来应对多播信道间的差异性。在SVC编码的基础上，为了改善用户观看体验及提升用户所接收的视频质量，所提出的协作式视频传输方案引入有效吞吐量这一概念，在一定时延约束下，根据信道反馈信息灵活地对不同信道上的不同SVC视频层进行码率调整。仿真结果表明，所提出的方案能够有效地减小端到端时延，有效丢失率，提高有效吞吐量。      

NOMA-D2D协作无线系统的物理层安全

针对未来海量用户设备接入的物理层安全（physical layer security，PLS）应用需求，提出了一种组合非正交多址接入（non-orthogonal multiple access，NOMA）、终端直通（device-to-device，D2D）与中继协作的NOMA-D2D协作无线系统PLS模型。该模型由采用发射天线选择（transmit antenna selection，TAS）的基站、分别作为基站和D2D发射端的NOMA远端蜂窝用户、D2D接收端以及被动窃听者组成，其中的D2D发射端承担基站的 NOMA 近端用户和解码转发中继两种角色。利用高斯—切比雪夫正交定理推导两种 TAS 方案下NOMA-D2D协作无线系统的安全中断概率、非零安全容量概率以及渐近安全中断概率的近似表达式。数值计算和仿真实验验证了NOMA-D2D协作无线系统PLS性能分析的准确性；在基站总功率恒定时增大分配给远端蜂窝用户的功率能有效提升NOMA-D2D协作无线系统的PLS性能。     

Distributed detection for cellular CDMA

A distributed scheme is proposed for a cellular code division multiple access system in which each base station is served by three widely spaced sectored antennas, with each antenna site performing separate detection. At worst the error probability achieved at the base station is always equal to the minimum of those at each antenna site. With coherent signalling employed, the distributed detection has significantly increased system capacity over simple sectored antennas     

Distributed closed-loop spatial multiplexing for uplink multiuser systems

Focusing on the uplink, where mobile users (each with a single transmit antenna) communicate with a base station with multiple antennas, we treat multiple users as antennas to enable spatial multiplexing across users. Introducing distributed closed-loop spatial multiplexing with threshold-based user selection, we propose two uplink channel-assigning strategies with limited feedback. We prove that the proposed system also outperforms the standard greedy scheme with respect to the degree of fairness, measured by the variance of the time averaged throughput. For uplink multi-antenna systems, we show that the proposed scheduling is a better choice than the greedy scheme in terms of the average BER, feedback complexity, and fairness. The numerical results corroborate our findings.     

Trade-off of multiplexing streams in MIMO broadcast channels

A soft antenna trade-off scheme between different quality of service (QoS) groups in multiple input multiple output (MIMO) broadcast channels (BCs) is proposed. With the proposed scheme, a higher priority group takes precedence in using multiplexing streams for satisfying its QoS requirement. The number of streams allocated to each QoS group is gracefully adapted to the wireless channel conditions and QoS requirements. Numerical results demonstrate that the proposed scheme can improve the probability of QoS-guaranteed transmission while simultaneously maximizing the total throughput.     

MIMO OFDM排序判决反馈检测方法

一般多天线系统的判决反馈检测方法中,是顺序检测每用户天线端接收信号,其缺点是若从前面若干幅接收天线检测出的信号误差较大,则会降低后续天线端检测性能.为解决该问题,本文提出一种排序判决反馈检测方法,其思想先对待检测的用户天线支路排序,利用计算预测误差获得的待检测天线支路对应信道逆阵向量是已检测天线支路对应信道逆阵向量的线性组合特征,应用正交变换求解获得用户天线支路的检测顺序,使当前待检测用户天线端的迫零向量与已检测用户天线支路的迫零向量正交,这样可使由已检测用户天线支路带来的判决反馈误差最小,进而提高了检测性能.     

Downlink scheduling of multiuser MIMO systems with transmit beamforming

This article deals with downlink scheduling for multiuser multiple-input multiple-output （MIMO） systems, where the base station communicates with multiple users simultaneously through transmit beamforming. Most of the existing transmission schemes for multiuser MIMO systems focus on optimizing sum rate performance of the system. The individual quality of service （QoS） requirements （such as packet delay and minimum transmission rate for the data traffic） are rarely considered. In this article, a novel scheduling strategy is proposed, where we try to optimize the global system performance under individual QoS constraints. By performing scheduling into two steps, namely successive user selection and power allocation, the scheduler can achieve efficient resource utilization while maintaining the QoS requirements of all users. Extensive simulations and analysis are given to show the effectiveness of the proposed scheduler.     

User-relay Assignment based Antenna Selection Scheme in Multi-user Multi-relay AF Cooperative Communication Network

In this paper, the user-relay assignment based antenna selection scheme for multi-user cooperative network is proposed to improve the transmission bit rate and system performance. The low complexity suboptimal user-relay assignment algorithm is first proposed to simultaneously improve both the system performance and individual user’ performance in the multi-user amplify-and-forward (AF) cooperative user-relay network, where each user can fairly select its best relay according to the instantaneous channel state information (CSI). Based on the user-relay assignment scheme and space shift keying (SSK), we further present the antenna selection scheme to further enhance the system performance for the multi-user AF cooperative user-relay network. In order to minimize the system bit error rate (BER), three antenna selection criteria based on the maximum likelihood (ML) detection have been proposed. Numerical results and theoretical analysis show that the proposed schemes can effectively improve the average system sum rate, the outage probability of the worst user, and the system BER performance, compared with the other schemes.     

On the study of cognitive two-way relaying using a denoise-and-forward relay

In this paper, we consider a cognitive cooperative wireless system consisting of two primary users, a cognitive base station (CBS) and a secondary user. A cognitive two-way relaying (CTWR) scheme is proposed, where the secondary user tries to assist the primary users to meet their target quality of service (QoS) by acting as a cognitive two-way relay, when the direct link of the two primary users cannot meet the target QoS. As a return, the primary network shares its transmission time-slot with the secondary user. First, the system outage probabilities of traditional direct transmission and CTWR scheme are analyzed, giving both the exact and approximated expressions. Second, a transmission time-slot splitting algorithm between the primary and secondary users is proposed. Third, the ergodic capacity of the secondary user under the CTWR scheme is examined. Finally, simulation results show that the proposed scheme cannot only help the primary users to meet their target QoS, but also improve the communication opportunities of the secondary user.