基于深度强化学习的反向散射网络资源分配机制

为了提升反向散射网络中物联网设备的平均吞吐量,提出了一种资源分配机制,构建了用户配对和时隙分配联合优化资源分配模型。由于该模型直接利用深度强化学习(Deep Reinforcement Learning,DRL )算法求解导致动作空间维度较高且神经网络复杂,故将其分解为两层子问题以降低动作空间维度：首先,基于深度强化学习算法,利用历史信道信息推断当前的信道信息以进行最优的用户配对;然后,在用户固定配对的情况下,基于凸优化算法,以最大化物联网设备总吞吐量为目标进行最优的时隙分配。仿真结果表明,与其他资源分配方法相比,所提资源分配方法能有效提升系统吞吐量,且有较好的信道适应性和收敛性。     

OFDM系统Femtocell网络下行联合资源分配

为进一步提高家庭基站(Femtocell)网络中频谱利用率并优化功率分配,在基于正交频分复用技术(OFDM)系统网络中,提出一种子载波联合优化的多用户资源分配算法,即以最大化频谱利用率作为目标函数,加入基站选择因子对家庭基站进行待机模式选择优化,再对用户的子载波资源进行公平分配,最后利用线性封顶注水算法对小区基站用户功率进行优化分配。仿真结果表明,多用户资源分配算法不仅使频谱和功率利用率都得到显著增长,而且提高了系统吞吐量和用户公平性。该家庭基站资源寻优模型有效地改善了频谱紧缺和功率浪费现状,降低了家庭基站之间的干扰。      

面向窃听用户的RIS-MISO系统鲁棒资源分配算法

针对信道不确定性影响、用户信息泄露和能效提升等问题,该文提出一种基于不完美信道状态信息的可重构智能反射面(RIS)多输入单输出系统鲁棒资源分配算法。首先,考虑能量收集最小接收功率约束、合法用户最小保密速率约束、基站最大发射功率约束及RIS相移约束,基于有界信道不确定性,建立一个联合优化基站主动波束、能量波束、RIS相移矩阵的多变量耦合非线性资源分配问题。然后,利用Dinkelbach,S-procedure和交替优化方法,将原非凸问题转换成确定性凸优化问题,并提出一种基于连续凸近似的交替优化算法。仿真结果表明,与传统非鲁棒算法对比,所提算法具有较低的中断概率。     

密集异构网络中基于多目标优化的资源分配策略

密集异构网络(Dense Heterogeneous Network, DHN)通过部署小基站可以提升网络容量和用户速率,但小基站的密集部署会产生巨大的能耗和严重的干扰,进而影响系统的能量效率(Energy Efficiency, EE)和频谱效率(Spectral Efficiency, SE)。在保证用户服务质量(Quality of Service, QoS)需求的前提下,为了联合优化系统的能量效率和频谱效率,研究了密集异构网络中下行链路的资源分配(Resource Allocation, RA)问题。首先,将频谱和小基站发射功率分配问题建模为联合优化系统能量效率和频谱效率的多目标优化问题;其次,提出了基于单策略多目标强化学习(Single-strategy Multi-objective Reinforcement Learning, SMRL)的资源分配算法求解所建立的多目标优化问题。仿真结果表明,与基于单目标强化学习的资源分配算法相比,所提算法可以实现系统能量效率和频谱效率的联合优化,与基于群体智能算法的资源分配算法相比,所提算法的系统能量效率提高了1%～1.5%,频谱效率...     

面向用户体验的多小区混合非正交多址接入网络资源分配方法

该文研究了多小区混合非正交多址接入(MC-hybrid NOMA)网络的资源分配.为满足异构用户的服务体验,以最大化全网综合平均意见评分(MOS)累加和为目标,考虑基站选择、信道接入和功率资源分配的联合优化问题,该文提出一种用户、基站和信道3方的2阶段转移匹配算法,并根据用户MOS进行子信道功率优化.仿真结果表明所提多小区混合NOMA网络资源分配方案能有效提升全网用户服务体验和公平性.     

云雾混合网络下基于多智能体架构的资源分配及卸载决策研究

针对D2D辅助的云雾混合架构下资源分配及任务卸载决策优化问题,该文提出一种基于多智能体架构深度强化学习的资源分配及卸载决策算法.首先,该算法考虑激励约束、能量约束以及网络资源约束,联合优化无线资源分配、计算资源分配以及卸载决策,建立了最大化系统总用户体验质量(QoE)的随机优化模型,并进一步将其转化为MDP问题.其次,该算法将原MDP问题进行因式分解,并建立马尔可夫博弈模型.然后,基于行动者-评判家(AC)算法提出一种集中式训练、分布式执行机制.在集中式训练过程中,多智能体通过协作获取全局信息,实现资源分配及任务卸载决策策略优化,在训练过程结束后,各智能体独立地根据当前系统状态及策略进行资源分配及任务卸载.最后,仿真结果表明,该算法可以有效提升用户QoE,并降低了时延及能耗.     

云雾混合网络下基于多智能体架构的资源分配及卸载决策研究

针对D2D辅助的云雾混合架构下资源分配及任务卸载决策优化问题,该文提出一种基于多智能体架构深度强化学习的资源分配及卸载决策算法。首先,该算法考虑激励约束、能量约束以及网络资源约束,联合优化无线资源分配、计算资源分配以及卸载决策,建立了最大化系统总用户体验质量(QoE)的随机优化模型,并进一步将其转化为MDP问题。其次,该算法将原MDP问题进行因式分解,并建立马尔可夫博弈模型。然后,基于行动者-评判家(AC)算法提出一种集中式训练、分布式执行机制。在集中式训练过程中,多智能体通过协作获取全局信息,实现资源分配及任务卸载决策策略优化,在训练过程结束后,各智能体独立地根据当前系统状态及策略进行资源分配及任务卸载。最后,仿真结果表明,该算法可以有效提升用户QoE,并降低了时延及能耗。     

H-CRAN网络下联合拥塞控制和资源分配的网络切片动态资源调度策略

针对异构云无线接入网络(H-CRAN)网络下基于网络切片的在线无线资源动态优化问题,该文通过综合考虑业务接入控制、拥塞控制、资源分配和复用,建立一个以最大化网络平均和吞吐量为目标,受限于基站(BS)发射功率、系统稳定性、不同切片的服务质量(QoS)需求和资源分配等约束的随机优化模型,并进而提出了一种联合拥塞控制和资源分配的网络切片动态资源调度算法。该算法会在每个资源调度时隙内动态地为性能需求各异的网络切片中的用户分配资源。仿真结果表明,该文算法能在满足各切片用户QoS需求和维持网络稳定的基础上,提升网络整体吞吐量,并且还可通过调整控制参量的取值实现时延和吞吐量间的动态平衡。     

双层网络中一种新的接入控制机制

在由宏小区和家庭小区构成的双层网络中,合理的进行资源分配可有效的控制干扰和优化容量.本文通过建立双层网络的下行干扰模型,分析了不同的资源分配方案和接入机制下的网络性能,提出了接入机制的优化算法,在保证用户通信概率要求的前提下对系统容量进行了优化.仿真结果表明,本文的接入控制机制在保证用户中断概率的条件下,提高了家庭基站允许的最大发射功率,优化了系统容量.     

基于两跳匹配的OFDMA中继网络联合资源分配算法

针对OFDMA中继网络的两跳特性,提出一种基于两跳匹配的中继网络联合资源分配算法。首先根据中继网络的两跳性建立两跳速率匹配模型,然后利用对偶分解理论将中继网络资源分配的主问题分解为：中继选择、子载波分配和功率分配三个主要的子问题并进行联合优化,同时基于中继网络两跳性在子载波分配的过程中考虑两跳子载波配对,以逼近系统最优解。最后为了保证算法的公平性,考虑子载波分配因子约束以优化子载波分配。仿真结果表明：所提算法将中继选择、两跳子载波配对与功率分配联合优化以进一步提升系统吞吐量,同时引入子载波分配因子约束,保证了算法的公平性。     

Resource allocation mechanism in the TWDM-PON and C-RAN joint architecture with hybrid energy supply

Aiming at the problems of low resource utilization rate,high energy consumption and poor user service quality in the existing virtualized Cloud Radio Access Network,an energy-aware virtualized resource allocation mechanism with hybrid energy supply was proposed.According to the energy sources and energy consumption of different network devices,energy arrival and energy consumption models were established.Furthermore,under the premise of guaranteeing the quality of user services,considering proportional fairness and energy consumption optimization,distributed algorithms based on asynchronous update were used to allocate resources and harvested energy for different types of virtual cloud radio access networks and user virtual base stations to effectively improve the energy efficiency of network.The simulation results show that the proposed resource allocation mechanism can reduce energy consumption while effectively reducing the latency and improving the throughput.     

能耗和时延感知的虚拟化云无线接入网络资源分配机制

针对现有虚拟化云无线接入网络(C-RAN)资源利用率低、能耗高、用户服务质量无法得到保证等问题,该文提出一种能耗和时延感知的虚拟化资源分配机制。根据虚拟化C-RAN的网络特点及业务流量特征,考虑资源约束和比例公平,建立能耗和时延优化模型。进而,利用启发式算法为不同类型虚拟C-RAN和用户虚拟基站分配资源,完成资源的全局优化配置。仿真结果表明,所提资源分配机制在提高网络资源利用率的同时,不但使能耗节省了62.99%,还使时延降低了32.32%。     

面向分布式缓存系统的无线资源管理:动机、挑战与方法

移动社交网络平台的多元化发展与推广,使得移动用户对数据传输的需求呈爆发式增长趋势,并对匮乏的频谱资源及高负荷基站的管理提出了新挑战.为了解决这些问题,将分布式缓存系统引入无线协作网络中,鼓励移动用户间协作缓存热门内容,以实现用户间低能耗、高可靠、低时延的内容共享,同时为基站和回程链路减负.分析了无线分布式缓存系统中无线资源管理所面临的机遇和挑战,针对频谱高效、能量高效及安全传输等关键问题,围绕图论和匹配理论对无线资源管理方法展开了讨论和总结.此外,讨论和分析了挖掘和利用社交信息有效提升无线资源效率的可行性,同时对社会学和无线网络领域的交叉研究进行了展望.     

接入与回传一体化小基站的接入控制与资源分配联合优化算法

针对全双工无线接入与回传一体化小基站场景下长期的频谱效率和能效同时最大化问题,该文提出一种基于近似动态规划理论的接入与回传一体化小基站接入控制与资源分配联合优化算法。该算法首先联合考虑当前基站的资源使用和功率配置情况,在任一用户需求动态到达以及平均时延、小基站回传速率和传输功率约束下,使用受限马尔科夫决策过程(CMDP)建立频谱效率最大化和功率消耗最小化的多目标优化模型,其次运用切比雪夫理论将多目标优化问题转化为单目标问题,并使用拉格朗日对偶分解法进一步转化为非受限的马尔科夫决策过程(MDP)问题。最后,为了解决其求解时存在的“维度灾”爆炸问题,该文提出基于近似动态规划的无线接入与回传一体化小基站资源动态分配算法进行求解,得到此时的接入与资源分配策略。仿真结果表明,所提算法能在保证平均时延约束、小基站回传速率约束和传输功率约束的同时最大化长期平均频谱效率和能效。     

Energy‐efficient resource allocation for device‐to‐device communication through noncooperative game theory

Wireless network with high data rate applications has seen a rapid growth in recent years. This improved quality of service (QoS) leads to huge energy consumption in wireless network. Therefore, in order to have an energy‐efficient resource allocation in cellular system, a device‐to‐device (D2D) communication is the key component to improve the QoS. In this paper, we propose a noncooperative game (NCG) theory approach for resource allocation to improve energy efficiency (EE) of D2D pair. A three‐tier network with macrocell base station (MBS), femtocell base station (FBS), and D2D pair is considered, which shares the uplink resource block. A resource allocation strategy with constraints is arrived, which maintains minimum throughput for each user in the network. The proposed resource allocation strategy optimizes the EE of D2D pair in the three‐tier network, which achieves Nash equilibrium (NE) and Pareto optimality (PO). Simulation results validate that EE is uniform and optimum for all D2D pair, which converges to NE when channel is static and it converges to PO when the channel is dynamic.     

User association and resource allocation in green mobile edge networks using deep reinforcement learning

In order to meet the emerging requirements for high computational complexity, low delay and energy consumption of the 5th generation wireless systems (5G) network, ultra-dense networks (UDNs) combined with multi-access edge computing ( MEC) can further improve network capacity and computing capability. In addition, the integration of green energy can effectively reduce the on-grid energy consumption of system and realize green computation. This paper studies the joint optimization of user association (UA) and resource allocation (RA) in MEC enabled UDNs under the green energy supply pattern, users need to perceive the green energy status of base stations (BSs) and choose the one with abundant resources to associate. To minimize the computation cost for all users, the optimization problem is formulated as a mixed integer nonlinear programming (MINLP) which is NP-hard. In order to solve the problem, a deep reinforcement learning ( DRL)-based association and optimized allocation (DAOA) scheme is designed to solve it in two stages. The simulation results show that the proposed scheme has good performance in terms of computationcost and time out ratio, as well achieve load balancing potentially.     

Joint energy efficiency and spectral efficiency optimization algorithm for UDN under the restriction of interference threshold and backhaul capacity

Aiming at the scenarios which consider the constraint of backhaul capacity restriction and interference threshold in ultra-dense networks (UDN),an integer linear programming (ILP) and Lagrangian dual decomposition (LDD) based joint optimization algorithm of energy efficiency and spectrum efficiency was proposed.In the proposed algorithms,the user association problem with the constraint of limited backhaul capacity was modelled as an ILP problem and then finished the connection between the user and the base station of microcell by solving this problem with dynamic programming method.Therefor,Lagrangian dual decomposition (LDD) was applied in an iteration algorithm for spectrum resource allocation and power allocation.The simulation results show that compared with traditional schemes,the proposed algorithm can significantly improve the energy efficiency and spectrum efficiency of system and use the microcell’s load capacity more efficiently.     

Deep reinforcement learning based resource allocation algorithm in cellular networks

In order to solve multi-objective optimization problem,a resource allocation algorithm based on deep reinforcement learning in cellular networks was proposed.Firstly,deep neural network (DNN) was built to optimize the transmission rate of cellular system and to complete the forward transmission process of the algorithm.Then,the Q-learning mechanism was utilized to construct the error function,which used energy efficiency as the rewards.The gradient descent method was used to train the weights of DNN,and the reverse training process of the algorithm was completed.The simulation results show that the proposed algorithm can determine optimization extent of optimal resource allocation scheme with rapid convergence ability,it is obviously superior to the other algorithms in terms of transmission rate and system energy consumption optimization.     

基于信道定价的无线虚拟网络资源分配策略:匹配/Stackelberg分层博弈

针对分层匹配博弈不能跟踪信道变化以及循环迭代收敛慢等问题,该文提出一种基于信道定价的无线虚拟网络资源分配策略:匹配/Stackelberg分层博弈。分别以基于流带宽的用户满意度、系统带宽及切片功率作为报酬函数建立3级联合优化模型,并采用匹配/Stackelberg分层博弈求解。在博弈下层,定义移动虚拟网络操作者(MVNOs)m—切片n对m_n及其与用户(UEs)的1对1匹配博弈以代替UEs与MVNOs的多对1匹配,对\begin{document}${m_n}$\end{document}定义基于信道平均信息的切片功率价格,加速上、下一致收敛并使UEs适应信道选择最优,证明均衡点存在并给出了低复杂度的分布式拒绝-接收算法;在博弈上层,基于UEs与已匹配关系,形成基础资源提供者(InPs)与的Stackelberg博弈,给出了基于局部信道信息的功率定价和分配策略,使系统效用及频谱效率基于信道最优。最后定义了双层循环稳定条件及过程。仿真表明,该策略在信道跟踪、频谱效率、效用方面均优于随机定价的匹配/Stackelberg分层博弈以及传统分层匹配博弈。     

A resource allocation and cell association algorithm for energy efficiency in HetNets

Aiming at further improving the energy efficiency of system while meeting the data rate requirement of the users, in this paper, a network energy efficiency optimization problem by jointly optimizing the resource allocation of macrocells and small cells is formulated and proven to be an NP‐hard problem, which cannot be worked out by direct methods in polynomial time. Considering the effectiveness of the modified particle swarm optimization (MPSO) algorithm, an MPSO‐based resource allocation and cell association algorithm is employed to solve the proposed joint optimization problem where the frequency resource partitioning of the macrocells, the transmission power, and the cell‐association bias of the small cells are jointly optimized. During the implementation of the optimization algorithm, the cell association alters with the transmission power and cell‐association bias adjustment, and the small cells where there are no active users associating with them will be turned off. The data rate requirement of the users is an indispensable metric in the communication system. Taking no account of the data rate requirement of the users, the frequency resource partitioning and cell association alteration will deteriorate the data rate of a few of users. Under such circumstances, we take the users' data rate requirement as a constraint of the joint optimization problem. The system level simulation results show that, by jointly optimizing the resource allocation of macrocells and small cells while guaranteeing the users' data rate requirement, the energy efficiency and system throughput can be further improved, and energy consumption can be further reduced.