基于NOMA的超密集MEC网络任务卸载和资源分配方案

为了提高移动边缘计算(mobile edge computing,MEC)网络的频谱效率,满足大量用户的服务需求,建立了基于非正交多址接入(non-orthogonal multiple access,NOMA)的超密集MEC系统模型。为了解决多个用户同时卸载带来的严重通信干扰等问题,以高效利用边缘服务器资源,提出了一种联合任务卸载和资源分配的优化方案,在满足用户服务质量的前提下最小化系统总能耗。该方案联合考虑了卸载决策、功率控制、计算资源和子信道资源分配。仿真结果表明,与其他卸载方案相比,所提方案可以在满足用户服务质量的前提下有效降低系统能耗。     

基于非正交多址接入的移动边缘计算安全节能联合资源分配

为提高基于非正交多址接入(NOMA)的移动边缘计算(MEC)系统中计算任务部分卸载时的安全性,该文在存在窃听者情况下研究MEC网络的物理层安全,采用保密中断概率来衡量计算卸载的保密性能,考虑发射功率约束、本地任务计算约束和保密中断概率约束,同时引入能耗权重因子以平衡传输能耗和计算能耗,最终实现系统能耗加权和最小。在满足两个用户优先级情况下,为降低系统开销,提出一种联合任务卸载和资源分配机制,通过基于二分搜索的迭代优化算法寻求问题变换后的最优解,并获得最优的任务卸载和功率分配。仿真结果表明,所提算法可有效降低系统能耗。     

无人机辅助MEC系统中基于最优SIC顺序的能耗优化方案

在基于上行非正交多址接入（NOMA）的无人机（UAV）辅助移动边缘计算（MEC）系统中,NOMA的连续干扰消除（SIC）顺序已成为限制上行任务卸载链路传输性能的瓶颈,为降低系统能耗,对SIC顺序进行了讨论,提出了联合信道增益与任务时延约束的最优SIC顺序。在满足设备给定任务时延、设备最大发射功率约束以及UAV轨迹的约束下,基于最优SIC顺序提出了最小化系统能耗的问题。由于该问题是个复杂的非凸问题,采取交替优化的方法求解该优化问题,以实现功率分配和UAV轨迹的优化;利用匹配理论,提出了低复杂度算法来得到不同时隙的最优设备分组。仿真结果表明,与其他SIC顺序相比,最优SIC顺序能够在相同的任务时延约束下实现更小的系统能耗;所提的低复杂度设备分组算法能够得到最优设备分组。     

车联网中基于NOMA-MEC的卸载策略研究

随着车联网(IoV)的迅猛发展,请求进行任务卸载的汽车终端用户也逐渐增长,而基于移动边缘计算(MEC)的通信网络能够有效地解决任务卸载在上行传输时延较高的挑战,但是该网络模型同时也面临着信道资源不足的问题。该文引入的非正交多址(NOMA)技术相较于正交多址(OMA)能够在相同的信道资源条件下为更多的用户提供任务卸载,同时考虑到任务卸载过程中多方面的影响因子,提出了混合NOMA-MEC卸载策略。该文设计了一种基于深度学习网络(DQN)的博弈算法,帮助车辆用户进行信道选择,并通过神经网络多次迭代学习,为用户提供最优的功率分配策略。仿真结果表明,该文所提出的混合NOMA-MEC卸载策略能够有效地优化多用户卸载的时延以及能耗,最大限度保证用户效益。     

车联网中基于NOMA-MEC的卸载策略研究

随着车联网(IoV)的迅猛发展,请求进行任务卸载的汽车终端用户也逐渐增长,而基于移动边缘计算(MEC)的通信网络能够有效地解决任务卸载在上行传输时延较高的挑战,但是该网络模型同时也面临着信道资源不足的问题。该文引入的非正交多址(NOMA)技术相较于正交多址(OMA)能够在相同的信道资源条件下为更多的用户提供任务卸载,同时考虑到任务卸载过程中多方面的影响因子,提出了混合NOMA-MEC卸载策略。该文设计了一种基于深度学习网络(DQN)的博弈算法,帮助车辆用户进行信道选择,并通过神经网络多次迭代学习,为用户提供最优的功率分配策略。仿真结果表明,该文所提出的混合NOMA-MEC卸载策略能够有效地优化多用户卸载的时延以及能耗,最大限度保证用户效益。     

上行免调度NOMA系统中基于遗传算法的扩频矩阵优化方法

为了提升基于压缩感知(Compressive Sensing,CS)框架下的免调度非正交多址接入(Non-Orthogonal Multiple Access,NOMA)系统的信道估计和多用户检测性能,本文提出了一种基于遗传算法(Genetic Algorithm,GA)的扩频矩阵优化方法.该方法以最小化扩频矩阵的互相...     

基于强化学习的非正交多址接入和移动边缘计算联合系统信息年龄更新

物联网发展对信息时效性的需求越来越高,信息新鲜度变得至关重要。为了维持信息新鲜度,在非正交多址接入(NOMA)和移动边缘计算(MEC)的联合系统中,对多设备单边缘计算服务器的传输场景进行了研究。在该场景中,如何分配卸载任务量和卸载功率以最小化平均更新代价是一个具有挑战性的问题。该文考虑到现实中的信道状态变化情况,基于多代理深度确定性策略梯度(MADDPG)算法,考虑信息新鲜度影响,建立了最小化平均更新代价的优化问题,提出一种寻找最优的卸载因子和卸载功率决策。仿真结果表明,采用部分卸载的方式可以有效地降低平均更新代价,利用MADDPG算法可以进一步优化卸载功率,经比较,MADDPG算法在降低平均更新代价方面优于其他方案,并且适当地减少设备数量在降低平均更新代价方面效果更好。     

一种基于Max-Log-MPA算法的改进方法

非正交多址接入技术(Non-Orthogonal Multiple Access,NOMA)具备高频谱效率和大连接的特性.随着移动数据和用户数量的爆炸式增长,NOMA技术的代表之一——稀疏码多址接入(Sparse Code Multiple Access,SCMA)技术具有愈发重要的研究意义.为了降低SCMA系统的检测...     

基于NOMA的协作中继网络安全性能分析

由于非正交多址接入(Non-orthogonal Multiple Access,NOMA)能够显著提升系统的频谱资源利用率,在下一代移动通信中得到广泛应用。对NOMA环境下多中继协作网络的最优中继选择方案和系统安全性能进行了分析和讨论,其中包含窃听者仅窃听中继和窃听者同时窃听中继及源节点这2种情况下的系统安全性能表现,并与相同场景下正交多址接入(Orthogonal Multiple Access,OMA)网络进行了对比。最终的理论分析和仿真结果表明,在所提出系统模型中,当信道条件相同时,NOMA网络总能取得相较于OMA网络更好的安全性能,同时随着系统中继节点数目的增多,NOMA网络在物理层安全性能上获得更大的优势。     

面向后5G的非正交多址技术综述

在后5G时代,作为一种候选方案,非正交多址技术(Non-Orthogonal Multiple Access,NOMA)正在被5G的演进技术标准讨论,该技术可以满足大规模连接和高吞吐量的要求。阐述了NOMA技术的基本原理、技术提案、性能仿真和潜在的研究方向。首先概述了NOMA技术的发展和原理,比较了NOMA相比于正交多址技术(Orthogonal Multiple Access,OMA)的优势;根据NOMA技术在资源块(Resource Block,RB)上复用方式的不同,从比特级和符号级的层面讨论了NOMA技术不同的技术路线。其次着重阐述了NOMA技术的仿真实验,以多用户共享接入(Multi-User Shared Access,MUSA)和稀疏码多址接入技术(Sparse Code Multiple Access,SCMA)为例进行了性能分析。最后给出了NOMA在未来的潜在研究方向,包括与多入多出(Multiple-Input Multiple-Output,MIMO)结合、与认知无线电结合和全双工NOMA等。     

数字孪生使能的智能超表面边缘计算网络任务卸载

针对新兴的计算密集型应用对移动用户高计算性能需求问题,该文提出一种数字孪生(DT)结合智能反射面(RIS)辅助的移动边缘计算(MEC)任务卸载方案。首先,在满足用户传输功率、用户和资源设备能耗、计算资源限制条件下,通过联合优化用户卸载决策、用户传输功率、RIS 相移、波束成形矢量、计算资源分配,建立一个系统能耗最小化问题;其次,将该非凸组合优化问题分解为3个子问题,使用深度双Q网络(DDQN)方法确定用户卸载策略;然后对每个训练时间步进行一次求解,基于交替迭代方法得到问题的优化解。仿真结果表明,基于DDQN的算法训练速度较快,有效降低了系统总能耗。     

Joint resource allocation for hybrid NOMA-assisted MEC in 6G networks

Multi-access Edge Computing (MEC) is an essential technology for expanding computing power of mobile devices, which can combine the Non-Orthogonal Multiple Access (NOMA) in the power domain to multiplex signals to improve spectral efficiency. We study the integration of the MEC with the NOMA to improve the computation service for the Beyond Fifth-Generation (B5G) and the Sixth-Generation (6G) wireless networks. This paper aims to minimize the energy consumption of a hybrid NOMA-assisted MEC system. In a hybrid NOMA system, a user can offload its task during a time slot shared with another user by the NOMA, and then upload the remaining data during an exclusive time duration served by Orthogonal Multiple Access (OMA). The original energy minimization problem is non-convex. To efficiently solve it, we first assume that the user grouping is given, and focuses on the one group case. Then, a multilevel programming method is proposed to solve the non-convex problem by decomposing it into three subproblems, i.e., power allocation, time slot scheduling, and offloading task assignment, which are solved optimally by carefully studying their convexity and monotonicity. The derived solution is optimal to the original problem by substituting the closed expressions obtained from those decomposed subproblems. Furthermore, we investigate the multi-user case, in which a close-to-optimal algorithm with low-complexity is proposed to form users into different groups with unique time slots. The simulation results verify the superior performance of the proposed scheme compared with some benchmarks, such as OMA and pure NOMA.     

基于NOMA的移动边缘计算系统公平能效调度算法

将移动边缘计算技术(MEC)与非正交多址技术(NOMA)结合,同时考虑公平性,该文研究了采用NOMA上行部分卸载的MEC系统公平能效问题。首先将基于公平函数的用户速率与功耗比值定义为公平能效函数,随后提出了两种公平能效调度准则下的能效调度算法,即最大化最小速率准则下DK-SCA算法及最大化系统能效准则下DK-SCALE算法,通过算法实现分别得到两种公平能效调度准则下用户最佳本地CPU处理频率及最佳传输功率。最后通过仿真表明,与基准方案相比,所提基于NOMA的部分卸载方案能够有效地将本地计算和基于NOMA的边缘卸载结合,达到最佳的公平能效性能。     

超密集网络中基于移动边缘计算的任务卸载和资源优化

移动边缘计算(MEC)通过在无线网络边缘为用户提供计算能力,来提高用户的体验质量。然而,MEC的计算卸载仍面临着许多问题。该文针对超密集组网(UDN)的MEC场景下的计算卸载,考虑系统总能耗,提出卸载决策和资源分配的联合优化问题。首先采用坐标下降法制定了卸载决定的优化方案。同时,在满足用户时延约束下采用基于改进的匈牙利算法和贪婪算法来进行子信道分配。然后,将能耗最小化问题转化为功率最小化问题,并将其转化为一个凸优化问题得到用户最优的发送功率。仿真结果表明,所提出的卸载方案可以在满足用户不同时延的要求下最小化系统能耗,有效地提升了系统性能。     

Computation energy efficiency maximization based resource allocation scheme in wireless powered mobile edge computing network

For wireless powered mobile edge computing (MEC) network,a system computation energy efficiency (CEE) maximization scheme by considering the limited computation capacity at the MEC server side was proposed.Specifically,a CEE maximization optimization problem was formulated by jointly optimizing the computing frequencies and execution time of the MEC server and the edge user(EU),the transmit power and offloading time of each EU,the energy harvesting time and the transmit power of the power beacon.Since the formulated optimization problem was a non-convex fractional optimization problem and hard to solve,the formulated problem was firstly transformed into a non-convex subtraction problem by means of the generalized fractional programming theory and then transform the subtraction problem into an equivalent convex problem by introducing a series of auxiliary variables.On this basis,an iterative algorithm to obtain the optimal solutions was proposed.Simulation results verify the fast convergence of the proposed algorithm and show that the proposed resource allocation scheme can achieve a higher CEE by comparing with other schemes.     

基于无人机无线能量传输的边缘计算系统能耗优化方法研究

无线能量传输(WPT)和移动边缘计算(MEC)可以为无线设备提供能量供应和任务计算,有效提高设备的能量效率。该文提出一种基于无人机无线能量传输的边缘计算系统能耗优化方法,在所提方法中,通过联合优化能量收集(EH)时间、用户发射功率和卸载决策,最小化系统总能耗。利用块坐标下降法(BCD),将优化问题分解为两个子问题,通过交替优化来获得最优能量收集时间、用户发射功率和卸载决策。仿真结果表明,该文提出的系统能耗优化方法优于其他基准方案,并且系统所需能量可以显著减少。     

Energy consumption optimization for self‐powered IoT networks with non‐orthogonal multiple access

Energy harvesting (EH) has been considered as one of the promising technologies to power Internet of Things (IoT) devices in self‐powered IoT networks. By adopting a typical harvest‐then‐transmit mode, IoT devices with the EH technology first harvest energy by using wireless power transfer (WPT) and then carry out wireless information transmission (WIT), which leads to the coordination between WPT and WIT. In this paper, we consider optimizing energy consumption of periodical data collection in a self‐powered IoT network with non‐orthogonal multiple access (NOMA). Particularly, we take into account time allocation for the WPT and WIT stages, node deployment, and constraints for data transmission. Moreover, to thoroughly explore the impact of different multiple access methods, we theoretically analyse and compare the performance achieved by employing NOMA, frequency division multiple access (FDMA), and time division multiple access (TDMA) in the considered IoT network. To validate the performance of the proposed method, we conduct extensive simulations and show that the NOMA outperforms the FDMA and TDMA in terms of energy consumption and transmission power.     

Users' experience matter: Delay sensitivity-aware computation offloading in mobile edge computing

As a promising computing paradigm, Mobile Edge Computing (MEC) provides communication and computing capability at the edge of the network to address the concerns of massive computation requirements, constrained battery capacity and limited bandwidth of the Internet of Things (IoT) systems. Most existing works on mobile edge task ignores the delay sensitivities, which may lead to the degraded utility of computation offloading and dissatisfied users. In this paper, we study the delay sensitivity-aware computation offloading by jointly considering both user's tolerance towards delay of task execution and the network status under computation and communication constraints. Specifically, we use a specific multi-user and multi-server MEC system to define the latency sensitivity of task offloading based on the analysis of delay distribution of task categories. Then, we propose a scoring mechanism to evaluate the sensitivity-dependent utility of task execution and devise a Centralized Iterative Redirection Offloading (CIRO) algorithm to collect all information in the MEC system. By starting with an initial offloading strategy, the CIRO algorithm enables IoT devices to cooperate and iteratively redirect task offloading decisions to optimize the offloading strategy until it converges. Extensive simulation results show that our method can significantly improve the utility of computation offloading in MEC systems and has lower time complexity than existing algorithms.