共查询到20条相似文献,搜索用时 109 毫秒
1.
2.
3.
4.
5.
随着区块链技术的不断发展,区块的传输延迟成为区块链系统可扩展性的性能瓶颈。远程直接内存访问(RDMA)技术能够支持高带宽和低时延的数据传输,为低延迟区块传输提供了新的思路。因此,结合RDMA原语的特性,设计了用于区块信息共享的区块目录结构,并在此基础上设计并实现了区块传输的基本工作过程。实验结果表明,相较于基于TCP的方案,在1 MB大小的区块上基于RDMA的区块传输机制将节点间的区块传输延迟降低了44%,全网络的区块传输延迟降低了24.4%,在10 000节点规模的区块链上,区块链发生临时分叉的数量降低了22.6%。可见,基于RDMA的区块传输机制充分发挥了高速网络的性能优势,降低了区块传输延迟及临时分叉的数量,提高了现有区块链系统的可扩展性。 相似文献
6.
控制网络中传输延迟的存在,导致确定性闭环网络控制系统转化成一时变不确定性系统.针对网络控制系统中存在小于一个采样周期的传输延迟,提出控制方案,分析建立数学模型,设计出满足要求的时延网络控制器.并通过仿真验证了该结果的正确性,使系统在能控性和稳定性方面得到明显改善. 相似文献
7.
8.
董明忠 《计算机技术与发展》2009,19(7)
IEEE802.11分布式协调机制(DCF)是颇具影响力的无线网络标准,协议基于载波检测/碰撞退避多址接人协议和时隙退避算法.针对IEEE802.11DCF饱和状态不能解决无线网络复杂的工作机制,提出三维Markov链(3M)模型,分析IEEE802.11DCF的非饱和性能,各站点在发送成功后随机产生新的数据分组,且到达率满足Poisson分布.三维Markov模型的仿真与理论分析吻合,这表明3M对系统的吞吐率、延迟性能、稳定性等参数有明显的提高与改善. 相似文献
9.
根据Ad Hoc网络MAC层DCF的指数退避机制及二维Markov链分析模型,分析退避计数器在退避过程当中每一次递减的时间间隔,提出逻辑时隙的概念,并得出逻辑时隙的3个组成部分,推导出饱和负载条件下DCF的平均介质访问延迟等于逻辑时隙的平均长度乘以成功发送数据所用的平均退避逻辑时隙数.实验表明,该方法能较准确地计算出Ad Hoc网络的DCF介质访问延迟. 相似文献
10.
一种带有动态输出反馈控制器的网络控制系统的Markov跳变模型 总被引:3,自引:0,他引:3
对于网络控制系统(Networked control systems, NCS)中的随机延迟时变有界, 其上界大于一个采样周期的情况, 建立了带有动态输出反馈控制器的连续时间系统模型. 以此为基础, 应用连续时间系统模型离散化与增广状态空间方法, 建立了受控于有限状态Markov链的离散时间跳变系统模型. 在该系统的稳定性控制器设计中, 给出了一种求内点法的初始可行解的方法, 从而解决了所建模型的可解性问题. 理论分析和车载倒立摆上的仿真计算表明, 由上述模型与改进算法所给出的动态输出反馈控制器能使系统稳定. 相似文献
11.
网络控制系统随机稳定性研究 总被引:7,自引:2,他引:7
研究了具有随机网络诱导时延及数据包丢失的网络控制系统随机稳定性问题. 本文用一个具有两个状态的马尔可夫链来描述数据通过网络传输时随机数据包丢失过程, 利用马尔可夫跳变线性系统理论, 将网络控制系统建模为一个具有两种运行模式的马尔可夫跳变线性系统, 给出了在状态反馈控制下网络控制系统随机稳定的线性矩阵不等式形式的充分条件, 最后用一个仿真示例验证了该方法的有效性. 相似文献
12.
This paper considers robust stochastic stability and PI tracking control problem for Markov jump systems with both input delay and an unknown nonlinear function. Based on the traditional PI control strategy, a new controller design scheme is proposed for nonlinear time-delay Markov jump systems which can realize multiple control objectives including robust stochastic stability and tracking performance. By using the Lyapunov stability theory and LMI algorithms, a sufficient condition for the solution to robust stochastic stability and tracking control problem is obtained. Then, the desired controller with PI structure is designed, which ensures the resulting closed-loop system is robust stochastically stable and the system state has favorable tracking performance. Finally, a numerical example is provided to illustrate the effectiveness of the proposed results. 相似文献
13.
14.
15.
相较于传统的地面认知网络,星地认知网络链路传输时延较长,因此基于实时信道感知的认知用户中断概率较高。以离散时间马尔科夫链描述授权用户的动态及衰落信道,同时考虑信道转移概率的不确定性,建立基于似然不确定性模型的信道状态马尔科夫链,利用鲁棒的向后递归方法得到离线存储的功率分配矩阵,并对网络中多个用户采用博弈论的方法,进而提出基于中断概率的功率控制算法。仿真结果表明,与RRAP和TS-RS-PA算法相比,该算法在网络延迟存在的情况下能够更有效地降低中断概率并节约能耗。 相似文献
16.
采用滑动扇区方法,研究了不确定随机马尔可夫跳变系统的变结构控制设计问题。首先给出随机马尔可夫跳变系统滑动扇区的定义,然后基于线性矩阵不等式技术,提出一种滑动扇区及变结构控制律设计方法。经过理论证明该控制律能够确保随机马尔可夫跳变不确定系统二次稳定,并有效地抑制抖振。最后数值仿真算例验证了控制方案的有效性。 相似文献
17.
The deregulation of the electricity market made the open communication infrastructure an exigent need for future power system. In this scenario dedicated communication links are replaced by shared networks. These shared networks are characterized by random time delay and data loss. The random time delay and data loss may lead to system instability if they are not considered during the controller design stage. Load frequency control systems used to rely on dedicated communication links. To meet future power system challenges these dedicated networks are replaced by open communication links which makes the system stochastic. In this paper, the stochastic stabilization of load frequency control system under networked environment is investigated. The shared network is represented by three states which are governed by Markov chains. A controller synthesis method based on the stochastic stability criteria is presented in the paper. A one-area load frequency control system is chosen as case study. The effectiveness of the proposed method for the controller synthesis is tested through simulation. The derived proportion integration (PI) controller proves to be optimum where it is a compromise between compensating the random time delay effects and degrading the system dynamic performance. The range of the PI controller gains that guarantee the stochastic stability is determined. Also the range of the PI controller gains that achieve the robust stochastic stability is determined where the decay rate is used to measure the robustness of the system. 相似文献
18.
针对一类具有随机时延的网络控制系统(NCSs)周期反馈控制问题,提出一个新颖的马尔可夫随机时滞系统模型.该模型包括一个分段常数广义采样保持函数(PCGSHF)以及一个用于描述网络诱导时延和数据包丢失的马尔可夫链.应用李亚普诺夫方法以及线性矩阵不等式技术得到了闭环 NCSs 随机稳定的充分条件,并给出了状态反馈控制器和 PCGSHF 的设计方法.最后,通过仿真算例验证了设计方法的有效性. 相似文献
19.
20.
In this paper, adaptive NN control is proposed for bilateral teleoperation system with dynamic uncertainties, unknown external disturbances, and unsymmetrical stochastic delays in communication channel to achieve transparency and robust stability. Compared with previous passivity‐based teleoperation framework, the communication delays are unsymmetrical and stochastic. By partial feedback linearization using nominal dynamics, the nonlinear dynamics of the teleoperation system are transformed into two subsystems: local master/slave dynamics control and time‐delay motion tracking. By integrating Markov jump systems and adaptive parameters updating, adaptive NN control strategy is developed. The stability of the closed‐loop system and the boundedness of tracking errors are proved using Lyapunov–Krasovskii functional synthesis under specific linear matrix inequalities conditions. The proposed adaptive NN control is robust against motion disturbances, parametric uncertainties, and unsymmetrical stochastic delay, which effectiveness is validated by extensive simulation studies. Copyright © 2013 John Wiley & Sons, Ltd. 相似文献