Call Admission Control (CAC) is one of the key traffic management mechanisms that must be deployed in order to meet the strict requirements for dependability imposed on the services provided by modern wireless network...Call Admission Control (CAC) is one of the key traffic management mechanisms that must be deployed in order to meet the strict requirements for dependability imposed on the services provided by modern wireless networks. In this paper, we develop an executable top-down hierarchical Colored Petri Net (CPN) model for multi-traffic CAC in Orthogonal Frequency Division Multiple Access (OFDMA) system. By theoretic analysis and CPN simulation, it is demonstrated that the CPN model is isomorphic to Markov Chain (MC) assuming that each data stream follows Poisson distribution and the corresponding arrival time interval is an exponential random variable, and it breaks through MC's explicit limitation, which includes MC's memoryless property and proneness to state space explosion in evaluating CAC process. Moreover, we present four CAC schemes based on CPN model taking into account call-level and packet-level Quality of Service (QoS). The simulation results show that CPN offers significant advantages over MC in modeling CAC strategies and evaluating their performance with less computational complexity in addition to its flexibility and adaptability to different scenarios.展开更多
软件定义网络(software-defined networks,SDN)流量调度提升网络性能和资源利用率、实现节能和负载均衡至关重要.传统的多目标优化算法在高流量和网络动态性增加的情况下显著影响算法的收敛速度,难以满足复杂网络环境的多样化需求.针对...软件定义网络(software-defined networks,SDN)流量调度提升网络性能和资源利用率、实现节能和负载均衡至关重要.传统的多目标优化算法在高流量和网络动态性增加的情况下显著影响算法的收敛速度,难以满足复杂网络环境的多样化需求.针对此问题,提出了一种基于深度强化学习的流量预测在线路由算法——OTPR-DRL:根据流量特征预测关键流和普通流,结合网络状态和流量信息建立线性规划问题获得关键流路由的最优解.为满足普通流不同服务质量(quality of service,QoS)需求,引入通用效用函数实现多目标优化,通过多智能体和优先级经验回放机制为普通流选择路由.实验结果表明,在高流量强度下,OTPR-DRL与现有的算法相比,提高了收敛速度,至少降低了10.26%的网络传输时延,3.09%的丢包率,提高了1.70%的吞吐率.展开更多
多交路运营是中国城市轨道交通网络化运营组织的重要组成部分,研究乘客在多交路运营条件下的出行选择行为,对把握乘客出行规律、满足多样化出行需求具有重要意义.基于随机后悔最小化模型,引入乘客对路径属性感知的异质性,构建融合效用...多交路运营是中国城市轨道交通网络化运营组织的重要组成部分,研究乘客在多交路运营条件下的出行选择行为,对把握乘客出行规律、满足多样化出行需求具有重要意义.基于随机后悔最小化模型,引入乘客对路径属性感知的异质性,构建融合效用与后悔机制的多尺度混合模型,克服了传统模型未考虑路径熟悉度导致的乘客出行行为与实际出行行为之间的决策偏差.通过整合容忍阈值与决策惯性,提出一种多交路出行选择建模方法,基于典型案例的陈述偏好(stated preference,SP)调查数据,完成模型参数估计与性能验证.研究结果表明,乘客对出行时间属性的容忍阈值为6.98 min;相较于基准模型,考虑决策惯性的模型在似然值、贝叶斯信息准则(Bayesian information criterion,BIC)及命中率指标上均表现更优,表明其具备更强的数据拟合能力;支付意愿分析进一步揭示乘客愿意为服务提升承担额外时间成本,从而验证了所提模型的有效性与实用性.展开更多
基金Supported by the National Natural Science Foundation of China (No. 61271421)the Education Department of Henan Province (No. 2011GGJS-002 and No. 12A510023)
文摘Call Admission Control (CAC) is one of the key traffic management mechanisms that must be deployed in order to meet the strict requirements for dependability imposed on the services provided by modern wireless networks. In this paper, we develop an executable top-down hierarchical Colored Petri Net (CPN) model for multi-traffic CAC in Orthogonal Frequency Division Multiple Access (OFDMA) system. By theoretic analysis and CPN simulation, it is demonstrated that the CPN model is isomorphic to Markov Chain (MC) assuming that each data stream follows Poisson distribution and the corresponding arrival time interval is an exponential random variable, and it breaks through MC's explicit limitation, which includes MC's memoryless property and proneness to state space explosion in evaluating CAC process. Moreover, we present four CAC schemes based on CPN model taking into account call-level and packet-level Quality of Service (QoS). The simulation results show that CPN offers significant advantages over MC in modeling CAC strategies and evaluating their performance with less computational complexity in addition to its flexibility and adaptability to different scenarios.
文摘软件定义网络(software-defined networks,SDN)流量调度提升网络性能和资源利用率、实现节能和负载均衡至关重要.传统的多目标优化算法在高流量和网络动态性增加的情况下显著影响算法的收敛速度,难以满足复杂网络环境的多样化需求.针对此问题,提出了一种基于深度强化学习的流量预测在线路由算法——OTPR-DRL:根据流量特征预测关键流和普通流,结合网络状态和流量信息建立线性规划问题获得关键流路由的最优解.为满足普通流不同服务质量(quality of service,QoS)需求,引入通用效用函数实现多目标优化,通过多智能体和优先级经验回放机制为普通流选择路由.实验结果表明,在高流量强度下,OTPR-DRL与现有的算法相比,提高了收敛速度,至少降低了10.26%的网络传输时延,3.09%的丢包率,提高了1.70%的吞吐率.
文摘多交路运营是中国城市轨道交通网络化运营组织的重要组成部分,研究乘客在多交路运营条件下的出行选择行为,对把握乘客出行规律、满足多样化出行需求具有重要意义.基于随机后悔最小化模型,引入乘客对路径属性感知的异质性,构建融合效用与后悔机制的多尺度混合模型,克服了传统模型未考虑路径熟悉度导致的乘客出行行为与实际出行行为之间的决策偏差.通过整合容忍阈值与决策惯性,提出一种多交路出行选择建模方法,基于典型案例的陈述偏好(stated preference,SP)调查数据,完成模型参数估计与性能验证.研究结果表明,乘客对出行时间属性的容忍阈值为6.98 min;相较于基准模型,考虑决策惯性的模型在似然值、贝叶斯信息准则(Bayesian information criterion,BIC)及命中率指标上均表现更优,表明其具备更强的数据拟合能力;支付意愿分析进一步揭示乘客愿意为服务提升承担额外时间成本,从而验证了所提模型的有效性与实用性.
