The need to provide different service levels through the Internet has stimulated the research on differentiated services (DiffServ). This architecture is based on a set of mechanisms, deployed along an aggregated flow...
详细信息
The need to provide different service levels through the Internet has stimulated the research on differentiated services (DiffServ). This architecture is based on a set of mechanisms, deployed along an aggregated flow path, designed to offer differentiated treatments. Some approaches have appeared with the goal to decrease the implementation complexity of these mechanisms. C. L. Lee et al. have proposed a scalable architecture based on the DiffServ philosophy, where core routers do not need to support any specific function for service differentiation. In our work, we analyze, through simulation, the performance of that architecture regarding disordered packet loss. We also consider a different structure for queueing occupancy control that significantly reduces the rate of packet loss in the core of the domain. Therefore in our approach, the whole DiffServ functionality could efficiently be executed only in the edge routers, including the packet discard control function.
Quality of service provisioning in wireless ad hoc networks plays an integral part in determining the success of network-centric warfare as envisioned in future military operations. It requires good scalability of the...
详细信息
Quality of service provisioning in wireless ad hoc networks plays an integral part in determining the success of network-centric warfare as envisioned in future military operations. It requires good scalability of the QoS architecture since ad hoc networks in the battlefield tend to be large. Previous work attacking QoS in ad hoc networks seldom considers the scalability issues. In this paper, we propose a scalable QoS architecture for such networks. Our scheme draws upon the positive aspects of both IntServ and DiffServ, and extends upon the scalable LANMAR routing protocol to support QoS. The scheme is also capable of incorporating mobile backbone networks (MBNs) to further improve the scalability. Simulation results show that our proposed QoS architecture can achieve good scalability in terms of large network size and mobility.
In this paper a novel algorithm is proposed which combines the merits of explicit congestion notification (ECN) and backward explicit congestion notification (BECN) mechanisms for congestion control in TCP/ipnetworks...
详细信息
In this paper a novel algorithm is proposed which combines the merits of explicit congestion notification (ECN) and backward explicit congestion notification (BECN) mechanisms for congestion control in TCP/ipnetworks. A comparative performance evaluation of the combined ECN and BECN mechanism is carried out using both long-lived FTP flows and short-lived Web traffic. The simulation results show that the combined ECN+BECN mechanism benefits from BECN's early notification under heavy congestion and ECN's reliable delivery of congestion notification. It is observed that the ECN+BECN mechanism significantly reduces queue fluctuations due to early congestion indication compared to ECN. The loss of BECN Internet control message protocol (ICMP) source quenches on the reverse path does not adversely impact performance due to this combined approach. It is also shown that the ECN+BECN scheme can significantly reduce the ICMP source quench reverse traffic in a network compared to a BECN only network. Experiments with Web traffic workloads show measurable improvement in both average object transfer delay and fairness over ECN.
In current research on mobile ad-hoc networks, test-beds are always needed. However, real test-beds are costly, and sometimes unfeasible, especially for large-scale mobile ad-hoc networks. Therefore, emulation systems...
详细信息
ISBN:
(纸本)0780378229
In current research on mobile ad-hoc networks, test-beds are always needed. However, real test-beds are costly, and sometimes unfeasible, especially for large-scale mobile ad-hoc networks. Therefore, emulation systems are provided. Unfortunately, exiting emulation systems have obvious defects. They can hardly support scalability or real-time simultaneously, neither adding nor removing mobile node dynamically. In this paper, we propose a test system named ManTS. With distributed architecture and some novel approaches, such as direct transmittal and virtual application traffic, and others. ManTS is able to build test-beds for large-scale mobile ad-hoc networks and guarantee real-time emulation at the same time. Furthermore, ip or upper layer protocols and applications implemented on Linux can run on ManTS without any modification. And with ManTS, simulated mobile nodes can be added into and deleted from mobile ad-hoc networks freely, which is quite useable to test routing and transport protocols. In a word, ManTS is a more flexible test system, and really suitable to test ip and upper layer protocols and applications for large-scale mobile ad-hoc networks.
Musical collaboration over telecommunication networks has marvellous possibilities when it comes to musical education, practise and performance. The deployment and use of high-capacity digital networks makes it possib...
详细信息
Musical collaboration over telecommunication networks has marvellous possibilities when it comes to musical education, practise and performance. The deployment and use of high-capacity digital networks makes it possible to obtain end-to-end audio and video latency of 5-20 milliseconds, which is similar to the audio time delay typically experienced between musicians on a stage. The main challenge of reaching this latency budget over digital packet switched network such as the Internet, is how to control the queuing delay of the ip packet experienced at each router on the network paths between each participating musician. Other important aspects are video and audio codec latency, and error resilient tools needed to cope with situations where ip packets arrive too late or have been lost in the network. We give an overview of resent research (work-in-progress) in this field, including network simulations, conducted at the NTNU, and suggest how these techniques can be utilized under live concert performances.
This paper provides a reference way to realize bandwidth control, a basic and important part of QoS, in WLAN. Using iptables and TC (trafficcontrol) of Linux, AC (Access control, the boundary node of a mini WLAN) con...
详细信息
ISBN:
(纸本)0780377028
This paper provides a reference way to realize bandwidth control, a basic and important part of QoS, in WLAN. Using iptables and TC (trafficcontrol) of Linux, AC (Access control, the boundary node of a mini WLAN) controls all packets pass by. Packets can be dropped, be deferred or be delivered at once. Then, AC can control any user's bandwidth according to his ip address, MAC address, user name and so on. Besides that, it can set a user's priority with similar ways and a user with high priority will be served preferentially. At the end of this paper, we will talk about some possible improvement.
In previous years, the Internet using a conventional ip mechanism had a lot of problems such as slow packet forwarding, complex routing and inefficient usage of network resources. Multi-Protocol Label Switching (MPLS)...
详细信息
In previous years, the Internet using a conventional ip mechanism had a lot of problems such as slow packet forwarding, complex routing and inefficient usage of network resources. Multi-Protocol Label Switching (MPLS) is one of the mechanisms that are proposed to solve these problems. In addition, multicast is one of the big issues in the Internet society because it can efficiently support real-time multimedia traffic. Therefore, the next generation Internet will be a system which have an MPLS backbone and can support multicast. An MPLS multicast mechanism is considered under Internet broadcasting systems. Specially, the scalability problem of common problems in MPLS and multicast is solved using label aggregation. In the proposed algorithm, the Label Edge Router (LER) of edge router in MPLS has a new table containing information of each tree node. The table is called the multicast Tree Node Table. Extending the Constraint-based Routed Label Distribution Protocol (CR-LDP), the label allocation is performed by using this table. The scalability problem can be solved using this algorithm because it reduces label consumption. So, the MPLS multicast mechanism can be easily supported in a large Internet backbone network.
Web service has become the largest consumer of Internet resources and is a promising 3G wireless service. Constructing an accurate model of Web traffic is essential for networking experiments involving end-to-end perf...
详细信息
ISBN:
(纸本)7563506861
Web service has become the largest consumer of Internet resources and is a promising 3G wireless service. Constructing an accurate model of Web traffic is essential for networking experiments involving end-to-end performance. In this paper, we employ an approach of analyzing TCP/ip header to obtain Web traffic model. By comparing Web traffic of wireless and wire network, we present some different characteristics between them that reflect the influences of wireless limited bandwidth and high tariff. Furthermore, Hurst parameter shows that the self-similar nature of wireless Web traffic still exists.
One ip terminal can occupy a multiple number of slots within time frames in the GPRS networks. A limited number of radio resources (slots) are allocated in a base station for such ip terminals. If one ip terminal can ...
详细信息
One ip terminal can occupy a multiple number of slots within time frames in the GPRS networks. A limited number of radio resources (slots) are allocated in a base station for such ip terminals. If one ip terminal can occupy only one slot discontinuously in a time frame, for a real time packet transmission, the number of accepted calls is obtained by dividing the number of radio resources during that time by the maximum allowed number of ip calls. The real time iptraffic flow over the GSM-GPRS network is discussed in [J.H. Sarker and S.J. Halme, 2002]. An analysis for the non-real time bursty ip packet transmission over the GPRS air interface that measures the channel efficiency and the probability that a packet is forced for queuing and the header/payload ratio are executed in this paper. Results show that the channel efficiency as well as the probability that a packet is forced for queuing increases with increasing call intensity, maximum number of admitted ip calls and the burstiness factor. The channel utilization decreases with the increase of header/payload ratio.
暂无评论