Abstract: In this paper we perform a detailed analysis of point-to-point packet delay in an operational tier-1 network. The point-to-point delay is the time between a packet entering a router in one PoP (an ingress point) and its leaving a router in another PoP (an egress point). It measures the one-way delay experienced by packets from an ingress point to an egress point across an ISP's network and provides the most basic information regarding the delay performance of the ISP's network. Using packet traces captured in the operational network, we obtain precise point-to-point packet delay measurements and analyze the various factors affecting them. Through a simple, step-by-step, systematic methodology and careful data analysis, we identify the major network factors that contribute to point-to-point packet delay and characterize their effect on the network delay performance. Our findings are: 1) delay distributions vary greatly in shape, depending on the path and link utilization; 2) after constant factors dependent only on the path and packet size are removed, the 99th percentile variable delay remains under 1 ms over several hops and under link utilization below 90% on a bottleneck; 3) a very small number of packets experience very large delay in short bursts

Abstract: Sensor networks have many potential applications in biology, physics, medicine, and the military. One major challenge in sensor networks is to maximize network life under the constraint of limited power supply. The paper addresses energy-efficiency in the context of routing and data gathering. A new protocol is proposed: Hybrid Indirect Transmission (HIT). HIT is based on a hybrid architecture that consists of one or more clusters, each of which is based on multiple, multi-hop indirect transmissions. In order to minimize both energy consumption and network delay, parallel transmissions are used both among multiple clusters and within a cluster. This is made possible by having each sensor independently compute a medium access controlling TDMA schedule. The computation within each sensor is intelligent yet simple. Formal analysis shows that it requires O(n) space and O(n x logn) time complexities, and O(1) setup messages prior to the computation, where n is the total number of sensors. HIT does not require sensor nodes with CDMA capability, or the remote base station to compute a data gathering schedule. Performance is evaluated by simulating and comparing HIT with three other existing protocols, including Low Energy Adaptive Clustering Hierarchy (LEACH), Power Efficient Gathering for Sensor Information System (PEGASIS), and Direct Transmission. Results have shown that HIT greatly reduces both energy consumption and network delay; it also maintains longer network life compared to these three existing protocols. Security issues and a potential application of HIT in biomedical sensing technology are also rigorously discussed. This work is significant to the advancement of energy-efficient micro sensor networks; the proposed protocol is promising and would contribute to the use of wireless micro sensor networks in future biomedical sensing technologies.

Abstract: Reducing the channel zapping time is critical problem to resolve for the spread of the IPTV service. In general, channel change times depends on several Parameters: command processing time, network delay time, STB layer delay time, STB jitter buffer delay time and MPEG decoder time. We focus on the network delay time, which does not exist in the traditional broadcasting service such as terrestrial, cable, or satellite broadcasting service. This could weaken the spread of IPTV service. In this paper we present the method of improving the channel zapping time by making adjacent multicast groups of current channel come to home gateway in advance. When IP STB requests join to new channel by sending IGMP membership report message, home gateway sends not only IGMP membership report message for the channel but also for the adjacent channels. Therefore multicast streams for the adjacent channels of current one always come to home gateway. When IP STB requests join to an adjacent channel, home gateway can forward the corresponding multicast stream immediately.

Abstract: Distributed multi-player games use dead reckoning vectors to intimate other (at a distance) participating players about the movement of any entity by a controlling player. The dead reckoning vector contains the current position of the entity and the velocity components. When a participating player receives a vector, traditionally it puts the entity at the current position specified by the vector and starts projecting the path of the entity from that point using the local clock of the receiver. In this paper we show that this traditional method of usage of dead reckoning vector brings in inaccuracy in the receivers' rendering of the entity. This inaccuracy can be substantial even with low network delay between the sender-receiver pairs and increases with network delay. We propose the use of globally synchronized clocks among the participating players and a time-stamp augmented dead reckoning vector that enables the receiver to render the entity accurately. We modified the popular game BZFlag with this technique, and compared the accuracy seen in game playing using the traditional method and the proposed technique. We conducted several types of experiments varying the frequency of generation of dead reckoning vectors and the delay between the sender and the receivers. The experiments show significant quantitative improvement in accuracy even for 100ms delay between the sender-receiver pairs and appreciable qualitative improvement in game playing experience.

Abstract: For assessment of a coverage of a cellular network system, in particular a CDMA cellular network system, at least one present quality value of the network system is measured at a predetermined location, under a present load of the network system. As well, at least one required quality value of the network system is determined which is required for offering a predetermined service under a predetermined load, in particular a target load, of the network system. Finally, the required quality value is compared with the measured quality value for determining whether the predetermined location is covered by the network system for the predetermined service under the predetermined load of the network system. By employing these steps, large network areas may be analyzed without influencing the traffic in the network. Reliable results are obtained without using expensive simulations depending from theoretical models and assumptions.

Abstract: Computer networks have progressed from a simple store-and-forward medium to a complex communication infrastructure. Routers in the network need to implement a variety of functions ranging from simple packet classification for forwarding and firewalling to complex payload modifications for encryption and content adaptation. As these functions increase in number and complexity, more processing time is required, and packets experience a significant processing delay. In most network simulations, this delay has not been addressed because it was considered negligible. However, we show that this network processing delay can reach the magnitude of long-distance propagation delay and thus becomes a significant contributor to the overall packet delay. We evaluate different network applications and develop a model that characterizes packet processing cost with only a few parameters that can easily be derived from our simulations. To validate our simulation and our model, we compare them to actual network measurements. The contributions of this work can be used to increase the accuracy of network simulations and improve network performance estimations.

Abstract: An arrangement which jointly exploits an agent component installed in a server-side portion of client-server architecture for emulating requests of actual clients of the client-server architecture and measuring a server execution time indicative of a time elapsed between reception of an emulated end-user request at a server component and generation of a result of said emulated end-user request at said server component; information stored in a server log file about a hit end to end response time, i.e., the time elapsed between the instant in which the end-user sends a request to the server component and the instant in which result of the request reaches the end-user. From the server execution time and the hit response time, the delay due to the network connecting the server-side portion of the client-server architecture to a client-side portion may be determined. The network delay and the server execution time are then used for monitoring performance of the client-server architecture by distinguishing network related problems from server-related problems or even from client-related problems.

Abstract: It is widely assumed that highly interactive online games require a minimum quality of service (QoS) from the network, which suggests that offering premium IP service quality for game players may be a new source of revenue for Internet service providers (ISPs). In order to offer such services, ISPs must know the upper bounds of the performance metrics players are willing to tolerate. This paper reports on our preliminary attempts to experimentally establish performance bounds for different types of online games. First we placed a group of players in a controlled network environment, where artificial network delay and loss is introduced during their games. We logged objective measures (game performance indicators from the game server) and used a questionnaire to establish a subjective measure of user perceived quality as a function of different network conditions. Our paper concludes with an analysis of our results and a comparison with previous work based on indirect measurements.

Abstract: End-to-end measurement is a common tool for network performance diagnosis, primarily because it can reflect user experience and typically requires minimal support from intervening network elements. However, pinpointing the site of performance degradation from end-to-end measurements is a challenging problem. In this paper, we show how end-to-end delay measurements of multicast traffic can be used to infer the under-lying logical multicast tree and the packet delay variance on each of its links. The method does not depend on cooperation from intervening network elements; multicast probing is bandwidth efficient. We establish desirable statistical properties of the estimator, namely consistency and asymptotic normality. We evaluate the approach through simulations, and analyze its failure modes and their probabilities.

Abstract: Network selection methods and apparatus with home network prioritization after network signal recovery and/or power-on are disclosed. In one illustrative example involving automatic network selection, a mobile station (200) selects and operates with a non-home communication network (406). The mobile station (200) then experiences an out-of-coverage condition (or a power down condition) but subsequently regains signal coverage (or is powered back on). In response, the mobile station (200) scans to identify a plurality of communication networks in its coverage area. If a home network (402) is identified as being available, the mobile station (200) selects and operates with the home network (402). Otherwise, if the previous non-home network (406) (e.g. the RPLMN) is identified as being available, the mobile station (200) continues operation with the previous non-home network (406).

Abstract: A method and apparatus for providing wireless data packet call regulation and call access for wireless Voice over Internet Protocol (VoIP) communications in a wireless network. A first and second time stamp is generated upon the reception and successful transmission of a data packet. A packet delay is determined based upon the first and second time stamps. At least one average delay metric is used to determine whether to provide new call access based on the at least one average delay metric. Another aspect of the present invention is to block a class of users based, in part, on the average sector delay for each user subscriber class. If the average sector delay is greater than that allowed for a class of users, users of that class are denied access. A delay quality of service (QoS) block metric is used to determine if an individual user should be force dropped.

Abstract: A method for operating a network and a network architecture implementing the method are described. The method, in one embodiment, begins by grouping a plurality of nodes into zones, where the network includes the plurality of nodes. At least one of the nodes in each one of the zones is one of a plurality of boundary nodes, and each of the boundary nodes in each one of the zones is coupled to a boundary node in another of the zones by one of a first number of inter-zone optical links. One of the zones includes a number of nodes, each of which is coupled to at least one other of the nodes by one of a second number of intra-zone optical links. Next, a non-boundary node is configured to transmit network information to other of the nodes. The non-boundary node is a node in the zone that is not a boundary node. Finally, a boundary node in the zone is configured to limit transmission of the network information through itself to other of the boundary nodes. The network information so limited can be restoration information (in the event of a failure), network topology information, and/or other network information.

Abstract: The popularity of network-based control systems (NBCS) is continuously growing. One of the most intriguing aspects is the transportation of control network data over IP-based networks using accepted standards such as EIA-852. To a large extent the actual quality of control (QoC) in such systems depends on the network timing such as delay and delay jitter. This paper presents a classification of relevant quality of service parameters and identifies application classes. Subsequently, the paper focuses on the effect of delay jitter at a fixed mean delay on the QoC. Two sources of delay jitter are identified in IP-based control systems: 1) network traffic induced and 2) protocol induced. As an example of a simple control loop implemented over an EIA-852-based system we investigate how the induced jitter affects the QoC using a time-discrete simulation model. Conclusions are drawn as to how the findings in the EIA-852 system can be interpreted and extended to a generalized NBCS.

Abstract: One of the major hurdles limiting IPv6 adoption is the existence of poorly managed experimental IPv6 sites that negatively affect the perceived quality of the IPv6 Internet. To assist network operators in improving IPv6 networks, we are exploring methods to identify wide-area IPv6 network problems. Our approach makes use of parallel IPv4 and IPv6 connectivity to dual-stacked nodes.We identify the existence of an IPv6 path problem by comparing IPv6 delay measurements to IPv4 delay measurements. Our test results indicate that the majority of IPv6 paths have delay characteristics comparable to those of IPv4, although a small number of paths exhibit a much larger delay with IPv6. Thus, we hope to improve the quality of the IPv6 Internet by identifying the worst set of problems.Our methodology is simple. We create a list of systems with IPv6 and IPv4 addresses in actual use by monitoring DNS messages. We then measure delay to each address in order to select a few systems per site based on their IPv6:IPv4 response-time ratios. Finally, we run traceroute with Path MTU discovery to the selected systems and then visualize the results for comparative path analysis. This paper presents the tools used to support this study, and the results of our measurements conducted from two locations in Japan and one in Spain.

Abstract: A method and apparatus are disclosed for generating routing information in a data communications network. A first network element (such as a router) receives information relating to a second network element, such as another node or a network link. In response, the first network element determines whether the information relating to the second network element indicates a change in the network. When information relating to a second network element indicates a change in the network, the first network element determines a new shortest path through the network from the first network element for each network element in the network. After a delay, the first network element updates routing information for the first network element based on the new shortest path for the first network element. Preferably the delay is proportional to the distance of the first network element from the second network element.

Abstract: One of the most critical issues in wireless ad hoc networks is represented by the limited availability of energy within network nodes. Thus, making good use of energy is a must in ad hoc networks. In this paper, we define as network lifetime the time period from the instant when the network starts functioning to the instant when the first network node runs out of energy. Our objective is to devise techniques to maximize the network lifetime in the case of cluster-based systems, which represent a significant sub-set of ad hoc networks. Cluster-based ad hoc networks comprise two types of nodes: cluster-heads and ordinary nodes. Cluster-heads coordinate all transmissions from/to ordinary nodes and forward all traffic in a cluster, either to other nodes in the cluster or to other cluster-heads. In this case, to prolong the network lifetime we must maximize the lifetime of the cluster-heads because they are the critical network element from the energy viewpoint. We propose an original approach to maximize the network lifetime by determining the optimal assignment of nodes to cluster-heads. Given the number of cluster-heads, the complexity of the proposed solution grows linearly with the number of network nodes. The network topology is assumed to be either static or slowly changing. Two working scenarios are considered. In the former, the optimal network configuration from the energy viewpoint is computed only once; in the latter, the network configuration can be periodically updated to adapt to the evolution of the cluster-heads energy status. In both scenarios, the presented solution greatly outperforms the standard assignment of nodes to cluster-heads, based on the minimum transmission power criterion.

Abstract: A wireless communication unit (101, 200), network switch (112, 300) and corresponding methods facilitate routing and handover or handin of calls from a second network (105) to a first network (103) when the unit has been operating in second network and is being handed in to the first network. The communication unit comprises a transceiver (203) and a controller (209) arranged to operate with the transceiver to determine (503) that the wireless communication unit is within a coverage area of the first network; to exchange messages (505, 507) corresponding to call appearance information with the switch in the first network; to decide whether a call within the second network can be handed in to the first network based on the exchange of the messages (507); and to initiate (519, 619) a handin of the call to the first network only when the decision is affirmative.

Abstract: A system and method is provided for using network asset management data including attributes such as device type, location, and link speed for modeling network performance and identifying performance issues. The invention provides measured and observed performance data for one or more segments of a network (e.g., one or more hops or end to end) so that identification of diminished performance in the network might be accomplished. Additionally, the invention may provide projections of performance impacts based upon proposed changes to one or more components in the network. Performance metrics, ping times, MIB data, link speeds, propagation delays, device latency, serialization rates, conversational speeds, conversational utilization and other data are used to model a network for determining or isolating problematic issues and to forecast performance impacts of potential changes to the network.

Abstract: The present invention is a delay based model and in fact uses queueing delay as a congestion measure, providing advantages over prior art loss based systems. One advantage is that queueing delay can be more accurately estimated than loss probability. This is because packet losses in networks with large bandwidth-delay product are rare events under TCP Reno and its variants , and because loss samples provide coarser information than queueing delay samples. Indeed, measurements of delay are noisy, just as those of loss probability. Thus, another advantage of the present invention is that each measurement of queueing delay provides multi-bit information while each measurement of packet loss (whether a packet is lost) provides only one bit of information for the filtering of noise. This makes it easier for an equation-based implementation to stabilize a network into a steady state with a target fairness and high utilization.

Abstract: An approach for pre-activating network devices includes using prior usage of a network device to estimate times when the network device will be used in the future. The network device is then pre-activated, i.e., transitioned to an active operational state, prior to the estimated times so that the network device will be ready when needed. This reduces or eliminates delays in processing attributable to having to wait for the network device to transition to the active operational state.

Abstract: Peering allows service providers to handle traffic surges without over-provisioning, reduce the cost of dedicated infrastructure, and leverage the specialization and prices of partner providers. We develop a peering system for multi-provider content delivery based on a cost-optimized peer selection algorithm. We formulate a cost model for evaluating competing peering strategies, and use measurement data collected from globally distributed network probe stations, large-scale Web sites, and existing service provider infrastructures to empirically evaluate proposed peering strategies. Our analysis shows that our peer selection algorithm is significantly more efficient than greedy alternatives, in terms of minimizing service cost and respecting network delay and server capacity thresholds, over a broad range of real-world scenarios.

Abstract: Methods and systems for correlating network traffic between non-translative network systems are provided. Generally, time-based offset data, or transmission latency, is first determined between devices in non-translative network segments by injecting, with a time stamp, a known network pattern at a first end of the network topology. Traces are then recorded, with time stamps, of the network traffic over one or more nodes throughout the non-translative network. The generate network traffic is then compared to the traced network traffic in a best fit to thereby determine the time latency in traffic throughout the network. Later, when it is desired to determine causality of network activity between non-translative network segments, the determined latency between different network devices can be compared to traced patterns of network traffic to determine the origin of a network operation that created an observed event.

Abstract: A communication network with controlled routing operations, including a set of nodes, wherein topological information on the network is broadcast to all the nodes in the set. The network is associated with an external network, such as, a standard public network, and certain nodes are configured to perform a gateway function between the network and the associated external network. Each node is configured for building two sets of addresses including one primary address, which is fixed and independent from the prefixes announced by the gateways and one or more secondary addresses which are valid global IP addresses, based on the global prefixes announced by the gateways. Each node in the network is configured for using one of the secondary addresses as the source address of outgoing traffic, and for generating and broadcasting multiple interface messages containing all its secondary addresses in the place of interface addresses.

Abstract: We propose a two dimensional probabilistic cellular automata for the description of traffic flow in a small city network composed of two intersections. The traffic in the network is controlled by a set of traffic lights which can be operated both in fixed-time and a traffic responsive manner. Vehicular dynamics is simulated and the total delay experienced by the traffic is evaluated within specified time intervals. We investigate both decentralized and centralized traffic responsive schemes and in particular discuss the implementation of the green-wave strategy. Our investigations prove that the network delay strongly depends on the signalisation strategy. We show that in some traffic conditions, the application of the green-wave scheme may destructively lead to the increment of the global delay.

Abstract: With a view on improving user-perceived performance on networks supporting best effort flows, e.g., multimedia/data file transfers, we propose a family of bandwidth allocation criteria that depends on the residual work of on-going transfers. Analysis and simulations show that allocating bandwidth in this fashion can significantly improve the user-perceived delay, bit transmission delay, and throughput over traditional approaches, e.g., by 58% on an 80% loaded linear network. A simple implementation based on TCP Reno, exemplifies how one might approach practically realizing such gains. We discuss several other advantages of incorporating such differentiation at the transport level. In particular we make the case that favoring small transfers combined with user impatience or peak rate constraints, both of which are natural mechanisms for users to express the utility of completing transfers, offers a lightweight approach to achieving good overall network goodput and/or utility for best effort networks.

Abstract: Location-aware applications require a geographic location service of Internet hosts. We focus on a measurement-based service for the geographic location of Internet hosts. Host locations are inferred by comparing delay patterns of geographically distributed landmarks, which are hosts with a known geographic location, with the delay pattern of the target host to be located. Results show a significant correlation between geographic distance and network delay that can be exploited for a coarse-grained geographic location of Internet hosts.

Abstract: A network node in the communication network is a target network node, connected by a first communication path with at least one first and one second network node, connected to each other by means of a further communication path. A routing table is held in each network node. The direct communication path from the first network node to the target network node is entered in the routing table of the first network node as the primary routing path and the path leading from the first network node to the second network node and then to the target network node is entered as the reserve routing path. An analogous entry is made in the routing table of the second network node. The relevant reserve routing path is only used on disruption of the primary routing path. A disruption in the primary communication path, provided between the first network node and the target network node is reported by the first network node, by means of a message sent to the second network node. This controls the second network node such that, on a drop-out of the primary connection path thereof to the target network node, a transmission of data packets to the target network node by means of the reverse routing path, leading from the second network node to the target network node via the first network node, is prevented.

Abstract: Disclosed is a method and apparatus for aligning clock domains over an asynchronous network between a source controlled by a first clock and a destination controlled by a second clock. The predicted delay is estimated for transmitting packets between a source and destination over the network. The time-stamped synchronization packets are sent to the destination, each time-stamped synchronization packet carries timing information based on a master clock at the source. A set of synchronization packets are received at the destination to create a set of data points, and the set of data points is weighted so that synchronization packets exhibiting a delay further from the expected delay are accorded less weight than synchronization packets exhibiting a delay closer to the expected delay. The expected delay is updated to create a current delay estimate based on the set of data points taking into account the different weighting of the data points. These steps are continually repeated on new sets of data points created from newly received synchronization packets using the current delay estimate for the expected delay. And a clock domain at the destination is continually aligned with a clock domain at the source based on the current delay estimate for packets traversing the network between the source and destination.