IoT devices have significantly altered the methods of interaction, operation, and functionality within home environments. However, individuals, particularly those with limited technical proficiency who stand to gain the most from these advancements, likely encounter challenges stemming from the intricate setup processes, a critical stage with the potential to limit their widespread adoption. Thus, we focus on the user experience during the setup phase of mainstream smart home devices and conduct an empirical study of 15 representative smart home IoT devices. We scrupulously examine their setup processes, as well as accompanying instructions and user manuals, to assess multi-faceted usability concerns. Our findings reveal 19 usability issues, indicating notable barriers, inconsistencies, and a lack of intuitiveness, which may deter consumers from successfully configuring and using these devices.

IoT devices are susceptible to botnet malware due to inherent system limitations. Once compromised, these devices can be exploited to leak private user information or launch DDoS attacks. In this paper, we present Wital, an adaptable mitigation strategy that curtails the malicious use of IoT devices through a stringent whitelist-based approach. This method combines static manufacturer usage profiling and dynamic traffic monitoring into a client-side firewall. Our strategy ensures that, even if devices are compromised, their impact is minimized. Using the P4 language for programmable data planes, we prototype the firewall solutions and demonstrate that this solution can significantly reduce potential device exploitation.

IoT devices benefit from nuanced policy enforcement to ensure safe, reliable, and efficient operations. However, the practical implementation of IoT policy enforcement systems involves multifaceted complexities. This paper delves into three-dimensional challenges: the intricacies of architecture and interoperational support, the expressiveness and safety concerns of policy languages, and the steep learning curve associated with policy creation. We investigate these areas, evaluate options, and prototype systems, considering the WoT, Pkl, and LLMs as potential solutions to address each challenge, respectively.

Synthetic traffic generation can produce sufficient data for model training of various traffic analysis tasks for IoT networks with few costs and ethical concerns. However, with the increasing functionalities of the latest smart devices, existing approaches can neither customize the traffic generation of various device functions nor generate traffic that preserves the sequentiality among packets as the real traffic. To address these limitations, this paper proposes IoTGemini, a novel framework for high-quality IoT traffic generation, which consists of a Device Modeling Module and a Traffic Generation Module. In the Device Modeling Module, we propose a method to obtain the profiles of the device functions and network behaviors, enabling IoTGemini to customize the traffic generation like using a real IoT device. In the Traffic Generation Module, we design a Packet Sequence Generative Adversarial Network (PS-GAN), which can generate synthetic traffic with high fidelity of both per-packet fields and sequential relationships. We set up a real-world IoT testbed to evaluate IoTGemini. The experiment result shows that IoTGemini can achieve great effectiveness in device modeling, high fidelity of synthetic traffic generation, and remarkable usability to downstream tasks on different traffic datasets and downstream traffic analysis tasks.

We propose and implement Directory-Based Access Control (DBAC), a flexible and systematic access control approach for geographically distributed multi-administration IoT systems. DBAC designs and relies on a particular module, IoT directory, to store device metadata, manage federated identities, and assist with cross-domain authorization. The directory service decouples IoT access into two phases: discover device information from directories and operate devices through discovered interfaces. DBAC extends attribute-based authorization and retrieves diverse attributes of users, devices, and environments from multi-faceted sources via standard methods, while user privacy is protected. To support resource-constrained devices, DBAC assigns a capability token to each authorized user, and devices only validate tokens to process a request.

With over 20 million units sold since 2015, Amazon Echo, the Alexa-enabled smart speaker developed by Amazon, is probably one of the most widely deployed Internet of Things consumer devices. Despite the very large installed base, surprisingly little is known about the device's network behavior. We modify a first generation Echo device, decrypt its communication with Amazon cloud, and analyze the device pairing, Alexa Voice Service, and drop-in calling protocols. We also describe our methodology and the experimental setup. We find a minor shortcoming in the device pairing protocol and learn that drop-in calls are end-to-end encrypted and based on modern open standards. Overall, we find the Echo to be a well-designed device from the network communication perspective.

To scale the Internet of Things (IoT) beyond a single home or enterprise, we need an effective mechanism to manage the growth of data, facilitate resource discovery and name resolution, encourage data sharing, and foster cross-domain services. To address these needs, we propose a GlObaL Directory for Internet of Everything (GOLDIE). GOLDIE is a hierarchical location-based IoT directory architecture featuring diverse user-oriented modules and federated identity management. IoT-specific features include discoverability, aggregation and geospatial queries, and support for global access. We implement and evaluate the prototype on a Raspberry Pi and Intel mini servers. We show that a global implementation of GOLDIE could decrease service access latency by 87% compared to a centralized-server solution.

As IoT services scale up from single homes to smart cities, directories and mapping services are needed to manage potentially millions of devices. However, directory service providers will likely struggle to accommodate the increasing number of IoT devices, made more challenging by their heterogeneous metadata and the large volume of queries. One of the critical challenges, the high heterogeneity of IoT, is being addressed by a working standard of W3C, which formalizes a physical or virtual device as a formatted Thing Description (TD).We propose a local directory service architecture with a series of design requirements. With a focus on query performance, we build a proof-of-concept system to store metadata of IoT devices as TDs in terms of the working standard. A Raspberry Pi is configured to investigate the query performance of relational database and non-relational database as the classic choices for internal directories. Evaluation results demonstrate that compared with relational database, non-relational database can achieve 2.9 times higher resilience on property query and 2.35 times faster processing on spatial query, with mild loss on aggregation query.

With the proliferation of online video, measuring quality of experience (QoE) has become a pivotal aspect for the analysis of today's over-the-top (OTT) video streaming. To monitor video QoE, we introduce YouSlow that can detect various playback events (e.g., start-up latency, rebufferings and bitrate changes) from video players while a video is being played. Using YouSlow, we have collected more than 400,000 YouTube views from more than 100 countries. We measured the impact of these playback events on video abandonment and found that rebufferings incur abandonment rates six times higher than start-up latency, mostly caused by pre-roll ads. A single rebuffering event has three times the impact of a bitrate change. Even increasing the bitrate can raise abandonment rates by a factor of four compared to keeping the bitrate constant.

Container-based clouds have recently received great attention from the industry. However, we notice that this new type of cloud inevitably requires complex network setups and configurations from both providers and customers when deployed on an existing cloud system; Providers need to install additional network elements such as proxy servers and Network Address Translation (NAT), and customers need to use subdomain names and randomly assigned port numbers to access their services. Thus, we propose a new network architecture that performs M-to-N mapping between network addresses and containers in order to simplify the network setup and configuration. To achieve our goals, we adopt a software-defined networking (SDN) approach. We discuss the benefits and use cases of our approach, and present detailed designs and implementation.

On the Internet, most traffic is generated by content delivery applications such as video streaming, P2P and web surfing. With the proliferation of mobile applications that offer on-demand access to services, the ability to track end-user's perceived service quality has become a pivotal aspect for future networks such as 5G. Software-defined networking (SDN) and network functions virtualization (NFV) in the mobile core offer unprecedented flexibility for opportunistic traffic steering, leading to advanced levels of quality-of-service (QoS) control. We leverage the SDN concept to dynamically control network traffic over wide area networks (WANs) from edge nodes of wireless networks (e.g., a packet data network gateway in LTE), depending on changing network conditions and application types. Our SDN-based application-aware routing system allows mobile network operators to achieve better utilization of their networks, service providers to improve customer satisfaction, and end-users to experience desirable service quality for various network applications. As for a prototype, we attach 802.11g to Mininet's virtual networks in order to show the feasibility of our approach.

Today's over the top (OTT) video service providers take advantage of content distribution networks (CDNs) and adaptive bitrate (ABR) streaming where a video player adjusts resolutions based on end-to-end network conditions. Although the mechanisms are useful to improve user-perceived video quality, they do not resolve the root causes of congestion problems. To pinpoint a bottleneck and improve video quality-of-experience (QoE), we leverage a software-defined networking (SDN) platform from OTT video service provider's point of view. Our proposed SDN application is designed to monitor network conditions of streaming flow in real time and dynamically change routing paths using multi-protocol label switching (MPLS) traffic engineering (TE) to provide reliable video watching experience. We use an off-the-shelf SDN platform to show the feasibility of our approaches.

Existing failure diagnostic techniques for end users are insufficient to pinpoint the root causes of network failures due to their limited capabilities to probe other network elements. We present DYSWIS, an automatic network fault detection and diagnosis system for end-users. DYSWIS leverages user collaboration to distinguish important network faults from false positive indications, and diagnoses the root cause of the fault using diagnostic rules that consider diverse information from multiple nodes. Our rule system is specially designed to support crowdsourcing and distributed probes. We have implemented DYSWIS and compared its performance with other tools to prove that several network failures which are difficult to be diagnosed by the single-user probe can be detected and diagnosed successfully with our approach.

Slow Internet connectivity is often caused by poor Wi-Fi performance. The main reasons of such performance degradation include channel contention and non-Wi-Fi interference. Although these problem sources can be easily removed in many cases once they are discovered, it is difficult for end users to identify the sources of such interference. We investigated the characteristics of different sources that can degrade Wi-Fi performance, and developed WiSlow, a software tool that diagnoses the root causes of poor Wi-Fi performance using user-level network probes, and leveraging peer collaboration to identify the physical location of these causes. WiSlow uses two main methods: packet loss analysis and 802.11 ACK number analysis. The accuracy of WiSlow exceeds 90% when the sources are close to Wi-Fi devices. Also, our experiment proves that the collaborative approach is feasible for determining the relative location of an interfering device.

We investigate video server selection algorithms in a distributed video-on-demand system. We conduct a detailed study of the YouTube Content Delivery Network (CDN) on PCs and mobile devices over Wi-Fi and 3G networks under varying network conditions. We proved that a location-aware video server selection algorithm assigns a video content server based on the network attachment point of a client. We found out that such distance-based algorithms carry the risk of directing a client to a less optimal content server, although there may exist other better performing video delivery servers. In order to solve this problem, we propose to use dynamic network information such as packet loss rates and Round Trip Time (RTT) between an edge node of a wireless network (e.g., an Internet Service Provider (ISP) router in a Wi-Fi network and a Radio Network Controller (RNC) node in a 3G network) and video content servers, to find the optimal video content server when a video is requested. Our empirical study shows that the proposed architecture can provide higher TCP performance, leading to better viewing quality compared to location-based video server selection algorithms.

Adaptive bitrate (ABR) technologies are being widely used in today's popular HTTP-based video streaming such as YouTube and Netflix. Such a rate-switching algorithm embedded in a video player is designed to improve video quality-of-experience (QoE) by selecting an appropriate resolution based on the analysis of network conditions while the video is playing. However, a bad viewing experience is often caused by the video player having difficulty estimating transit or client-side network conditions accurately. In order to analyze the ABR streaming performance, we developed YouSlow, a web browser plug-in that can detect and report live buffer stalling events to our analysis tool. Currently, YouSlow has collected more than 20,000 of YouTube video stalling events over 40 countries.

Troubleshooting network problems on networked home devices is not easy, because most devices have insufficient computing power to run sophisticated diagnostic tools and have no user interfaces to debug the problem directly. We propose MoT, a network problem diagnosis platform that leverages the collaboration of smart objects, smartphones, and computers. We take advantage that recent devices have multiple communication interfaces. Therefore, when a device has a problem with an interface, it can send a probe request to other devices using an alternative interface. We use collaborative mechanisms to diagnose the root cause of a network problem. It can use cooperation from internal nodes or send requests to external nodes. We demonstrate the feasibility of this approach by implementing an Android application and an algorithm that diagnoses a push notification failure.

We present a study of traffic behavior of two popular over-the-top (OTT) video streaming services (YouTube and Netflix). Our analysis is conducted on different mobile devices (iOS and Android) over various wireless networks (Wi-Fi, 3G and LTE) under dynamic network conditions. Our measurements show that the video players frequently discard a large amount of video content although it is successfully delivered to a client. We first investigate the root cause of this unwanted behavior. Then, we propose a Quality-of-Service (QoS)-aware video streaming architecture in Long Term Evolution (LTE) networks to reduce the waste of network resource and improve user experience. The architecture includes a selective packet discarding mechanism, which can be placed in packet data network gateways (P-GW). In addition, our QoS-aware rules assist video players in selecting an appropriate resolution under a fluctuating channel condition. We monitor network condition and configure QoS parameters to control availability of the maximum bandwidth in real time. In our experimental setup, the proposed platform shows up to 20.58% improvement in saving downlink bandwidth and improves user experience by reducing buffer underflow period to an average of 32 seconds.

Accurately determining a user's floor location is essential for minimizing delays in emergency response. This paper presents a floor localization system intended for emergency calls. We aim to provide floor-level accuracy with minimum infrastructure support. Our approach is to use multiple sensors, all available in today's smartphones, to trace a user's vertical movements inside buildings. We make three contributions. First, we present a hybrid architecture that combines sensor-based dead reckoning and beacon-based infrastructure. Second, we present the elevator module for tracking a user's movement in an elevator. The elevator module addresses three core challenges that make it difficult to accurately derive displacement from acceleration. Third, we present the stairway module which determines the number of floors a user has traveled on foot. Unlike previous systems that track users' foot steps, our stairway module uses a novel landing counting technique.

Cloud computing is great for scaling applications but the latency in a guest VM can be unpredictable due to resource contention between neighbors. For telephony applications, which are latency-sensitive, we propose a system to monitor telephony server latencies and adapt the server load based on the measured latencies. We implemented the system and evaluated it on an Amazon EC2 test bed. We show indirectly by comparing our server on EC2 and on a local VM, that there may be contention between EC2 VMs in the wild that leads to higher server latency. While there is some overhead due to constant monitoring of the server, our system manages to lower latency by reducing the load to the server.

This paper shows an innovative solution for distributing dynamic sensor data by using distributed caches. Our proposal is based on the concepts of service modularization and virtualization of network nodes made available by the NetServ hosting environment, which has been defined and implemented with the aim of extending the functions of the network nodes. Through a lab experiment involving tens of nodes, we have demonstrated a significant performance improvements in term of traffic saving and download time in comparison with a legacy, Internet-based, approach. Beyond this performance improvements, the proposed solution holds also functional improvements, in terms of dynamic deployment and easy integration with services making use of sensor data.

Content-centric networks promise to address content networking issues in a better way than today's host-based networking architecture. But content-centric networking does not inherently address the issue of services, particularly service scalability and mobility. We present our work on CCNxServ, a system that allows for dynamic service deployment and scalability in a content-centric networking implementation (CCNx) through an intuitive use of the content naming scheme. It thus extends the concept of content-centric networking towards services.

In this paper, we propose a new gossip-based signaling dissemination method for the Next Steps in Signaling protocol family. In more detail, we propose to extend the General Internet Signaling Transport (GIST) protocol, so as to leverage these new dissemination capabilities from all NSIS Signaling Layer Protocol applications using its transport capabilities. The extension consists of two main procedures: a bootstrap procedure, during which new GIST-enabled nodes discover each other, and a dissemination procedure, which is used to effectively disseminate signaling messages within an Autonomous System. To this aim, we defined three dissemination models, bubble, balloon, and hose, so as to fulfill requirements of different network and service management scenarios. An experimental campaign carried out on GENI shows the effectiveness of the proposed solution.

Autonomic management capabilities of the Future Internet can be provided through a recently proposed service architecture called NetServ. It consists of the interconnection of programmable nodes which enable dynamic deployment and execution of network and application services. This paper shows how this architecture can be further improved by introducing the OpenFlow architecture and implementing the OpenFlow controller as a NetServ service, thus improving both the NetServ management performance and its flexibility. These achievements are demonstrated experimentally on the GENI environment, showing the platform self-protecting capabilities in case of a SIP DoS attack.

We present a mobility solution for stateless applications, where the mobile host can change its IP address as well as the content servers used by ongoing client sessions. This allows content retrieval to always use the locally optimal source when the host moves between networks. We refer to this approach as “stateless mobility” since neither the network nor the content servers hold mobility-related state information. Our mobility solution, referred to as MOSAIC, is applied to HTTP sessions using GET method invocation, which represent a large fraction of mobile Internet traffic. By moving MOSAIC underneath the socket interface, we create a generic application-independent feature. We realized a lightweight implementation using an L3 packet filter, which promises easy portability to other platforms. MOSAIC's overall concept is evaluated via measurements using public Internet services.

Securing Voice over IP (VoIP) is a crucial requirement for its successful adoption. A key component of this is securing the signaling path, which is performed by the Session Initiation Protocol (SIP). Securing SIP can be accomplished by using Transport Layer Security (TLS) instead of UDP as the transport protocol. However, using TLS for SIP is not yet widespread, perhaps due to concerns about the performance overhead. This paper studies the performance impact of using TLS as a transport protocol for SIP servers. We evaluate the cost of TLS experimentally using a testbed with OpenSIPS, OpenSSL, and Linux running on an Intel-based server. We analyze TLS costs using application, library, and kernel profiling and use the profiles to illustrate when and how different costs are incurred. We show that using TLS can reduce performance by up to a factor of 17 compared to the typical case of SIP-over-UDP. The primary factor in determining performance is whether and how TLS connection establishment is performed due to the heavy costs of RSA operations used for session negotiation. This depends both on how the SIP proxy is deployed and what TLS operation modes are used. The cost of symmetric key operations such as AES, in contrast, tends to be small. Network operators deploying SIP-over-TLS should attempt to maximize the persistence of secure connections and will need to assess the server resources required. To aid them, we provide a measurement-driven cost model for use in provisioning SIP servers using TLS. Our cost model predicts performance within 15% on average.

While VoIP spread widely, its deployments are isolated islands interconnected via the PSTN and E.164 phone numbers are still the universal addresses for phones. Mapping phone number to SIP address is essential for SIP peering based on phone numbers. The public ENUM designed for this has not been successful due to the complexity of phone number ownership delegation. This paper proposes a secure and reliable mechanism for automatic phone number mapping verification, addressing the problem of the public ENUM and also enabling phone number based P2P SIP peering establishment.

A handful of proposals address the problem of bootstrapping a large DHT network from scratch, but they all forgo the standard DHT join protocols in favor of their own distributed algorithms that build routing tables directly. Motivating their algorithms, the proposals make a perfunctory claim that the standard join protocols are not designed to handle the huge number of concurrent join requests involved in such a bootstrapping scenario. Moreover, the proposals assume a pre-existing unstructured overlay as a starting point for their algorithms. We find the assumption somewhat unrealistic. We take a step back and reexamine the performance of the standard DHT join protocols. Starting with nothing other than a well-known bootstrap server, when faced with a large number of nodes joining nearly simultaneously, can the standard join algorithms form a stable DHT overlay? If so, how quickly? Our simulation results show that Chord and Kademlia's join protocols can actually handle the bootstrapping scenario quite well. For 10,000 nodes joining at a rate of 1,000 nodes per second, Chord and Kademlia took less than 20 and 15 seconds, respectively, to form a stable overlay. The Chord join protocol, however, requires a slight modification for fast bootstrapping. We elucidate the reason why the modification is necessary.

An important group of location-based services (LBS), including 9-1-1 service, rely on the mapping between a user's location and a service boundary in order to select the appropriate service provider. In such cases, mobile clients can cache the mapping information to reduce service latency and server load. However, caching such a mapping can be burdensome on mobile devices because representing the polygon that defines a service boundary requires a large amount of data. We present GeoPS-PD, a polygon simplification algorithm designed for LBS applications. Unlike existing algorithms, GeoPS-PD never produces a false positive, is tunable at runtime for the desired balance between target polygon size and area coverage, and optionally takes into account the population density. We demonstrate the efficacy of GeoPS-PD using the US state boundary data. For New York, GeoPS-PD produces a simplified polygon which is only 3% of the original size, yet covers 95% of the original area, and makes the LBS queries 3.17 times faster.

Content-centric networks have been proposed as a new network paradigm that is centered around the distribution of content. A key idea of content-centric networks is to address content by name and to enable nodes in the network to respond to content requests. Most proposals for content-centric networks require a "clean slate" approach and a replacement of today's TCP/IP protocol stack, which raises questions about a feasible deployment path. In this paper, we ask the question to which extent the ideas of content-centric networks can be realized on top of today's IP protocol suite. We explore an approach for name-based addressing that extends today's TCP/IP protocols in a fully standard compliant way. We implement our new method in order to demonstrate its feasibility and evaluate the performance of the system using both latency and processing overhead as measures. The obtained results demonstrate that name-based addressing on IP is feasible. We also acknowledge that content-centric networks designed from the ground up go beyond what can be achieved on top of IP.

The NGN (Next Generation Network), which can provide advanced multimedia services over an all-IP based network, has been the subject of considerable attention for several years. While there have been tremendous efforts to develop its architecture and protocols, especially for IMS, a key technology of the NGN, its wide deployment is still a long way off. However, efforts to create an advanced signaling infrastructure able to meet many requirements have resulted in a large number of functional components and interactions between these components. Thus, the carriers are trying to explore effective ways to deploy IMS while offering value-added services. As one such approach, we have proposed a self-organizing IMS. A self-organizing IMS enables IMS functional components and corresponding physical nodes to adapt dynamically and automatically based on actual conditions such as network load and available system resources while continuing IMS operation. To realize this, service continuity for users is an important requirement when a reconfiguration occurs during operation. In this paper, we propose a mechanism that will provide service continuity to users without any impact on clients by extending the SBC (Session Border Controller). Furthermore, we implement it, show its behavior and evaluate its processing time.

Store-carry-forward communication, which is set as the heart of all routing protocols for mobile disruption-tolerant networks (DTNs), exploits nodes' mobility to bring messages closer to their destinations by exchanging messages across mobile nodes when they meet in close proximity. Understanding the subtle characteristics of human mobility leads to better service and application provisioning for mobile DTNs. We use GPS traces collected from multiple mobile users to empirically study different aspects of human mobility. Various Markov models (first, second and third-order) are estimated from users' mobility data. Based on empirical evidence, second-order Markov models are deemed sufficient to estimate mobile users' future locations accurately. These Markov models permit the design of a new routing algorithm for mobile DTNs capable of more efficiently routing data objects to their destination locations. The relay selection in this routing algorithm is based on mobile users' absorption times to the destination location. Simulations show that the proposed routing algorithm consumes less energy than legacy epidemic routing algorithms without excessive transmission delays.

Many applications in VANETs use broadcast communications to disseminate information or keep proper functionality in a network. When the broadcast is multi-hop, it tends to cause unnecessary resource consumption due to the excessive redundant packets on the network, this is known as the broadcast storm problem. Recently, the broadcast storm problem has gained attention due to the impact on VANETs. Many studies emphasize the importance of the forwarding area as a factor for choosing forwarder vehicles. But as important as the use of the area is knowing how many candidates there are to forward a broadcast message. This paper proposes a technique to mitigate that problem through a game theory mechanism: “the volunteer's dilemma”. With this mechanism, the probability of a vehicle forwarding a message is computed, taking into account the cost and the benefit of message propagation. The mechanism's behavior was evaluated by simulations using NS-2. Results showed that when comparing it to a protocol that uses only the area which will be covered as factor for calculating the forwarding probability, packet loss ratio was reduced, as well as latency.

We present a one-semester transition course intended to bridge the gap between a Java-based introductory sequence and advanced systems courses. We chose to structure our course as a series of lab assignments that, while independent, are also milestones in a single main project, writing a web server from scratch. By anchoring the course on a single real-world application, we were able to provide depth, instill good programming practices, give insight into systems, and generate excitement.

Capability-based systems that use explicit authorization (permission) for flows have been proposed in order to prevent Denial-of-Service (DoS) attacks. Even though the performance analyses of these systems show that they are efficient in preventing the attacks, they suffer from the difficulty of obtaining permission, incompatibility with current network architecture, and attacks that circumvent the permission rules. We propose a signaling architecture for network traffic authorization, called Permission-Based-Sending (PBS), aiming to prevent DoS attacks. PBS uses the concept similar to existing capability-based systems in the manner that the sender should get authorization (permission) from a receiver for flows. However, PBS introduces new and practical approaches to overcome the deficiencies of those systems. On-path signaling enables easy installation and management of the permission state. Working on current network protocols supports compatibility and allows PBS to be deployed in existing networks. In addition, a monitoring mechanism provides a second line of defense against attacks. Our analysis and performance evaluation show that PBS is an effective and scalable solution to prevent several kinds of attacks, and improves the resilience of the system against network failure by using soft-state mechanisms.

The 11 papers in this special issue on mission critical networking are divided into three categories: quality of service issues (three papers); security issues (four papers); and configuration and data collection issues (four papers).

TCP has traditionally been considered inappropriate for real-time applications. Nonetheless, popular applications such as Skype use TCP since UDP packets cannot pass through restrictive network address translators (NATs) and firewalls. Motivated by this observation, we study the delay performance of TCP for real-time media flows. We develop an analytical performance model for the delay of TCP. We use extensive experiments to validate the model and to evaluate the impact of various TCP mechanisms on its delay performance. Based on our results, we derive the working region for VoIP and live video streaming applications and provide guidelines for delay-friendly TCP settings. Our research indicates that simple application-level schemes, such as packet splitting and parallel connections, can reduce the delay of real-time TCP flows by as much as 30% and 90%, respectively.

This paper presents mDHT, a novel architectural enhancement to DHT using multicast service discovery. In mDHT, a group of host computers in a subnet participate in a DHT overlay as a single node. A query is routed from subnet to subnet until it reaches the final destination subnet, where it is resolved among the hosts using link-local multicast. Under a reasonable deployment assumption, mDHT offers many benefits over standard DHTs, such as locality, easy bootstrapping, high availability, and near imperviousness to node churn.

In today's mobile networks, devices often move from one network to the next, forming transitory associations without a fixed support infrastructure. The state in this network frequently changes due to node transitions, and nodes are always meeting new peers. In this scenario, traditional models for writing networking software, such as the client-server model or even the peer-to-peer model, turn out to be unsuitable for writing such "ad-hoc applications". A new framework needs to be developed for this class of applications to be aware of node transitions as well as metadata (properties) that the nodes possess. In this paper, we present a framework we have built for this class of applications. Our library, called BonAHA (Bonjour for Ad-Hoc Applications) aims to be easy and intuitive, abstract the ad-hoc networking details and at the same time provide flexibility to the developer to develop powerful and rich ad-hoc applications.

Applications using multicast service discovery protocols, such as iTunes, have become increasingly popular. However, multicast service discovery protocols generate significant network traffic overhead, especially in a wireless network. We measure and analyze the traffic of one of the most widely deployed multicast service discovery protocols, DNS-SD/mDNS, in a campus wireless network where a single multicast domain serves a large number of users. We define three service discovery models which correspond to different service discovery behaviors. We compare these three models in terms of packet overhead and service discovery delay for different network sizes and service lifetime assumptions. The measurement shows that mDNS traffic consumes about 13 percent of the total bandwidth of the network.

Voice over WiFi (VoWiFi) will soon be the main alternative to cellular phones. Providing a satisfactory user experience remains difficult, however. We focus on call admission control (CAC) for both constant bit rate (CBR) and variable bit rate (VBR) VoIP traffic. Our approach is based on measuring the time between idle times. It requires no infrastructure changes, adds no probing traffic and has low complexity. We demonstrate through extended simulations that our approach achieves very good accuracy for both delay estimation and CAC when only VoIP sources are present and when both VoIP sources and data sources are present. Furthermore, we confirm TBIT performance through experiments.

We present vDelay, a tool for measuring the capture-to-display latency (CDL) and frame-rate of real-time video applications such as video chat and conferencing. vDelay allows measuring CDL and frame-rate without modifying the source code of these applications. Further, it does not require any specialized hardware. We have used vDelay to measure the CDL and frame-rate of popular video chat applications such as Skype, Windows Live Messenger, and GMail video chat. vDelay can also be used to measure the CDL and frame-rate of these applications in the presence of bandwidth variations.

We present vDelay, a tool for measuring the capture-to-display latency (CDL) and frame-rate of real-time video applications such as video chat and conferencing. Real-time video chat applications have three key software components: a video encoder that compresses the video captured from the camera, a video decoder that decompresses the video received over the network, and a playout buffer that smooths the playout of received video due to network delay variations. These software components impact capture-to-display latency (CDL) and frame-rate of the real-time video played at a receiver application. Capture-to-display latency is the total time to encode and decode a video frame, playout buffer time, and latency of the network path. Along with bit-rate, these two metrics provide quick insights into the performance of a real-time video application.

We study the performance of four popular IM clients focusing our attention on video-chat. In particular, we analyze how Skype, Windows Live Messenger, Eyebeam and X-Lite react to changes in available bandwidth, presence of HTTP and bit-torrent traffic and random packet losses.

Most of the routing protocols for mobile disruption-tolerant networks (DTNs), are designed to route data from a single source to a single destination. Many real-world applications, however, need to transfer data to a group of recipients rather than individuals. Proposed architectures for multicast communication in mobile DTNs rely on a global knowledge of multicast group memberships that makes them infeasible for infrastructure-less networks. Our interest-aware data distribution protocol distributes data among a community of mobile users in DTNs based on their interests. Our interest-aware algorithm automatically learns users' interests from the history of their cached data and provides them with their preferred data content. We implemented this protocol in the ONE simulator and evaluated its performance using reality-mining mobile traces. Our simulations show that our protocol delivers 30% more relevant data to mobile users' interests than epidemic routing. Furthermore, this superior performance is achieved with 35% lower distribution of irrelevant data content among mobile users.

We measured the capacity for VoIP traffic in an IEEE 802.11b wireless testbed and compared it with the theoretical capacity and our simulation results. We identified factors that have been commonly overlooked in past studies but affect experiments and simulations. We found that in many papers, the capacity for VoIP traffic has been measured via simulations or experiments without considering these factors, showing different capacity in each paper. After these corrections, simulations and experiments yielded a capacity estimate of 15 calls for 64 kb/s CBR VoIP traffic with 20 ms packetization interval and 38 calls for VBR VoIP traffic with a 0.39 activity ratio. Furthermore, we measured the capacity for VoIP traffic using each access category introduced in the 802.11e standard and the effect of the TCP traffic on VoIP traffic. We found that while the 802.11e standard can protect the QoS of VoIP against TCP traffic, it does not improve the capacity due to the significant retransmissions during TXOP.

We propose two features for peer-to-peer SIP in order to eliminate a sign-up process so that users can use the SIP user agent software right after installation. One is a locally generated SIP-URI which eliminates the sign-up process. Another is a feature to exchange contact addresses. It helps users to create new entries in the local address book without typing SIP-URIs. We implemented these features on SIP-Communicator with a peer-to-peer client protocol to work with a peer-to-peer SIP overlay network.

The stream control transmission protocol (SCTP) is a relatively recent transport protocol, offering features beyond TCP. Although SCTP is an alternative transport protocol for the session initiation protocol (SIP), we do not know how SCTP features influence SIP server scalability and performance. To estimate this, we measured the scalability and performance of two servers, an echo server and a simplified SIP server on Linux, for both SCTP and TCP. Our measurements found that using SCTP does not significantly affect data transfer latency. However, the number of sustainable associations drops to 17-21% or to 50% of the TCP value if we adjust the acceptable gap size of packet reordering.

As IEEE 802.11 networks in a BSS are increasingly used to carry VoIP, concerns about QoS arise. The overall delay of all VoIP flows drastically increases when the number of VoIP sources approaches the capacity, due to the characteristics of CSMA/CA. We propose a novel call admission control with the Queue size Prediction using Computation of Additional Transmission (QP-CAT) to avoid admitting an excessive number of simultaneous calls. In QP-CAT, an AP can accurately predict the effect of new VoIP flows on the delay of existing VoIP flows by predicting its queue size, before the new VoIP flow is actually admitted. It can be easily extended to support 802.11e so that the AP can predict the effect even when background traffic exists with VoIP traffic under 802.11e.

The next generation Internet needs to support multimedia services such as voice/video over IP (VoIP) and IP-based television (IPTV) and media distribution. The goal of the project is to develop a testbed for understanding and analysis of vulnerabilities of IP-based multimedia networks. This is a collaborative project between University of North Texas, Columbia University, Purdue University, and University of California at Davis. This project was awarded in 2006 and in one year, we have developed a multi-university testbed with variety of PBX solutions including Asterisk, a multitude of voice and video IP phones, and multiple universities were connected through a secure VPN and Internet!. As with any endeavor, experiences and we have many unique issues, which sometimes cause setbacks. We have dealt with issues concerning interoperability between phones and servers, network address translation (NAT) connectivity issues, and many other collaboration issues concerning people and technology. An interworking of students from multiple universities, faculty, system administrators and support personnel has brought the testbed resources together for a working environment. In this paper we described our progress and experiences in more detail and how to fulfill our mission statement while supporting the various collaborative efforts involved with this project. In addition, we described the current research activities based on the testbed. Finally we described the next steps in the testbed development.

Application and desktop sharing allows sharing of any application with one or more people over the Internet. The participants receive the screen-view of the shared application from the server. Their mouse and keyboard events are delivered and regenerated at the server. Application and desktop sharing enables collaborative work, software tutoring, and e-learning over the Internet. We have developed an application and desktop sharing platform called BASS which is efficient, reliable, independent of the operating system, scales well via heterogeneous multicast, supports all applications, and features true application sharing.

Over the years, IEEE 802.11b/g wireless networks have been deployed in various locations such as hotels, airports and enterprises. Although IEEE 802.11b/g can be considered a mature technology, its deployment still presents challenges due to the limited number of non-overlapping channels available. This is particularly true in scenarios with a high density of users where a large number of APs covering roughly the same area is required. Through measurements we investigate different deployment scenarios, trying to provide a set of guidelines for the deployment of IEEE 802.11b/g networks so to minimize co-channel interference and maximize throughput. This, when the number of APs required to cover an area is larger than the number of non-overlapping channels available. In particular, we show how using partially overlapping channels causes lower retry rate and higher throughput than if deploying multiple APs on each of the non-overlapping channels.

As the complexity of networked systems increases, we need mechanisms to automatically detect failures in the network and diagnose the cause of such failures. To realize true self-healing networks, we also need mechanisms to fix these failures and ensure service availability by providing alternatives when a failure is detected. In this paper, we address the detection and diagnosis of network failures. We introduce DYSWIS ("Do you see what I see ") which observes the system from multiple points in the network. Our system consists of detection nodes and diagnosis nodes. Detection nodes detect failures by passive traffic monitoring and active probing. Diagnosis nodes determine the cause of failures using historical information about similar failures and by performing active tests. They are based on a rule engine and represent network dependency relationship encoded as rules. We present our prototype system which considers network components present in a VoIP network and show the feasibility of our solution.

TCP has traditionally been considered unfriendly for real-time applications. Nonetheless, popular applications such as Skype use TCP since UDP packets cannot pass through many NATs and firewalls. Motivated by this observation, we study the delay performance of TCP for real-time media flows. We develop an analytical performance model for the delay of TCP. We use extensive experiments to validate the model and to evaluate the impact of various TCP mechanisms on its delay performance. Based on our results, we derive the working region for VoIP and live video streaming applications and provide guidelines for delay-friendly TCP settings. Our research indicates that simple application-level schemes, such as packet splitting and parallel connections, can reduce the delay of real-time TCP flows by as much as 30% and 90%, respectively.

Protocol can be used as an enabling platform for delivering communication and providing event-based services. We present a mobile workforce management solution which integrates fleet management with presence-based communication to show how presence technology can be used to realize location-based communication services. In particular, our solution integrates a user's presence information with their vehicle’s status and location information to enable safe communications with a mobile user. Our solution can provide services for both consumer (residential) and an enterprise scenario, for example, E911, rental car management, delivery companies (e.g., Fedex), for purposes like scheduling and automatic dispatching of mobile crews. We discuss our solution in the context of Verizon’s mobile workforce (mobile employee) management scenario. The solution incorporates integration of the mobile employee’s location obtained from a GPS device in a vehicle; the mobile employee’s presence information obtained from his/her cell-phone and laptop as well as the status and diagnostic information of the vehicle. A prototype system and its realization is presented to show the main concepts, feasibility and usefulness of such a solution.

Application and desktop sharing allows sharing any application with one or more people over the Internet. The participants receive the screen-view of the shared application from the server. Their mouse and keyboard events are delivered and regenerated at the server. Application and desktop sharing enables collaborative work, software tutoring and e-learning over the Internet. We have developed an application and desktop sharing platform called ADS which is efficient, reliable, operating system independent, scales well, supports all applications and features true application sharing.

The recent disruption of the Skype voice-over-IP system, triggered by a massive reboot of the hosts around the world, brought to light the importance of efficient bootstrapping in a large-scale peer-to-peer network. Thus far, the problem of bootstrapping a DHT network from a near-total failure has received limited attention from the research community. We present an outline of our plan to investigate DHT bootstrapping mechanisms for large-scale deployments on the Internet.

In typical single-link 802.11 ad-hoc networks, devices often move from one network to the next, forming transitory associations without a fixed support infrastructure. The state in this network frequently changes due to node transitions. In such a scenario, traditional models for writing networking software, such as the client-server model or even the peer-to-peer model, turn out to be unsuitable for writing "ad-hoc applications".

Traditionally, signal-to-noise ratio of a mobile determines the handoff dynamics of the mobile. But in certain cases, precise location of the mobile augmented by information services, such as IEEE 802.21 MIS, can expedite the handoff with similar performance results. We illustrate an experimental system that takes advantage of the mobile's relative location with the neighboring access point to perform proactive handoff. It keeps track of the current location of the mobile and then uses the information from the neighboring networks to help perform the proactive handoff. Proactive handover technique helps the mobile to communicate with these networks before the handover is complete thereby reducing the delay and packet loss. In some cases, location-assisted handover could prove to be more useful compared to the handover technique based on signal-noise-ratio.

Zero Configuration Networking (Zeroconf) assigns IP addresses and host names, and discovers service without a central server. Zeroconf can be used in wireless mobile ad-hoc networks which are based on IEEE 802.11 and IP. However, Zeroconf has problems in mobile ad-hoc networks as it cannot detect changes in the network topology. In highly mobile networks, Zeroconf causes network overhead while discovering new services. In this paper, we propose an algorithm to accelerate service discovery for mobile ad-hoc networks. Our algorithm involves the monitoring of network interface changes that occur when a device with IEEE 802.11 enabled joins a new network area. This algorithm allows users to discover network topology changes and new services in real-time while minimizing network overhead.

ENUM is a DNS-based protocol standard for mapping E.164 telephone numbers to Internet Uniform Resource Identifiers (URIs). It places unique requirements on the existing DNS infrastructure, such as data scalability, query throughput, response time, and database update rates. This paper measures and evaluates the performance of existing name server implementation as ENUM servers. We compared PowerDNS (PDNS), BIND and Navitas. Results show that BIND is not suitable for ENUM due to its poor scaling property. Both PDNS and Navitas can serve ENUM. However, Navitas turns out to be highly optimized and clearly outperforms PDNS in all aspects we have tested. We also instrumented the PDNS server to identify its performance bottleneck and investigated ways to improve it.

When the node density in a network decreases below the level necessary to sustain ad-hoc and mesh networks, communication can succeed only by leveraging node mobility and transitioning to message-based communications. In the 7DS (Seven Degrees of Separation) project, we have been investigating how to emulate two core Internet services, namely Web access for information retrieval and email for delivering messages from mobile nodes to the Internet. We have implemented and evaluated a 7DS prototype system that leverages search, feedback and propagation limits to build a scalable system that can deliver data to and from mobile nodes. 7DS makes data exchange in disconnected networks possible by providing an application-level set of protocol services that will enable exchange of information between peer devices. It enables dynamic information exchange by using a proxy server, a multicast query system, a search engine, and a transport entity. With these entities, 7DS can perform efficient and transparent data exchange among peers in the absence of a network connection. Data exchange with the larger Internet occurs when peers enter or exit the peer network.

In a wireless network, mobile nodes (MNs) repeatedly perform tasks such as layer 2(L2) handoff, layer 3(L3) handoff and authentication. These tasks are critical, particularly for real-time applications such as VoIP. We propose a novel approach, namely Cooperative Roaming (CR), in which MNs can collaborate with each other and share useful information about the network in which they move. We show how we can achieve seamless L2 and L3 handoffs regardless of the authentication mechanism used and without any changes to either the infrastructure or the protocol. In particular, we provide a working implementation of CR and show how, with CR, MNs can achieve a total L2+L3 handoff time of less than 16 ms in an open network and of about 21ms in an IEEE 802.11i network. We consider behaviors typical of IEEE 802.11 networks, although many of the concepts and problems addressed here apply to any kind of mobile network.

In recent years VoIP and IEEE 802.11 networks have seen a rapid growth. In IEEE 802.11 networks, an Access Point (AP) and the stations (STAs) it serves form a Basic Service Set (BSS). Each AP can support a limited number of concurrent voice calls; we refer to this number as the AP capacity. After the number of concurrent voice calls in the BSS surpasses the AP capacity, the communication of all users at that AP suffers from high delay, and therefore, poor quality. These are the problems we address with Call Admission Control (CAC). We propose a mobile-station-based CAC mechanism. One of the strengths of the proposed approach is that it is very simple and yet accurate, while not requiring any probing of the medium. Our approach is entirely client-based, thus not requiring any changes in the infrastructure and the protocol.

We measured the capacity for VoIP traffic in an 802.11b test-bed and compared it with the theoretical capacity and our simulation results. We identified factors that have been commonly overlooked in past studies but affect experiments and simulations. We found that in many papers the capacity for VoIP traffic has been measured via simulations or experiments without considering those factors, showing different capacity in each paper. After these corrections, simulations and experiments yielded a capacity estimate of 15 calls for 64 kb/s CBR VoIP traffic with 20 ms packetization interval and 34 calls to 36 calls for VBR VoIP traffic with 0.39 activity ratio.

Public safety answering points (PSAPs) serve limited geographic areas, so emergency callers must be directed to the most appropriate PSAP. As part of the overall Internet Engineering Task Force (IETF) emergency services architecture, we have developed a new protocol, LoST (location-to-service translation), that allows end systems and voice-over-IP (VoIP) proxies to map location data into URLs representing either PSAPs or other session initiation protocol (SIP) proxies that perform a more fine-grained mapping. LoST is designed to operate globally, with a highly-distributed authority.

Recently, there has been increased interest, shown by the research community and by industry fora, to enable emergency services for voice-over-IP based architectures. In order to provide emergency services functionality over the Internet, several aspects like location determination, dial string configuration, location conveyance, routing to the appropriate public safety answering point (PSAP) and security aspects have to be considered. This paper discusses security threats for the IP-based emergency architecture concentrating on denial-of-service attacks against public safety answering points (PSAPs) by using faked calls that aim to exhaust limited PSAP resources, such as the emergency personnel. We illustrate a number of different approaches that have been mentioned in the past to tackle these threats, including location signing, location-by-reference and authenticated emergency calls.

Presence has taken shape as a way to present a comprehensive view of the communications capabilities of a user or resource. Standardized protocols allow for receiving presence data from a variety of sources, such as a user's communication devices, cellular provider, online calendar and sensors in his environment. This data can describe many different aspects of his communication capability, such as his device characteristics, activities, and even physical locations. The large number of presence sources may lead to irrelevant or inconsistent data. Also, the data may be incomplete and not as usable as it could be to the presence watcher. This paper proposes the use of presence composition to remove unusable data and create new more usable data. We discuss the details of this composition and present a format with which a user can specify a policy for composition.

Secured and seamless mobility across heterogeneous access networks needs optimization at all layers. Authentication and security association at the link layer is one of the major components of delay during handoff. We propose a network- layer assisted proactive handoff scheme that helps to optimize the handoff process involving link-layer security across multiple subnets. We demonstrate this proactive scheme and analyze the results for IEEE 802.11-based networks for both roaming and non-roaming scenarios. We then compare these results with the pre-authentication techniques offered by IEEE 802.11i.

A mobility event is the result of one network connection path being replaced by another via the rebinding of common system properties. The rebinding is a sequential process that may potentially involve multiple protocol layers of the mobile and require multiple network interactions. This overall process results in a period of time in which network service is degraded by transient data loss and increased end-to-end delay. Optimizations of the handover process mitigating these service degradations have been developed without a formal or systematic framework for mobility solutions. We develop a systematic systems model of the basic properties associated with a mobility event and design a framework around these properties that can provide methodologies for optimizing the handoff components. We then summarize the experimental results from a 3GPP2-based mobility testbed and highlight the delays associated with the functional components of the handoff event. We apply two types of optimization techniques and compare the results.

Real-time services such as VoIP and multimedia streaming are affected during a mobile's rapid handoff due to the associated delay resulting from associated handoff operations of discovery, detection, configuration, registration, media redirection, and processing at different network nodes. Redundant communication path of any signaling and media adds to the delay during session setup and media delivery. This paper highlights the overall problem associated with redundant routing in a 3GPP2-based MMD (multimedia domain) environment and proposes several mechanisms to mitigate these problems. We also compare these mechanisms and select one for existing IMS/MMD networks that use MIPv4 as the mobility protocol. We present the implementation details of the selected redundant routing mitigation technique in the experimental MMD environment and verify the effectiveness by analyzing the measurement results.

Service usage in emerging ubiquitous environments includes seamless and personalized usage of public and private devices discovered in the vicinity of a user. In our work, we describe an architecture for device discovery, device configuration, and the transfer of active sessions between devices. The presented architecture uses the Session Initiation Protocol (SIP) as a standardized, widely used signaling protocol for IP-based multimedia services. Our solution includes support of simple existing devices, split of sessions between devices, user-control of location-based behavior, and handling of security and privacy concerns. We present the implementation and show the feasibility of our work with analytical evaluation and measurements.

Abstract The original Mobile IP (MIP) protocol does not perform Route Optimisation but uses Home Agents to forward traffic. Thus, it does not have problems with simultaneous mobility, that is, the special case when both end hosts are mobile and move at about the same time. However, MIP for IPv6 (MIPv6) uses binding updates that are sent directly to a correspondent node. Session Initiation Protocol based mobility management (SIPMM) and MIP with location registers (MIP‐LR) also use direct binding updates between a mobile host and a correspondent node. Thus, MIPv6, MIP‐LR and SIPMM are vulnerable to the simultaneous mobility problem. In this paper, we analyse the simultaneous mobility problem and solution mechanisms, and propose new ways for MIPv6, MIP‐LR and SIPMM to handle simultaneous mobility. Copyright © 2006 John Wiley & Sons, Ltd.

In this paper, we examine the feasibility of sending real-time CBR workloads over TCP. This is motivated by the friendliness of NATs and firewalls towards TCP as opposed to UDP as well as by recent improvements in Internet's bandwidth and loss rates. Traditionally, TCP has been considered undesirable for real-time CBR workloads. We evaluate this assertion by developing a novel analytical tool that yields TCP's sender-to-receiver socket delay distribution for CBR workloads. A key insight gained is that the use of smaller than MSS-sized packets in CBR workloads can exploit the TCP's ACK counting mechanism thereby limiting the delay impact of congestion window variations. We leverage this insight to provide a heuristic and system-level guidelines for reducing TCP transport delays.

During a layer-3 (L3) handoff, address acquisition via DHCP is often the dominant source of handoff delay, with duplicate address detection (DAD) responsible for most of the delay. We propose a new DAD algorithm, passive DAD (pDAD), which we show to be effective, while introducing only a few milliseconds of delay. Unlike traditional DAD, pDAD also detects the unauthorized use of an IP address before it is assigned to a DHCP client.

We examine query based (QB) and Bloom filter based (BF) schemes for data retrieval in a partially connected ad hoc network. For the QB scheme, a node retrieves data from another node by querying. Search efficiency deteriorates when the query success probability is small. We propose a Bloom filter based (BF) scheme to improve search efficiency. A data holder summarizes its content by a Bloom filter and broadcasts it to the data querier, which sends a query only if the requested data is available. We derive the bandwidth utilization in the single neighbor scenario and search efficiency as a function of node density in the multiple neighbors scenario. Numerical results show that the BF scheme offers significant gain in search efficiency. Our performance evaluation makes use of the M/M/oo queue to model mobility and is novel. It is useful for studying the impact of mobility in other mobile networks.

We characterize the resource usage of an upload application for partially connected ad hoc networks. Time-based (TB) and hop-based (HB) TTL schemes are proposed and evaluated via both analysis and simulation. A HB(b, 2) scheme is near optimal, where 2 and b denote depth and breadth of the data dissemination tree. b is chosen such that the probability of message delivery approaches one. We also consider feedback schemes to notify mobile nodes to purge message replicas upon successful message delivery. With feedback, we obtain a 100-fold and 5-fold cost reduction for TB and HB schemes, respectively.

Skype is a peer-to-peer VoIP client developed by KaZaa in 2003. Skype claims that it can work almost seamlessly across NATs and firewalls and has better voice quality than the MSN and Yahoo IM applications. It encrypts calls end-to-end, and stores user information in a decentralized fashion. Skype also supports instant messaging and conferencing. This report analyzes key Skype functions such as login, NAT and firewall traversal, call establishment, media transfer, codecs, and conferencing under three different network setups. Analysis is performed by careful study of Skype network traffic.

The General Internet Signaling Transport (GIST) protocol is currently being developed as the base protocol component in the IETF Next Steps In Signaling (NSIS) protocol stack to support a variety of signaling applications. We present our study on the protocol overhead and performance aspects of GIST. We quantify network-layer protocol overhead and observe the effects of enhanced modularity and security in GIST. We developed a first open source GIST implementation at the University of Göttingen, and study its performance in a Linux testbed. A GIST node serving 45,000 signaling sessions is found to consume average only 1.1 ms for processing a signaling message and 2.4 KB of memory for managing a session. Individual routines in the GIST code are instrumented to obtain a detailed profile of their contributions to the overall system processing. Important factors in determining performance, such as the number of sessions, state management, refresh frequency, timer management and signaling message size are further discussed. We investigate several mechanisms to improve GIST performance so that it is comparable to an RSVP implementation.

In a mobile environment, as a mobile node moves from one point of attachment to another during an ongoing application session it is subjected to packet loss due to network and link layer transition. Such packet loss affects the quality of ongoing communication session such as interactive VoIP traffic and streaming media. We provide a solution to this scenario by buffering packets for the mobile node at an access router or network node near the edge of the network where mobile may be moving away from or moving towards. The buffered packets are then forwarded to the mobile node once the handoff process completes. The buffering scheme is used in conjunction with existing mobility protocols, access protocols or as an independent network or link layer mechanism. Ability to control the buffer dynamically provides a reasonable trade-off between delay and packet loss which is within the threshold limit for real-time communication. The overview and mechanisms of such schemes are described and comparisons on existing buffering schemes are also provided

In IEEE 802.11 wireless networks, the downlink delay rises as the number of VoIP nodes increases while the uplink delay remains small due to the same chance of media access between nodes and the Access Point (AP). This degrades the capacity and QoS of VoIP significantly. Therefore, we introduce Adaptive Priority Control (APC) to balance the downlink and uplink delay of VoIP traffic at the MAC layer, by giving to the AP a higher transmission priority, which is adaptively decided according to the uplink and downlink traffic volume. And, we verify through theoretical analysis that APC is an optimal method to balance the uplink and downlink delay.

Many applications used in the Internet today benefit from using location information. To better handle location information in Internet telephony applications, we did a comprehensive application-layer analysis of location information and location-based communication services. We first summarize and categorize end-user-oriented location description and location detection approaches. We then summarize and categorize how to use location information to provide communication services and introduce several interesting location based communication services. Based on the analysis, we have incorporated location-based service handling in our session initiation protocol (SIP) based Internet telephony infrastructure and our language for end system services (LESS).

Internet telephony can introduce many novel communication services, however, this novelty puts learning burden on users. We have developed an intelligent service creation environment which can automatically create services by learning from communication behavior. The service creation environment models communication services as decision trees and uses the incremental tree induction (ITI) algorithm for learning. We use our language for end system services (LESS) to describe learned results and have implemented a simulation environment to verify the learning algorithm. We noticed that automatically generated services may introduce undesirable side-effects that can cause users to lose calls, money, or privacy. Thus, we did an analysis on service risk management for LESS-based services and proposed several approaches for fail-safe services.

Providing emergency services in VoIP networks are vital to the success of VoIP. It not only presents design and implementation challenges, but also gives an opportunity to enhance the existing emergency call handling infrastructure. We propose an architecture to deliver emergency services in SIP-based VoIP networks, which can accommodate PSTN calls through PSTN to SIP gateways. Our architecture addresses the issues of identifying emergency calls, determining callers' locations, routing emergency calls to appropriate public safety answering points (PSAPs), and presenting required information to emergency call takers. We have developed a prototype implementation to prove our architecture's feasibility and scalability. We expect to undertake a pilot project at a working PSAP with our implementation once it is thoroughly tested.

We propose DotSlash, a self-configuring and scalable rescue system, for handling web hotspots at dynamic content Web sites. To support load migration for dynamic content, an origin Web server sets up needed rescue servers drafted from other Web sites on the fly, and those rescue servers retrieve the scripts dynamically from the origin Web server, cache the scripts locally, and access the corresponding database server directly. We have implemented a prototype of DotSlash for the LAMP configuration, and tested our implementation using the RUBBoS bulletin board benchmark. Experiments show that by using DotSlash a dynamic content web site can completely remove its web server bottleneck, and can support a request rate constrained only by the capacity of its database server.

Despite recent advancements in cellular phone network infrastructure, survivability of the calls is still an open issue. Important business calls and tele-conferences cannot benefit from mobility because landlines are preferred owing to their reliability and tolerance to failures. In this paper, we present a framework to improve call survivability by monitoring the user's location and intimating him about the possibility of loss of connectivity, interference, and room schedules that might disrupt conversation. In addition, if a call drops because of unforeseeable circumstances like battery failure or cell-phone damage then the call is switched to the nearest available device capable of streaming voice

In order to support session-based real-time communication in a highly mobile environment it is desirable to limit end-to-end delay, jitter and packet loss at a certain threshold level. This paper describes a framework of media-independent pre-authentication (MPA), a new handover optimization mechanism that has a potential to address issues on existing mobility management protocols and mobility optimization mechanisms to achieve these values. MPA is a mobile-assisted, secure handover optimization scheme that works over any link-layer and with any mobility management protocol. This paper also presents an initial implementation of MPA and performance results to show how existing protocols could be leveraged to realize the functionalities of MPA and provide the desired results.

P2P systems inherently have high scalability, robustness and fault tolerance because there is no centralized server and the network self-organizes itself. This is achieved at the cost of higher latency for locating the resources of interest in the P2P overlay network. Internet telephony can be viewed as an application of P2P architecture where the participants form a self-organizing P2P overlay network to locate and communicate with other participants. We propose a pure P2P architecture for the Session Initiation Protocol (SIP)-based IP telephony systems. Our P2P-SIP architecture supports basic user registration and call setup as well as advanced services such as offline message delivery, voice/video mails and multi-party conferencing. We also provide an overview of practical challenges for P2P-SIP such as firewall, Network Address Translator (NAT) traversal and security.

The integrated networking technology prototype demonstrates the capability of several emerging networking technologies to operate together seamlessly to enhance network services targeted for dynamic mobile environments such as those found in the battlefield. The prototype consists of four technologies: (1) autoconfiguration technology supports autonomous and rapid network deployment and configuration; (2) self-managed virtual network (SMVN) technology provides virtual networking capabilities for networks which do not natively support these functions; (3) integrated mobility management technology supports session continuity in the presence of node mobility; (4) assured IP QoS technology supports guarantees for mission-critical applications. The prototype is capable of operating over the various types of equipment and protocols to be utilized in battlefield networks. The prototype is designed such that network services and functions are survivable and reconfigurable. In this paper, we describe the four technologies as well as the integrated prototype in the laboratory environment.

With the deployment of IEEE 802.11 networks, supporting real-time traffic with stringent quality of service (QoS) requirements on these networks becomes a critical issue. We propose two new media access schemes, namely dynamic point coordination function (DPCF) and modified DPCF (DPCF2). These can improve the capacity for voice-over-IP (VoIP) traffic by up to 20% in IEEE 802.11b networks. We show how we can also achieve a drastic improvement in the end-to-end delay with mixed VoIP and data traffic. Delay is kept around 100 ms in heavily loaded traffic conditions with an average value under 60 ms in normal traffic conditions.

We present a location-based, ubiquitous service architecture, based on the session initiation protocol (SIP) and a service discovery protocol that enables users to enhance the multimedia communications services available on their mobile devices by discovering other local devices, and including them in their active sessions, creating a "virtual device." We have implemented our concept based on Columbia University's multimedia environment and we show its feasibility by a performance analysis.

The rapid deployment of new applications and the interconnection of networks with increasing diversity of technologies and capacity make it more challenging to provide end-to-end quality assurance to the value-added services, such as the transmission of real-time multimedia and mission critical data. In a network with enhancements for QoS support, pricing of network services based on the level of service, usage, and congestion provides a natural and equitable incentive for multimedia applications to adapt their sending rates according to network conditions. We have developed an intelligent service architecture that integrates resource reservation, negotiation, pricing and adaptation in a flexible and scalable way. In this paper, we present a generic pricing structure that characterizes the pricing schemes widely used in the current Internet, and introduce a dynamic, congestion-sensitive pricing algorithm that can be used with the proposed service framework. We also develop the demand behavior of adaptive users based on a physically reasonable user utility function. We introduce our multimedia testbed and describe how the proposed intelligent framework can be implemented to manage a video conference system. We develop a simulation framework to compare the performance of a network supporting congestion-sensitive pricing and adaptive reservation to that of a network with a static pricing policy. We study the stability of the dynamic pricing and reservation mechanisms, and the impact of various network control parameters. The results show that the congestion-sensitive pricing system takes advantage of application adaptivity to achieve significant gains in network availability, revenue, and user-perceived benefit relative to the fixed-price policy. Congestion-based pricing is stable and effective in limiting utilization to a targeted level. Users with different demand elasticity are seen to share bandwidth fairly, with each user having a bandwidth share proportional to its relative willingness to pay for bandwidth. The results also show that even a small proportion of adaptive users may result in a significant performance benefit and better service for the entire user population-both adaptive and nonadaptive users. The performance improvement given by the congestion-based adaptive policy further improves as the network scales and more connections share the resources. Finally, we complement the simulation with experimental results demonstrating important features of the adaptation process.

Resource management mechanisms for large-scale, globally distributed network services need to assign groups of clients to servers according to network location and expected load generated by these clients. Current proposals address network location and traffic modeling separately. We develop a novel clustering technique that addresses both network proximity and traffic modeling. Our approach combines techniques from network-aware clustering, location inference, and spatial analysis. We conduct a large, measurement-based study to identify and evaluate Web traffic clusters. Our study links millions of Web transactions collected from two world-wide sporting event Websites, with millions of network delay measurements to thousands of Internet address clusters. Because our techniques are equally applicable to other traffic types, they are useful in a variety of wide-area distributed computing optimizations, and Internet modeling and simulation scenarios.

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.

Identifying the origins of a distributed denial-of-service (DDoS) attack is among the hardest research topics in the Internet security area, due to the stateless nature of IP networks and to the confusing and diffusing effect of DDoS attacks. We select ten essential factors to analyze a newly-designed DoS-resistant ICMP messaging scheme and demonstrate its feasibility, effectiveness, security, and immunity to denial-of-service (DoS) attacks.

Due to the growth in ubiquitous computing technology in the past few years, the need for context-aware service discovery across wide area networks is becoming prevalent. We propose GloServ, which is a global service discovery architecture that locates services throughout wide and local area networks. It supports services encompassing different domains such as events, people or places. Services can be described semantically using the resource description framework (RDF) and can be queried using the RDF query language (RQL). GloServ hierarchically defines services using RDF schemas and assigns each service a URI according to its location within the hierarchy. The hierarchical architecture for GloServ is similar to how domain names are categorized in DNS. Service discovery can either be initiated by the user or by the system. For automated service discovery, users are detected with sensors and are presented with services available according to their preferences. Graphical user interfaces for querying data is dynamically generated through the processing of RDF data.

In order to ensure proper quality of service for real-time communication in a mobile wireless Internet environment it is essential to minimize the transient packet loss when the mobile is moving between different cells (subnets) within a domain. Network layer mobility management schemes have been proposed to provide optimized fast-handoff for multimedia streams during a client's frequent movement within a domain. This paper introduces application layer techniques to achieve fast-handoff for real-time RTP/UDP based multimedia traffic in a SIP signaling environment. These techniques are based on standard SIP components such as user agent and proxy which usually participate to set up and tear down the multimedia sessions between the mobiles. Unlike network layer techniques, application layer techniques do not have to depend upon any additional components such as home agent and foreign agent. It thus provides a network access independent solution suitable for application service providers.

No abstract available.

Currently, denial-of-service (DoS) attacks remain among the most critical threats on the Internet. Ironically, although a major objective of Internet protocol design was to prevent DoS attacks and to maintain network connections in order to provide essential services to the masses uninterruptedly, the Internet is suffering significantly from distributed DoS (DDoS) attacks at present. In recent years, numerous research studies have focused on this topic. However, after deliberately analyzing the proposed defensive measures and supposedly secure systems, we figure that most are themselves extremely vulnerable to DoS/DDoS attacks. This paper contributes a detailed analysis of various existing and proposed defensive measures, as well as their vulnerabilities.

We evaluate the QoS of a number of VoIP end-points, in terms of mouth-to-ear (M2E) delay, clock skew, silence suppression behavior and robustness to packet loss. Our results show that the M2E delay depends mainly on the receiving end-point. Hardware IP phones, when acting as receivers, usually achieve a low average M2E delay (45-90 ms) under low jitter conditions. Software clients achieve an average M2E delay from 65 ms to over 400 ms, depending on the actual implementation. All tested end-points can compensate for clock skew, although some suffer from occasional playout buffer underflow. Only a few of the tested end-points support silence suppression. We find that these silence detectors have a relatively long hangover time (> 0.5 sec), and they may falsely detect music as silence. All hardware IP phones we tested support some form of packet loss concealment better than silence substitution. The concealment generally works well for two to three consecutive losses at 20 ms packet intervals, but voice will quickly deteriorate beyond that.

This paper describes the session initiation protocol (SIP) based mobility in IPv6 and its performance in our laboratory testbed. For real-time mobile multimedia communication, we use SIP for signaling protocol as well as for supporting terminal mobility. While performance study for real-time mobile communication refers to several factors and their measurements, we analyze here only the handoff delay due to node mobility. In particular, we are interested to examine the delay incurred when a mobile node moves to a new link and perform the duplicate address detection (DAD) and router selection, and analyze the delay in each case. We notice that the IPv6 Linux implementation provides substantial amount of delay during node movement, which severely affects the performance of real-time applications. Therefore, we modify the Linux kernel and compare with the unaltered one. Finally, we conclude that a faster handoff is achievable in SIP based terminal mobility with intelligent modifications to unaltered Linux kernel.

We describe our session initiation protocol (SIP)-based VoiceXML browser, sipvxml, that allows programming interactive voice response applications that are accessible from telephones as well as IP phones. We also describe how we have used sipvxml in our multi-party multimedia conferencing server. We propose other applications and extensions that can benefit from this technology in our IP telephony test bed.

In Internet telephony, end systems can take a much larger role in providing services than in traditional telephone systems. We analyze the importance of end system services and describe the services and the service logic execution environment (SLEE) implemented in out SIP user agent, SIPC. Since we believe that end system services differ in their requirements from network services, we define a new service creation scripting language call language for end system services (LESS). Compared with other service creation languages, LESS is extensible, can be easily understood by non-programmers and contains commands and events for direct user interaction and the control of media applications.

In the past decade, there have been numerous efforts in ubiquitous computing, making computational resources or communication more widely available. We believe that it is time to move to a global-scale ubiquitous computing system that is securable, administered by multiple independent administrators and integrates off-the-shelf hardware and software. We are developing such a system based on the Session Initiation Protocol (SIP), with Bluetooth devices for location sensing and Service Location Protocol (SLP) for service discovery. We also introduce context-aware location information to augment device discovery and user communication. The system builds on our CINEMA infrastructure and can support a range of activities, from home-based settings to collaboration between distant sites.

Many approaches and topologies --- including multicast and media mixing --- have been proposed for distributed Internet conferencing. While existing solutions can work well for large or pre-arranged conferences, they can be less appropriate for smaller, impromptu ones. We present an alternative, full mesh conferencing, which allows any number of parties to communicate in a conference without a central point of control. The protocol allows parties to join and leave the conference at any time, and ensures that all members of the conference are always informed of new members. The paper gives an overview of the protocol, analyzes it, describes a simulation environment for it, and discusses its applicability to the Session Initiation Protocol (SIP) and to other forms of decentralized communication.

Internet server selection mechanisms attempt to optimize, subject to a variety of constraints, the distribution of client requests to a geographically and topologically diverse pool of servers. Research on server selection has thus far focused primarily on techniques for choosing a server from a group administered by single entity, like a content distribution network provider. In a federated, multi-provider computing system, however, selection must occur over distributed server sets deployed by the participating providers, without the benefit of the full information available in the single-provider case. Intelligent server set selection algorithms will require a model of the expected performance clients would receive from a candidate server set.In this paper, we study whether the complex policies and dynamics of intelligent server selection can be effectively modeled in order to predict client performance for server sets. We introduce a novel server set distance metric, and use it in a measurement study of several million server selection transactions to develop simple models of existing server selection schemes. We then evaluate these models in terms of their ability to accurately predict performance for a second, larger set of distributed clients. We show that our models are able to predict performance within 20ms for over 90% of the observed samples. Our analysis demonstrates that although existing deployments use a variety of complex and dynamic server selection criteria, most of which are proprietary, these schemes can be modeled with surprising accuracy.

The paper presents two new mechanisms, global attributes and preference filters, that improve the efficiency and extendability of the service location protocol (SLP). Global attributes describe common service properties, which can be used to perform queries across different service types, and can be standardized for service management. Preference filters facilitate processing of search results, such as finding the best match, in SLP servers. They can reduce the amount of data transferred to the client, and better support thin clients with limited resources and capabilities.

While there has been extensive research into wide-area load balancing for Web servers, dynamic server selection issues for rich media have largely been ignored We argue that streaming media imposes new challenges that are not met by current proposals for collecting and disseminating server and network path metrics. We define a model for quantitatively evaluating the network overhead of competing Internet measurement strategies, propose a novel strategy to better meet requirements for rich media, and use topologies representative of the Internet to show our strategy reduces total and average per link probe overhead by as much as 98%. We discuss additional advantages to our approach, including the ability to represent multiple, potentially streaming media specific, metrics; the ability to effect policy-based selection; and protection against denial of service attacks.

In earlier work, we had proposed a pricing model in which service prices are based on QoS (resources consumed) and long-term user demand, and also have a congestion-sensitive component to motivate rate and service adaptation by applications with elastic demand. The network is provisioned to provide multiple services, with short-term, dynamic configuration of network resources. Congestion pricing schemes in the network literature fall into two basic categories: tatonnement and bandwidth auctions. As far as we know, there has been no work comparing these two schemes in the same environment. The goal of this paper is to develop pricing schemes based on the above two approaches, in an environment with short-term resource allocation and demand adaptation. We address some important practical issues related to making the schemes work in such an environment. We compare the tatonnement, and auction-based schemes with respect to network utilization, connection blocking rate, user satisfaction and network revenue, and draw some general conclusions about the relative benefits of the two approaches.

Packet loss degrades the perceived quality of voice over IP (VoIP). In addition, packet loss in the Internet tends to come in bursts, which may further degrade audio quality. Using the Gilbert loss model, we infer that changing the packet interval affects loss burstiness, which in turn influences forward error correction (FEC) performance. Next, we perform subjective listening tests based on Mean Opinion Score (MOS) to evaluate the effect of bursty loss on VoIP perceived quality. Then, we compare the perceived quality achieved by two major loss repair methods: FEC and low bit-rate redundancy (LBR). Our MOS test results show that FEC is much preferred over LBR. In addition, our MOS results reveal that, under bursty loss, FEC quality is much better with a moderately large packet interval. Finally, because FEC introduces an extra delay proportional to the packet interval, we present a method of optimizing the packet interval to maximize FEC MOS by considering the delay impairment in ITU's E-model standard.

Conference control has been an area of intensive research over the years but widely accepted robust and scalable solutions and standards are still lacking. The main conference control components are conference management and floor (resource) control. We identify the requirements for conference control and propose a component-based scalable conference control framework employing the Session Initiation Protocol (SIP) and the Simple Object Access Protocol (SOAP). The framework assumes a single control point, but our architecture can scale to large groups by distributing media via a tree-shaped hierarchy of conference servers.

No abstract available.

We propose a new, media-independent protocol for including program timing, structure and identity information in Internet media streams. The protocol uses signaling messages called cues to indicate events whose timing is significant to receivers, such as the start or stop time of a media program. We describe the implementation and operation of a prototype Internet radio station which transmits program cues in audio broadcasts using a real-time transport protocol (RTP). A collection of simple yet powerful stream processing applications we implemented demonstrate how application creation is greatly eased when media streams are enriched with program cues.

In an effort to realize wireless Internet telephony and multimedia streaming in a highly mobile environment a testbed emulating a wireless Internet has been built. This allows the setting up of multimedia calls between IP mobiles and integration between IP and PSTN end-points in a wireless environment. Different functionalities and components involved with the wireless Internet streaming multimedia have been prototyped and experimented in the testbed. These include signaling, registration, dynamic binding, location management as well as supporting the QoS features for the mobile users. This paper describes some of the components of the testbed and highlights the experiences while building this testbed which could be beneficial to some who plan to build a similar testbed to realize several features and capabilities of Mobile Wireless Internet, before actually bringing to the market.

With the influx of multi-media streaming content over the Internet, building a streaming infrastructure that would support flexible next generation Internet application in a scalable may on heterogenous access technology is most desirable. This paper presents a streaming architecture which provides some innovative methods and technique that would help build an IP based radio/TV network. This would provide tools for E-commerce and would benefit the broadcasting stations, local affiliates, ISPs, and the end users. Additionally this paper highlights the mechanism behind some of the components associated with this architecture such as local content, global content, program management with local control, payment model, security, advertisement insertion, seamless mobility within a domain.

We propose a new, media-independent protocol for including program timing, structure and identity information in Internet media streams. The protocol uses signaling messages called cues to indicate events whose timing is significant to receivers, such as the start or stop time of a media program. We describe the implementation and operation of a prototype Internet radio station which transmits program cues in audio broadcasts using the Real-Time Transport Protocol. A collection of simple yet powerful stream processing applications we implemented demonstrate how application creation is greatly eased when media streams are enriched with program cues.

The differentiated services framework (DiffServ) has been proposed to provide multiple quality of service (QoS) classes over IP networks. A network supporting multiple classes of service also requires a differentiated pricing structure. We propose a pricing scheme in a DiffServ environment based on the cost of providing different levels of quality of service to different classes, and on long-term demand. Pricing of network services dynamically based on the level of service, usage, and congestion allows a more competitive price to be offered, allows the network to be used more efficiently, and provides a natural and equitable incentive for applications to adapt their service contract according to network conditions. We develop a DiffServ simulation framework to compare the performance of a network supporting congestion-sensitive pricing and adaptive service negotiation to that of a network with a static pricing policy. Adaptive users adapt to price changes by adjusting their sending rate or selecting a different service class. We also develop the demand behavior of adaptive users based on a perceptually reasonable user utility function. Simulation results show that a congestion-sensitive pricing policy coupled with user rate adaptation is able to control congestion and allow a service class to meet its performance assurances under large or bursty offered loads, even without explicit admission control. Users are able to maintain a stable expenditure. Allowing users to migrate between service classes in response to price increases further stabilizes the individual service prices. When admission control is enforced, congestion-sensitive pricing still provides an advantage in terms of a much lower connection blocking rate at high loads.

This paper presents 7DS, a novel peer-to-peer data sharing system. 7DS is an architecture, a set of protocols and an implementation enabling the exchange of data among peers that are not necessarily connected to the Internet. Peers can be either mobile or stationary. It anticipates the information needs of users and fulfills them by searching from information among peers. We evaluate via extensive simulations the effectiveness of our system for data dissemination among mobile devices with a large number of user mobility scenarios. We model several general data dissemination approaches and investigate the effect of the wireless converage range, 7DS, host density, query interval and cooperation strategy among the mobile hosts. Using theory from random walks, random environments and diffusion of controlled processes, we model one of these data dissemination schemes and show that the analysis confirms the simulation results for scheme

We describe the architecture and implementation of our Internet teleph ony test-bed intended to replace the departmental PBX (telephone switch). It interworks with the traditional telephone networks via a PSTN/IP gateway. It also serves as a corporate or campus infrastructure for existing and future services like web, email, video and streaming media. Initially intended for a few users, it will eventually replace the plain old telephones from our offices, due to the cost benefit and new services it offers. We also discuss common inter-operability problems between the PBX and the gateway.

We present an architecture that enables the sharing of information among mobile, wireless, collaborating hosts that are intermittently connected to the Internet. Participants in the system obtain data objects from Internet-connected servers, cache them and exchange them with others who are interested in them. The system exploits the fact that there is a high locality of information access within a geographic area. It aims to increase the data availability to participants with lost connectivity to the Internet. We investigate how user mobility and query patterns affect data dissemination in such an environment. We discuss the main components of the system and possible applications. Finally, we present simulation results that show that the ad hoc networks can be very effective in distributing popular information.

We present an experimental analysis of on-off patterns in voice over IP (VoIP), where we study the talk-spurt/gap distribution produced by two modern silence detectors: ITU G.729 Annex B Voice Activity Detector (VAD) and NeVoT Silence Detector (SD). The results indicate that spurt/gap distributions are fairly sensitive to both the sound volume and the type of silence detectors, but all showed that the traditional assumption of exponential distribution does not always fit well with the audio sessions we recorded. Both the spurt and gap distributions are more "heavy-tailed" than the exponential curve. In particular, the gap distribution deviates much more strongly from the exponential model, even when "hangover" is applied. To estimate how such deviation affects VoIP applications, we investigate the performance of voice traffic multiplexing. In particular, we look at the probability of having a out-of-profile packet (p/sub o/) when a token bucket filter is placed at the multiplexing end. We ran a series of simulations under three increasingly accurate settings: the exponential model, the real CDF, and the raw silence detector outputs. The token bucket results are robust with regard to the details of the distribution. This is particularly true when the multiplexing factor N is large and the token buffer size B is not too big. When N is small and/or B is big, however, the estimated p/sub o/ under the real CDF is about 30% to 200% larger than under the exponential model. The relative difference between the raw silence detector outputs and the real CDF is generally much smaller than between the real CDF and the exponential model. Therefore, the data traffic in VoIP has a small temporal correlation and a secondary effect on the multiplexer performance.

The service location protocol (SLP), a proposed IETF standard, provides a flexible and scalable service discovery framework for IP networks. It can be deployed with or without a directory service. This paper presents mSLP, the mesh-enhanced service location protocol. mSLP enhances SLP with a fully-meshed peer directory agent (DA) architecture. Peer DAs exchange service registration information, and maintain the same consistent data for shared scopes. mSLP improves the reliability and consistency of SLP directory services. It also greatly simplifies service agent (SA) registrations in systems with multiple DAs. mSLP is backward compatible with SLPv2 and can be deployed incrementally.

Little attention has been given to understanding the fault recovery characteristics and performance tuning of native IP multicast networks. This paper focuses on the interactions of the component protocols to understand their behavior in network failure and recovery scenarios. We consider a multicast environment based on the Protocol Independent Multicast (PIM) routing protocol, the Internet Group Management Protocol (IGMP) and the Open Shortest Path First (OSPF) protocol. Analytical models are presented to describe the interplay of all of these protocols in various multicast channel recovery scenarios. Quantitative results for the recovery time of IP multicast channels are given as references for network configurations, and protocol development. Simulation models are developed using the OPNET simulation tool to measure the fault recovery time and the associated protocol control overhead, and study the influence of important protocol parameters. A testbed with five Cisco routers is configured with PIM, OSPF, and IGMP to measure the multicast channel failure and recovery times for a variety of different link and router failures. In general, the failure recovery is found to be lightweight in term of control overhead and recovery time. Failure recovery time in a WAN is found to be dominated by the unicast protocol recovery process. Failure recovery in a LAN is more complex, and strongly influenced by protocol interaction and implementation specifics. Suggestion for improvement of the failure recovery time via protocol enhancements, parameter tuning, and network configuration are provided.

Transport of real-time voice traffic on the Internet is difficult due to packet loss and jitter. Packet loss is handled primarily through a variety of different forward error correction (FEC) algorithms and local repair at the receiver. Jitter is compensated for by means of adaptive playout buffer algorithms at the receiver. Traditionally, these two mechanisms have been investigated in isolation. In this paper, we show the interactions between adaptive playout buffer algorithms and FEC, and demonstrate the need for coupling. We propose a number of novel playout buffer algorithms which provide this coupling, and demonstrate their effectiveness through simulations based on both network models and real network traces.

We present a novel approach for realizing adaptive QoS control in the Internet called the adaptive load service (ALS). ALS was designed in a similar fashion to the available bit rate service (ABR) proposed for ATM. That is, the senders transmit control messages indicating their requirements, intermediate routers adjust the indicated values in accordance with their available resources and the receivers send the contents of the control messages back to the senders. The transmission behavior of the senders is then adjusted in accordance with the indicated values in the returned control messages. Unlike ABR, ALS is IP-based and was designed to scale for large multicast groups and accommodate the needs of heterogeneous receivers through the integration of sender-based and receiver-based adaptation mechanisms in addition to relying on the network feedback for adjusting the transmission behavior. Performance tests of ALS under different simulation topologies show its efficiency in terms of bandwidth utilization and loss reduction. The tests also show ALS to be fair towards competing TCP connections. Additionally, even though ALS uses a simple algorithm for determining the fair bandwidth share a connection should be using, ALS manages to fairly distribute available resources in accordance with the max-min fairness criterion.

No abstract available.

No abstract available.

We study a dynamic, usage- and congestion-dependent pricing system in conjunction with price-sensitive user adaptation of network usage. We first present a resource negotiation and pricing (RNAP) protocol and architecture to enable users to select and dynamically renegotiate network services. We develop mechanisms within the RNAP architecture for the network to dynamically formulate prices and communicate pricing and charging information to the users. We then outline a general pricing strategy in this context. We discuss candidate algorithms by which applications (singly, or as part of a multi-application system) can adapt their rate and QoS requests, based on the user-perceived value of a given combination of transmission parameters. Finally, we present experimental results to show that usage- and congestion-dependent pricing can effectively reduce the blocking probability, and allow bandwidth to be shared fairly among applications, depending on the elasticity of their respective bandwidth requirements.

Internet telephony offers the opportunity to design a global multimedia communications system that may eventually replace the existing telephony infrastructure. We describe the upper-layer protocol components that are specific to Internet telephony services: the Real-Time Transport Protocol (RTP) to carry voice and video data, and the Session Initiation Protocol (SIP) for signaling. We also mention some complementary protocols, including the Real Time Streaming Protocol (RTSP) for control of streaming media, and the Wide Area Service Discovery Protocol (WASRV) for location of telephony gateways.

Internet telephony must offer the standard telephony services. However the transition to Internet-based telephony services also provides an opportunity to create new services more rapidly and with lower complexity than in the existing public switched telephone network (PSTN). The Session Initiation Protocol (SIP) is a signaling protocol that creates, modifies and terminates associations between Internet end systems, including conferences and point-to-point calls. SIP supports unicast, mesh and multicast conferences, as well as combinations of these modes. SIP implements services such as call forwarding and transfer placing calls on hold, camp-on and call queueing by a small set of call handling primitives. SIP implementations can re-use parts of other Internet service protocols such as HTTP and the Real-Time Stream Protocol (RTSP). We describe SIP, and show how its basic primitives can be used to construct a wide range of telephony services.

RTP, the Real Time Transport Protocol, has gained widespread acceptance as the transport protocol for voice and video on the Internet. Its companion control protocol, the Real Time Control Protocol (RTCP), is used for loose session control, QoS reporting, and media synchronization, among other functions. The RTP specification describes an algorithm for determining the RTCP packet transmission rate at a host participating in a multicast RTP session. This algorithm was designed to allow the RTP to be used in sessions with anywhere from one to a million members. However, we have discovered several problems with this algorithm when used with very large groups with rapidly changing group membership. One problem is the flood of RTCP packets which occurs when many users join a multicast RTP session at nearly the same time. To solve this problem, we present a novel adaptive timer algorithm called reconsideration. We present a mathematical analysis of this algorithm, and demonstrate that it performs extremely well, reducing the congestion problem by several orders of magnitude. We also back up these results with simulation.

Although the Internet was designed to handle non-real time data traffic, it is being used increasingly to carry voice and video. One important class of contributors to this growth are Internet telephones. Critical to more widespread use of Internet telephony is smooth interoperability with the existing telephone network. This interoperability comes through the use of Internet telephony gateways (ITGs) which perform protocol translation between an IP network and the public switched telephone network (PSTN). In order for an IP host to call a user on the PSTN, the IP host must know the IP address of an appropriate gateway. We consider the problem of finding these gateways. An analysis of a number of protocol architectures is presented, including hierarchical databases, multicast advertisement, routing protocols, and centralized databases. We propose a new protocol architecture, called Brokered Multicast Advertisements (BMA) which serves as a lightweight, scalable mechanism for locating ITGs. The BMA architecture is general, and can be applied to location of any service across a wide area network.

Recent advances in distributed systems and transportable software and increasing demand for better quality-of-service (QOS) control in multiservice networks are driving a re-examination of network software architectures. We established the COMET Group (Control Management and Telemedia) at Columbia University's Center for Telecommunications Research to provide a comprehensive understanding of network software architecture of the 1990s and beyond. In this paper, we present an overview of our activities, focusing on new research initiatives, international forum participation and on-going research projects. Collectively, these activities shape our vision of a new era driven by open programmable networking.

Currently, a variety of the MBONE video tools provide video conferencing capabilities on different platforms and with a variety of compression algorithms. However most of these tools lack the ability to interact with other media agents that might be used during a conferencing session. Such interaction is required, for example, for achieving lip synchronisation between audio and video streams or for quality of service control. In this paper we present a new video tool, NEVIT. This tool provides the basic capabilities needed for video conferencing services such as video capturing, compression and decompression engines and multicasting and ATM network interfaces. To ease the interaction with other media agents, NEVIT incorporates a message handling facility to interact over a local conference bus with other media agents, a floor controller of the conference controller. Currently, we are working on adding lip synchronisation and quality-of-service control using this conference bus.

Recent interest in supporting packet-audio applications over wide area networks has been fueled by the availability of low-cost, toll-quality workstation audio and the demonstration that limited amounts of interactive audio can be supported by today's Internet. In such applications, received audio packets are buffered, and their playout delayed at the destination host in order to compensate for the variable network delays. The authors investigate the performance of four different algorithms for adaptively adjusting the playout delay of audio packets in an interactive packet-audio terminal application, in the face of such varying network delays. They evaluate the playout algorithms using experimentally-obtained delay measurements of audio traffic between several different Internet sites. Their results indicate that an adaptive algorithm which explicitly adjusts to the sharp, spike-like increases in packet delay which were observed in the traces can achieve a lower rate of lost packets for both a given average playout delay and a given maximum buffer size.>

The authors distinguish between four types of service that may be provided to real-time traffic by packet-switched networks, ranging from "need-blind" and "need-based best-effort" to "guaranteed throughput" and "bounded delay jitter" services. They evaluate a number of scheduling policies that offer need-based, best-effort service. They introduce hop-laxity (HL) scheduling which is based on the time remaining until the packet must reach its destination as well as the number of hops separating it from the destination. HL scheduling is evaluated through simulation and has been implemented within a BSD-based kernel and tested on the DARTnet network. The results indicate that HL scheduling tends to equalize delays between calls with large and small number of hops as compared to a FIFO discipline, reducing the 99.9% percentile of delay and the fraction of late packets. They compare HL scheduling to the FIFO+ discipline suggested by Clark et al., (see SIGCOMM Symposium on Communications Architectures and Protocols, p.14-26, 1992, and Computer Communication Review, vol.22, no.4) and find that their delay properties are similar. Other disciplines, such as minimum laxity or transit priority, may actually do more harm than good.>

The dramatically increased bandwidths and processing capabilities of future high-speed networks make possible many distributed real-time applications, such as sensor-based applications and multimedia services. Since these applications will have traffic characteristics and performance requirements that differ dramatically from those of current data-oriented applications, new communication network architectures, and protocols will be required. In this paper we discuss the performance requirements and traffic characteristics of various real-time applications, survey recent developments in the areas of network architecture and protocols for supporting real-time services, and develop frameworks in which these, and future, research efforts can be considered.>

The stochastic properties of time-out loss periods are characterized for infinite queues, that is, uninterrupted intervals during which the virtual wait is at or above some fixed threshold. Analytic expressions and numerical techniques presented are for computing both time-based measures such as the distribution of periods during which all arriving packets are lost due to excessive delay as well as packet-based measures such as the distribution of the number of consecutively lost packets and the number of successful packets between such periods of loss. Both continuous and discrete-time systems are examined. It is shown that the assumption of random packet loss severely underestimates the number of consecutively lost packets.>

An investigation is conducted of the possibility of locally controlling short-term congestion for loss-tolerant but delay-sensitive traffic (such as packet voice) through selective discarding of packets based on the virtual work found by a packet on arrival to a queue (local deadlines). By analysis and simulation of a multistage virtual circuit, it is shown that this approach can cut voice-tolerable loss rates in half for high loads. It is also shown that the simple case of using the same local deadline throughout the network performs nearly as well as taking reduced interior traffic into consideration and optimizing loss performance over a set of heterogeneous local deadlines. As an example, the issue of establishing control parameters at call-setup time is also considered.>

The simulation of parallel digital signal processing algorithms and hardware structures can be improved by modeling processing elements as processes executing concurrently on a single CPU or a local-area network. The 'DSP Workbench' strives to hide the location of processes and their communication details from the user.