Although GPS spoofing of individual devices has been extensively examined, little systematic research on swarm spoofing has been conducted. In general, swarm missions may allow each device to navigate independently for different tasks, and it is much more complicated to build corresponding spoofing signals for such general cases. To address this issue, we formulate a general swarm spoofing method to explore the theoretical capabilities and limitations of common cases. We then propose a basic swarm spoofing model to show that, if we try to spoof each receiver precisely, we can only attack a small number of receivers ($\leq 9$) simultaneously in theory. However, in practice, we often need to deal with many receivers. Therefore, we develop a method that can spoof more receivers with acceptable errors. We present a method to construct spoofing messages and evaluate its effectiveness in practical settings with simulations. Although this work focuses on the GPS system, the proposed ideas can be applied to other GNSSs.

As many mobile devices use Global Navigation Satellite Systems (GNSSs) to determine their locations for control, compromising such systems can result in serious consequences, as shown by existing GPS spoofing attacks. However, most such spoofing attacks focus on the effect of a single spoofer attacking a single receiver. In this paper, we investigate the impacts of a single spoofer on multiple receivers, motivated by research on attacking drone swarms. Our analysis independently shows that, using a single spoofer, multiple receivers at different locations in a spoofing area will see the same location reading. We consider the base case of spoofing four satellites and also the generic case when more satellites are involved in the spoofing attack. More importantly, we conduct real-world experiments to validate our analysis and demonstrate the potential threats to many practical applications. We use off-the-shelf SDR cards for spoofing and consumer GPS receivers for obtaining spoofed location readings. While this method can enable various attacks on mobile devices depending on GPS, it is also applicable to all existing GNSSs, because they use similar principles to determine locations.

IoT devices fundamentally lack built-in security mechanisms to protect themselves from security attacks. Existing works on improving IoT security mostly focus on detecting anomalous behaviors of IoT devices. However, these existing anomaly detection schemes may trigger an overwhelmingly large number of false alerts, rendering them unusable in detecting compromised IoT devices. In this paper we develop an effective and efficient framework, named CUMAD, to detect compromised IoT devices. Instead of directly relying on individual anomalous events, CUMAD aims to accumulate sufficient evidence in detecting compromised IoT devices, by integrating an autoencoder-based anomaly detection subsystem with a sequential probability ratio test (SPRT)-based sequential hypothesis testing subsystem. CUMAD can effectively reduce the number of false alerts in detecting compromised IoT devices, and moreover, it can detect compromised IoT devices quickly. Our evaluation studies based on the public-domain N-BaIoT dataset show that CUMAD can on average reduce the false positive rate from about 3.57% using only the autoencoder-based anomaly detection scheme to about 0.5%; in addition, CUMAD can detect compromised IoT devices quickly, with less than 5 observations on average.

While existing methods can disrupt drone invasions in a protected airspace, none of them is able to accurately guide a drone to a desired location for safe handling. To address this issue, theDrone Position Manipulation (DPM)attack is proposed in this article to explore weaknesses in common navigation algorithms of consumer drones. The main advantage of DPM is that it can accurately redirect an invading drone to a desired location, by carefully crafting the spoofed guidance inputs of the drone to exploit the adjustment of path-following navigation algorithms. Compared with existing methods, this is the first work to achieve such accurate quantitative control. Another main contribution of this work is that it explores three fundamental components (i.e., guidance sensing, state estimation, and navigation control) together to enable the quantitative manipulation of flight paths, different from all existing methods. Furthermore, a formal analysis of the attack range is presented for investigating where a drone can be redirected from its target under given constraints. The evaluation on the Software-in-the-loop (SITL) module ofArduPilotsystem shows that the proposed attack is able to not only accurately lead a drone to a redirected destination but also make it fairly far away from its target. Lastly, the proposed attack can be applied to many autopiloted systems, because it exploits common weaknesses in these systems.

Although some existing counterdrone measures can disrupt the invasion of certain consumer drone, to the best of our knowledge, none of them can accurately redirect it to a given location for defense. In this paper, we proposed a Drone Position Manipulation (DPM) attack to address this issue by utilizing the vulnerabilities of control and navigation algorithms used on consumer drones. As such drones usually depend on GPS for autopiloting, we carefully spoof GPS signals based on where we want to redirect a drone to, such that we indirectly affect its position estimates that are used by its navigation algorithm. By carefully manipulating these states, we make a drone gradually move to a path based on our requirements. This unique attack exploits the entire stack of sensing, state estimation, and navigation control together for quantitative manipulation of flight paths, different from all existing methods. In addition, we have formally analyzed the feasible range of redirected destinations for a given target. Our evaluation on open-source ArduPilot system shows that DPM is able to not only accurately lead a drone to a redirected destination but also achieve a large redirection range.

IoT device identification plays an important role in monitoring and improving the performance and security of IoT devices. Compared to traditional non-IoT devices, IoT devices provide us with both unique challenges and opportunities in detecting the types of IoT devices. Based on critical insights obtained in our previous work on understanding the network traffic characteristics of IoT devices, in this paper we develop an effective machine-learning based IoT device identification scheme, named iotID. In developing iotID, we extract 70 features of TCP flows from three complementary aspects: remote network servers and port numbers, packet-level traffic characteristics such as packet inter-arrival times, and flow-level traffic characteristics such as flow duration. Different from existing work, we take into account the imbalance nature of network traffic generated by various devices in both the learning and evaluation phases of iotID. Our performance studies based on network traffic collected on a typical smart home environment consisting of both IoT and non-IoT devices show that iotID can achieve a balanced accuracy score of above 99%.

Understanding network traffic characteristics of IoT devices plays a critical role in improving both the performance and security of IoT devices, including IoT device identification, classification, and anomaly detection. Although a number of existing research efforts have developed machine-learning based algorithms to help address the challenges in improving the security of IoT devices, none of them have provided detailed studies on the network traffic characteristics of IoT devices. In this paper we collect and analyze the network traffic generated in a typical smart homes environment consisting of a set of common IoT (and non-IoT) devices. We analyze the network traffic characteristics of IoT devices from three complementary aspects: remote network servers and port numbers that IoT devices connect to, flow-level traffic characteristics such as flow duration, and packet-level traffic characteristics such as packet inter-arrival time. Our study provides critical insights into the operational and behavioral characteristics of IoT devices, which can help develop more effective security and performance algorithms for IoT devices.

Because cloud storage services have been broadly used in enterprises for online sharing and collaboration, sensitive information in images or documents may be easily leaked outside the trust enterprise on-premises due to such cloud services. Existing solutions to this problem have not fully explored the tradeoffs among application performance, service scalability, and user data privacy. Therefore, we propose CloudDLP, a generic approach for enterprises to automatically sanitize sensitive data in images and documents in browser-based cloud storage. To the best of our knowledge, CloudDLP is the first system that automatically and transparently detects and sanitizes both sensitive images and textual documents without compromising user experience or application functionality on browser-based cloud storage. To prevent sensitive information escaping from on-premises, CloudDLP utilizes deep learning methods to detect sensitive information in both images and textual documents. We have evaluated the proposed method on a number of typical cloud applications. Our experimental results show that it can achieve transparent and automatic data sanitization on the cloud storage services with relatively low overheads, while preserving most application functionalities.

When been threatened by hijacking or suicide-bypilots, the airliner may either crash itself or be shot down due to the potential of the suicide attack. There exist some solutions that allow air traffic controllers or federal agents to take over pilots' authority in the emergency. Though rarely, an air traffic controller may abuse this privilege to mishandle airliners that leads to catastrophic events. In this paper, to mitigate such risks, we propose a multisecret value access control framework based on new designed and existing cryptographic techniques such as XOR-based secret sharing schemes (SSSs). It not only satisfies the efficiency requirement but also assures that each nation owns a unique secret value. We further develop and implement the XOR-based SSS on Linux system. Both experimental results and performance evaluation demonstrate that our solution is not only efficient and bust also secure by design for the multinational air traffic management.

Freenet is a popular peer to peer anonymous content-sharing network, with the objective to provide the anonymity of both content publishers and retrievers. Despite more than a decade of active development and deployment and the adoption of well-established cryptographic algorithms in Freenet, it remains unanswered how well the anonymity objective of the initial Freenet design has been met. In this paper we develop a traceback attack on Freenet, and show that the originating machine of a content request message in Freenet can be identified; that is, the anonymity of a content retriever can be broken, even if a single request message has been issued by the retriever. We present the design of the traceback attack, and perform both experimental and simulation studies to confirm the feasibility and effectiveness of the attack. For example, with randomly chosen content requesters (and random contents stored in the Freenet testbed), Emulab-based experiments show that, for$24$to$43$percent of the content request messages, we can identify their originating machines. We also briefly discuss potential solutions to address the developed traceback attack. Despite being developed specifically on Freenet, the basic principles of the traceback attack and solutions have important security implications for similar anonymous content-sharing systems.

As many malicious apps have been distributed in Android markets, it is urgent to fix the vulnerabilities exploited by these apps and develop effective mitigation methods. In this paper, we identify a common vulnerability in many Android apps. This vulnerability is rooted in an unprotected Android component, called "Activity". Utilizing this vulnerability, a malicious app can stealthily and cannily collect sensitive user data. To better understand this issue, we have built "ActivityHijacker", an app that can detect the right moment to hijack the Activity component and intercept a user's password while it is being inputted in real time. To assess the prevalence of this vulnerability, we further analyzed 8 Android OSs (version 2.x to 5.x) and 22 banking apps collected in January 2016 from various Android markets. Our results show that all these Android OSs and apps are susceptible to the vulnerability; 14 out of 22 banking apps can be hijacked without any prompts. We further present a mitigation mechanism that restricts the Activity component to authorized apps.

Abstract The emerging smart grid incorporates modern information systems with traditional power delivery systems to achieve better control, efficiency, and reliability. However, it also brings potential cyber security risks into the essential power system. Therefore, we need a large‐scale framework to address the basic security issues such as identity management in the smart grid. While public key infrastructures (PKIs) have been developed for large‐scale distributed information systems, the existing PKI solutions cannot meet the unique requirements of the electrical power grid. In this paper, we first identify those unique requirements and discuss how the existing PKI trust models may be adapted to meet these requirements; we further evaluate the existing certificate revocation models; we then discuss various PKI adaptations for the smart grid. Copyright © 2014 John Wiley & Sons, Ltd.

Abstract Spamming botnets present a critical challenge in the control of spam messages because of the sheer volume and wide spread of the botnet members. In this paper, we advocate the approach for recipient mail servers to filter messages directly delivered from remote end‐user (EU) machines, given that the majority of spamming bots are EU machines. We develop a support vector machine (SVM)‐based classifier to separate EU machines from legitimate mail server (LMS) machines, using a set of machine features that cannot be easily manipulated by spammers. We investigate the efficacy and performance of the SVM‐based classifier using a number of real‐world data sets. Our performance studies show that the SVM‐based classifier is indeed a feasible and effective approach in distinguishing EU machines from LMS machines. For example, training and testing on an aggregated data set containing both EU machines and LMS machines, on average, we found that the SVM‐based classifier can achieve a 99.25 % detection accuracy, with very small false positive rate (0.35%) and false negative rate (1.27%), significantly outperforming eight Domain Name System‐based blacklists widely used today. Copyright © 2012 John Wiley & Sons, Ltd.

Retweet has become one of the most prominent feature on social networks and an important mean for secondary content promotion. Most existing investigations of retweet behaviors on social networks are conducted based on empirical studies or information diffusion models (such as stochastic process or cascading model). To the best of our knowledge, such a retweet process has not been investigated as a chaotic process. In this paper, we have first examined that retweet time series by 0-1 test where the results provide identification of chaotic behaviors. Furthermore, taking into account of the proven chaotic characteristic, chaos LS-SVM prediction method is applied to form predictions using only a small fraction of the retweet time series. Our evaluation on Sina Weibo dataset and comparisons with a bayesian model and strawman modal show that this nonlinear prediction method can translate to good step ahead forecasts and perform high accuracy in retweet prediction.

Although supporting Quality of Service (QoS) on the traditional Internet has been extremely challenging, QoS supports are still highly desirable for many real-time applications. While existing QoS schemes often show poor scalability, low efficiency, and limited QoS supports, the fast development of Software Define Networks (SDNs) provides new opportunities to address QoS support with its centralized network control. In this paper, we propose a synchronized architecture to exploit the unique features of SDNs without conducting static reservations. Assume all routers are synchronized in time, we can precisely determine the upstream delay of a packet at a router. Meanwhile, because we also know the fairly accurate resource availability on its downstream routers (from the SDN controller), we can determine the packet's service priority on its current router based on its end-to-end (e2e) delay requirement and the expected delay that the packet may experience in its downstream. In particular, we propose a synchronized multi-hop scheduling (SMS) scheme to exploit both upstream information and downstream resource availability to speed up or slow down a packet. This is completely different from all existing per-hop or multi-hop schemes that mostly utilize upstream information. Furthermore, we propose to selectively drop a packet that is unlikely to meet its deadline. Our simulation results show that the proposed scheme outperforms existing schemes in packet missing rates.

Although cascading failures have occurred on many real-world networks, to our best knowledge, no one has clearly identified this issue on a social network. In this paper, we identify this potential issue on social networks, and develop a theoretical model to analyze related issues. Note that highly-influential “super” users play critical roles on a social network. When they are suddenly unavailable, a large portion of the social network may be seriously disrupted. The proposed model captures this dynamic process and helps us better understand related issues. Furthermore, we evaluate the proposed model under four attack strategies based on real social network datasets collected on Twitter and Sina Weibo. We also analyze the connectivity, the persistent time, and the cascade effect of a social network under these attacks. Our results show that social network service providers have to pay closer attention to super users to avoid dramatic failures.

Supporting real time traffic on the Internet has always been a challenge due to undesired network dynamics. In this paper, we introduce a synchronized routing architecture to address this issue from a new perspective. As routers are synchronized in time, we can easily find accurate upstream delays. When we can also obtain downstream delays with a measurement framework, we can then determine the dynamic service priority for a packet at a hop to speed up or slow down as necessary, to enable efficient resource sharing among multiple flows across different paths. Consequently, we can achieve better overall system performance without making explicit reservations for each flow. Assume we have the downstream measurement framework in hand, in this paper, we focus on synchronized multi-hop scheduling schemes and illustrate its basic capacity via both analysis and simulation.

In this paper, we propose a new SPNR model and identify the concrete propagation relationships and obtain the spreading threshold. We evaluate the proposed model with simulations and compare the simulation results with real data on Sina Weibo, the largest micro-blogging tool in China. The results show that the new model is effective for capturing the rumor spreading in real social networks. To obtain effective rumor control strategy, we further analyze the key factors that affect the maximum value of steady state, the point of decline, and the life cycle of a rumor. These results help us develop new rumor control strategies.

As privacy becomes an increasingly important issue on today's Internet, attacks on anonymous tools gain more attention. In this paper, we investigated countermeasures to a routing table insertion (RTI) attack on Freenet. We analyzed the key factors supporting the RTI attack, and proposed a simple randomized routing method to mitigate the RTI or similar attacks. We implemented the proposed method on the Freenet Thynix simulator. Our results show that the proposed method works well on small world topologies and thus a feasible choice to deal with the RTI attack on Freenet. We are further investigating more effective countermeasures based on these results.

A traceback attack was recently developed on Freenet, which can identify the originating machine of a content request message, even if a single content request message has been issued by a content retriever. The traceback attack exploited a few fine-grained design and development decisions made in Freenet, including the unique identifier (UID) based mechanism to prevent routing loops of content request messages. In this paper we develop a simple yet effective scheme named dynID to thwart the traceback attack on Freenet. In dynID, the UID associated with a content request message is dynamically changed at the beginning portion of the message forwarding path. As a consequence, an attacker can only trace back a content request message to the node where the UID value is last changed; it cannot uniquely determine the originating machine of the message. Importantly, dynID only has negligible impacts on the performance of Freenet in locating content on the network. For example, our simulation studies based on the original Freenet source code show that, for all content requests, we can successfully locate the corresponding requested content.

Cloud computing is broadly recognized as one of major factors in achieving more flexible, scalable, and efficient systems. However, as customers lose the direct control of their data and applications hosted by cloud providers, the trustworthiness of cloud services is a main issue that hinders the deployment of cloud applications. In this paper, we have developed a novel framework to detect compromises on physical servers in cloud services, via remote attestation with a Trusted Third Party (TTP). Furthermore, to avoid the TTP becoming a bottleneck, we have designed a cloud based TTP platform, using a small private cloud to audit large clouds. We have implemented a prototype system, and evaluated it with several common benchmarks to demonstrate its efficiency. Our experimental results show that the proposed framework is effective in detecting compromise and adds little overhead to a common IaaS cloud environment.

Freenet is a popular peer to peer anonymous network, with the objective to provide the anonymity of both content publishers and retrievers. Despite more than a decade of active development and deployment and the adoption of well-established cryptographic algorithms in Freenet, it remains unanswered how well the anonymity objective of the initial Freenet design has been met. In this paper we develop a traceback attack on Freenet, and show that the originating machine of a content request message in Freenet can be identified; that is, the anonymity of a content retriever can be broken, even if a single request message has been issued by the retriever. We present the design of the traceback attack, and perform Emulab-based experiments to confirm the feasibility and effectiveness of the attack. With randomly chosen content requesters (and random contents stored in the Freenet testbed), the experiments show that, for 24% to 43% of the content request messages, we can identify their originating machines. We also briefly discuss potential solutions to address the developed traceback attack. Despite being developed specifically on Freenet, the basic principles of the traceback attack and solutions have important security implications for similar anonymous content sharing systems.

As one of the most popular low-latency anonymous communication systems, Tor has been a great success but still faces some challenges, e.g., subject to low-resource attacks and no explicit mechanisms to explore tradeoffs between anonymity and performance. In this paper, we propose a Relay Recommendation System (RRS) for Tor to provide reliable relay information for building paths with better performance, mitigate low-resource attacks, and enable users to explore the tradeoffs between performance and anonymity based on their needs. We first present the design of RRS, and show its performance improvement and resistance to low-resource, high-resource, and reputation attacks. We further analyze the potential anonymity decrease, and propose new path selection schemes to enable selective anonymity based on user needs. We have evaluated RRS via both analysis and experiments on a Tor simulation platform. Our results show that RRS achieves significant performance improvement with a small decrease of anonymity.

Compromised machines are one of the key security threats on the Internet; they are often used to launch various security attacks such as spamming and spreading malware, DDoS, and identity theft. Given that spamming provides a key economic incentive for attackers to recruit the large number of compromised machines, we focus on the detection of the compromised machines in a network that are involved in the spamming activities, commonly known as spam zombies. We develop an effective spam zombie detection system named SPOT by monitoring outgoing messages of a network. SPOT is designed based on a powerful statistical tool called Sequential Probability Ratio Test, which has bounded false positive and false negative error rates. In addition, we also evaluate the performance of the developed SPOT system using a two-month e-mail trace collected in a large US campus network. Our evaluation studies show that SPOT is an effective and efficient system in automatically detecting compromised machines in a network. For example, among the 440 internal IP addresses observed in the e-mail trace, SPOT identifies 132 of them as being associated with compromised machines. Out of the 132 IP addresses identified by SPOT, 126 can be either independently confirmed (110) or highly likely (16) to be compromised. Moreover, only seven internal IP addresses associated with compromised machines in the trace are missed by SPOT. In addition, we also compare the performance of SPOT with two other spam zombie detection algorithms based on the number and percentage of spam messages originated or forwarded by internal machines, respectively, and show that SPOT outperforms these two detection algorithms.

Tor (the second generation onion routing) is arguably the most popular low-lateney anonymous communication system now. In this paper, we reexamine the anonymity of Tor based on our observation of "super nodes". These nodes are more available and reliable than other nodes and provide high bandwidth for assisting the system in both performance and stability. We first confirm their existence by analyzing the life cycles of node IP addresses and node bandwidth contributions via two correlation approaches, on a set of self-collected data and a set of real data from the Tor official collection. We then analyze the effect of super nodes on the anonymity of Tor, discuss attacks that exploit such knowledge, and verify our analysis with real data to show potential damages. Furthermore, we investigate new attacks that exploit the knowledge of super nodes. Our simulation results show that these attacks can greatly damage the anonymity of Tor.

Flow-based traffic processing systems have been widely deployed on multi-core systems for performance purpose. However, the current flow table design cannot balance parallel processing performance and memory usage of locks, and is unable to automatically adapt to different network settings. To address these issues, in this paper we propose a Hypercube Flow Table (HFT) based on multi- dimension hash tables. HFT supports different lock granularities to allow multiple cores to process packets in parallel and reduce memory usage by assigning a lock for a group of buckets, instead of one lock per bucket as many current designs. Furthermore, HFT can adapt to various network situations by dynamically adjusting buckets assignment. We have tested HFT on a traffic classification system with two Intel Xeon E5504 processors. The results show that HFT can effectively balance memory usage and parallel processing performance, and further adapt to various network settings by adjusting buckets assignment.

Network security assessment is critical to the survivability and reliability of distributed systems. In this paper, we propose a novel assessment approach that supports automatic vulnerability assessment utilizing Bayesian attack graphs. We also integrate several major vulnerability database into a comprehensive database and build a customized vulnerability scanner to assist attack graph generation. Different from existing solutions that manually assign probabilities to a Bayesian attack graph, we design a set of quantitative metrics to automatically analyze vulnerability and evaluate the proposed approach with real-world examples. Our results show the promising capability of the proposed approach in further improving assessment quality.

Thunder (also called Xunlei) is the most popular P2P file sharing application in China and probably the most popular P2P software in term of traffic volume and number of users. Precisely identifying Thunder traffic can help network administrators to efficiently manage their networks. Traditional methods of identifying P2P traffic such as port-based or content-based approaches are ineffective to Thunder traffic, because of its dynamic packet format, flexible port numbers, and payload encryption. In this paper, we developed a novel Heuristic Message Clustering approach (HMC) to identify Thunder traffic, and obtain its state machine and key transaction cycles, thus identifying Thunder traffic fast and accurately. We first evaluate our method in a controlled environment and then with real campus traces. The results show that HMC is able to identify Thunder flows with high precision and low error rate.

Classifying encrypted traffic is critical to effective network analysis and management. While traditional payload- based methods are powerless to deal with encrypted traffic, machine learning methods have been proposed to address this issue. However, these methods often bring heavy overhead into the system. In this paper, we propose a hybrid method that combines signature-based methods and statistical analysis methods to address this issue. We first identify SSL/TLS traffic with signature matching methods, and then apply statistical analysis to determine concrete application protocols. Our experimental results show that the proposed method is able to recognize over 99% of SSL/TLS traffic and achieve 94.52% in F-score for protocols identification.

Compromised machines are one of the key security threats on the Internet; they are often used to launch various security attacks such as DDoS, spamming, and identity theft. In this paper we address this issue by investigating effective solutions to automatically identify compromised machines in a network. Given that spamming provides a key economic incentive for attackers to recruit the large number of compromised machines, we focus on the subset of compromised machines that are involved in the spamming activities, commonly known as spam zombies. We develop an effective spam zombie detection system named SPOT by monitoring outgoing messages of a network. SPOT is designed based on a powerful statistical tool called Sequential Probability Ratio Test, which has bounded false positive and false negative error rates. Our evaluation studies based on a two- month email trace collected in a large U.S. campus network show that SPOT is an effective and efficient system in automatically detecting compromised machines in a network. For example, among the 440 internal IP addresses observed in the email trace, SPOT identifies 132 of them as being associated with compromised machines. Out of the 132 IP addresses identified by SPOT, 126 can be either independently confirmed (110) or highly likely (16) to be compromised. Moreover, only 7 internal IP addresses associated with compromised machines in the trace are missed by SPOT.

Delay guarantees are critical to real-time applications. To ensure the end-to-end delay guarantee of a flow, the previous approaches usually statically divide the end-to-end delay requirement into per-hop delay requirements. However, such static resource allocations may cause unbalanced usage in a network, limit the total number of sessions that can be accommodated, and result in low network utilization. To address this issue, we propose a dynamic allocation adjustment (DAA) approach to evenly spread traffic over flow paths and related links, and improve the system utilization. Our simulation results show that DAA is able to significantly increase the total number of flows admitted; it reduces the average link reservations, decreases the total resource usage per flow, and balances link loads.

Efficiently utilizing limited peer upstream bandwidth is one of key challenges in P2P streaming. Many mechanisms have been proposed to address this issue by considering peer availability, data redundancy, and priority encoding schemes. In this paper, we investigate this issue from a unique aspect: how to place video data at peers with various upstream capacities in order to better utilize their upstream bandwidth. We propose several effective placement solutions. Our analysis and simulation results show the efficacy of the proposed solutions.

We propose an overlay filtering framework that enables protected servers to actively defend themselves and filter out specific attack flows as early as possible close to attack sources without collateral damages, different from current filtering approaches that provide passive protection at or close to destinations. This framework is designed to protect many distributed network applications that do not require pre-existing trust relationships, different from the previous overlay schemes designed for pre-authenticated known clients. We present the basic architecture and performance analysis in this paper.

In this paper, we propose a Loopback approach in a two-level streaming architecture to exploit collaborative client/proxy buffers for improving the quality and efficiency of large-scale streaming applications. At the upper level we use a content delivery network (CDN) to deliver video from a central server to proxy servers. At the lower level a proxy server delivers video with the help of collaborative client caches. In particular, a proxy server and its clients in a local domain cache different portions of a video and form delivery loops. In each loop, a single video stream originates at the proxy, passes through a number of clients, and finally is passed back to the proxy. As a result, with limited bandwidth and storage space contributed by collaborative clients, we are able to significantly reduce the required network bandwidth, I/O bandwidth, and cache space of a proxy. Furthermore, we develop a local repair scheme to address the client failure issue for enhancing service quality and eliminating most required repairing load at the central server. For popular videos, our local repair scheme is able to handle most of single-client failures without service disruption and retransmissions from the central server. Our analysis and simulations have shown the effectiveness of the proposed scheme.

In this paper, we propose a hybrid architecture to integrate peer-to-peer (P2P) streaming approaches with content distribution networks (CDNs). We further utilize multiple description (MD) coding in this architecture to address challenging issues in P2P approaches such as low peer upstream bandwidth and low quality assurance. The proposed schemes take advantage of the high availability of CDNs, the flexibility of layered MD, and the high scalability of P2P approaches to better utilize peer upstream bandwidth, reduce server load, and assure service quality. In addition, we evaluate the performance of the proposed approach through analysis and simulation

Bandwidth contention is still a challenging problem in providing efficient IP-based video-on-demand (VOD) service on broadband cable networks (BCNs), due to the lack of effective approaches to exploit the unique characteristics of BCNs. To address this issue, we have designed an optimal full-sharing (Dong Y et al., 2003) for VOD service over a single channel, which maximizes the number of simultaneous video sessions on the single channel. In this paper, we further extend our optimal scheduling for VOD service over multiple channels. We first analyze the expected session bandwidth of a video and then develop an efficient video assignment mechanism for maximizing the profit of a VOD system.

In this paper, we examine and characterize effects of communication interactions of parallel and sequential jobs on a nondedicated network of workstations. We quantitatively model the interaction process and measure communication delays of parallel jobs caused by the interaction. Measurement results on an ATM network of workstations support our analytical models.