Large networks are often structured hierarchically by grouping nodes into different domains in order to deal with he scaling problem. In such networks, it is infeasible to maintain the detailed network information at every router. Therefore, topology information of domains are summarized before broadcasted. This process is called topology aggregation. Hierarchical routing protocols are then used to find a route among the domains. We study several basic problems associated with hierarchical QoS routing, including (1) how to make QoS-aware topology aggregation, (2) how to represent the aggregated network state, and (3) how to find an end-to-end route based on aggregated information. The novelty in this research is our new network QoS representation which is line segments on the delay-banadwidth plane. We also present a distributed routing mechanism that works with our representation. Our theoretical and simulation results show that the protocol achieves scalability and improved routing performance.

King Shan Lui;Klara Nahrstedt;式刚 陈

Conference on Local Computer Networks

2000

We demonstrate that CSMA/CA networks, including IEEE 802.11 networks, exhibit severe fairness problem in many scenarios, where some hosts obtain most of the channel's bandwidth while others starve. Most existing solutions require nodes to overhear transmissions made by contending nodes and, based on the overheard information, adjust local rates to achieve fairness among all contending links. Their underlying assumption is that transmissions made by contending nodes can be overheard. However, this assumption holds only when the transmission range is equal to the interference range, which is not true in reality. As our study reveals, the overhearing-based solutions, as well as several nonoverhearing AIMD solutions, cannot achieve MAC-layer fairness in various settings. We propose a new rate control protocol, called Proportional Increase Synchronized multiplicative Decrease (PISD). Without relying on overhearing, it provides fairness in CSMA/CA networks, particularly IEEE 802.11 networks, by using only local information and performing localized operations. It combines several novel rate control mechanisms, including synchronized multiplicative decrease, proportional increase, and background transmission. We prove that PISD converges and achieves (weighted) fairness. We further introduce Queue Spreading (QS) to achieve MAC-layer fairness when there are multiple contention groups, in which case PISD will fail.

Ying Jian;Ming Zhang;式刚 陈

IEEE/ACM Transactions on Networking

2011-10

We study a fundamental problem of how to schedule complex workflows in the cloud for applications such as data analytics. One of the main challenges is that such workflow scheduling problems involve many constraints, requirements and varied objectives and it is extremely difficult to find high-quality solutions. To meet the challenge, we explore using mixed integer programming (MIP) to formulate and solve complex workflow scheduling problems. To illustrate the MIP-based method, we formulate three related workflow scheduling problems in MIP. They are fairly generic, comprehensive and are expected to be useful for a wide range of workflow scheduling scenarios. Using results from numerical experiments, we demonstrate that, for problems up to certain size, the MIP approach is entirely applicable and more advantageous over heuristic algorithms.

Yi Wang;Ye Xia;式刚 陈

2017-9-8

While a considerable amount of research has been done for detecting the binary worms exploiting the vulnerability of buffer overflow, very little effort has been spent in detecting worms that consist of only text, i.e., printable ASCII characters. We show that the existing worm detectors often either do not examine the ASCII stream or are not well suited to efficiently detect worms in the ASCII stream due to the structural properties of the ASCII payload. In this paper, we analyze the potentials and constraints of the ASCII worms vis-a-vis their binary counterpart, and devise a detection technique that would exploit those limitations. We introduce DAWN, a novel ASCII worm detection strategy that is fast, easily deployable, and has very little overhead. Unlike many signature - based detection methods, DAWN is completely signature-free and therefore capable of detecting zero-day outbreak of ASCII worms.

Parbati Kumar Manna;Sanjay Ranka;式刚 陈

2008

Traffic measurement is a critical function in transportation engineering. We consider privacy-preserving point-to-point traffic measurement in this paper. We measure the number of vehicles traveling from one geographical location to another by taking advantage of capabilities provided by the intelligent cyberphysical road systems (CPRSs) that enable automatic collection of traffic data. The challenge is to allow the collection of aggregate point-to-point data while preserving the privacy of individual vehicles. We propose a novel measurement scheme, which utilizes bit arrays to collect "masked" data and adopts maximum-likelihood estimation (MLE) to obtain the measurement result. Both mathematical proof and simulation demonstrate the practicality and scalability of our scheme.

Yian Zhou;Zhen Mo;Qingjun Xiao;式刚 陈;Yafeng Yin

IEEE Transactions on Vehicular Technology

2016-5

Radio frequency identification has been gaining popularity in a variety of applications from shipping and transportation to retail industry and logistics management. With a limited reader-tag communication range, multiple readers (or reader antennas) must be used to provide full coverage to any deployment area beyond a few meters across. However, reader contention can seriously degrade the performance of the system or even block out some tags from being read. Most prior work on this problem requires hardware and protocol support that is incompatible with the EPC Gen2 standard. Moreover, they assume the knowledge of a reader network that precisely describes the contention relationship among all readers, but the efficient acquisition of the reader network in a practical system with commercial-off-the-shelf (COTS) tags is an open problem. This study fills the gap by proposing a novel protocol Retwork, which works under the limitations imposed by commercial Gen2-compatible tags and identifies all possible reader contentions efficiently through careful protocol design that exploits the flag-setting capability of these tags. We have implemented a prototype with 8,000 commercial tags. Extensive experiments demonstrate that Retwork can reduce communication overhead by an order of magnitude, in comparison to an alternative solution.

Jia Liu;Xingyu Chen;式刚 陈;Wei Wang;Dong Jiang;Lijun Chen

2020

Traffic volume estimation is critical to the intelligent transportation engineering. Previous state-of-the-art studies mainly focus on measuring two types of traffic volume: 'point' traffic (i.e., the number of vehicles passing a given location) and 'point-to-point' traffic (i.e., the number of vehicles traversing between two given locations) during each measurement period. In this paper, we extend this line of research from single-period to multiple periods and study new problems of estimating the number of k -persistent vehicles that pass a location or two different locations in at least k -out-of- t predefined measurement periods. We propose two novel k -persistent traffic estimators with privacy-preserving for the point and point-to-point traffic models, respectively. Through theoretical analysis, we prove that our solution can solve more general traffic measurement problems and employ stronger privacy preserving, i.e., \epsilon -differential privacy, than the existing studies. We also demonstrate the effectiveness and the accuracy of the proposed estimators through extensive experiments based on real transportation traffic flows in Shenzhen, China for five consecutive working days. The numerical results show that the estimators can achieve a tradeoff between the estimation accuracy and privacy preservation through proper parameter setting.

Yu E. Sun;河 黄;式刚 陈;You Zhou;Kai Han;Wenjian Yang

IEEE Internet of Things Journal

2019-10

Many Internet multicast applications such as teleconferencing and remote diagnosis have Quality-of-Service (QoS) requirements. The requirements can be additive (end-to-end delay), multiplicative (loss rate), or of a bottleneck nature (bandwidth). Given such diverse requirements, it is a challenging task to build QoS-constrained multicast trees in a large network where no global network state is available. This paper proposes a scalable QoS multicast routing protocol (SoMR) that supports all three QoS requirement types. SoMR is scalable due to small communication overhead. It achieves favorable tradeoff between routing performance and routing overhead by carefully selecting the network sub-graph in which it searches for a path that can support the QoS requirements. The scope of search is automatically tuned based on the current network conditions. An early-warning mechanism helps detect and route around the long-delay paths in the network. The operations of SoMR are completely decentralized. They rely only on the local state stored at each router.

式刚 陈;Yuval Shavitt

Journal of Parallel and Distributed Computing

2008-2

An origin-destination (OD) flow between two routers is the set of packets that pass both routers in a network. Measuring the sizes of OD flows is important to many network management applications such as capacity planning, traffic engineering, anomaly detection, and network reliability analysis. Measurement efficiency and accuracy are two main technical challenges. In terms of efficiency, we want to minimize per-packet processing overhead to accommodate future routers that have extremely high packet rates. In terms of accuracy, we want to generate precise measurement results with small bias and standard deviation. To meet these challenges, we design a new measurement method that employs a compact data structure for packet information storage and uses a novel statistical inference approach for OD-flow size estimation. We perform simulations to demonstrate the effectiveness of our method.

Tao Li;式刚 陈;Yan Qiao

2012

Cloud storage has been gaining popularity because its elasticity and pay-as-you-go manner. However, this new type of storage model also brings security challenges. This paper studies the problem of ensuring data integrity in cloud storage. In the Proof of Retrievability (PoR) model, after outsourcing the preprocessed data to the server, the client will delete its local copies and only store a small amount of meta data. Later the client will ask the server to provide a proof that its data can be retrieved correctly. However, most of the prior PoR works apply only to static data. The existing dynamic version of PoR scheme has an efficiency problem. In this paper, we extend the static PoR scheme to dynamic scenario. That is, the client can perform update operations, e.g., insertion, deletion and modification. After each update, the client can still detect the data losses even if the server tries to hide them. We develop a new version of authenticated data structure based on a B+ tree and a merkle hash tree. We call it Cloud Merkle B+ tree (CMBT). By combining the CMBT with the BLS signature, we propose a dynamic version of PoR scheme. Compared with the existing dynamic PoR scheme, our worst case communication complexity is O(logn) instead of O(n).

Zhen Mo;Yian Zhou;式刚 陈

2012