A5014376333- Network Traffic and Congestion Control
- Software-Defined Networks and 5G
- Network Security and Intrusion Detection
- Peer-to-Peer Network Technologies
- Caching and Content Delivery
- Advanced Optical Network Technologies
- Internet Traffic Analysis and Secure E-voting
- Energy Efficient Wireless Sensor Networks
- Advanced Wireless Network Optimization
- Mobile Ad Hoc Networks
- Wireless Communication Networks Research
- Interconnection Networks and Systems
- IPv6, Mobility, Handover, Networks, Security
- Wireless Networks and Protocols
- Software System Performance and Reliability
- Network Time Synchronization Technologies
- Image and Video Quality Assessment
- Mobile Agent-Based Network Management
- Opportunistic and Delay-Tolerant Networks
- Network Packet Processing and Optimization
- Advanced Queuing Theory Analysis
- Security in Wireless Sensor Networks
- Cooperative Communication and Network Coding
- Cloud Computing and Resource Management
- Wireless Body Area Networks
Purdue University West Lafayette
2016-2025
University of California, Riverside
2018
University of California, Santa Barbara
2018
University of Kentucky
2018
Osaka University
2018
Chinese University of Hong Kong
2018
Bell (Canada)
2018
Laboratoire d'Informatique de Paris-Nord
2007
Rensselaer Polytechnic Institute
2006
The Ohio State University
1998-2003
Topology control in a sensor network balances load on nodes and increases scalability lifetime. Clustering is an effective topology approach. We propose novel distributed clustering approach for long-lived ad hoc networks. Our proposed does not make any assumptions about the presence of infrastructure or node capabilities, other than availability multiple power levels nodes. present protocol, HEED (Hybrid Energy-Efficient Distributed clustering), that periodically selects cluster heads...
Prolonged network lifetime, scalability, and load balancing are important requirements for many ad-hoc sensor applications. Clustering nodes is an effective technique achieving these goals. In this work, we propose a new energy-efficient approach clustering in networks. Based on approach, present protocol, HEED (hybrid distributed clustering), that periodically selects cluster heads according to hybrid of their residual energy secondary parameter, such as node proximity its neighbors or...
This paper describes the "explicit rate indication for congestion avoidance" (ERICA) scheme rate-based feedback from asynchronous transfer mode (ATM) switches. In ERICA, switches monitor their load on each link and determine a factor, available capacity, number of currently active virtual channels. information is used to advise sources about rates at which they should transmit. The algorithm designed achieve high utilization with low delays fast transient response. It also fair robust...
We propose an application level multicast approach, Topology Aware Grouping (TAG), which exploits underlying network topology information to build efficient overlay networks among group members. TAG uses about path overlap members construct a tree that reduces the relative delay penalty, and number of duplicate copies packet on same link. study properties TAG, model experiment with its economies scale factor quantify benefits compared unicast IP multicast. also compare approach ESM in...
Energy efficiency is critical for wireless sensor networks. The data gathering process must be carefully designed to conserve energy and extend the network lifetime. For applications where each continuously monitors environment periodically reports a base station, tree-based topology often used collect from nodes. In this work, we study construction of tree maximize lifetime, which defined as time until first node depletes its energy. problem shown NP-complete. We design an algorithm starts...
Energy efficiency is critical for wireless sensor networks. The data-gathering process must be carefully designed to conserve energy and extend network lifetime. For applications where each continuously monitors the environment periodically reports a base station, tree-based topology often used collect data from nodes. In this work, we first study construction of tree when there single station in network. objective maximize lifetime, which defined as time until node depletes its energy....
Network Function Virtualization (NFV) brings a cloud service automation paradigm to demand-driven elastic flexing of infrastructure resources. Thus, it is essential characterize the impact hardware and virtualization options on virtual network function (VNF) performance, load underlying infrastructure. In this paper, we present VNF characterization case studies with three sample open-source platforms, Clearwater IMS two intrusion detection system VNFs (Snort Suricata). We demonstrate that...
Novel routing paradigms based on policies, quality of service (QoS) requirements, and packet content have been proposed for the Internet over last decade. Constraint-based algorithms select a path satisfying constraints that are either administrative-oriented (policy routing) or service-oriented (QoS routing). The routes, in addition to constraints, selected reduce costs, balance network load, increase security. In this article, we discuss several constraint-based approaches explain their...
Serving multimedia content over the Internet with negligible delay remains a challenge. With advent of Web 2.0, numerous video sharing sites using different storage and delivery models have become popular. Yet, little is known about these from global perspective. Such an understanding important for designing systems which can efficiently serve to users all world. In this paper, we analyze compare underlying distribution frameworks three services - YouTube, Dailymotion Metacafe based on...
Wi-Fi is an integral part of today's Internet infrastructure, enabling a diverse range applications and services. Prior approaches to resource allocation optimized Quality Service (QoS) metrics, which often do not accurately reflect the user's Experience (QoE). To address gap between QoS QoE, we introduce Maestro, adaptive method that formulates problem as partially observable Markov decision process (PO-MDP) maximize overall system QoE fairness. Maestro estimates without using any...
The Available Bit Rate (ABR) service has been developed to support data applications over Asynchronous Transfer Mode (ATM) networks. network continuously monitors its traffic and provides feedback the source end systems. This paper explains rules that sources have follow achieve a fair efficient allocation of resources.
Several sensor network applications, such as environmental monitoring, require data aggregation to an observer. For this purpose, a tree, rooted at the observer, is constructed in network. Node clustering can be employed further balance load among nodes and prolong lifetime. In paper, we design implement system, iHEED, which node integrated with multi-hop routing for TinyOS. We consider simple operators, AVG or MAX. use energy consumption model keep track of battery cluster heads regular...
We study sleep/wake scheduling for low duty cycle sensor networks. Our work is different from prior in that we explicitly consider the effect of synchronization error design algorithm. In our previous work, have studied single hop communications, e.g., intra-cluster communications between a cluster head and members. showed there an inherent trade-off energy consumption message delivery performance (defined as capture probability). proposed optimal algorithm, which satisfies probability...
Researchers in the denial-of-service (DoS) field lack accurate, quantitative, and versatile metrics to measure service denial simulation testbed experiments. Without such metrics, it is impossible severity of various attacks, quantify success proposed defenses, compare their performance. Existing DoS equate with slow communication, low throughput, high resource utilization, loss rate. These are not because they fail monitor all traffic parameters that signal degradation. They quantitative...
Overlay networks among cooperating hosts have recently emerged as a viable solution to several challenging problems, including multicasting, routing, content distribution, and peer-to-peer services. Application-level overlays, however, incur performance penalty over router-level solutions. This paper quantifies explains this for overlay multicast trees via: 1) Internet experimental data; 2) simulations; 3) theoretical models. We compare number of protocols with respect tree structure,...
In this paper, we present RoCC, a robust congestion control approach for datacenter networks based on RDMA. RoCC leverages switch queue size as an input to PI controller, which computes the fair data rate of flows in queue, signaling it flow sources. The parameters are self-tuning guarantee stability, rapid convergence, and near-optimal throughput wide range scenarios. Our simulation DPDK implementation results show that can achieve up 7× reduction PFC frames generated under high average...
Creating an experimental infrastructure for developing next-generation information security technologies.
Distance prediction algorithms use O(N) round trip time (RTT) measurements to predict the N <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> RTTs among nodes. can be applied improve performance of a wide variety Internet applications: for instance, guide selection download server from multiple replicas, or construction overlay networks multicast trees. Although accuracy existing has been extensively compared using relative error metric,...
ATM-UBR service responds to congestion by dropping cells when switch buffers become full. TCP connections running over UBR experience low throughput and high unfairness. For 100% throughput, each needs equal the sum of window sizes all connections. Intelligent drop policies can improve performance with limited buffers. The UBR+ proposes enhancements for intelligent drop. early packet discard scheme improves but does not attempt fairness. selective based on per-connection buffer occupancy...