- Energy Efficient Wireless Sensor Networks
- Mobile Ad Hoc Networks
- Network Traffic and Congestion Control
- Caching and Content Delivery
- Software-Defined Networks and 5G
- Peer-to-Peer Network Technologies
- Opportunistic and Delay-Tolerant Networks
- Indoor and Outdoor Localization Technologies
- Wireless Networks and Protocols
- Network Security and Intrusion Detection
- Network Packet Processing and Optimization
- Context-Aware Activity Recognition Systems
- Internet Traffic Analysis and Secure E-voting
- Mobile Crowdsensing and Crowdsourcing
- Complex Network Analysis Techniques
- IoT and Edge/Fog Computing
- Cooperative Communication and Network Coding
- Security in Wireless Sensor Networks
- Software System Performance and Reliability
- Human Mobility and Location-Based Analysis
- Structural Health Monitoring Techniques
- Energy Harvesting in Wireless Networks
- Modular Robots and Swarm Intelligence
- Mobile Agent-Based Network Management
- Green IT and Sustainability
University of Southern California
2016-2025
Southern California University for Professional Studies
2014-2024
University of Massachusetts Amherst
2020
Google (United States)
2016
LAC+USC Medical Center
2007-2015
University of California, Riverside
2010-2013
Hanyang University
2013
National Yang Ming Chiao Tung University
2013
Institute of Electrical and Electronics Engineers
2010
University of California, Los Angeles
2004-2009
Advances in processor, memory and radio technology will enable small cheap nodes capable of sensing, communication computation. Networks such can coordinate to perform distributed sensing environmental phenomena. In this paper, we explore the directed diffusion paradigm for coordination. Directed is datacentric that all named data. All a diffusion-based network are application-aware. This enables achieve energy savings by selecting empirically good paths caching processing data in-network....
Article Free Access Share on Next century challenges: scalable coordination in sensor networks Authors: Deborah Estrin USC/Information Sciences Institute, 4676 Admiralty Way, Marina del Rey, CA CAView Profile , Ramesh Govindan John Heidemann Satish Kumar Authors Info & Claims MobiCom '99: Proceedings of the 5th annual ACM/IEEE international conference Mobile computing and networkingAugust 1999 Pages 263–270https://doi.org/10.1145/313451.313556Published:01 August 1999Publication History...
Advances in processor, memory, and radio technology enable small cheap nodes capable of sensing, communication, computation. Networks such can coordinate to perform distributed sensing environmental phenomena. We explore the directed diffusion paradigm for coordination. Directed is data-centric that all communication named data. All a directed-diffusion-based network are application aware. This enables achieve energy savings by selecting empirically good paths caching processing data...
Wireless sensor networks promise fine-grain monitoring in a wide variety of environments. Many these environments (e.g., indoor or habitats) can be harsh for wireless communication. From networking perspective, the most basic aspect communication is packet delivery performance: spatio-temporal characteristics loss, and its environmental dependence. These factors will deeply impact performance data acquisition from networks.In this paper, we report on systematic medium-scale (up to sixty...
Previously proposed sensor network data dissemination schemes require periodic low-rate flooding of in order to allow recovery from failure. We consider constructing two kinds multipaths enable energy efficient failure the shortest path between source and sink. Disjoint multipath has been studied literature. propose a novel braided scheme, which results several partially disjoint schemes. find that are viable alternative for energy-efficient isolated patterned failures.
Structural monitoring---the collection and analysis of structural response to ambient or forced excitation--is an important application networked embedded sensing with significant commercial potential. The first generation sensor networks for monitoring are likely be data acquisition systems that collect at a single node centralized processing. In this paper, we discuss the design evaluation wireless network system (called Wisden acquisition. incorporates two novel mechanisms, reliable...
Using detailed traces from 255 users, we conduct a comprehensive study of smartphone use. We characterize intentional user activities -- interactions with the device and applications used impact those on network energy usage. find immense diversity among users. Along all aspects that study, users differ by one or more orders magnitude. For instance, average number per day varies 10 to 200, amount data received 1 1000 MB. This level suggests mechanisms improve experience consumption will be...
Making effective use of the vast amounts data gathered by large-scale sensor networks will require scalable, self-organizing, and energy-efficient dissemination algorithms. Previous work has identified data-centric routing as one such method. In an asso-ciated position paper [23], we argue that a companion method, storage (DCS), is also useful approach. Under DCS, sensed are stored at node determined name associated with data. this paper, describe GHT, Geographic Hash Table system for DCS on...
Mercator is a program that uses hop-limited probes-the same primitive used in traceroute-to infer an Internet map. It informed random address probing to carefully exploring the IP space when determining router adjacencies, source-route capable routers wherever possible enhance fidelity of resulting map, and employs novel mechanisms for resolving aliases (interfaces belonging router). This paper describes design these heuristics our experiences with Mercator, presents some preliminary analysis
In-network data aggregation is essential for wireless sensor networks where energy resources are limited. In a previously proposed dissemination scheme (directed diffusion with opportunistic aggregation), opportunistically aggregated at intermediate nodes on low-latency tree. this paper, we explore and evaluate greedy aggregation, novel approach that adjusts points to increase the amount of path sharing, reducing consumption. Our preliminary results suggest that, under investigated...
In most distributed systems, naming of nodes for low-level communication leverages topological location (such as node addresses) and is independent any application. this paper, we investigate an emerging class systems where does not rely on network location. Rather, based attributes that are external to the topology relevant When combined with dense deployment nodes, kind named data enables in-network processing aggregation, collaborative signal processing, similar problems. These approaches...
Wireless sensor networks promise fine-grain monitoring in a wide variety of environments. Many these environments (e.g., indoor or habitats) can be harsh for wireless communication. From networking perspective, the most basic aspect communication is packet delivery performance: spatio-temporal characteristics loss, and its environmental dependence. These factors will deeply impact performance data acquisition from networks.In this paper, we report on systematic medium-scale (up to sixty...
Many emerging smartphone applications require position information to provide location-based or context-aware services. In these applications, GPS is often preferred over its alternatives such as GSM/WiFi based positioning systems because it known be more accurate. However, extremely power hungry. Hence a common approach periodically duty-cycle GPS. duty-cycling trades-off accuracy for lower energy. A key requirement then, system that provides accurate while spending minimal
C. Faloutsos et al. (see Proc. ACM SIGCOMM, 1999) found that the inter autonomous system (AS) topology exhibits a power-law vertex degree distribution. This result was quite unexpected in networking community and stirred significant interest exploring possible causes of this phenomenon. The work A.-L. Barabasi R. Albert Science, p.509-512, its application to network generation A. Medina MASCOTS, 2001) have explored promising class models yield strict distributions. We re-examine BGP (border...
article Share on Data-centric storage in sensornets Authors: Scott Shenker ICSI, Berkeley, CA CAView Profile , Sylvia Ratnasamy ICSI/UCB, Brad Karp Ramesh Govindan USC Comp. Sci., LA, Deborah Estrin UCLA Authors Info & Claims ACM SIGCOMM Computer Communication ReviewVolume 33Issue 1January 2003 pp 137–142https://doi.org/10.1145/774763.774785Published:01 January 2003Publication History 251citation1,528DownloadsMetricsTotal Citations251Total Downloads1,528Last 12 Months24Last 6 weeks6 Get...
In many sensor networks, data or events are named by attributes. Many of these attributes have scalar values, so one natural way to query interest is use a multi-dimensional range query. An example is: "List all whose temperature lies between 50° and 60°, light levels lie 10 15." Such queries useful for correlating occurring within the network.In this paper, we describe design distributed index that scalably supports queries. Our (or DIM) uses novel geographic embedding classical structure,...
Following the long-held belief that Internet is hierarchical, network topology generators most widely used by research community, Transit-Stub and Tiers, create networks with a deliberately hierarchical structure. However, in 1999 seminal paper Faloutsos et al. revealed Internet's degree distribution power-law. Because distributions produced Tiers are not power-laws, community has largely dismissed them as inadequate proposed new attempt to generate graphs power-law distributions.Contrary...
Geographic routing has been widely hailed as the most promising approach to generally scalable wireless routing. However, correctness of all currently proposed geographic algorithms relies on idealized assumptions about radios and their resulting connectivity graphs. We use testbed measurements show that these are grossly violated by real radios, violations cause persistent failures in routing, even static topologies. Having identified this problem, we then fix it proposing Cross-Link...
The efficacy of data aggregation in sensor networks is a function the degree spatial correlation sensed phenomenon. While several (i.e., routing with compression) techniques have been proposed literature, an understanding performance various schemes across range correlations lacking. We analyze compression wireless using application-independent measure (an empirically obtained approximation for joint entropy sources as distance between them) to quantify size compressed information, and...
Resource constrained mobile devices need to leverage computation on nearby servers run responsive applications that recognize objects, people, or gestures from real-time video. The two key questions impact performance are what offload, and how structure the parallelism across device server. To answer these questions, we develop evaluate three interactive perceptual applications. We find offloading choices should be dynamic, even for a given application, as depends scene complexity well...
The ubiquity of smartphones and their on-board sensing capabilities motivates crowd-sensing, a capability that harnesses the power crowds to collect sensor data from large number mobile phone users. Unlike previous work on wireless sensing, crowd-sensing poses several novel requirements: support for humans-in-the-loop trigger actions or review results, need incentives, as well privacy security. Beyond existing crowd-sourcing systems, exploits processing devices. In this paper, we design...
Mobile app ecosystems have experienced tremendous growth in the last six years. This has triggered research on dynamic analysis of performance, security, and correctness properties mobile apps ecosystem. Exploration execution using automated UI actions emerged as an important tool for this research. However, existing largely developed analysis-specific automation techniques, wherein logic exploring is intertwined with analyzing properties. PUMA a programmable framework that separates these...
Various sensor network measurement studies have reported instances of transient faults in readings. In this work, we seek to answer a simple question: How often are such observed real deployments? We focus on three types faults, caused by faulty readings that appear abnormal. To understand the prevalence first explore and characterize four qualitatively different classes fault detection methods. Rule-based methods leverage domain knowledge develop heuristic rules for detecting identifying...