- Green IT and Sustainability
- Caching and Content Delivery
- Network Traffic and Congestion Control
- Wireless Communication Networks Research
- Advanced MIMO Systems Optimization
- Advanced Wireless Network Optimization
- Privacy-Preserving Technologies in Data
- Wireless Networks and Protocols
- Vehicular Ad Hoc Networks (VANETs)
- Mobile Ad Hoc Networks
- Cooperative Communication and Network Coding
- Speech and Audio Processing
- Internet Traffic Analysis and Secure E-voting
- Opportunistic and Delay-Tolerant Networks
- Software-Defined Networks and 5G
- Speech Recognition and Synthesis
- Embedded Systems Design Techniques
- Spam and Phishing Detection
- Blockchain Technology Applications and Security
- Music and Audio Processing
- Human Mobility and Location-Based Analysis
- Vehicle License Plate Recognition
- Energy Harvesting in Wireless Networks
- Parallel Computing and Optimization Techniques
- Interconnection Networks and Systems
Hanyang University
2011-2022
Carnegie Mellon University
2012
University of California, Los Angeles
2011-2012
UCLA Health
2007-2011
This paper describes an unconventional way to apply wireless networking in emerging technologies. It makes the case for using a two-tier hybrid wireless/wired architecture interconnect hundreds thousands of cores chip multiprocessors (CMPs), where current technologies face severe scaling limitations excessive latency, long wiring, and complex layout. We propose recursive structure called WCube that features single transmit antenna multiple receive antennas at each micro router offers...
Base stations (BSes) in the 3G cellular network are not energy proportional with respect to their carried traffic load. Our mea- surements show that exhibits high fluctuations both time and over space, thus incurring waste. In this paper, we propose a profile-based approach green infrastruc- ture. We profile BS approximate network-wide proportionality using non-load-adaptive BSes. The instrument is leverage temporal-spatial diversity node deployment heterogeneity, power off under-utilized...
Vehicular ad hoc networks (VANETs) are envisioned to provide us with numerous interesting services in the near future. One of most promising applications is dissemination commercial advertisements via car-to-car communication. However, due non-cooperative behavior selfish nodes or even malicious ones real-world scenario, such vehicular advertisement system cannot be realized unless proper incentives and security mechanisms taken into consideration. This paper presents Signature-Seeking Drive...
This paper introduces a promising application over vehicular ad hoc networks (VANETs), where advertisers use VANETs to disseminate their commercial ads via car-to-car communication, targeting large number of potential customers inside cars. However, due the noncooperative behavior selfish or even malicious nodes in real-world scenarios, such advertisement system cannot be realized unless proper incentives and security mechanisms are place. presents signature-seeking drive (SSD), which is...
Connected vehicles are at risk of exposing their location history when using potentially untrusted location-based services (LBSs) in the driving process. We propose a method called mutually obfuscating paths (MOP) that enables to provide highly accurate realtime updates LBS while preventing from tracking vehicles. The instrument is leverage connected vehicles' two network interfaces: in-car LTE Internet (for accessing LBS) and car-to-car Dedicated short-range communications...
Base station (BS) networks in 3G cellular infrastructure do not consume energy proportion to their carried traffic load. Our measurements show that the exhibits high fluctuations both time and over space, thus incurring waste. In this paper, we propose Green base (GreenBSN) approximate network-wide proportionality using non-load-adaptive BSes. The instrument is a traffic-driven approach. By leveraging inherent temporal-spatial dynamics node deployment heterogeneity, power off under-utilized...
In this paper, we use real experiments to study multiple-input-multiple-output (MIMO) 802.11n rate adaptation (RA) on a programmable access point (AP) platform. Our case shows that existing RA solutions offer much lower throughput than even fixed-rate scheme. It is proven all such algorithms are MIMO-mode oblivious; they do not differentiate spatial diversity and multiplexing modes. We first design MiRA, novel MIMO scheme zigzags between intra- inter-MIMO modes address dynamics. Second,...
We study the problem of scalable monitoring operational 3G wireless networks. Threshold-based performance in large networks is very challenging for two main factors: network scale and dynamics both time spatial domains. A fine-grained threshold setting (e.g., perlocation hourly) incurs prohibitively high management complexity, while a single static fails to capture dynamics, thus resulting unacceptably poor alarm quality (up 70% false/miss rates). In this paper, we propose solution, called...
Rate adaptation is a mechanism unspecified by the IEEE 802.11 standards, yet critical to system performance exploiting multirate capability at physical layer. In this paper, we conduct systematic experimental study on rate over wireless networks. Our key contributions are as follows: First, present critique popular design guidelines adopted many practical algorithms and uncover their limitations. reveals that these seemingly correct can be misleading in practice, thus incurring significant...
We study the challenging problem of strategic content placement in a dynamic MANET. Existing techniques cannot cope with such network dynamics since they are designed for fixed networks. Opportunistic caching approaches insufficient as do not actively manage contents certain goals. In this paper, we present novel management approach called LACMA, which leverages location information available to mobile devices via GPS. The main idea LACMA is bind data geographic (as opposed nodes). This...
With various IoT cameras today, sharing of their video evidences, while benefiting the public, threatens privacy individuals in footage. However, protecting visual without losing authenticity is challenging. The conventional post-process blurring would open door for posterior fabrication, whereas realtime results poor quality, low-frame-rate videos due to limited processing power commodity cameras. This paper presents Pinto, a software-based solution producing privacy-protected,...
This paper addresses the problem of frequency domain packet scheduling (FDPS) incorporating spatial division multiplexing (SDM) multiple input output (MIMO) techniques on 3GPP Long-Term Evolution (LTE) downlink. We impose LTE MIMO constraint selecting only one mode (spatial or transmit diversity) per user transmission time interval (TTI). First, we address optimal selection (multiplexing in each TTI order to maximize proportional fair (PF) criterion adapted additional and domains. prove that...
Existing anti-censorship and anonymizing systems like Tor rely on volunteers who run relays. In reality, the most challenging part, however, is to gather enough beyond altruism. There have been a few incentive schemes for volunteering, but they require external monetary resources or e-cash infrastructure. This paper presents SymBiosis, web-browsing ecosystem that makes case direct benefit trading between censored users uncensored their own resources. The two-way interaction in SymBiosis...
eXpressive Internet Architecture (XIA) [1] is an architecture that natively supports multiple communication types and allows networks to evolve their abstractions functionality accommodate new styles of over time. XIA embeds elegant mechanism for handling unforeseen legacy routers.
This work experimentally studies the energy consumption of multiple-antenna MIMO 802.11 devices. Our measurements reveal an increase in power and speed with number antennas. State art proposals have limitations to save networks. First, they focus on either maximizing or minimizing consumption. Second, only seek minimize for receiver side mobile As a result, present utilize gains infrastructure. To this end, we design Collaborative Energy Save (CMES), which seeks identify transmitter-receiver...
eXpressive Internet Architecture (XIA) [1] is an architecture that natively supports multiple communication types and allows networks to evolve their abstractions functionality accommodate new styles of over time. XIA embeds elegant mechanism for handling unforeseen legacy routers. In this demonstration, we show overcomes three key barriers in network evolution (outlined below) by (1) allowing end-hosts applications start using (e.g., service content) before the them, (2) ensuring upgrading...
The use of smartphones as voice recorders has made it easy to record audios proof conversations, but sharing such audio evidence incurs speech and privacy risks. However, protecting speech/voice without losing authenticity is challenging. conventional post-process redaction conversion recordings, which invalidate their original signatures, make the unverifiable prone tampering. In this paper, we present Overo, an recording/sharing solution that supports processing authenticity. Overo records...
This article shows potential of using DSRC radios for vehicle tracking protection. We focus on traffic data collection as our target application where vehicles send their locations to a server while driving. Such are easily trackable-revealing location history-as the often requires frequent and accurate updates. presents PathCloak, method that enables report preventing from properly vehicles. PathCloak leverages vehicles' two network interfaces: in-car Internet (for accessing server)...
Threshold-based performance monitoring in large 3G networks is very challenging for two main factors: network scale and dynamics both time spatial domains. There exists a fundamental tradeoff between the size of threshold settings alarm quality. In this paper, we propose scalable solution, called threshold-compression that characterizes via intelligent aggregation. The insight behind our solution to identify groups elements with similar behaviors across location dimensions, thus forming...
Threshold-based performance monitoring in large 3G networks is very challenging for two main factors: network scale and dynamics both time spatial domains. There exists a fundamental tradeoff between the size of threshold settings alarm quality. In this paper, we propose scalable solution, called threshold-compression that characterizes via intelligent aggregation. The insight behind our solution to identify groups elements with similar behaviors across location dimensions, thus forming...