Research

ELEMENT: a latency diagnosis framework that decomposes end-to-end TCP latency into endhost and network delays Implement a user-level library that uses ELEMENT to minimize delays

FluidMem: a new approach to memory disaggregation that leverages the user-fault mechanism to achieve full memory disaggregation in software

ECHO: a cloud resource management system that overbooks backup VMs by optimizing the overbooking rate tradeoff between availability and utilization

FLARE: A coordinated HAS solution for the fog computing that optimizes the total utility while maintaining stable video quality and supporting user-/device-specific needs

SNN: a practical machine learning-based relation analysis system, which can be used in different areas SNNCache: leverage SNN to utilize the inter-relationships among sequenced requests in caching decision

Spoofing Alerts in 4G LTE Networks: investigate the details of Wireless Emergency Alert (WEA) protocol and develop and demonstrate the first practical spoofing attack

CASTLE: a fully distributed scheduling framework that jointly optimizes the spectral efficiency and battery consumption

An effective phase noise calibration technique which can be broadly applicable to COTS WiFi based motion sensing

AMUSE: a practical, cost-aware WiFi offloading system that exploits a user’s delay tolerance and offloads satisfying her throughput-delay tradeoffs and data budget constraints

TUBE: an end-to-end system for offering time-dependent pricing to users, offer lower prices in less congested periods, encouraging users to shift some traffic to less congested periods