Graduate Projects in UT Austin
Research on Acoustic Imaging
AIM: Acoustic Imaging on a Mobile [BEST PAPER AWARD in Proceedings of ACM MobiSys 2018].
AIM: Acoustic Imaging on a Mobile [BEST PAPER AWARD in Proceedings of ACM MobiSys 2018].
|
The popularity of smartphones has grown at an unprecedented rate, which makes smartphone based imaging especially appealing. In this paper, we develop a novel acoustic imaging system using only an off-the-shelf smartphone. It is an attractive alternative to camera based imaging under darkness and obstruction. Our system is based on Synthetic Aperture Radar (SAR). To image an object, a user moves a phone along a predefined trajectory to mimic a virtual sensor array. SAR based imaging poses several new challenges in our context, including strong self and background interference, deviation from the desired trajectory due to hand jitters, and severe speaker/microphone distortion. We address these challenges by developing a 2-stage interference cancellation scheme, a new algorithm to compensate trajectory errors, and an effective method to minimize the impact of signal distortion. We implement a proof- of-concept system on Samsung S7. Our results demonstrate the feasibility and effectiveness of acoustic imaging on a mobile. [pdf] [video] [slides] |
Undergraduate Projects in SJTU
Research on Indoor Localization
Temporal Correlation of RSS Improves Accuracy of Fingerprinting Localization [Infocom 2016]
|
This work mainly focus on studying the temporal correlation effects on RSS fingerprinting based indoor localization method. From theoretical modeling and formula derivation, we found that temporal correlation of signals can definitely correct the localization criteria for MLE method, so that it can significantly improve the fundamental limits like accuracy and reliability. With expected experiment results, the paper was accepted by INFOCOM 2016, TMC 2017.
[pdf] |
Foxconn Cooperation Project
Posiiot Location Based Service System in iOS Development Part
|
We build a indoor positioning system named as "Posiiot". The whole system includes iOS/Android smartphone clients, website and server. I lead my team members to develop an iOS localization application for Foxconn. The iOS development consists of the BLE scanning, Map displaying, Pedometer, Info and Communication components. We also designed and implemented the location determination algorithms, both online k Nearest Neighborhood for Wi-Fi RSS fingerprints and online gradient descent method for Bluetooth RSS. The accuracy of our system can be within 2 meters.
[video] |
|
Ericsson Cooperation Project
Dallas Load Testing Tool to Simulate Real User and Network Behaviors
|
During the second half of last year, I assisted in a cooperation project between IWCT (Institute of Wireless Communication and Technology) andEricsson. I renovated the traffic model and wrote a simulation software by C++ to model the user behavior propagation in 3GPP communication network. In particular, I read tens of thousands of Erlang codes to build the traffic model as a state machine for user activities in WCDMA network, as well as wrote a simulation software to model the user behaviors of the cellular network with error within 0.1%. In addition, I also simulated the traffic packages and user activity transitions in MATLAB to prove its stability.
|
Intelligent Transportation System
Lane-level Localization System Leveraging Sensors in Smartphone
|
We built a crowdsourcing based lane localization system to identify the number of lanes in a road and locate which lane the vehicles are in real time, so that it can facilitate the travel programming of the pilotless automobile and high precision vehicle navigation. The key idea of the system is leveraging the drivers’ smartphones to collect information from embedded GPS module, accelerator and gyroscope sensors, which are then fused by Integrated Multiple Model (IMM), and finally, identify the lane-level locations with constraint k-means clustering algorithm in server side.
|
|