Xiaoyang Zhong-CS@PU

EDUCATION

2011.8 ~ 2018.5	PhD in Computer Science, Purdue University
	Research Domain: Internet of Things (IoT) and Wireless Sensor Networks (WSNs). GPA: 3.93
2007.8 ~ 2011.7	BS in Electronic Science and Technology, University of Science and Technology of China (USTC) GPA: 3.45

RESEARCH

Wireless Sensor Networks

WSN Deployment
Developed the core components of a sensor node application for a WSN testbed of 100 nodes running RTOS, which integrated complex functionalities including sampling, routing, wireless reprogramming, compressed sensing, etc.

- The testbed has about 100 nodes of different platform types, including ATmega128 based devices (MicaZ and IRIS) and TI MSP430 based device (TelosB).
- Enabled remote accessing by implementing multi-hop networking based on IEEE 802.15.4. (the underlying technology of ZigBee)
- Solved the MAC layer mismatch problem between heterogeneous node devices using a packet sniffer.
- Everything fits in the tiny nodes of 4 KB RAM (MicaZ) or 48 KB ROM (TelosB).
Downward Routing in IoT/WSNs
Designed and implemented a reliable, energy efficient, and extremely scalable routing protocol for node actuation in large-scale IoT systems with resource-constraint devices.

- Improved the scalability by 5x by developing a Bloom filter based source routing algorithm, leading to the minimization of the packet overhead and the energy overhead.
- Achieved > 98% reliability by implementing opportunistic routing into the packet forwarding process.
- Minimized energy overhead by implementing an efficient method to build the routing table based on upward traffic.
- Implemented in TinyOS/nesC for various sensor platforms and deployed on a real-world WSN testbed.
MobileDeluge
Designed a mobile over-the-air programming tool to assist the node reprogramming process in out-door, heterogeneous, and duty-cycled WSN deployments. Significantly reduced the labor for real-world WSN maintenance.

- Designed a control layer on top of existing approaches to coordinate the behaviors of the mobile base station and target nodes.
- Enabled 2x faster deployment of new applications and reduced the labor for maintaining WSN testbed.
- Implemented in TinyOS/nesC for various sensor platforms;
- Developed PC side gateway in Java.
Network Analysis and Benchmarking
Perform comprehensive network analysis for a real-world WSN testbed and devise a benchmark data suite for the research community.

- Analyzed the network performance and routing dynamics of a real-world WSN testbed from three aspects including link level characteristics, routing level characteristics, and temporal characteristics.
- Demonstrated an effective metric based on entropy theory to measure the amplitude of topological dynamics.
- Devised the first WSN benchmark data suite with full topological information for the research community.
Topology Reconstruction for Dynamic WSNs
Designed a topology reconstruction algorithm based on compressed sensing in cooperation with another student for very dynamic WSNs.

- Reduced packet overhead by implementing a path encoding mechanism using a 4-byte field in each packet.
- Achieved > 96% reconstruction accuracy by implementing a decoding algorithm based on compressed sensing.
- Implemented in TinyOS/nesC for various sensor platforms.
- Implemented PC side decompression algorithm in Python.
Sensor Board and Driver for TelosB (TI MSP430) Platform
Designed a sensor board and driver for TelosB mote to drive external sensors.

- Designed a 2-layered sensor board using Eagle.
- Developed sensor driver based on TinyOS/nesC for TI MSP430 based devices.
- Worked for both analog and digital sensors (UART), e.g., Decagon EC-5 soil moisture sensor, MSP-1 water potential sensor, and MPS-2 digital water potential sensor.
- Solved the clock drifting problem of TelosB during UART communication using a fridge and oscilloscope.
- Reduced the cost of unit sensor board to be less than $10.
Quality of Service Control for the Internet of Things Systems
Designed a distributed QoS control algorithm based on Gur Game to control nodes’ sleep/active states.

- Improved energy efficiency of the whole network for about 30%; achieved fast system convergence.

Internet of Things

Smart Home
- Designed a system using Raspberry Pi/Arduino to control home device with CoAP protocol.
- Developed Python server to send notifications using email (SMTP) and SMS (Twillio).
Greenhouse
- Designed a system using TelosB platform (with Contiki OS) to monitor temperature, humidity, light, and soil moisture for home plants.
- Applied RPL routing protocol to build a low power IPv6 multi-hop network.
- Developed PC side server using Python, which parses the sensor data and sends notification using email (SMTP) and SMS (Twillio).

SELECTED PUBLICATIONS

Xiaoyang Zhong and Yao Liang, Scalable Downward Routing for Wireless Sensor Networks and Internet of Things Actuation. arXiv preprint arXiv:1802.03898, 2018.

G. Villalba, F. Plaza, X. Zhong, T. W. Davis, M. Navarro, Y. Li, T. A. Slater, Y. Liang, and X. Liang, A Networked Sensor System for the Analysis of Plot-Scale Hydrology. Sensors, 2017, 17(3), 636.

Zhong, Xiaoyang and Yao Liang. Raspberry Pi: An Effective Vehicle in Teaching the Internet of Things in Computer Science and Engineering. Electronics (Basel) 2016.

Liu, Rui, Yao Liang, and Xiaoyang Zhong. Monitoring Routing Topology in Dynamic Wireless Sensor Network Systems. ICNP 2015.

Navarro, Miguel, Tyler W. Davis, German Villalba, Yimei Li, Xiaoyang Zhong, Newlyn Erratt, Xu Liang, and Yao Liang. Towards long-term multi-hop WSN deployments for environmental monitoring: an experimental network evaluation. Journal of Sensor and Actuator Networks 3, no. 4 (2014): 297-330.

Zhong, Xiaoyang, Miguel Navarro, German Villalba, Xu Liang, and Yao Liang. MobileDeluge: Mobile Code Dissemination for Wireless Sensor Networks. In Mobile Ad Hoc and Sensor Systems (MASS), 2014 IEEE 11th International Conference on, pp. 363-370. IEEE, 2014. (Code)

IoT TUTORIALS

I worked as TA for several network related courses: Data Communication & Computer Networks (53600, 43600), Wireless Sensor Networks (59000), Internet of Things (49000).

Internet of Things is a relatively new area consisting of many newly developed hardware and software, resulting in insufficient support for beginners. I developed tutorials to help students to get hands-on experiences on IoT-related hardware and software. (some links in the tutorials may be outdated due to the updates in the websites)