ComSenTer's First Doctoral Graduate

Menglei Zhang became ComSenTer's very first doctoral graduate after defending his PhD thesis at New York University on April 18 in front of an audience of his thesis committees and invited students.

In his thesis "Will TCP work in Low Latency Networks", Zhang evaluated the performance of traditional TCP and new congestion control algorithms in the low latency networks.

Zhang demonstrated how the traditional TCP must be improved to adapt the low latency networks. He also proposed a dynamic receive window scheme that utilizes cross layer information, and a novel way to detect congestion by using linear regression.

Zhang's doctoral thesis built on the following works:

[1] Zhang, Menglei, et al. "Transport layer performance in 5G mmWave cellular." 2016 IEEE Conference on Computer Communications Workshops (INFOCOM WKSHPS). IEEE, 2016.

[2] Zhang, Menglei, et al. "ns-3 implementation of the 3GPP MIMO channel model for frequency spectrum above 6 GHz." Proceedings of the Workshop on ns-3. ACM, 2017.

[3] Zhang, Menglei, et al. "TCP dynamics over mmWave links." 2017 IEEE 18th International Workshop on Signal Processing Advances in Wireless Communications (SPAWC). IEEE, 2017.

[4] Zhang, Menglei, et al. "Will TCP Work in mmWave 5G Cellular Networks?." IEEE Communications Magazine 57.1 (2019): 65-71.

[5] Zhang, Menglei, et al. "Determining a receive window of a receiving device that reduces bufferbloat in a wireless communications system, such as that caused by tcp dynamics over millimeter wave links." U.S. Patent Application No. 15/952,116.

The thesis offers a thorough study of the congestion control protocols in the low latency 5G network. This analysis helps operators to improve their network configuration when the low latency links are deployed.

The most challenging part of this work was building the simulator that models the end-to-end performance of the 5G low latency links. "To fully understand the behavior of applications, we have to incorporate the interact between congestion control protocols and physical layer link variations," Zhang said.

"It took several years for Zhang and other students to build this mmWave module in ns-3. Thanks to this mmWave module, we can be the first one to explore the end-to-end performance in the 5G deployments," said Zhang's doctoral advisor professor Sundeep Rangan proudly. "I am also happy to see a lot of other researchers from both academia and industry are using our simulators."

Zhang has also worked as research assistant in NYU WIRELESS research center since September 2015 which has allowed him to build a strong foundation in channel modelling, networking, and system level simulations.

He says he is now ready for new challenges and will start a research job in Intel after graduation.