Home > Events > Oral Candidacy - Lixing Song

Oral Candidacy - Lixing Song

Start: 5/12/2017 at 10:30AM
End: 5/12/2017 at 2:00PM
Location: 100 Stinson Remick
Attendees: Faculty and staff are welcome to attend the presentation portion of the defense.
Add to calendar:
iCal vCal

Lixing Song
Oral Candidacy
May 12, 2017        10:30 am        100 Stinson Remick
Adviser:  Dr. Aaron Striegel
Dr. Richard Billo        Dr. Christian Poellabauer          Dr. Dong Wang


Active and Passive Techniques for Mobile Network Characterization


WiFi has emerged as a pivotal technology for mobile devices offering the potential for exceptional connectivity speeds. However, due to the dynamics of the wireless channel, the performance of WiFi may vary significantly making it as the bottleneck to decide the Quality of Experience (QoE) of users. WiFi network characterization emerges as a tool to understand the WiFi performance. The key challenge is how to characterize performance in an accurate and efficient manner. Fortunately, with the introduction of frame aggregation from 802.11e, it is interesting to notice that the statistics revealed from frame assembling innately embodies rich set information about network condition. By exploiting frame aggregation as the enabler, this proposal consists of three WiFi characterization approaches with different target metrics and application scenarios. In the first work (Aggregation Intensity based WiFi Characterization), frame aggregation is used as the link congestion indicator to facilitate available bandwidth estimation on WiFi path. Inspired by the fact that the frame aggregation can be impacted by queue length and other AP conditions, the second work (A-MPDU-based AP Load Mechanism) devises a compact AP load measurement scheme. Finally, by manipulating the control packets that are related to frame aggregation, the third work (Passive WiFi Characterization via Scan) proposes an efficient passive client side WiFi environment traffic characterization method. In terms of the degree of completion, I have conducted extensive experiments to evaluate the first two works. The results show significant performance improvement over prior methods. As the proposed work, I will focus on completing the third one as the future work of my Ph.D. study.