Title:  Proactive methods to maximize mmWave WLAN performance

Committee: 

Dr. Blough, Advisor      

Dr. Sundaresan, Chair

Dr. Owen

Abstract: The objective of the proposed research is to develop proactive methods to extend millimeter wave (mmWave) signal availability in WLAN to avoid significant drops in performance caused by signal blockage. To accommodate rapidly increasing application bandwidth demands, mmWave WLAN has been identified as a promising technology with the potential to achieve Gbps throughput. However, the small wavelength of mmWave brings the innate problem of having both high penetration loss and weak diffraction abilities around objects. Directional high gain antennas are therefore adopted for mmWave communication, making mmWave extremely sensitive to any blockage that lies between the access point (AP) and user equipment (UE). Therefore, in this proposal, we focus on using proactive approaches to mitigate potential network outage caused by dynamic blockages. Our approach thus far consists of two parts. The first part considers maximizing signal coverage in the the spatial aspect in the interior planning phase after router deployment, and the second part studies the possibility of temporally maximizing best case transmission conditions with proactive scheduling approaches using predicted blockage information. For the first part, we study the use of dedicated flat passive reflectors to improve coverage in indoor mmWave WLANs through a reflector placement scheme that accommodates any general indoor scenario with pre-deployed ceiling-mounted APs.For the second part, we first formulate and solve an optimal scheduling problem in order to investigate the potential performance improvements of proactive schedulers that make use of blockage prediction. For future work, we propose to follow up the scheduling problem with an efficient approximate or heuristic algorithm to give a practical solution to the proactive scheduling problem. Also, we propose an integrated approach to combine mobility prediction and known indoor environments to produce longer term blockage predictions, which could be used in proactive scheduling algorithm to provide more temporal freedom or help with other proactive resource allocation problem such as handovers.