In this project we design and validate a transformational approach to Cellular 1000X (i.e., increasing the capacity of cellular networks by 1000-fold), focusing on dense urban environments where today’s cellular networks are facing a capacity crisis. Our architecture is based on outdoor picocells using the 60 GHz unlicensed band, focusing initially on the downlink bottleneck, with standard 4G LTE used for feedback and uplink data.
Figure 1 plots our proposed 60GHz picocell architecture for outdoor mobile broadband, which augments existing LTE networks for significant increase in capacity. Each basestation contains many antenna arrays, each with a large number of elements. Leveraging LTE uplink as the feedback loop, each basestation can track users, form highly directional beams simultaneously to multiple users. The beams follow users through direct or reflected paths, creating high-bandwidth downlink transmissions. Users can beam in reverse directions to build 60GHz uplink, or fall back to LTE uplink.
Collaborators: Prof. Ben Zhao
Year 1: We have focused on measurement-based validation of our basic concept. In particular, we focused on convincing the larger networking community that, while our approach to cellular 1000X is radical, there are no fundamental physical limitations to achieving it. This was done by a series of measurements in controlled as well as natural urban environments, which indicate that there is adequate link budget even when using reflected paths, that blockage can be addressed using reflections and picoclouds, that interference can be significantly reduced by using directional transmission, and that steerable arrays on off-the-shelf 60 GHz IEEE 802.11ad transceivers designed for indoor environments can keep up with pedestrian mobility. These results, together with a preliminary simulation-based interference analysis, were published in MobiCom 2014.
Year 2: We focused on three research items. First, we have been working on quantifying inter-cell interference using a combination of analysis and simulation for 60 GHz picocells deployed in typical urban environments. In particular, we carry out an interference analysis for mm wave picocells in an urban canyon, accounting for the geometry associated with the sparse multipath characteristic of this band. We show that the inter-cell interference caused by the main beam directed at the desired user vanishes after a number of cells determined by the heights of the BS and the user, and the attainable network capacity to be of the order of Terabits/sec per km along the canyon, using 2 GHz of spectrum. Overall, our analysis provides a strong indication of the very large capacity offered by the 60Ghz picocells. Second, we propose and investigate a compressive architecture for estimation and tracking of sparse spatial channels in millimeter (mm) wave picocellular networks. Finally, we develop a 60Ghz mobile imaging algorithm that leveraging 60GHz communication hardware in mobile devices, and capturing signals reflected by the target surface.
Demystifying 60GHz Outdoor Picocells
Yibo Zhu, Zengbin Zhang, Zhinous Marzi, Chris Nelson, Upamanyu Madhow, Ben Y. Zhao, Haitao Zheng
Proceedings of 20th Annual International Conference on Mobile Computing and Networking (MobiCom)
Maui, HI, Sept 2014. [Abstract, Paper: PDF (975KB)]
- Interference Analysis for mm-Wave Picocells
Zhinus Marzi, Upamanyu Madhow, Haitao Zheng
Proceedings of IEEE Globecom
San Diego, CA, Dec 2015. [Paper]
- Compressive channel estimation and tracking for large arrays in mm wave picocells
Zhinus Marzi, Dinesh Ramasamy and Upamanyu Madhow
Submitted for Journal Publication, Under Review. [Paper]
- Reusing 60GHz Radios for Mobile Radar Imaging
Yanzi Zhu, Yibo Zhu, Ben Y. Zhao and Haitao Zheng
Proceedings of 21st Annual International Conference on Mobile Computing and Networking (MobiCom)
Paris, France, Sept 2015 [ Paper ]
- 60GHz Mobile Imaging Radar
Yibo Zhu, Yanzi Zhu, Zengbin Zhang, Ben Y. Zhao and Haitao Zheng
Proceedings of 16th Workshop on Mobile Computing Systems and Applications (HotMobile)
Santa Fe, NM, Feb 2015.[ Paper ]
Graduate students working on this program are exposed to an interdisciplinary mix of concepts from signal processing, communication theory, network protocols and capacity analysis, and measurement-based modeling.
Our systems vision is based on the power of highly directional networking, and includes approaches to adapt highly directional beams subject to hardware constraints such as coarse phase control. We believe that demonstration of the potential gains will stimulate further innovations in design of radio frequency integrated circuits (RFICs) and antenna arrays, and are working actively with circuit designers to make this happen.
We are in close touch with major companies in this space, including Google, Qualcomm, and Samsung, and will strive to use these connections to try to promote technology transfer of the ideas that we develop. Both PIs have been recently awarded Google Faculty Research awards for research in mm wave communication.