Development of Channel Models:
In the millimeter-wave regime, signals propagate poorly in non-line-of-sight conditions. In particular, mmWave signals suffer from shadowing due to foliage or the interposition of the user's body between the transmitter and the receiver, and the path loss over a given distance is more severe than it is at lower frequencies. As a result, links will be shorter, leading to smaller cell sizes and in turn even denser networks. Measurement data and channel models are almost non-existent for these frequencies. Development of Communication Protocols:To date there are only a handful of protocols that were developed for mmWave communications, including, IEEE 802.11ad and IEEE 802.15.3c for short distance (few meters) and very high speed links. These standards specify operation in 60 GHz band and mainly include indoor scenarios. In September 2014, the IEEE 802.11 working group has formed a new study group (NG60) to develop revisions to IEEE 802.11ad for increased indoor capacity and to cover outdoor backhaul scenarios. More recent demonstrations by manufacturers like Ericsson and Samsung suggest the development of next generation cellular systems to operate in the 28 and 38 GHz bands for outdoor and dense urban deployments. It is clear that current protocols cannot be used as is at higher frequencies and detailed system simulation models with accurate models characterizing propagation in the mmWave bands are needed to enable the rapid development of novel protocols for mmWave communications in the 28, 38, and 60 GHz bands.
Millimeter-wave communications systems are a natural fit for fifth generation mobile networks. Specifically, systems that combine mmWaves, massive MIMO and ultra-dense networks marry new spectrum with greater spectral and spatial efficiency and promise a long-term solution for the spectrum crunch. mmWave antennas are also small enough to support electronically steerable antenna arrays not only at the base stations, but also in the handsets, the larger bandwidths are well-suited to support broadband device-to-device communications in meshed networks, and the shorter wavelengths lead naturally to small cell sizes. However, technology challenges are greater at mmWave frequencies, and bringing mmWaves to the marketplace will require not only understanding channel and network models, but addressing problems inherent in new hardware development. The problems which fall within the scope of this project are described below.
Development of Channel Models:
We have implemented an 83 GHz channel sounder and have the hardware in house to complete a 28 GHz channel sounder and construct initial channel models. However, there is also interest from industry in 38 GHz bands. We plan to extend our system and channel models to that frequency range. This project has shown us the difficulties involved in hardware development at mmWave frequencies. Even so, there are still be a number of real-world problems to address before mmWave systems can be deployed at the frequencies of current industrial interest; for example, extending our models to examine signals beyond 100 GHz appears to be of interest to industry, which is already exploring hardware at 140 GHz.