**5. Conclusions**

New mmWave technologies, including E-band, multi-band boosters, and LOS-MIMO, have been recently introduced to meet the global demand on microwave backhaul capacity and 5G network build-out. The large bandwidth available at mmWave frequencies in the 30-300 GHz range will enable very high connection speed and capacity. With large available bandwidth, E-bands (71–76 and 81–86 GHz) are considered a strategic solution for 5G heterogeneous networks. While E-band links are generally used up to 3 km, a multi-band technology combining E-band with traditional lower frequency bands is suggested for longer distance deployment. Supporting transmission of multiple data streams simultaneously, LOS-MIMO is the latest wireless backhaul technology to significantly increase the capacity of short-range, point-to-point, mmWave line-of-sight connections. In this paper, these new wireless backhaul technologies are studied using outdoor test links deployed in the same region. The trial was performed in Gothenburgh, Sweden, over the Ericsson premises in 2017.

Real time path attenuation caused by changing weather conditions were monitored and recorded for a 32 GHz 2 × 2 LOS-MIMO link and a 38 GHz SISO link over 6.87 km, as well as an E-band microwave backhaul link over 3 km. The measurement records showed that the rain-induced attenuation of all the three test links are closely related to the variation of the rain rate, with an average correlation value greater than 0.8. The 32 GHz LOS-MIMO link and 38 GHz link were deployed in the same location; the rain attenuation was similar for the two links, but the LOS-MIMO has greater capacity compared to a SISO link. While the 32 and 38 GHz links were built over a much longer distance than the E-band link, but they were less a ffected by atmospheric attenuation. For light rainfall, the di fference in the signal attenuation observed in three test links is less significant. As the rain intensity increases, over a relatively short deployment of 3 km, for the 82 GHz link, the rain attenuation could be greater than 40 dB for a heavy rain event. The high rain attenuation restricts the use of E-band links over longer distances. While E-band backhaul links can achieve high throughput, the coverage is limited; lower frequency links are more robust and can be deployed over much longer distances. Multi-band booster technology pairs a higher frequency link with a lower frequency link, meaning the capacity and coverage requirements for the next-generation microwave backhaul links can be met compared to a single frequency microwave link.

Accurate rain monitoring of precipitation is of grea<sup>t</sup> importance to many applications, including meteorology, hydrology, agriculture, and flood monitoring. Microwave backhaul link is considered as a new tool for near-ground rainfall monitoring. We examined the accuracy of using these new mmWave backhaul technologies for rain rate estimation. Additional attenuation due to e ffects of water film on the antenna surface and other atmospheric conditions, such as humidity, needs to be considered for rain rate estimation for improved accuracy. In real deployments, the measured rain attenuation of the test links is found to be 1–1.75 dB higher than calculated rain attenuation based on the ITU model. We have applied data post-processing and attenuation correction to the received signal level measurement. The derived rain rate from all the links have been shown to be very good compared to the rain rate recorded by the weather stations located in the measurement site. These additional weather data obtained from commercial cellular networks will be particularly useful for big data analysis. Furthermore, mmWave backhauls are expected to be widely used for 5G and smart city networks in cities and densely populated areas, and there is a grea<sup>t</sup> potential to use these links for precipitation and flood monitoring in urban areas.

**Author Contributions:** Conceptualization, C.H.; methodology, C.H.; software, C.H.; validation, G.S., H.W.; formal analysis, C.H.; investigation, J.H.; resources, C.H.; data curation, Q.G.; writing—original draft preparation, C.H.; writing—review and editing, C.H.; visualization, C.H.; supervision, C.H.; project administration, J.H.; funding acquisition, C.H. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by National Natural Science Foundation of China, gran<sup>t</sup> number 41605122, 41775032; Chinese Academy of Sciences President's International Fellowship Initiative for Visiting Scientists, gran<sup>t</sup> number 2018VTA0013.

**Acknowledgments:** The authors would like to thank Ericsson gratefully for giving them access to data. The authors would also like to thank Hagit Messer and Pinhas Alpert's research groups for helpful advice.

**Conflicts of Interest:** The authors declare no conflict of interest.
