This paper studies the use of optical feeder links in very high throughput satellites (VHTS) networks with emphasis on gateway diversity techniques to mitigate the inherent propagation losses in optical frequencies. Focusing on a GEO scenario, the paper considers a system-wide approach investigating various challenges of optical feeder links. These include transmission schemes amenable for transparent on-board processing, optical channel models taking into account blockage by clouds and fading caused by atmospheric turbulence in addition to complexity of on-board and on-ground processing. The channel models are then used to dimension the ground segment towards ensuring a given availability percentage (e.g., 99.9%). The channel model and payload complexity further influence the choice of link layer techniques used for counteracting fading due to atmospheric turbulence in the absence of blockage. An elaborate end-to-end simulator incorporating the proposed channel models capturing the nuances of various processing blocks like optical-electrical conversion is developed. The system performance results provide interesting insights and a framework for assessing the feasibility and advantages of optical feeder links in VHTS systems.
K E Y W O R D Sanalog and digital transparent payloads, channel models, GEO networks, microdiversity and macrodiversity, optical feeder links, optical ground station network selection, optical received time series, optical SNR
| INTRODUCTIONA key challenge towards future generation very high throughput satellites (VHTS) is the limited spectrum of about 2 GHz available in the Ka band (20/30 GHz). A potential solution for resolving this issue could be to move the feeder links from the Ka band to the Q/V band (40/50 GHz) 1-4 or even to the W band (70/80 GHz) 5 where bandwidths of up to 5 GHz are available. While this allows for the use of radio frequency (RF) domain expertise, the bandwidth available is limited compared with requirement estimates and the spectrum is also solicited in the emerging 5G paradigm. 6 Another revolutionary solution is to move the feeder link from RF frequencies to optical frequencies. [7][8][9] Both these approaches are challenging due to the attenuation by atmospheric phenomena (e.g., rain and clouds) whose severity increases with the frequency. In both cases, a network of multiple gateways with smart switching capabilities is thus envisaged. 10,11 While RF at high frequencies is less impaired than the optical counterpart, larger number of gateways is nonetheless required to achieve very high throughputs due to the limited available bandwidth per