Ground Application System Design for Future Q/V-band Very High-Throughput Satellite

LI Xinhua, LI Jinlin, WANG Xinrong, ZHANG Lei Space Star Technology Co. Ltd., CAST, Beijing 100094

Abstract: Over the next decade,Very High-Throughput Satellite (VHTS) will bring enough capacity to serve high speed internet and in-flight connectivity markets at scale, offering fiber-like services both in terms of price and speed.In this context, a generic VHTS mission utilizing Q/V-band feeder links and Ka-band user links is described. However,the rain attenuation on the feeder links becomes a limiting issue because of the higher frequencies. Toward this, an exploitation of multiple gateways (GWs) as a transmit diversity measure for overcoming severe propagation effects are being considered. Ground Application System (GAS) design of VHTS is illustrated including N+P GWs (N active and P redundant GWs) diversity, frequency plan, link budget and system capacity.

Key words: ground application system, Q/V-band, VHTS, gateway diversity, N+P scheme

1 GENERAL TREND OF GEO-HTS DEVELOPMENT

At present, personal internet access, enterprise data transmission, 3G/4G base station data backhaul, aircraft communications, maritime communications, military communications and other communications have demonstrated significant demand for high-throughput satellites. The application scenarios are becoming more and more extensive. Through technological innovation of high-throughput satellites, market applications will continue to develop.

In order to meet the fast growing internet access demands,satellite systems need to reduce their cost/bit to be comparable with terrestrial systems. Terrestrial systems are unable to satisfy all demands in all geographical areas and thus broadband by satellite is a key potential service provision platform. Current satellite systems' capacities vary from 10 Gbps to 300 Gbps. For example, Spaceway-3 offers a throughput of 10 Gbps, KaSAT of 70 Gbps, Viasat-I of 140 Gbps and Viasat-II of 300 Gbps.Considering the future traffic demands, the raw capacity should approach Terabit/s by 2020 to meet these demands. So, the challenge is to reduce the cost/bit and to increase the capacity from today's 300 Gbps to 1 Tbps by 2020[1].

A key challenge to achieve a Terabit/s broadband satellite communications (SatCom) system is the limited spectrum of about 2 GHz available in the current Ka-band. Following the traditional trend, this can be tackled by the gradual shift to a higher frequency band whenever the appropriate technology is mature enough. Therefore, an attractive solution for resolving the issue is moving the feeder link from the Ka-band to the Q/V-band (40/50 GHz) where larger bandwidths, up to 5 GHz,are available[2]. Further, this move can free up the whole Kaband spectrum for the user link. This is a very interesting solution for satellite operators since the feeder link requires almost the same spectrum as the user link, but it does not provide any direct revenue. By moving the feeder link to the unused spectrum, satellite operators can use the freed bandwidth for commercial purposes. Moreover, it allows locating gateways (GWs)within the service area minimizing the interference between the feeder link and user link.

2 CURRENT SITUATION OF Q/V-BAND HTS

The available Q/V frequency band for satellite communications from ITU is shown in Table 1. The Q/V-band is the most suitable frequency band for millimeter wave frequencies (30 -300 GHz) to carry out the satellite communications business.The communications payload for this frequency band has gradually entered into the commercial satellite market.

Eutelsat Communications and Space Systems Loral (SSL)have successfully carried out transmissions at Extremely High Frequencies (EHF). The experimental payload was flown into space on the Eutelsat 65 West A Satellite.

In May 2016, the two companies analyzed the potential of the Q/V-band (40 - 50 GHz) as an enabler of future Terabit satellite broadband programs. The collected data will help steer the design of adaptive techniques and hub architectures that will shape the blueprint of future broadband communication systems. Extremely High Frequencies stand to enhance the performance of the next generation of High Throughput Satellite program. By offloading backhaul links between a satellite and its hubs from the Ka-band to the Q/V significantly more bandwidth can be made available for users and the number of hubs can be reduced, helping drive down the cost per bit. For SSL, the Q/V-band demonstration is a step forward on its roadmap for implementing new technologies that will support next generation telecommunications solutions.

Table 1 Q/V-band frequency allocation for fixed satellite services

Figure 1 EUTELSAT 65 West A satellite launched in 2016

On April 5, 2018, Eutelsat Communications announced the order of a next-generation VHTS satellite system named KONNECT VHTS utilizing Q/V-band feeder links and Ka-band user links, to support the development of its European fixed broadband and in-flight connectivity businesses. The satellite, which is due to enter into service in 2021, will be built by Thales Alenia Space which will develop a satellite and ground segment solution that is the most competitive on the market today. With a weight of 6.3 ton and a Ka-band capacity of 500 Gbps, KONNECT VHTS will embark the most powerful on-board digital processor ever put in orbit, offering capacity allocation flexibility,optimal spectrum use, and progressive ground network deployment. The project will be launched with firm multi-year distribution commitments from Orange and Thales, two key European players and leaders in their businesses.

3 ARCHITECTURE OF A VHTS SYSTEM

Figure 2 shows the framework of the generic of VHTS scenario with feeder links in Q/V-band and user links in Ka-band .

Figure 2 The VHTS system overall framework

Table 2 V-band attenuation components for several Chinese sites

· VHTS satellite

Bent pipe payloads are deployed VHTS utilizing Q/V-band feeder links and Ka-band user links.

· Gateway

A number of GWs are interconnected via terrestrial links.Each gateway station consists of antenna & RF subsystem, baseband processing subsystem, routing and switching subsystem,and equipment monitoring subsystem.

· User station

A user station is served by just one gateway at a time and switches to a redundant gateway when the signal becomes faded. User stations in a different subnet communicate through affiliated gateway transfer.

· Network Operation Control Center (NOCC)

NOCC is the management and control center of the ground system, having the business support and operation support functions, GW network management function and, equipment monitoring management function. In addition, NOCC performs the gateway switching operation and optimizes the handover process.

4 GROUND APPLICATION SYSTEM DESIGN

4.1 Gateway Diversity Design

4.1.1 Rain fade analysis

VHTS GAS is thus most vulnerable to rain over the gateways, as the loss of gateway communications affects all the users who are connected through it. Services such as teleconference and telemedicine which are subject to high availability requirements are more susceptible. V-band attenuation components for several Chinese sites (satellite position: 125°E) is depicted in Table 2. These highlight that the extra power margins necessary to achieve typical feeder link availabilities of 99.9% would be unbearable at V-band.

In Figure 3, an example of the Cumulative Distribution Function (CDF) of total atmospheric attenuations at 40/50 GHz for an Earth station located in Beijing is illustrated.

Figure 3 CDF of total atmospheric attenuation for a GW Earth station located in Beijing

From Figure 3, it can be seen that in order to provide an availability of 99.9%, a margin of 40 dB would be required.Uplink Power Control and Adaptive Coding Modulation (ACM)techniques are unable to cope the high degradations, thereby motivating the use of multiple GWs for transmit diversity to achieve the required availability on the feeder link.

4.1.2 Gateway diversity scheme

The traditional single site diversity (1:1) scheme, where one GW is supported by another redundant GW, can be an acceptable solution for low/medium throughput systems. On the other hand, for high capacity satellite systems where tens of GWs are required, it is not efficient to use the traditional approach.For single site diversity concepts high investment into ground infrastructure is necessary, however, simple payload solutions on-board the satellite can be employed and routing schemes remain easy.

An interesting GW transmit diversity technique is the N+P diversity scheme, which was studied by D. Mignolo[3]. In this scheme, there are N active GWs and P redundant or idle GWs.When there is an outage at one of the active GWs, switching occurs and traffic of the active GW is rerouted to one of the idle GWs. N+P concepts may be regarded as the next level of complexity in terms of payload and routing.

Smart GW diversity is another technique which was first studied by H. Skinnemoen for Ka-band[4], and has been further studied and developed for Q/V-band by M. Muhammad[5]. In this scheme, a User Terminal (UT) beam is connected not just to one gateway but to a pool of gateways. If a gateway experiences very deep fading, then its traffic can be routed to another gateway that has available capacity. The benefit of this scheme is that all N GWs are active and there is no need for redundant GWs but its disadvantage is that the throughput of users served by GWs sharing the traffic from their affected counterpart will be reduced or each GW should have some spare capacity available in order to support other GWs in case of outage； hence GWs need to be oversized in capacity. Also, the complexity and the on board equipment is further increased[6].

In this paper, we will focus on the N+P scheme and present our contributions for this scenario.

4.1.3 The availability for N+P diversity

In the N+P diversity scheme, the user beam traffic is supported by N operational gateways, each user beam being served by a single gateway, while P is the number of back-up gateways. When an active gateway link experiences rain fading,one of the redundant gateways replaces it. In Figure 4, the principle of this scheme is illustrated. In a system where only one redundant station exists, if two of the gateways-hubs become unavailable, then only the UT beams from one gateway will be served by the redundant gateway.

This scheme takes advantage of the statistical decorrelation of the gateway rain events. The further the distance between gateways, the lower the probability that two or more links are simultaneously experience fading, that is, spatial diversity is more efficient. The gateways are interconnected through fiber optic to the NOCC that performs the gateway switching operation and optimizes the handover process.

For the N+P scheme, a methodology to evaluate the feeder link availability was studied by A. Kyrgiazos[6], which assumes equal outage probabilities p for each gateway.

According to[6], P(kgatewaystooutage)approaches the binomial distribution.

P(1gatewaytooutage)means that only one gateway is in outage while the rest are operational. P(kgatewaytooutage)means that only k gateway is in outage while the rest are operational.

The probability of a user inside a spot beam to be in outage(ideal UT link) is the probability of the gateway that serves it to experience the outage (or deep fading), plus the probability to belong to one of the k gateways that experience outage.[6]

The availabilities that can be achieved with this configuration are subject to the network size and the number of redundant gateways.

From Figure 5, it can be seen that for a gateway network of 7 or 10 stations plus 1 extra, when a gateway is available for the 99.0% of the time, the availability for a user is 99.96%or 99.95% of the time (ideal UT link) respectively. Assuming,that one of GWs is located in Beijing, 99.0% time availability of the feeder links means 18.76 dB total (rain + atmospheric)attenuation. In Figure. 6, the availability for the user versus the feeder link outage is shown for a gateway network of 20 active GW plus 0, 1, and 2 redundant. In the case of 0 redundant, the availability for the user equals the availability of a GW. When more redundant GWs are added, the availability improves significantly, as shown in Table 3.

Figure 4 N+P diversity principle

Figure 5 Beam outage probability vs gateway outage for N+P diversity, 7 and 10 gateways network,1 or 2 redundant

From Figure 7, it can be seen that for a network of 30 gateways, the addition of 2 redundant gateways results in 99.9%time availability of the feeder links, when each gateway is available operationally 98% of time. For an earth station in Beijing,98% can be interpreted as 13.31 dB total (rain + gas + cloud)attenuation.

Figure 6 Beam outage probability vs gateway outage for N+P diversity, 20 gateways network, 0, 1, or 2 redundant

Table 3 Computation of feeder link availability (N=20)

Figure 7 Beam outage probability vs gateway outage for N+P diversity, 20, 30 and 40 gateways network, 2 or 3 redundant

4.1.4 Gateway switching process

The re-routing of traffic or the handover process is of paramount importance and it is a fundamental feature of N+P scheme. Figure 8 illustrates the switching strategy in detail and the different steps are explained in detail below.

· SNR Acquisition: In first step, NOCC collects the forward link SNR of all GWs, both active and idle. The forward link SNR of each GW is acquired by receiving outbound carrier at a user station co-addressed with GW.

·

GW Sorting: After collecting all SNRs, the NOCC sorts the active and the idle GWs based on their SNR in decreasing order (this is same as sorting the GWs based upon their rain attenuation in increasing order).The mth largest SNR amongst the active GWs and the corresponding GW index are denoted by βmand Gm,respectively. Similarly, for the idle GW, the kth largest SNR and the corresponding GW index are depicted by βkand, respectively.

Figure 8 Flowchart of the N+P gateway switching scheme

· Two (2) GW Pairing: After the sorting step, NOCC initiates pairing the active Gkand idle Gk', where k=1,2,…P and k'=N-k+1. Thus, P switch pairs will be forced such that the weakest active GW, GNwill have the best chance to switch to the strongest idle GW, G1.

· Two (2) GW Switching: The switching will take place based on a scheme similar to the modified switch and stay combining (MSSC) technique introduced for two GW in [7]. Based on this switching method, if βk'is lower than βthand βkis higher than βth, switching occurs between two GWs. Here βthis switching threshold and its selection is introduced in [8].

4.2 Frequency Plan

An overview of the generic VHTS scenario with feeder links in Q/V-band and user links in Ka-band making use of a 4 color frequency/polarization reuse method can be found below.

User Link:

· 160 Ka-band spot beams.

· 4 frequency/polarization reuse.

They turned to the east and fell to bowing till their bills touched the ground, but, oh horror--the magic word was quite forgotten, and however often the Caliph bowed and however touchingly46 his Vizier cried Mu

· Uplink: 250 MHz circular polarization/beam (illustrated in Figure 9).

· Downlink: 500 MHz circular polarization/beam (illustrated in Figure 10).

· Terminals are able to tune to more than one carrier simultaneously.Feeder Link:

· N = 10 + P GWs, where P shall be minimized.

· Diversity concept.

· Uplink: 8 GHz of bandwidth per GW (4 GHz per polarization, illustrated in Figure 11).

4.3 Link Analysis

Figure 9 Frequency plan for user uplink

The reference return link budget for clear sky user uplink and feeder downlink is illustrated in [9]. Ka-band user uplink and Q-band feeder downlink is assumed. The assumed user carrier bandwidth is 10 MHz, and the transponder bandwidth for the feeder downlink is 500 MHz, thus, the feeder downlink non-linear power amplifier handles 50 carriers. Although guard bands are not assumed, the adjacent channel interference is considered to be negligible. Concerning the link budget, it is important to note that the clear sky feeder downlink is stronger than the clear sky user uplink by 7.2 dB[9]. The total return link is mainly dominated by the user uplink in clear sky, and 16-QAM modulation and Turbo coding 3/4 is available.

Figure 10 Frequency plan for user downlink

Figure 11 Frequency plan for V-Ka forward link

Figure 12 Frequency plan for Ka-Q return link

This paper focuses on the forward link. Since the described system scenario is generic and the number of parameters to be tuned is vast, the explanatory value lies in the relative comparison of the results rather than in their magnitude. However, if not stated otherwise, the following assumptions apply in addition to the parameters listed for user link, feeder link and network in section 4.2.

· N=10 GWs, each serving 16 beams and P=1 backup GW.

· Considering 5 carriers per beam.

· All user terminals feature 75 cm dish antenna at an efficiency of 66%.

· 10% carrier spacing and 5% roll-off apply.

· DVB-S2x is adopted by the forward link.

· It is considered that fading events on feeder links are de-correlated between different GWs and that fading events in one spot beam are fully correlated.

· On a rainy day, GW in Beijing uplink availability is 99.0%and the user in Beijing downlink availability is 99.5%.

· A 7.3 m Q/V-band parabolic antenna and 250 W TWTA shall be deployed at GW in Beijing.

· The satellite G/T towards GW is 27.5 dB/K, satellite saturation EIRP towards user terminal is 67.1 dBW.

A clear sky link budget calculation considering 5 carriers per beam is provided in Table 4. It can be observed that the link can be closed at a total efficiency of 2.31 bits/s/Hz including spacing and roll-off.

A rain fade link budget calculation considering 5 carriers per beam is provided in Table 5. It can be observed that the link can be closed at GW uplink availability 99.0% and user downlink availability 99.5%. For a network of 10 gateways, the addition of 1 redundant results in 99.95% time availability of the feeder links,when each gateway is operational 99% of the time. The total forward link availability of a network of 10+1 gateways is equal to GW uplink availability (99.95%) times user downlink availability(99.5%). Thus, the total forward link availability is 99.45%.

Table 4 (a) V-band clear sky feeder uplink

Table 4 (b) Transponder operating point

Table 4 (c) Ka-band clear sky user downlink

Table 4 (d) Forward link margin in clear sky

Table 5 Rain fade link budget

4.4 System Capacity

It can be observed from link analysis that the forward link can be closed at a total efficiency of 2.31 bits/s/Hz including spacing and roll-off. In the same way, the return link can be closed at a total efficiency of 2.27 bits/s/Hz including spacing and roll-off. The total throughput of the assumed VHTS is about 248 Gbps, which is shown in Table 6.

Table 6 VHTS throughput

5 CONCLUSION

To cope with demand of the modern information hungry society for growing data-rates and ubiquitous coverage, satellite communications system performance and related cost effectiveness need to improve constantly. This can be done by increasing the efficiency of bandwidth utilization up to certain limits or by resorting to frequency bands providing larger allocations of the precious frequency resource for the service envisaged. In modern VHTS systems, this can be achieved with the utilization of the Q/V-band for the feeder links. However, the Q/V-band feeder link is sensitive to atmospheric impairments. This paper described operational concepts using N active and P redundant GW diversity technique applied to a generic VHTS mission for rain fade mitigation. Performance evaluations in terms of simulation results were provided comparing the availability of the feeder links with some redundant gateways added to the same network. For a network of 10 gateways, the addition of 1 redundant gateway results in an availability of 99.95% of time of the feeder links, when each gateway is operational 99% of time. For a GW in Beijing, with an availability of 99% of time it can be interpreted as an attenuation of 18.76 dB total (rain + atmospheric). In addition, GWs switching scheme, frequency plan, link analysis and throughput of a generic VHTS were proposed.