Blog

January 14, 2018

This paper was given at the Ka and Broadband Communication Conference – 2015 – It is presented here as in the original proceedings’ submission and conference PDF distribution. [proceedings]

Abstract

The introduction of high throughput satellites (HTS), with multi-spot beams, as well as the emergence of LEO and MEO constellations, provide a significant increase in available satellite communication throughput but brings about new challenges in optimally utilizing the satellite bandwidth, beam and power resources in cases of varying demand between beams, non-uniform traffic patterns during the day (peak hours) and the effect of multiple time zones in traffic load distribution. Current and future satellites introduce flexible techniques, such as flexible power allocation, flexible bandwidth allocation and beam-hopping to cope with those challenges.

Of all those techniques, beam-hopping was shown to provide a level of flexibility that makes it possible to increase served traffic, reduce areas of unmet demand while enabling the reduction of power consumption on-board.

In [17] various issues regarding beam-hopping, from the terminal, payload and eco-system point of view, were addressed. In the paper, we discuss and analyze the system considerations for implementation of beam-hopping in a multi-beam environment and the trade-offs required for different applications. These include the effects of beam switching time, synchronization accuracy, revisit-time constraints, waveform constraints, receiver and payload synchronization.

Authors

Avraham Freedman
SatixFy Ltd., 12 Hamada st. Rehovot, Israel 74140, Tel: +972-89393203, Fax: +972-89393223, avi.freedman@satixfy.com
Doron Rainish
SatixFy Ltd., 12 Hamada st. Rehovot, Israel 74140, Tel: +972-89393210, Fax: +972-89393223 , doron.rainish@satixfy.com
Doron Elinav
SatixFy Ltd., 12 Hamada st. Rehovot, Israel 74140, Tel: +972-547479470, Fax: +972-89393223 , doron.elinav@satixfy.com

This post is the first in a three parts series. The following two parts will be published next. you can see part 2 and part 3 (follow links)
Part 1 – is an introduction and abstract and use cases of beam hopping system design.
Part 2 – covers beam hopping basic system consideration and illumination strategies.
Part 3 – covers beam hopping adaptation to demand and summary and conclusion.

1. Introduction

The introduction of high throughput satellites (HTS), with multi-spot beams, provides a significant increase in available satellite throughput but brings about new challenges in optimally utilizing the satellite bandwidth, beam and power resources in cases of varying demand between beams, non-uniform traffic patterns during the day (peak hours) and the effect of multiple time zones in traffic load distribution. Current and future satellites introduce flexible techniques, such as flexible power allocation, flexible bandwidth allocation and beam hopping to cope with those challenges.

Together with the HTS, new low earth orbit (LEO) and medium earth orbit (MEO) constellations are being discussed and introduced in the satellite market. For those types of satellites as well, flexibility could be a key feature to enable cost-effective provision of services to areas with variable demand.

In this paper, we concentrate on Beam Hopping (BH). BH as one of the most flexible techniques, is a technique in which the satellite resource- the transmission beam is shared in time among the users. Unlike conventional TDM, transmission takes place within a directional beam pointing at the destination, either as switching the transmission to a given beam within a bank of fixed multi-spot beams, or by means of a fast-steerable antenna. Obviously, several such transmitters can be installed in the satellite.

Beam Hopping was part of early experimental systems, such as the Advanced Technology Communication Satellite (ACTS) program [1]. In [2] and [3], a beam-hopping architecture in the context of the Teledesic Network is presented. The importance of beam hopping is emphasized in this article, saying “In general, the spot beam downlink architecture makes efficient use of the RF spectrum and satellite resources. First, it permits an increase in system capacity by allowing for frequency reuse through spatial isolation between beams. Second, it makes efficient use of satellite power by focusing the radiated RF power only where it is needed. Finally, the use of a number of beams operating over the entire allocated frequency band, which can hop from one cell to another on a per-packet basis, allows for statistical multiplexing of traffic destined to various destinations in the satellite footprint.”

[4] Describes architectures for beam hopping payload antenna and analyzes the problem of the beam hopping time planning. Results have shown that a capacity gain of 30% can be achieved with respect to a conventional system. An important conclusion from that article is: “… that, if the traffic demand is relatively sparse over the coverage region, the beam hopping system can provide up to a factor 3 gain in capacity with respect to the conventional system”.

[6]-[10] are a some of the publications resulted from a comprehensive study made under the support of the European Space Agency ([11],[12]). [6] compares beam hopping to other flexible payload techniques, such as flexible frequency allocation with and without Multi Port Amplifier (MPA) and flexible TWTA’s. Comparison was made in terms of capacity (reduction of unmet capacity and exceeding capacity and increase the usable capacity) and in terms of DC payload power consumption. The analyses and simulations performed were made against varying demand scenarios over Europe in the years 2010-2020. In all scenarios tested, pure beam -hopping showed advantage over conventional and other flexible payload methods. [7] and [8] refer to the same scenarios, albeit with a different algorithm for resource allocation. Those papers conclude that frequency domain and time domain flexibilities are equivalent in terms of performance, and claim that there is a duality both types of flexibilities, and actual implementation depends on cost and complexity.

[9] and [10] are more introductory showing the results of the study stating: “In general, the beam-hopped payload offers more throughput than the other two (Dynamic Bandwidth Allocation and Flexible Power Allocation) and better meets the traffic demand. The flexibility is limited to the maximum capacity a beam can offer. A beam-hopped payload has intrinsically more potential due to the fact that a beam can access the entire bandwidth in both polarizations.”

The problem of optimal allocation of resources in a beam-hopping system, including power, bandwidth, allocation time as well as gateway resource is also considered in [18], by decomposing the joint bandwidth and power allocation problem into two independent sub-problems.

In order to get some practical, order-of-magnitude number, [13] can be referred to. [13] is the guideline for implementation of the DVB-S2X standard, including the use of superframes for beam-hopping. The document summarizes the advantage of beam hopping, stating:

• Lower DC power consumption (<50%)
• Capacity increase by +15%
• Reduction of the unmet and excess capacity by 20%
• Better flexibility in allocating capacity to the beams with high traffic demand

Combined with antenna design, [15] shows that using beam hopping an advantage can be achieved by designing narrower spot beams, thus increasing gain and reducing interference between beams.

An interesting use of the flexibility beam-hopping provides is described in [16], where a secondary satellite, cognitive of a primary beam-hopping transmission plan, provides significant additional capacity to a served area with minor deterioration of the primary service.

In addition to those advantages, a beam hopping system presents some technical challenges:

• The terminals are required to receive burst transmissions in the forward link, whereas the current terminals for GEO and LEO satellites are mostly designed for continuous reception (always-on Forward Link)
• The beam hopping introduces additional delay, and in some instances, delay jitter.
• The payload should be capable to switch transmissions to the right beam, and, depending on the payload architecture, be synchronized to the gateways
• The beam-hopping time plan, or, more generally, the resource allocation per beam in terms of time, bandwidth, and power, should be correctly planned as to optimize the utilization of those resources and to provide for the required demand effectively.

Despite those challenges, there is a growing number of BH capable satellites designed to be launched in the near future. [17] looks at beam hopping from the point of view of the whole eco-system. It addresses
the technical as well as the non-technical issues that may hinder its adoption in the market, including availability of ground segment equipment, confidence from operators and required standardization to enable a multivendor open market for highly flexible systems.

Beam Hopping is a technique that may apply to a variety of implementations and applications. This paper surveys a number of use cases and presents some design considerations as well as performance that can be expected in those cases.

2. Beam Hopping Systems – Scope of Use Cases

Beam hopping may be applied to a variety of platforms, system architectures, applications, and user types. There are distinctions between:

• Broadcast vs. multi-cast applications
• Continuous vs, packetized data streams
• Fixed vs. mobile users
• High data rate vs. low data rate applications
• Delay tolerant vs. delay sensitive applications

And, regarding the implementation:

• GSO platform vs. LEO/MEO
• Transparent payload (“bent-pipe”) vs. regenerative payload
• Multi-beam switching vs. steerable antennas
• Predefined vs. data-driven illumination

Over a GEO satellite, being a large platform with large coverage area, a variety of applications might be carried, typically by different network operators sharing the bandwidth. The large footprint of a beam, would cover many users, which would mean a high degree of multiplexing, both in the frequency and time domain during the dwell time. The relatively long lifetime of a GEO satellite would typically lead to a design based on a transparent payload, to enable independence on future communication standards.

A smaller LEO/MEO satellite with a smaller footprint would typically be operated by a single operator, and provide a low number of beams. The smaller beam area entails a small number of terminals to be served, hence a low degree of multiplexing. Regenerative payload will be typically preferred as the shorter life time of operation allows the use of state-of the art payload, efficient use of uplink and downlink with smaller risk of being prematurely obsolete.