Towards Providing Network Connectivity via Satellite

Uri Bar-Zemer
Consultant Academic Satellite Gateway Bridgewater College, Providence, USA
E-mail: uri@ids.net


To date there have been several attempts, some more successful than others, to provide links between institutions in this part of the world and their counterparts in Europe and the Units States. By the nature of things, the Moscow area has been better provided than further outlying regions with no less intensive and deserving needs. For the purpose of this gathering I will attempt to outline an approach which ultimately might cater to the needs of a wider dispersion, hopefully within the footprint of the global beams of the Russian satellite system.

Most existing satellite links were installed to address the communications needs of a specific institution, usually they are linked to one or several counterparts on the other end. The examples which come to mind are the link between MSU and Helsinki, the HEP link with DESY in Germany and the IKI (Institute for Space Research)-Brown satellite connection. I am sure there are some I missed, no offense intended.

Given a state of affairs in which technology for the various modes of communication is fairly mature, and that includes all of data, voice and video, the issue becomes one of appropriate system design, and of course that of funding. I will limit my remarks to the options available using satellite communications and perhaps open for later discussion the subjects of hybrid systems incorporating fiber and radio LAN. Although extremely important from a funding perspective, I will only touch briefly on the cost of space segment as it is already the subject of ongoing discussions with the various Russian satellite authorities.

Experience with existing point to point links tends to show that the specific groups for whom the link was installed, benefit greatly from its existence while at the same time, excluding a great number of potential users who do not happen to be on the same fiber ring or otherwise have easy access to the facilities at either end.

Given the highly distributed nature of the user community which this workshop gathered to address, coupled with the outstanding satellite communications resources which are available in this country, and viewed in the context of the relative absence of mature infrastructure outside of major population centers.

Several possible solutions come to mind in the attempt to overcome the sheer size of the territory - eleven time zones, as compared to "only" four in the United states,

These may include: conventional point to point satellite links with feeds into both Local Area Networks and Wide Area Networks.

Very Small aperture Terminals hard wired into LANs or reaching the end user by way of various wireless devices.

Low data rate miniaturized satellite transceivers.

For the purpose of this discussion I will focus on the second and third options as the first is both self explanatory and is already well applied.

The real objective is to be able to land signal on the premises of participating institutions in Russia, and to enter LANs from these sites so as to maximize the utility and the cost effectiveness of the system. The intention is to use the smallest antennas possible, and communication equipment which could operate unattended once installed.

Very Small Aperture Terminals, VSAT for short, use antennas ranging from several centimeters to a maximum of 3.8 meter - Typically, these terminals are installed in close proximity to the end user, require little or no maintenance and cost less to buy, operate and maintain than larger more complex terminals.

In the United States, corporate VSAT systems such as those belonging to some major corporations may number several hundreds to several thousands of 1 meter antennas connected to a corporate or shared hub. In Europe, these systems tend to be smaller and can number from several nodes to several tens of nodes. Data rates can range from very short data bursts to a full T-1, they can operate in SCPC - Single Channel Per Carrier mode, DAMA, Demand assigned Multiple Access or TDMA Time Division Multiple Access. There are several other access scheme on which I will not dwell at this time.

SCPC mode within the confines of VSAT, is essentially analogous to the application of large antennas in that it can be used for point to point interactive traffic or point to multipoint non interactively. Typically it might deploy a small antenna, a low power transceiver located on the antenna and an indoor Monitoring & Control and modem unit. From an equipment perspective, this is a very cost effective way to connect one point to another. However, a system containing multiple sites may become costly on the space segment side of the equation. TDMA and DAMA systems known to most of you from other data applications, assign space by way of polling and by assigning packets to the proper slots respectively. A TDMA systems typically require an elaborate hub terminal where as DAMA systems are capable of managing the network from any assigned terminal and hence are less costly to install albeit limited in network size and efficiency.

The third option, the low data rate satellite transceivers can typically carry a maximum of 600 baud, however, it is extremely easy to install and operate, and if deployed on proprietary bandwidth - is most cost effective. There is little experience in using this equipment for the purposes we are discussing and this application will require testing. The manufacturer of one such system has already modified it to operate on the Gorizont satellites and indicated that he is willing to test applications in the field.

We are currently weighing the different VSAT options and network configurations against satellite services available, and the demand which will be generated by this group and other users.

Once a decision is reached, several demonstration sites are likely to be set up within a relatively short time to prove the concept. Once these have successfully operated for a period, the network will be expanded to the extent that available resources will allow. I should add that the demonstration should be modest in its goals and focus mostly on the transfer of data, but that voice and video applications should be part of the overall design and need to be tested as well.

On the US side, a full featured satellite terminal is scheduled for completion by December of this year. It will be equipped with an 11 meter C band Vertex antenna which will point to Atlantic satellites. A 6.1 meter Ku band antenna will serve to turn the signal around to and from domestic US satellites. This terminal is located 20 miles south of Boston, at Bridgewater State college and will be connected initially by microwave and eventually by fiber to the Woods Hole Oceanographic institute on Cape Cod, to MIT in Boston, and from there to the national fiber grid. A teaching center, television studio, teleconference studio and distance learning classes are part of this facility which will be completed by July of 1995.

The Bridgewater facility is the first of a three-terminal satellite system which is planned for installation in 1994-95. The second terminal will go up before the end of 1994 at a site between Baltimore and Washington DC. Its antenna will also point to the Atlantic part of the satellite arc. This terminal is designed to serve universities and governmental institutions in that region as well as be open to commercial traffic. It will rely on a 9 meter Vertex antenna and will have fiber connectivity to the national grid. The third antenna is scheduled for installation by mid 1995 at San Diego State University on the US West Coast. It too will be a 9 meter Vertex, and will point towards satellites over the Pacific. Bridgewater State College and San Diego State University are currently putting on the finishing touches on a Teaming Agreement which will provide for cross connectivity between the two colleges and by the same token, allow colleges and universities which need the connectivity over both Atlantic and Pacific regions to enter the system in the desired direction. This "distributed terminal system" will allow for full coverage of the United States from Russia and the Republics but will also cover European Russia to the Urals, including the Moscow region Atlantic Satellites and the East Coast of Russia via the Pacific satellites.

We hope that an agreement for a site in Moscow will emerge as a result of this workshop and some additional meetings so that coverage of Central Russia could be achieved by a second satellite hop using a domestic Russian satellite or possibly another system.

There are already several organizations which have indicated strong interest in using the facilities for interactive communication with Russia and East European countries. These include the World Bank, UNet, UNDP, USAID and a large number of colleges and universities which are already engaged in collaborative projects East to West. One would expect to see West European institutions become integrated in this network as it becomes available.

Having laid out the spread of possibilities, I suggest that a series of focused small-group sessions take place in the course of the workshop to discuss which of these options is relevant for each user group, explore in depth the cost factors and in-kind contribution of each side in order to asses the practicality of every proposed link.

One important issue which should dominate the work of each such group ought to be self sufficiency. In my judgment, one of the most positive outcomes of this workshop could and should be a path along which the contribution of the Russian side is at least equal to that of the US and European participants. I mean this in the broadest sense possible, with regard to Program material such as courseware intended for educational institutions outside of Russia, contribution of satellite resources such as space segment, and finally, the local production of the transceivers, modems and antennas necessary for such an elaborate and widespread scheme.

Choosing to use satellite communications interactively by definition means cooperation between a variety of organizations, user groups, institutions, and government agencies, all of which have at times divergent interests. Given the epitaph of the workshop I would like to express the hope that this conference will lay the foundation for the effort needed to achieve "A cooperative framework for networking" - I thank you for your attention.


Back to Index