SATELLITE COMMUNICATIONS

Why Satellites?

Satellite technology evolved from a limitation of RF behavior: line-of-sight transmission. If the earth were flat, this line-of-sight behavior would not be an issue (which is one of the reasons there were no satellites before Columbus.) A signal transmitted in any direction which has sufficient power will get to its intended receiver, and over (relatively) short distances, this is exactly what happens. The problem arises when an RF signal is transmitted over long distances (greater than 25 miles) and the receiver is obscured by the curvature of the Earth (see Figure 6-7).

The left side of Figure 6-7 demonstrates what happens to RF signals which have sufficient power to travel over 25 miles. They head toward outer space. This behavior, while a limitation, is also a benefit. Because of this behavior, the FCC can allocate the same frequency to different parties in different geographical locations. After about 25 miles, the signal is no longer Earthbound and it therefore no longer interferes with other signals which are at the same frequency.

An interesting thing to note is that there is an exception to the behavior depicted on the left side of Figure 6-7. As previously mentioned in the section on broadcasting, at low RF frequencies (less than 30 MHz), the RF energy gets reflected off the ionosphere and "bent" around the earth. Satellites are not needed for signals at these frequencies.

The right side of Figure 6-7 shows all you need to know about the utility of satellites. Satellites allow RF signals to overcome the curvature of the Earth and still obey their line-of-sight behavior.

Given the tremendous expense involved in launching and maintaining satellites, they are only used for long distance communications. You will probably never see people using satellite communications to call their next door neighbor.

How Satellites Work

Geosynchronous Orbit

The satellite systems used for intercontinental telecommunications and television re-transmission work for only one reason: the satellite does not move with respect to the Earth. This is a pretty good trick since the Earth is rotating at about 1000 miles per hour at the equator.

It just so happens that there is one specific orbit around the Earth, located 22,000 miles up from the equator, called geosynchronous orbit (GEO). (It is also referred to as geostationary orbit.) It is in this orbit where a satellite can rotate around the Earth at the same rotational speed as the Earth. A satellite rotating around the Earth in geosynchronous orbit appears to remain stationary when viewed from a point on the equator. (If the satellite did not remain fixed, the direction of the transmitted signal from the ground would have to be continually altered.)

How does geosynchronous orbit work? Without going back to Physics 101, there are two forces pulling the satellite in different directions, and at the GEO these two forces cancel each other out. One of these forces is the centrifugal force. You can envision centrifugal force by picturing yourself whipping a ball on a string around your head. If you let go, the ball goes flying outward. The other force acting on the satellite is gravity, which tries to pull the satellite inward (toward Earth). At 22,000 miles above the equator, these two forces cancel each other out and the satellite does not move with respect to the Earth.

As a visual appreciation for how high up 22,000 miles is, Figure 6. 8 is an approximate scale of the Earth and the satellite's distance above it. Because the satellite is so high above the Earth, it is a good news/bad news situation. The bad news is that because the satellite is so high up, it takes a lot of RF power to reach it from the Earth. The good news is that signals transmitted down from the satellite can reach almost an entire hemisphere. Therefore, a satellite used to forward telephone calls from the United States to Great Britain is located halfway across the Atlantic Ocean, at the equator, of course.

Uplink/Downlink

Once the satellite is comfortably situated in GEO, the next areas of concern are the uplink and the downlink. The uplink describes the signal which travels from the ground transmitter (called the Earth station) up to the satellite receiver, while the downlink describes the signal which travels from the satellite transmitter to the ground receiver.

The uplink and downlink frequency bands for satellite systems are allocated by the FCC in the United States (just like every other form of RF communication). Since some satellites are used for international communications, the downlink frequencies impact countries other than the United States. The responsibility for allocating these "international" frequencies falls on the International Telecommunications Union (ITU). Think of the ITU as the FCC for the rest of the world.

A really ingenious feature of satellite communications is that the uplink frequency band and the downlink frequency band (in a given system) are never the same. When signals traveling in different directions use different frequency bands, it is called duplex communications. A satellite used for telecommunications is an example of duplex communication. The implication of duplex communication is that when you can call somebody in another country, halfway around the world, using satellite communications, both of you can speak at the same time. (This advantage is somewhat diminished if you don't speak the other person's language.)

However, when talking to someone using satellite communications there is another issue to be considered called propagation delay. Recall how RF energy travels at the speed of light. The time it takes a signal to travel across town, using a cellular phone, is inconsequential and cannot be detected by human beings (or any other beings). Satellite communication is different. With a round trip of 44,000 miles, there is a one-quarter-second delay between the originating transmitter and the ending receiver. Of course, this delay is only in one direction. When trying to hold a conversation in which the other person also gets to speak, the round trip delay is half of one second. (Two round trips are made to the satellite.) This one half second delay is very noticeable when making an intercontinental telephone call.

The one-quarter-second delay is also present when viewing a sports event from another continent. Of course, in the case of broadcasting, the one-quarter-second delay is of little consequence, unless finding out the result of the World Cup final 250 milliseconds after the people in the stadium bothers you.

Satellite Dishes

Earth stations transmit and receive signals with dish antennas (see Figure 6-9) because they are highly directional (they have a narrow beamwidth). The size of the dish antenna used depends on the frequency and power of the uplink and downlink signals. High uplink power requires a large dish. On the other hand, high downlink power only requires a small dish (like that used for direct-to-home TV).

When receiving the downlink signal, dish antennas act as funnel reflectors. First, they funnel as much of the RF energy (from the satellite) as possible into the dish, which is made of an RF reflecting material. Obviously, the bigger the dish, the more RF energy it can funnel, which is why high power satellites, like the ones used in direct-to-home TV, only require a small dish. Once the RF energy enters the dish, it all gets reflected to a single "focal" point, slightly elevated from the surface of the dish (see Figure 6-9). This focal point contains the actual antenna element and the low noise amplifier (LNA).

Footprints

The antenna on a satellite projects an antenna pattern just like every other antenna. (Recall from Chapter 3 how an antenna pattern is a visual display of the direction which RF energy is radiating out from an antenna.) Because the antenna pattern generated from a satellite is projected onto the surface of the Earth, it is called an antenna footprint. The power radiated inside this footprint is called the effective isotropic radiated power, or EIRP. (I guess just plain "radiated power" was already taken.) A graphical depiction of an antenna footprint covering the United States is shown in Figure 6-10.

Notice from Figure 6-10 that the antenna pattern approximately covers the land area of the continental United States. This is not by accident. There is only a finite amount of RF energy being transmitted from the satellite, and while the satellite is high enough in GEO to transmit a signal which covers half the globe, why bother. (Other than television-watching whales, who would know the difference?) Covering half the globe results in less power reaching the United States. This leads to an interesting aspect of satellite footprints. They can be contoured, somewhat, to approximate the shape of the area they are designed to cover to make maximum use of the available RF power.

Satellite footprints are predominantly in the shape of a circle or oval. With respect to Figure 6-10, it might appear that certain parts of the United States receive no signal at all (those outside the footprint), which is not the case. Recall that an antenna pattern is a silhouette of an area which receives some minimum amount of RF energy. In this figure, the area outside of the oval still receives RF energy; it's just less than the area inside. There are some ramifications to a footprint not covering the entire United States. Under certain adverse conditions (e.g., heavy rain), it is entirely possible that those areas which lie outside the footprint will receive a degraded (or no) signal.

The footprint shown in Figure 6-10 is for a broadcast signal. It is obviously meant to be received by many different points in the United States. More than likely this is some kind of television broadcast. But what if the satellite carries intercontinental telephone calls; what will the footprint look like then? It will still be a circle or oval, only much smaller and focused only on the long distance telephone carrier's Earth station.

RF Electronics

By now you are probably wondering what the RF electronics on board a satellite consists of. (Even if you're not, that's what you're about to read.) At its most basic, the RF electronics on board a satellite, called a transponder, is just a simple combination of a few RF devices (see Figure 6-11).

Referring to Figure 6-11, the satellite antenna receives the uplink signal (at the uplink frequency) and sends it on to the duplexer. (Recall that a duplexer is nothing more than two bandpass filters in one box.) The duplexer then routes the signal to the low noise amplifier (LNA), which amplifies the signal. Next, the mixer frequency shifts the signal to the downlink frequency (source not shown) and, finally, the signal gets boosted by the high power amplifier (HPA) and sent through the duplexer on its way out the antenna. Figure 6-11 is a block diagram which describes the basic RF function of every satellite which has ever had the good fortune to make it all the way to GEO.

About the Spacecraft

GEO satellites do not last forever, but it is not because the RF electronics fail. (If nothing else, RF engineers build reliable satellites.) In fact, the RF electronics on board a satellite can last almost indefinitely, and because it gets its power from the sun (via solar panels), there should be no reason why a satellite does not last forever. Except for one thing. For GEO satellites to work properly, they need to stay fixed in GEO and, unfortunately, over time, satellites tend to drift a little. (Ok, so GEO isn't perfect.) To counteract this natural tendency to drift, satellites are equipped with something called station-keeping.

Station-keeping is nothing more than an onboard propulsion system which periodically releases small bursts of fuel. The momentum shift from these small bursts serves to reposition the satellite in its proper place. A satellite only carries a limited amount of this fuel, so when it runs out, the satellite can no longer be repositioned and quickly becomes useless. Just prior to using up the last of the fuel, the onboard propulsion system gives one big burst and sends the satellite to a different, out-of-the-way orbit, where it spends the remainder of its existence in the lonely, dark void of space.

Satellite Systems

Three Topologies

There are three basic topologies used in satellite communications: point-to-point, point-to-multipoint, and multipoint-to-point. Which topology is used is dictated by the application.

One example of point-to-point topology is intercontinental telecommunications. When a bunch of people in Great Britain want to call a bunch of people in the United States, their calls first go to an Earth station in Great Britain. All of the calls are combined into one signal and transmitted to a telecommunications satellite hanging out somewhere over the Atlantic Ocean (at the equator, of course). The signal is then retransmitted (by the satellite) to an Earth station in the United States (owned by one of the long distance carriers). The calls are then separated and routed along their way to their final destinations. In this example, the satellite is used to connect a single point (the Earth station in Great Britain) to a single point (the Earth station in the United States). In point-to-point topology, only two large satellite dishes are required and both are used to transmit and receive.

Point-to-multipoint topology is used in direct-to-home satellite television (called DBS for direct broadcast satellite). For broadcasts in the United States, a single Earth station, owned by the broadcast company, transmits its "entertainment" up to a satellite (situated somewhere over Kansas). The satellite then retransmits the signal in a footprint which covers the whole country, like the one shown in Figure 6-10. Anybody with a small dish antenna, located within (or near) the footprint, will be able to—for $29.99/month—receive the broadcast signal. In point-to-multipoint topology, there is one large dish for transmitting and many small dishes for receiving (only).

The most frequent use of multipoint-to-point topology is something called VSAT, which stands for very small aperture terminal. In this topology, many Earth stations with small to medium size dishes are used to relay information from a point of sale back to a single location, typically the central office. One of the big users of VSATs are gasoline companies. Quite often you will see gasoline stations with small satellite dishes on their roofs. This is a VSAT system. The system periodically relays sales information from the station, via satellite, back to the main office. In this way, the main office has an up-to-the-minute status on the sale of gasoline all over the country. In multipoint-to-point topology, there are many small dishes transmitting and only one dish receiving.

The Role of Frequency

The FCC has allocated many different frequency bands for what it calls Fixed Satellite Service, which is the government's way of saying GEO. By far, the two most popular frequency bands used for commercial Fixed Satellite Service are C-band and Ku-band. The specifics of C-band and Ku-band are shown in Figure 6-4.

Table 6-4. C-Band and Ku-Band Frequency Allocations

Band	Frequency Allocation
C-band downlink	3.7-4.2 GHz
C-band uplink	5.925-6.425 GHz
Ku-band downlink	11.7-12.2 GHz
Ku-band uplink	14.0-14.5 GHz

Wherever there is a home outfitted with a large dish antenna, it is probably receiving a C-band downlink (originally intended just for reception by the broadcast stations). If they have a small dish antenna, chances are it is Ku-band (direct-to-home satellite TV).

If you recall, the RF section of a satellite, called a transponder, is used to frequency shift the uplink signal to the frequency of the downlink signal. In today's satellites, there is more than one transponder in a given frequency band. For instance, C-band, with a downlink bandwidth of 500 MHz (3.7 GHz-4.2 GHz) has 12 transponders to cover the entire bandwidth. The 500 MHz bandwidth is divided into 12, equally sized bandwidths (42 MHz each) called channels. Each transponder is assigned a channel. There are many reasons for breaking up the total bandwidth into channels, but by far the most important is redundancy. If there are 12 transponders and one fails, there are still 11 left, but if there is only one transponder and it fails, you get the picture. (Besides, do you have any idea how hard it is to get an RF technician into GEO?)

The main goal of every satellite user is to make maximum use of the available bandwidth on the satellite. Transponder bandwidth, called capacity, is leased out at a very expensive hourly rate. Satellite users try to cram as much information as possible into every MHz of every transponder. Take the case of standard broadcast television. Recall from earlier in this chapter when I mentioned that a standard broadcast television signal requires 6 MHz of bandwidth. But transponders have 42 MHz of bandwidth. What do you suppose the TV stations do? If you guessed that they cram seven different channels into each transponder, you're on the right track, but you forgot about polarization. (You are forgiven if you didn't read that section.) Polarization allows TV broadcasters to cram 14 (7 horizontally polarized and 7 vertically polarized) different television channels into one transponder.

In more sophisticated satellite systems today, the 6 MHz television signal is digitized before it is transmitted up to the satellite. Why digitize it? Once a signal is digitized, it can be compressed in a process called digital compression (what else?) with the use of digital signal processors (DSP). Digital compression involves removing redundant information in the TV signal, resulting in less bandwidth required to transmit the signal. After compression, a 6 MHz signal may only take up 4 MHz, and you know what that means: more television channels into the same transponder (and more money into the satellite owner's pocket).

To give you an appreciation for the capacity of a 42 MHz transponder channel, an (uncompressed) telephone call requires 4 kHz of bandwidth, which means each transponder channel can carry over 10,000 different voice calls simultaneously (ignoring polarization).

A Special Satellite System-GPS

What is GPS?

GPS, or global positioning system, is a satellite system consisting of 21 satellites which do nothing but continuously transmit a strange, but useful, signal. GPS satellites offer no provision for receiving signals from their users. The signals transmitted from the GPS satellites are used for two things: navigation and timing. The signals allows anyone with a GPS receiver to know approximately where they are on the surface of the Earth. (I say approximately because there is always some error in the measurement, which ranges from inches to miles, depending on many things.) The GPS signals can also be used to tell what time it is, but then again so can a wristwatch.

When a GPS receiver is used to determine present location, it does not respond with something like "the corner of 5th and Market in downtown Pittsburgh." Instead, what it gives is a measurement in degrees of longitude and latitude, and while knowing your present location by longitude and latitude is interesting, by itself it isn't particularly useful (unless you happen to be Gilligan). Conveniently, today's GPS receivers come with a trip function, which enables it to remember your starting point and give you relative distance and direction measurements from that point, as you travel. For instance, if you are hiking in a densely wooded area which you are unfamiliar with, at your launching point you can command the GPS receiver to store that location in memory. Then, at any point in the trip, if you ask the GPS receiver how to get back, it will relay how far and in what direction you need to go to get there. Now that is useful.

An interesting thing about the GPS satellites is that they are not in geosynchronous orbit, which means they are not stationary, with respect to the rotation of the Earth. If you could see them from Earth (you would be Superman), you would see GPS satellites cruising by overhead periodically from one horizon to the other. Since each of the 21 satellites more or less transmits the same signal, and since there is no need for them to receive any signals, as long as some of the satellites are overhead at any moment in time—even though they are moving—the system works just fine.

Theory of Operation

GPS operates on the same basic principal as radar: if the time it takes for a signal to reach a certain point is known, then its distance can also be calculated. A GPS receiver receives specially encoded signals from the GPS satellites which enables it to determine how long it took for the signal to arrive. The GPS receiver then takes this time and converts it into a distance (just like radar).

GPS is also based on another concept called triangulation. In overly simplified terms, triangulation is used to determine where you are if you know how far you are from three different points. For reasons which are beyond the scope of this book, if a GPS receiver receives signals from at least four different GPS satellites, it can determine its location precisely.

Figure 6-12 is a two-dimensional explanation of the (three-dimensional) GPS position determination process. The left side of the figure depicts the situation with a single satellite located at point A. If the receiver knows its distance from point A, then it must be somewhere on the circle (the distance equals the radius). (In a three-dimensional world, this circle is sphere.) On the right side of the figure there is a second satellite at point B. If the receiver knows how far it is from point A and point B, then it must be on both circles, and therefore it must be at one of the two positions shown. When one more satellite is added, the receiver's position is known exactly, as one of the two points is eliminated. As mentioned before, in the real (three-dimensional) world, a fourth satellite is needed to determine position precisely.

Since there are only 21 satellites in the GPS system, they are concentrated around the equator, which means there are places on Earth, near the north and south poles, where less than four satellites are visible at all times. The result is that GPS does not work very well at the ends of the hemispheres (which is why you never hear about Santa Claus using it).

The Next Generation Satellites—LEO

Up until now, I have assumed that all communication satellites are in geosynchronous orbit, which was true until very recently. Today, there are several consortia (which is a fancy word for a group of rich people) working on a new kind of satellite system based on low Earth orbit satellites or LEOs. Low Earth orbit satellite systems (called constellations) use many small satellites, only a few hundred miles up, to relay RF information.

Why LEOs?

There are several ramifications to using satellites in low Earth orbit. First, the satellites are not stationary with respect to a fixed point on Earth. (If you could see them, they would fly by overhead very quickly.) These non-stationary satellites make communicating with them more complicated than with GEOs. First, RF contact must be made with a moving satellite (which is a challenge in itself). Then, after the RF link is established with the first satellite, it is only a short time before that same satellite is no longer "visible" and an RF connection is no longer possible.

The second ramification of LEO satellites is that because they are so low, they can only project a small antenna footprint (covering a small area), which means that many LEOs are required to fully cover a land area the size of the United States. People on opposite ends of the country, wishing to use LEOs to make contact, will not be able to use the same satellite to connect. All of these ramifications result in a satellite system which is much more complex than a standard, off-the-shelf GEO system.

If LEOs are so complex and require many satellites, why bother? The answer is surprisingly simple: power. LEOs require less RF power to reach than GEOs (which makes sense since they're closer). In fact, some LEOs are low enough to reach with the power in a handheld phone. This is why LEOs are being deployed at all, because they can be reached with the power in a handheld phone, unlike satellites in GEO. LEOs will be used strictly for mobile telephony. They will not be broadcasting your favorite television show any time soon.

Since telephony by LEO is very expensive, it is only meant to be used in places where no other form of mobile telephony (e.g., cellular phones) exists, like out in a desert or in the middle of the ocean. This niche use of an expensive technology has fueled a long running controversy over the soundness of the LEO business. There is one question which everybody is trying to answer: Are there enough customers out in the desert or in the middle of the ocean to pay for the expensive LEO satellite system? Stay tuned.

How LEOs Work

LEOs work by continuously relaying signals among each other as they move. Figure 6-13 shows generally how a LEO system works. In this example, the satellites are moving counterclockwise. The left side of the figure (Time = 0) shows how a person (located at point 1) on one side of the planet can talk to someone on the other side of the planet (located at point 2) using a LEO system. The phone call, using the LEO's frequency, is transmitted to whichever LEO satellite is closest overhead at that given moment (satellite A on the left side of the figure). That satellite (A) then relays the call to the next closest satellite (B) in the direction of the final destination. This relay process repeats itself until the call reaches the satellite (D) which is over the intended receiver. At that point, the signal is transmitted down with an antenna footprint which encircles the intended receiver. How the system knows which satellites to use at any moment in time is what makes the LEO systems so complex.

After a short period of time (Time = 1 on the right side of Figure 6-13), satellite A will no longer be over the person originating the call (1). It is at this point when satellite A "hands off" the call to satellite B and the process continues as before, only this time the last satellite utilized is no longer D but E. (All of the satellites have rotated counterclockwise.)

The other aspect to know about LEOs is the satellite-power tradeoff. If the LEO satellite system is used with very low-powered handheld phones, then the satellites must be close to Earth (but not so close that they fall on you). The closer to Earth they are, the smaller their footprints will be and, therefore, the more satellites that are required to blanket the whole Earth. And a lot of satellites—even small ones—translates into an expensive system. LEO system designers are constantly trading off between the fewest possible satellites (higher up) and the most user-friendly (low-powered) handheld phones.