Satellite Telephone Communications

To describe how satellite communications work, let's look in more detail at the process of making an intercontinental telephone call using a GEO communication satellite system. Just as it was in the 1870s when Alexander Graham Bell first invented it, the telephone receiver is an analogue device. It operates using continuously varying physical quantities, such as electrical current, without a single digital bit 0 or 1 to be seen anywhere.

In fact, making a telephone call many years ago was a completely analogue process. In this process, the voice produces pressure waves in the air—sound waves—that impinge on a small circular metal disk in the telephone mouthpiece. These waves then cause the disk (sometimes called a diaphragm) to vibrate in sympathy with the voice. Attached to the diaphragm is a lightweight coil of wire, which is positioned adjacent to a permanent magnet. As the wire coil moves up and down with the diaphragm in the magnetic field, a current is induced in the wire (see Chapter 6 for a brief explanation of electromagnetic induction), and this current can be thought of as an electrical representation of the information contained in the voice, that is, speech. This electrical version of the voice then propagates

Time (sec)

Figure 9.11: A rapidly varying and complex analogue signal is represented on the left. A small section of this is blown up on the right, with the time axis stretched, showing the process of conversion of the analogue signal to a digital one.

Time (sec)

Figure 9.11: A rapidly varying and complex analogue signal is represented on the left. A small section of this is blown up on the right, with the time axis stretched, showing the process of conversion of the analogue signal to a digital one.

down the cable to the destination telephone handset, where the voice current is passed through the diaphragm wire coil of its earpiece. The combination of the voice current in the coil and a magnet in the earpiece causes the diaphragm to move in sympathy with the current. This movement in turn produces sound waves in the air, re-creating an understandable representation of the voice for the recipient of the telephone call.

Because satellite communication is a digital process, at some point the analogue signal—in this case the voice current—needs to be converted into a sequence of zeros and ones to produce a digital signal. This process of converting an analogue signal into a digital one is referred to as digital encoding (the method of encoding described below is not currently the most commonly used technique, but it is probably one of the easiest to understand). How can we convert a rapidly varying voice current into a string of 0 and 1 values suitable for transmission to a satellite? The left-hand side of Figure 9.11 is a representation of the analogue signal produced by a telephone, that is, the variation of electrical voice current produced by the telephone mouthpiece. In general, this is a complex and rapidly varying signal, and at any precise moment in time it will have a particular numerical value. The first step, then, is to convert decimal numbers, representing the value of the current at a particular time, into binary numbers, which are simply a sequence of zeros and ones. Table 9.2 shows how the first eight numbers, including zero, can be written in binary digital language. To do so, we need a string of three bits (zeros or ones). The number of zeros and ones in the string required to represent a particular decimal number can be

Table 9.2: The first eight decimal numbers (including zero) represented in binary as strings of three zeros or ones

Decimal

Binary number

number

20

21

22

1

2

4

0

0

0

0

1

1

0

0

2

0

1

0

3

1

1

0

4

0

0

1

5

1

0

1

6

0

1

1

7

1

1

1

calculated by thinking about powers of 2. Since 8 is 23 (2 x 2 x 2), we require a string of three 0 and 1 values. Continuing in this way, the first 16 (24) decimal numbers require a string of four bits, the first 32 (25) require a string of five bits, the first 64 (26) require six bits, and so on. This is great for people like myself who have quite a few years behind them—instead of having to put fifty something candles on your birthday cake, you only need six candles to write your age in binary. Of course, you do need candles of two colours to distinguish the "0" and the "1". With these kind of jokes on offer you can see that, in my household, birthdays are a whole lot of fun!

As shown in Table 9.2, in binary digital language each bit represents a power of 2, so under the heading "Binary number" we have powers of 2 (20, 21, 22), which take the values of 1, 2, and 4, respectively. We do not often come across a number to the power 0, but any number to the power 0 (e.g., 20) takes the value 1. So for each decimal number, we put a 1 where the power of 2 contributes and a 0 where it does not. Here are a couple of examples: on Table 9.2, the decimal number 5 can be written in binary digital language as 1 0 1, because 5 in simple arithmetic is (1 x 1) + (0 x 2) + (1 x 4). Similarly, 6 is 0 1 1, because 6 is (0 x 1) + (1 x 2) + (1 x 4). Thus decimal numbers can be represented as strings of zeros and ones. This strange binary language is what a desktop computer is using all the time to perform its mathematical routines.

In Figure 9.11, the analogue signal—the voice current—is shown on the left-hand side, and we need to convert this into a long sequence of zeros and ones suitable for transmission to the satellite. To show how this is done, we have magnified a small part of the analogue signal on the right-hand side of the figure. At a particular moment, at point 1, the voice current takes a particular value I1, and this value can be converted into binary digital language. It is common to convert it into a string of eight bits, which means that the voice current value at this point can be assigned to any one of 256 (28) possible levels. Then a small fraction of a second later, called the sampling time, the value I2 of the voice current is measured again, and it too is converted to a string of 8 bits. This process continues throughout the telephone conversation, converting the voice current into a long string of 0 and 1 values, as shown in the boxes beneath each point. It is usual to sample the voice current about 8000 times a second (so that the sampling time is 0.000125 second). This short sampling time is needed so that all of the complex and rapidly varying detail of the original voice current is captured in the digital signal. Using this method, the data rate of one digital telephone voice is around 8 x 8000 or 64,000 bits per second—8 bits produced 8000 times each second. This is referred to as 64 kbps, where "k" stands for kilo, meaning a thousand. Computer-savvy readers will be familiar with data rates; for example, a broadband Internet connection might be, say, 5 Mbps, where the "M" stands for Mega, meaning a million.

There is still another step in the process of transmitting a voice to the satellite. The digital bit stream representing a voice now needs to be somehow put onto an EM wave so that it can be transmitted from the ground station antenna to the spacecraft. This step in the process is referred to as modulation (Fig. 9.12). Modulation entails putting the information

Figure 9.12: Three types of modulation, AM, FM, and PM, are used to put the digital bit stream onto the carrier wave.

contained in the digital bit stream onto a carrier wave, which can then be transmitted to the satellite. The carrier wave's job is simply to carry the information from ground to spacecraft. Initially, a carrier wave is an EM wave with a single frequency, as shown at the top of the figure. The next layer of the diagram shows the digital bit stream that needs to be carried by the wave, and there are three main ways of doing this. The first is amplitude modulation (AM), which is shown in the third layer in the figure. The amplitude of the wave is varied depending on the value of the bit; in this case, the amplitude is zero when a 0 bit is being carried, and nonzero when a 1 bit is transmitted. For frequency modulation (FM), as shown in the fourth layer of the diagram, the wave frequency is varied. When a 0 is carried, the frequency is low (long wavelength), and when a 1 is carried the frequency is high (short wavelength). Most conventional radios have AM and FM bands on their tuning dial. The third type, phase modulation (PM), is shown by the bottom line of the diagram. In this case, every time the value of a bit changes from a 0 to a 1, or vice versa, the phase of the carrier wave is changed by 180 degrees (see information on phase in Figure 9.10). In digital space communications, this is the most commonly used form of modulation.

When the carrier wave is received at the destination ground station, the process described above needs to be reversed so that the person at the other end of the phone can hear what the speaker said. The bit stream needs to be recovered by a process of demodulation of the carrier wave, and the analogue signal—the voice current—needs to be recovered by decoding the digital bit stream. It does seem to be a complex process, but it is all done routinely everyday, without anyone noticing!

It is also important to realize that there are thousands of these telephone conversations going on at the same moment, all of them sharing the same uplink to the spacecraft. To prevent one telephone conversation from bumping into another, each has a separate carrier wave at slightly different frequencies. So typically the ground station transmits a great wodge of carrier waves with frequencies ranging from, say, 6 to 6.5 GHz. On receipt of the uplink signal, the job of the satellite is to amplify the signal, change its frequency to, say, 4 to 4.5 GHz, and then retransmit it on the downlink to the destination ground station. The amplification is necessary as the signal strength after the uplink journey is small, about 10"8 W (or 0.000 000 01 W), a tiny fraction of the output of a domestic light bulb! The change in frequency is needed because the spacecraft usually uses the same antenna for both uplink and downlink, and the frequency shift prevents the information in the two links from becoming scrambled.

Was this article helpful?

0 0

Post a comment