Introduction To Satellite 2

Chapter 1
The Satellite
The Journey Into Space Orbits Subsystems Footprints Beams Frequencies Service Life
Rev D 07/04
The Journey Into Space

Satellites are transported into space by multistage rockets which place the satellite into an orbital slot. For most satellite launches, the scheduled launch rocket is initially aimed straight up. This allows the rocket to pass through the thickest part of the atmosphere most quickly and best minimizes fuel consumption.
After a rocket launches straight up, the rocket control mechanism uses the inertial guidance system to calculate necessary adjustments to the rocket's nozzles to tilt the rocket to the course described in the flight plan. In most cases, the flight plan calls for the rocket to head east because Earth rotates to the east, giving the launch vehicle a free boost. The strength of this boost depends on the rotational velocity of Earth at the launch location. The boost is greatest at the equator, where the distance around Earth is greatest and therefore rotation is fastest.
Rev D 07/04
Elements of an Orbit
A rocket must accelerate to at least 25,039 mph (40,320 kmph) to completely escape Earth's gravity and fly off into space. Earth's escape velocity is much greater than what is required to place a satellite in orbit. With satellites, the object is not to escape Earth's gravity, but to balance it. Orbital velocity is the velocity needed to achieve balance between gravity's pull on the satellite and the inertia of the satellite's motion -- the satellite's tendency to keep going. Even with gravity, if the intended satellite goes too fast, it will eventually fly away. On the other hand, if the satellite goes too slowly, gravity will pull it back to Earth. At the correct orbital velocity, gravity exactly balances the satellite's inertia, pulling down toward Earth's center just enough to keep the path of the satellite curving like Earth's curved surface, rather than flying off in a straight line. The orbital velocity of the satellite depends on its altitude above Earth. The closer the satellite is to Earth, the faster the required orbital velocity. For example, to maintain an orbit that is 22,223 miles (35,786 km) above Earth, the satellite must orbit at a speed of about 7,000 mph (11,300 kmph). That orbital speed and distance permits the satellite to make one revolution in 24 hours. Since Earth also rotates once in 24 hours, a satellite at 22,223 miles altitude stays in a fixed position relative to a point on Earth's surface. This kind of orbit is called geostationary. Geostationary orbits are ideal for weather satellites and communications satellites. Satellites usually start out in an orbit that is elliptical. The ground control station controls small onboard rocket motors to provide correction. The goal is to get the orbit as circular as possible. By firing a rocket when the orbit is at the apogee of its orbit (its most distant point from Earth), and applying thrust in the direction of the flight path, the perigee (closest point to Earth) moves further out. The result is a more circular orbit.
Rev D 07/04
Examples of Satellite Orbits

Geostationary Orbit Geostationary orbits (also called geosynchronous) are orbits in which the satellite appears to remain in a fixed positioned above the Earth, due to the rotational period of the satellite being the same as the earth (24 hours). Geostationary satellites are above a band along the equator, with an altitude of approximately 35,790 km.
Polar Orbit In a polar orbit, the satellite generally flies at a low altitude and passes over the planet's poles on each revolution. The polar orbit remains fixed in space as Earth rotates inside the orbit. As a result, much of Earth passes under a satellite in a polar orbit. Because polar orbits achieve excellent coverage of the planet, they are often used for satellites that do mapping and photography.
Rev D 07/04
Examples of Satellite Orbits (continued)

Asynchronous Orbit The Space Shuttle uses a much lower, asynchronous orbit, which means it will pass overhead at different times of the day. Other satellites in asynchronous orbits average about 400 miles (644 km) in altitude. Inclined Orbit A satellite in an inclined orbit drifts North and South of its dedicated orbital location while maintaining a synchronous orbit.
Satellite Subsystems
In order to maintain the "communication pipe" in the sky, communications satellites need six major on-board subsystems:

Antennas, Transponders, Power supplies, Command and telemetry, Thrust Stabilization
Antennas Geostationary satellites are far from earth and require directional antennas in order to communicate. A directional antenna uses a parabolic reflector (commonly referred to as a dish) to focus the radiated energy from the transmitter, and to focus the incoming energy to the receiver. This ability to focus energy is referred to as "antenna gain."
Rev D 07/04
Satellite Subsystems (continued)

Communications The communications subsystem is the reason communications satellites exist. Consisting of an array of transponders sharing uplink (Rx) and downlink (Tx) antennas, the communications subsystem is the "communication pipe" referred to in the previous page. Each transponder consists of three components:
Low Noise Amplifier (LNA): The uplink signal originated at an earth station that was approx. 36,000 km from the satellite. A distance that great will weaken and degrade the uplink signal to a considerable extent. In order to improve the received uplink signal level without significant further degradation, a sophisticated amplifier (LNA) receives the uplink signal from the Rx antenna and amplifies it without adding significant noise. Frequency Converter: The downlink signal must not be in the same frequency band as the uplink signal. This requirement is necessary to prevent damage to the LNA due to leakage from the downlink antenna to the uplink antenna. High Power Amplifier (HPA): It is a long way back to earth (another 36,000 km trip), and it is necessary to boost the output of the frequency converter with another sophisticated amplifier that feeds the downlink signal to the Tx antenna.
Rev D 07/04

Power Supply All these transponders need electricity to operate. Power is supplied by onboard batteries that are recharged by solar radiation.

Solar panels must be constantly exposed to light to produce electricity Solar panels are only effective if they will be receiving lots of sunlight Solar panels consist of many individual solar cells Each individual cell does not produce large amounts of electricity by itself Large surface area is needed in order to produce maximum electricity The need for large solar panels must be balanced with the need for the entire satellite to be relatively small Solar panels can be mounted on the body of the satellite Solar panels will not "run down" like a battery (completely renewable) Solar panels can be used together with a battery, which recharges when the satellite is in the sun and provides the satellite's power when the satellite is not
Command and Telemetry In order to be sure the communications subsystem is operating correctly it is necessary for ground based personnel to be able to monitor its operational status. This is the function of the telemetry subsystem. An onboard computer with links to all satellite systems provides this status data (called telemetry data) to a special telemetry downlink. Based on this information it may be necessary to take corrective action from the ground; e.g., if a transponder fails it is important to be able to switch in a spare. This is done from the ground via a command uplink that provides a communication path to the on-board computer.
Rev D 07/04

Thrust In order for a geostationary communications satellite to continue to function, it must remain stationary with respect to all the earth station antennas that are pointed at it. To correct for the orbital fluctuations that all satellites are subject to, each satellite carries a thrust subsystem to give it an occasional nudge to keep it "On Station."
Stabilization Remaining on-station is only half the battle. Additionally, the satellite's antennas must always be aimed at the same spot on the surface. This requires gyroscopic stabilization of the satellite body. This is accomplished with gyroscopes in one of two configurations:

Spin stabilization, in which the entire satellite body is spun (antennas are de-spun), or Internal gyroscopes
Rev D 07/04
Satellite Footprints
The geographic area of the Earth's surface over which a satellite can transmit to, or receive from, is called the satellite's "footprint." The footprint can be tailored to include beams with different frequencies and power levels.
Regional Coverage Eutelsat W1
Spotbeam Coverage Eutelsat W1
Rev D 07/04
Satellite Beams
Modern satellites are designed to focus on different ranges of frequency bands and different power levels at particular geographic areas. These focus areas are called beams. Intelsat offers four beam types:

Global: covering almost 1/3 of Earth's surface Hemi: covering almost 1/6 of Earth's surface Zone: covering a large landmass area Spot: covering a specific geographic area
Rev D 07/04
10
Satellite Frequencies
Satellites transmit information within radio frequency bands. The frequency bands most used by satellite communications companies are called C-band and the higher Ku-band. Over the next several years, the use of a higher frequency band known as Ka-band is expected to increase. C-band is the oldest allocation and operates in the frequency range around 6 GHz for transmission (uplink) and between 3.7 and 4.2 GHz for reception (downlink). Ku-band is the most common transmission format in Europe for satellite TV and uses around 14 GHz for uplink and between 10.9 and 12.75 GHz for downlink. Ka-band uses around 30 GHz up and between 18 and 20 GHz downlink frequency. C-band and Ku-band are becoming congested by an increasing amount of users, so satellite service operators are more and more turning to the use of Ka-band. The selection of the band is not something that individual service providers decide, but is rather chosen by large satellite operators based on different factors:
Availability: C-band is still the most widely available worldwide. Ku-band is becoming more available recently in regions which were less covered in the past (South America, Asia, Africa) C-band is more prone to interference from other transmission services that share the same frequencies (adjacent satellites or terrestrial transmissions) than the higher bands While the C-band technology is cheaper in itself, it requires larger dishes (1 to 3 m) than Ku- and Ka-band (0.6 to 1.8 m) and therefore imposes relatively higher (installation) costs on the end-user Ku- and especially Ka-band make better use of satellite capacity Higher frequency bands (Ku- and especially Ka-) suffer significantly more from signal deterioration caused by rainfall. To ensure availability in bad weather conditions, the signal has to be much stronger. Note that 0.1% of unavailability means that the service will be interrupted for almost 9 hours over a 1-year period. 1% unavailability represents 90 hours or almost 4 full days
Rev D 07/04
11
Service Life
In principle, geostationary satellites occupy a fixed position in space and consequently the ground-based antennas do not need to be constantly redirected to follow the satellites movements. The fact that the orientation of ground-based antennas is fixed is a major advantage of the geostationary satellite orbit used by satellite broadcasters. In practice however, the satellite wanders slightly around its nominal orbital position under the gravitational influence of bodies such as the Sun and the Moon, as well as other influences such as Sun radiation pressure and Earth asymmetry. It is therefore necessary to take corrective actions in order to keep the satellite within acceptable margins from its ideal position. This is achieved by activating the so-called thrusters that are mounted on the body of the satellite as part of its propulsion system. As long as the satellite has enough fuel left to operate its thrusters, it can be kept in the correct position. Typically this is 10 15 years. As soon as the satellite is out of fuel, it will drift out of control and into space, which brings an end to its operational life. The satellite service operator can decide to save on fuel (and by consequence extend the lifetime expectancy of a satellite) by allowing the satellite to drift a little bit. This is known as an inclined orbit. Although this may bring down the costs for the communication via this satellite, there is a consequence on the Earth station side. These stations have to be equipped for tracking (following the drift of) the satellite.
Rev D 07/04
12
Chapter 2
Modulation & RF
Modulation Types M-Phase Modulation Modulator Timing FEC Upconverters Amplifiers Frequency
Rev D 07/04
13
Modulator
All forms of satellite communications use a modulator. Modulation is the process in which the signal containing information is modified for transmission.
The main reason for using modulation is frequency translation (i.e. when a baseband signal modulates a radio frequency). Modulation is performed by causing a baseband signal to vary a parameter of a carrier wave. The modulator is the first component after baseband signal processing.
Rev D 07/04
14
Modulation Types
There are different types of modulation Amplitude Modulation (AM) Frequency Modulation (FM) Phase Modulation (PM)
Amplitude Modulation In amplitude modulation (AM), the amplitude of a high-frequency carrier signal is varied in proportion to the instantaneous state of the modulating message signal. AM is not used in satellite communications due to its inefficient use of power and susceptibility to noise interference. Frequency Modulation Frequency modulation (FM) is a method of impressing data onto a waveform by varying the frequency of the wave. This is used with analogue or digital data. The primary advantage over AM is its immunity to noise. Phase Modulation Phase modulation is a form of modulation in which the carrier waves phase is caused to vary in accordance with the modulating signal. Similar to FM, phase modulation is a type of angle modulation (in FM, the frequency is varied). In PM, the angle of a sine wave carrier is varied in accordance with the information contained in the modulating waveform.
Rev D 07/04
15
Modulation Types (continued)

Phase-Shift Keying (PSK) Phase-shift keying is a method of transmitting and receiving digital signals in which the phase of the signal is varied, in time, to convey information. Spectrally digital phase modulation, or phase-shift keying (PSK) is very similar to frequency modulation. It involves changing the phase of the transmitted waveform instead of the frequency, each phase change representing digital data. In its simplest form, a phase-modulated waveform can be generated by using the digital data ( ex. 1 and 0) to switch between two signals of equal frequency but opposing phase. It can be assumed that the number of phase shifts is not limited to only two states the transmitted carrier can undergo any number of phase changes. A simple way to view amplitude and phase is with the polar diagram. The carrier becomes a frequency and phase reference and the signal is interpreted relative to the carrier. The signal can be expressed in polar form as a magnitude and a phase. The phase is relative to a reference signal, the carrier in most communication systems. The magnitude is either an absolute or relative value. Both are used in digital communication systems. Polar diagrams are the basis of many displays used in digital communications, although it is common to describe the signal vector by its rectangular coordinates of I (In-phase) and Q (Quadrature).
Rev D 07/04
16
Modulation Types (continued)

More on PSK.. Amplitude modulation (AM) changes only the magnitude of the signal. Phase modulation (PM) changes only the phase of the signal. Amplitude and phase modulation can be used together. Frequency modulation (FM) looks similar to phase modulation, though frequency is the controlled parameter, rather than relative phase.
In digital communications, modulation is often expressed in terms of I and Q. This is a rectangular representation of the polar diagram. On a polar diagram as shown above, the I axis lies on the zero degree phase reference, and the Q axis is rotated by 90 degrees. The signal vectors projection onto the I axis is its I component and the projection onto the Q axis is its Q component.
Rev D 07/04
17
M-Phase Modulation
Two-state or Bi-phase PSK is typically known as BPSK and four-state or quadric-phase PSK is termed QPSK. Other combinations of phase modulation are possible and are increasing in popularity, including 8PSK and 16QAM. An important reason for the use of BPSK and QPSK is the high value of C/N required for acceptable bit error rates above QPSK. The Figure below shows how the three common modulation schemes perform in terms of Bit Error Rate (BER) vs. Carrier/Noise ratio (C/N) in a standard demodulator without any form of error correction. To achieve a given BER without any error correction coding, 8PSK requires approximately 5.5 dB higher C/N than QPSK, and 16QAM a further 2 dB higher than 8PSK. The higher order decoder needs a higher signal level to discriminate between smaller variations in amplitude and phase in the presence of noise.
Rev D 07/04
18
M-Phase Modulation (continued)

BPSK The simplest form of phase modulation is binary (two level) phase modulation. With BPSK the carrier phase has only two states, in-phase or 180 out of phase.
The balanced modulator acts like a phase reversing switch. Depending upon the logic condition of the digital input, the carrier is transferred to the output either in phase or 180 out of phase with respect to the reference oscillator. For BPSK, the output rate of change is equal to the input rate of change.
The prime advantages of BPSK include high tolerance to phase noise short acquisition time ideal for power limited systems
Rev D 07/04
19

QPSK A more common type of phase modulation is Quadrature Phase Shift Keying (QPSK). Quadrature means that the signal shifts between phase states which are separated by 90 degrees. The signal shifts in increments of 90 degrees from 45 to 135, 45, or 135 degrees. These points are chosen as they can be easily implemented using an I/Q modulator. Only two I values and two Q values are needed and this gives two bits per symbol. The I bit is modulated by a carrier that is in phase with the reference oscillator while the Q bit is modulated by a carrier that is 90 out of phase with the reference.
QPSK is therefore a more bandwidth-efficient type of modulation than BPSK, potentially twice as efficient.
Rev D 07/04
20

Consequently, since the QPSK output signal does not change phase until 2 bits have been clocked in, the fastest output rate of change is equal to the input bit rate.
The prime advantage of QPSK over BPSK is that it uses half the bandwidth of BPSK, since each symbol transmits 2 bits.
Since the demodulator needs to discern between a phase difference of only 90 for QPSK (as apposed to 180 for BPSK), the acquisition time is greater for QPSK.
Rev D 07/04
21
Constellation Diagram
It is possible to view the M-phase modulation energy pattern on an oscilloscope. The resulting display is called a constellation diagram. The QPSK diagram below shows the 4 points of energy clearly defined. This relates to an Eb/No of 20dB. With this much energy the demodulator would have no difficulty determining the bit state of each transmitted symbol.
Rev D 07/04
22
Constellation Diagram (continued)

The following diagram shows the same QPSK constellation with a lower energy per bit. Noise is clearly visible. The bottom diagram shows an unrecoverable state. The bit error rate would be unacceptably high.
Rev D 07/04
23
8PSK and 16QAM

While QPSK has been in use for many years, higher orders of modulation such as 8PSK and 16QAM have recently enabled the satellite industry to increase transponder efficiency. This increase in the information rate in a transponder comes at the expense of an increase in the carrier power to meet the threshold requirement into existing antennas. The figure below shows how the three common modulation schemes perform in terms of Bit Error Rate (BER) vs. Carrier/Noise ratio (C/N) in a standard demodulator without any form of error correction. To achieve a given BER without any error correction coding, 8PSK requires approximately 5.5 dB higher C/N than QPSK, and 16QAM a further 2 dB higher than 8PSK. The higher order decoder needs a higher signal level to discriminate between smaller variations in amplitude and phase in the presence of noise.
Rev D 07/04
24
8PSK and 16QAM (continued)

The 8PSK diagram is similar to the QPSK constellation with the difference that the carrier can exist in one of eight different states. As such, each state can represent three bits - 000 through to 111.
In 16-state Quadrature Amplitude Modulation (16QAM), there are four I values and four Q values. This results in a total of 16 possible states for the signal. It can transition from any state to any other state at every symbol time. Since 16 = 2, four bits per symbol can be sent. This consists of two bits for I and two bits for Q. The symbol rate is one fourth of the bit rate. So this modulation format produces a more spectrally efficient transmission. It is more efficient than BPSK, QPSK, or 8PSK.
Rev D 07/04
25
Bandwidth vs. Modulation
The following diagram shows the relative bandwidth required using different M-phase modulation techniques for the same data rate.
Rev D 07/04
26
Modulator Data Interface (DCE)

The data communications equipment (DCE) which acts as an interface between the baseband DTE equipment and the satellite RF terminal consists of a modulator, demodulator, and data interface. This data communications equipment is also known as a Modem. PSK modulation / demodulation is the choice for digital links over satellites. Why digital ? The move to digital modulation provides more information capacity, compatibility with digital data services, higher data security, better quality communications, and quicker system availability. Developers of communications systems face these constraints: available bandwidth permissible power inherent noise level of the system The RF spectrum must be shared, yet every day there are more users for that spectrum as demand increases. Digital modulation schemes have greater capacity to convey large amounts of information than analog modulation schemes.
Synchronous Data
Satellite modems have terrestrial interfaces which conform to internationally recognized standards. (ex. RS232, RS422, V.35 etc.). The modem data interface is the link between the earth station multiplexing / digital baseband equipment and the modulation / demodulation process. Serial communication normally consists of transmitting binary data across an electrical or optical link such as RS232 or V.35. The data, being binary, is usually represented by two physical states. For example, +5v may represent 1 and -5v represent 0. The accurate decoding of the data at the remote end is dependent on the sender and receiver maintaining synchronization during decoding. The receiver must sample the signal in phase with the sender. If the sender and receiver were both supplied by exactly the same clock source, then transmission could take place forever with the assurance that signal sampling at the receiver was always in perfect synchronization with the transmitter. This is seldom the case, so in practice the receiver is periodically brought into synch. with the transmitter. It is left to the internal clocking accuracy of the transmitter and receiver to maintain sampling integrity.
Rev D 07/04
27
Synchronous Data (continued)

Satellite modems have terrestrial interfaces which conform to internationally recognized standards. (ex. RS232, RS422, V.35 etc.). RS232 RS-232 has been around as a standard for decades as an electrical interface between Data Terminal Equipment (DTE) and Data Communications Equipment (DCE). The essential feature of RS-232 is that the signals are carried as single voltages referred to a common earth on pin 7. RS-232 is simple, universal, well understood and supported everywhere. However, it has some serious shortcomings as an electrical interface due to the influence of noise. The standards for RS-232 and similar interfaces usually restrict RS-232 to 20kbps or less and line lengths of 15m (50 ft) or less. These restrictions are mostly throwbacks to the days when 20kbps was considered a very high line speed, and cables were thick, with high capacitance. In practice, data speeds of up to 200k and cable lengths of 100m are possible if low capacitance, well shielded/grounded cable is used.
Rev D 07/04
28

RS422 / V35 (balanced) The limitations of RS-232 are largely eliminated by the balanced line interface.
A pair of wires is used to carry each signal. The data is encoded and decoded as a differential voltage between the two lines. As a differential voltage, in principle the interface is unaffected by differences in ground voltage between sender and receiver. Furthermore, if lines A and B are close together, they will be affected almost identically by external electromagnetic noise. If the lines are also twisted together, then neither line is permanently closer to a noise source than the other. Hence the well known "twisted pair" is extremely effective in eliminating noise from the signal. They can support line speeds over 100Mbps and work reliably at distances of several kilometers. There are several standards that incorporate balanced line signals into Dtype connectors. These include RS-449 (DB37), X.21 (DB15) and RS530 (DB25). RS449 Connector (DB-37)
Rev D 07/04
29

The V.35 interface was originally specified by CCITT as an interface for 48kbps line transmissions. It has been adopted for all line speeds above 20kbps, and seems to have acquired a life of its own. It was discontinued by CCITT in 1988, and replaced by recommendations V.10 and V.11. V.35 is a mixture of balanced (like RS422) and common earth (like RS232) signal interfaces. The control lines including DTR, DSR. DCD, RTS and CTS are single wire common earth interfaces, functionally compatible with RS-232 level signals. The data and clock signals are balanced, RS-422-like signals. The control signals in V.35 are common earth single wire interfaces because these signal levels are mostly constant or vary at low frequencies. The high frequency data and clock signals are carried by balanced lines. Thus single wires are used for the low frequencies for which they are adequate, while balanced pairs are used for the high frequency data and clock signals.
V.35 Connector
Rev D 07/04
30

The DMD20 modem has combined the RS232, RS422, and V.35 synchronous interface types onto one EIA-530 25-pin connector. The interface type is selected via front panel software control.
The EIA-530 interface has a maximum data rate of about 10 Mbits/sec over short distances.
Rev D 07/04
31

HSSI The High-Speed Serial Interface (HSSI) is a DTE/DCE interface that was developed by Cisco Systems and T3plus Networking to address the need for high-speed communication. Characteristic Maximum signaling rate Maximum cable length Number of connector points Interface Electrical technology Typical power consumption Topology Cable type Value 52 Mbps 50 feet 50 DTE-DCE Differential ECL 610 mW Point-to-point Shielded twisted-pair
HSSI uses differential emitter-coupled logic (ECL), which provides highspeed data transfer with low noise levels. HSSI pin connectors are significantly smaller than pin connectors for competing technologies. The HSSI cable uses the same number of pins and wires as the Small Computer Systems Interface 2 (SCSI-2) cable, but its electrical specification is more concise.
Rev D 07/04
32

G.703 G.703 is a CCITT standard for transmitting voice over digital carriers such as T1 and E1. It presents a standard method for encoding clock and data into a single signal. G.703 provides the specifications for pulse code modulation (PCM) at data rates from 64 Kbps to 2.048 Mbps. G.703 service is typically used for interconnecting data communications equipment such as bridges, routers, and multiplexers. G.703 is transported over balanced (120 ohm twisted pair presented on a single DB-15 or RJ-45 connector) or unbalanced (dual 75 ohm coax presented on a pair of BNC connectors) cable. Whether the G.703 is balanced or unbalanced depends on your geographic location and the carrier that supplies the service. Balanced service is the most common around the world with the exception of the U.K. and the Netherlands. In the picture below, J4 represents balanced G.703 while J6 and J9 represent unbalanced G.703.
Rev D 07/04
33

ASI Asynchronous Serial Interface (ASI) is, without a doubt, the most popular physical interface in the market today for transporting MPEG-2 program streams. Officially, the 75-ohm ASI interface is a DVB physical interface. There are two elementary streams that are addressed by the ASI interface, the 188 byte format and the 204 byte format. The 188 byte form is the more common ASI transport stream. From a linear view point the stream will look like this in serial packet form depending on whether or not it is 188 byte or 204 bytes (using the additional 16 bytes).
Regarding the optional 16 bytes, the most common usage is for forward error correction (FEC) such as the Reed-Solomon (RS) coded byte structure that is specified by ETS 300 429. The ASI physical interface is the method by which to transport the MPEG-2 stream from one device to another without disturbing the MPEG-2s intrinsic qualities. Although this can be accomplished by a number of physical means, we will use the 75-ohm coaxial cable interface type in our example.
Rev D 07/04
34
ASI Data Video Broadcast Example

The best place to begin is at the encoder level, where ASI signals are typically created. The incoming audio, video and data signals arrive in analog composite, component or digital means and are then encoded in the MPEG-2 format. The MPEG-2 video stream is built and is assigned a Packet Identifier or PID. The PID value is a two digit hex value such as 27. This PID value is important, as it will later play a larger role. Hidden in the PID are other pieces of fundamental information such as whether the stream is 4:2:0 or 4:2:2 and the compression data rate such as 8 Mb/second. Next are VBI layers where closed caption data is stored as well as other data that is carried along the input signal which are in common in lines 12 through 21. Next are audio PIDs that are also created, tagged a hex value and connected to the main video PID. Once these associations are made, they will cling together to help downstream devices to decode the video, audio and data information as one collective MPEG-2 signal. The modulator will be looking for an ASI signal input to modulate. The reason why is simple, it contains all the vital MPEG-2 data in its leanest form.
Rev D 07/04
35
A Few Words on DVB

Digital Video Broadcasting Group (DVB) is a group within ETSI (European Telecommunication Standards Institute) This group was created by experts in an effort to balance optimal performance versus practical and cost effective implementation Modulation and error correction standards for satellite (as well as terrestrial) broadcasting MPEG 2 / DVB combined standards provide interoperability between compliant digital transmission and receiving equipment DVB transport includes ASI (asynchronous serial interface) SSI ( synchronous serial interface, serial 422) SPI (synchronous parallel interface, 422 signal levels)
DVB transmission standards include DVB-S (satellite) and DVB-T (terrestrial) DVB-S QPSK standard Reed Solomon outer code (204, 188, 8) where 188 = packet size which consists of 184 bytes of data plus 4 byte header Viterbi inner code (1/2, 2/3, 3/4, 5,6, 7/8) Spectral shaping filter 0.35 QPSK modulation
Rev D 07/04
36
Ethernet Interface
Using the ethernet interface integrates expertise in satellite modem technology with speed-optimized IP processing. Using the DMD20 as an example, functionality such as automatic Learning and Aging, Auto-Crossover, Auto-Polarity, Auto-Negotiation and Embedded Quality of Service, offers true Plug-n-Play connectivity. Best of all, because it is a bridge, the ethernet interface allows all higherlevel protocols such as DHCP, UDP, TCP, HTTP, and FTP to pass transparently. Point-to-Point Applications The figure below illustrates a typical application connecting a remote office to the corporate headquarters via satellite. The IP addresses shown are for example purposes only. No restrictions are placed on the IP addresses or subnet masks chosen for a particular implementation. In this example, the DMD20s are simply connected to the WAN port on each router. There is no special setup or configuration required.
Rev D 07/04
37
Ethernet Interface (continued)

The Importance of Transparent Operation In some situations, the remote site may be nothing more than PCs that need to be connected to a central office for e-mail, file transfers, or Internet access. In these cases, it is often desirable to have all of the network IP addresses assigned dynamically on an as needed basis by a DHCP server. The DMD20 ethernet interface acts as a bridge instead of a router, the DHCP requests and responses required to bring the remote PCs online, are passed transparently across the satellite link.
Rev D 07/04
38
Ethernet Interface (continued)

Point-to-Multipoint Applications In many instances, customers need to connect multiple remote sites to a central hub. With unique port routing capabilities, you no longer need special static routes and ARP table entries. Simply daisy chain the hub modems as shown in the figure below, and let the routers do the work.
Rev D 07/04
39
Modulation Process
User data is accepted at the data interface (Sync Data I/O, G.703, ASI, HSSI etc.) and clocked into the modem by either internal timing (SCT, internal) loop timing (SCT, SCR) external timing (SCTE) external clock source (BNC)
Rev D 07/04
40
Modulator Timing
Both the Tx Clock and the Buffer Clock source may be independently locked to one of the following: SCT (Internal Oscillator) SCTE (External Tx Terrestrial Clock) EXC Clock (External Clock Source) Rx Satellite Clock (Loop Timing)
SCTE: Serial Clock Transmit External This clock is the Transmit Terrestrial Clock associated with the interface. With the G.703 Interface selected, SCTE is the clock that is recovered from the G.703 data stream. SCTE is sometimes referred to as Tx Terrestrial Timing and for Synchronous Interfaces such as RS-422, SCTE is sometimes referred to as TT (Terminal Timing).
Buffer Clock SCTE
Buffer Clock RXSAT
Rev D 07/04
41
Modulator Timing ( continued)

SCT: Serial Clock Transmit This clock is an internally generated clock that is output from the modem. The clock is generally used by the Terrestrial Terminal equipment for clocking the transmit data. The frequency of the clock is set the same as the Transmit Terrestrial Clock rate if internal is selected, or is the receive clock from the Demodulator if SCR is selected. SCT is sometimes referred to as Internal Timing or ST (Send Timing).
EXT CLK: External Clock This is an independent clock source. This clock is most often used if there is a station master clock. The EXT EXC can be selected, in the Interface/General Menu, to be None, BNC EXC, or IDI. IDI is used ONLY for D&I cases where external framing is selected. In this case the EXT EXC is set to IDI where the Receive Buffer Clock is derived from the external Receive T1/E1 trunk. BNC EXC: BNC External Clock Unbalanced external clock input into a BNC connector. This clock can be used to source the EXT CLK. Clock specification: Frequency: 1 MHz 20 MHz Level: 0.5 Vp-p 5 Vp-p Insert Data In This clock source is only used as an external frame source selected in D&I Mode. If External Frame Source is selected, then IDI must be selected for the buffer clock. External frame source is used when modems are daisychained, and prevents interruption to the E1 by any of the modems connected (i.e. if one modem loses lock it will not affect the E1 trunk).
Rev D 07/04
42
Data Processing
Scrambling is used to ensure a 50% duty cycle on the data so that clock can be reliably extracted at the distant end of the link. In digital format if all 1s or 0s are sent, clock slips will occur. Randomizing the data stream ensures frequent transitions and avoids the potential for transmitting a constant carrier with no phase change. The spectrum plots below show (in order from top to bottom) a modulator with no data input and scrambling turned on; a modulator with no data input and scrambling turned off; a modulator with no data input and scrambling turned off.
Rev D 07/04
43
Data Processing (continued)

Differential Coding ensures that data inversions do not occur by resolving phase ambiguity that may occur in the modulation process. It is mostly used in BPSK and QPSK. It can decrease coding gain performance by 1/3 of a dB. In normal QPSK, the receiver associates an absolute phase of the received signal with the I and Q channel data value. The problem is that phase is a relative quantity and the receiver does not (can not) know the absolute phase relative to the transmitter internal reference signal. This causes an I and Q channel ambiguity and/or an I and Q channel data inversion. Differential coding uses a phase shift or a relative phase to represent the absolute I and Q channel bit values. A phase change in the carrier signal from one symbol period to the next determines the I and Q channel data values. The encoder in the transmitter must generate the phase shifts based on the data to be transmitted. In higher order modulation schemes such as 8PSK and 16QAM differential coding is not used (disabled) since phase is already checked. For all Turbo applications, regardless of modulation used, differential coding must be disabled. Bit Error Rate (BER) In telecommunications transmission, the bit error rate (or BER as it is commonly called) is just one of many statistical measures of a communications link or channel. BER is the percentage of bits that have errors relative to the total number of bits received in a transmission, usually expressed as ten to a negative power. SENT BITS RECEIVED BITS 1101101101 1100101101
error BER = # of wrong bits = 1 = 0.1 # of total bits BER is normally displayed in scientific notation. For example, a transmission might have a BER 1E-6, meaning that out of 1,000,000 bits transmitted, one bit was in error. If the BER is 1E-12 we expect that there will be one bit in error for every 1 trillion bits. The more negative the exponent, the better the BER.
Rev D 07/04
44
Forward Error Correction (FEC)

Channel Coding is also known as forward error correction. The purpose of forward error correction (FEC) is to improve the capacity of a channel by adding some carefully designed redundant information to the data being transmitted through the channel. This allows errors in the transmission to be detected and corrected. Typically, better than 1E-6 is needed after FEC for the recovered data to be useable. The process of adding this redundant information is known as channel coding. Convolutional coding and block coding are the two major forms of channel coding. Convolutional codes (such as Sequential and Viterbi) operate on serial data, one or a few bits at a time. Block codes (such as Reed Solomon) operate on relatively large (typically, up to a couple of hundred bytes) message blocks. Errors as a result of noise or from intermittent causes can have the same BER but very different effects. Errors that are spread out are due to noise while errors that are grouped together are due to intermittent problems (example: loose connector, terrestrial interference). SPACED ERRORS BURST ERRORS 1101101011010011100 1111101011101101101
The above example shows the same error rate but the burst errors are more difficult to correct. Convolutional codes are best used for spaced errors while block codes are best used for burst errors. Viterbi coding is one of two types of coding algorithms used with convolutional encoding. The other type is Sequential coding. Sequential coding has the advantage that it can perform very well with long-constraintlength convolutional codes, but it has a variable coding time. Viterbi decoding has the advantage that it has a fixed coding time. The most common variant used in satellite networks is rate 1/2 convolutional coding using a code with a constraint length K=7 (industry standard). With this code, you can transmit binary or quaternary phaseshift-keyed ( BPSK or QPSK ) signals with at least 5 dB less power than you would need without it. This is very useful in reducing transmitter and/or antenna cost or permitting increased data rates given the same transmitter power and antenna sizes. The constraint length (K=7) refers to the size of the shift register used when performing the convolution. This means that the constraint length represents the number of k-bit shifts over which a single information bit (data bit) can influence the encoder output.
Rev D 07/04
45
Forward Error Correction (continued)

But of course there is a tradeoff when adding FEC to the data stream. The same data rate, with rate 1/2 convolutional coding requires twice the bandwidth of the same signal without it, given that the modulation technique is the same. FEC rates are typically stated as a fraction, for example 1/2 or 3/4. - 1 data bit enters the encoder 2 data bits are transmitted - 3 data bits enter the encoder 4 data bits are transmitted For example, given a data rate of 256kbps, when QPSK modulation is applied the symbol rate becomes 128ksps since QPSK transmits 2 bits per symbol. If Viterbi FEC is applied, this means adding one correction bit for every data bit and the symbol rate becomes 256ksps. If Viterbi is selected instead of , the symbol rate would be 170.667ksps If Viterbi 7/8 is selected instead of or , the symbol rate would be 146.286ksps By concatenating convolution coding and block coding an even greater BER can be achieved over convolution coding alone. Below is an example using Viterbi . Notice the improved BER.
Rev D 07/04
46

Reed Solomon takes a block of digital data and adds extra redundant bits. The advantage of using Reed Solomon is that the probability of an error remaining in the decoded data is usually much lower than the probability of an error if Reed Solomon is not used. This is often described as coding gain. Viterbi / Seqential coding is sometimes referred to as an inner code while Reed-Solomon is sometimes referred to as an outer code. Viterbi - Reed Solomon can add an additional 2dB coding gain over Viterbi alone. Below is an example using Viterbi . Notice the improved Eb/No.
Example: A satellite communication link is designed to operate at a Bit Error Rate (BER) of 1E-9 (no more than one bit in one million) is received in error). This can be achieved by boosting the power of the transmitter OR by adding Reed Solomon (or another type of FEC). Reed Solomon allows the system to achieve this target BER with a lower transmitter output power. The power saving given by Reed Solomon (in decibels) is the coding gain.
Rev D 07/04
47

A Reed Solomon code is specified at n,k with s-bit symbols. This means that the encoder takes k data symbols of s bits each and adds parity symbols to make an n symbol codeword. There are n-k parity symbols of s bits each. A Reed Solomon decoder can correct up to t symbols that contain errors in a codeword, where 2t=n-k. The following diagram shows a typical Reed-Solomon codeword (this is known as a Systematic code because the data is left unchanged and the parity symbols are appended):
Example: A popular Reed-Solomon code is RS(255,223) with 8-bit symbols. Each codeword contains 255 code word bytes, of which 223 bytes are data and 32 bytes are parity. For this code: n = 255, k = 223, s = 8 2t = 32, t = 16 The decoder can correct any 16 symbol errors in the code word: i.e. errors in up to 16 bytes anywhere in the codeword can be automatically corrected. Given a symbol size s, the maximum codeword length (n) for a Reed-Solomon code is n = 2s 1 For example, the maximum length of a code with 8-bit symbols (s=8) is 255 bytes.
Rev D 07/04
48

Reed Solomon is complex, as there are many n,k variations available. To overcome this there are several standards such as Intelsat IESS-308 and DVB (for example). Typically, 219,201 is used in IDR applications while 126,112 is used in IBS applications. Other factors may include the data rate, framing, and modulation types. For example, the DMD20 automatically sets the correct R-S code rate for IDR/IBS open network operation in accordance with the data shown in the table below. In Closed Net Mode, the DMD20 allows the following N and K settings: (126, 112), (219, 201), (194, 178), (225, 205). 219/201 uses less bandwidth than (for example) 126/112. The DMD20 allows for interleaving depths of 4 or 8 R-S Blocks. For Intelsat Network Modes, the DMD20 automatically sets the interleaving depth to 4 for QPSK or BPSK or 8 for 8PSK. In Closed Network Mode, the interleaver depth can be manually set to 4 or 8, and in DVB Network Mode, the DMD20 automatically sets the interleaver depth to 12.
Rev D 07/04
49
Turbo and Trellis Coding

As a result of the wide acceptance of convolutional codes, there have been many advances to extend and improve this basic coding scheme. This advancement resulted in two new coding schemes, namely, Trellis coded modulation (TCM) and Turbo codes. TCM adds redundancy by combining coding and modulation into a single operation (as the name implies). The unique advantage of TCM is that there is no reduction in data rate or expansion in bandwidth as required by most of the other coding schemes. TCM is only used with 8PSK and Turbo modulation. Recently, a near channel capacity error correcting code called turbo code was introduced. This error correcting code is able to transmit information across the channel with arbitrarily low (approaching zero) bit error rate. This code is a parallel concatenation of two convolutional codes separated by a random interleave.
One big advantage of turbo coding is that there is enough code structure (from the convolutional codes) to decode efficiently. Turbo Coding offers superior improvement in EB/No performance and reduced bandwidth as compared to the traditional Viterbi Reed Solomon Concatenated coding. Additionally, the Turbo Codec functions through a process called soft decision. The data is processed through the decoder, which loops it back again until the decoder determines it has reached a confident solution. This concept of feeding the output back into the input is similar to a turbo engine, and hence the name Turbo Codes.
Rev D 07/04
50
Turbo Coding (continued)

The graph bellows shows an example of the performance (coding gain) of Turbo coding. In relation to standard Viterbi FEC, Turbo coding offers a 3.5dB to 4dB coding gain advantage with no increase in bandwidth as well as reduced coding delay. Different manufacturers define their Turbo codes differently. For example, Radyne Comstream uses .795 and .493. Comtech refers to their codes as and . The Radyne Comstream DMD20 supports both Turbo coding schemes allowing interoperability between manufacturers.
It should be noted that there is greater throughput delay using Turbo as apposed to Viterbi-Reed/Solomon. This is due to the reiterative nature of the Turbo error correction algorithm.
Rev D 07/04
51
Symbol Rate and Bandwidth Revisited

Symbol rate is the rate at which data is clocked out of the modulator and is a sum of the user data rate, channel coding (FEC), and m-phase modulation index. The table below shows a data rate and bandwidth example using different convolutional FEC and modulation types. CODING Uncoded R1/2 R3/4 Uncoded R1/2 R3/4 Uncoded R1/2 R3/4 R7/8 MODULATION DATA RATE BPSK 128kbps BPSK 64kbps BPSK 96kbps QPSK QPSK QPSK 8PSK 8PSK 8PSK 16QAM 256kbps 128kbps 192kbps 512kbps 256kbps 384kbps 896kbps MATH Dr = Sr Dr x 2.0 Dr / 0.75 Dr x 0.5 Dr = Sr Dr / 1.5 Dr / 4.0 Dr / 2.0 Dr / 3.0 Dr / 7 OCCBW 128kHz 128kHz 128kHz 128kHz 128kHz 128kHz 128kHz 128kHz 128kHz 128kHz
To calculate the allocated bandwidth multiply the occupied bandwidth by 1.4 (128 x 1.4 = 179.2 kHz) When using 8PSK or 16QAM modulation with multiple carriers, it is advisable to add another 10% to the allocated carrier bandwidth in order to reduce the effect of adjacent channel interference. If more than 4 carriers are in the network it is also advisable to unequally space the carriers to avoid high levels of 3rd order intermodulation in the HPA and LNA of the transmit and receive earth station.
Rev D 07/04
52
So why doesnt everyone use 16QAM ?
The main reason is that as we increase the modulation index we need more energy per symbol to allow the demodulator to accurately extract the user data. Satellite operators sell capacity based on a PEB (power equivalent bandwidth) ratio. Whichever is the larger of the resource (as identified on the link budget) is what the customer will pay for. If, for example, 16QAM required more power than is available for that given bandwidth, it would be uneconomical.
Rev D 07/04
53
8PSK an example
In the first example below comparing 8PSK Turbo Code at rate 2/3 to QPSK at rate 5/6, the C/N threshold is slightly better, allowing the same satellite power and receiving antennas to be used. The information throughput has been increased by a full 35%, as a result of switching to 8PSK.
The second example shows that simply switching to 8PSK modulation and retaining RSV coding would still increase the throughput (30%) but degrade the C/N threshold by 1.7 dB. This may be enough to make an existing network unreliable or even unusable. Comparing the two examples, we see a Turbo Coding gain of around 2.0 dB over RSV.
Rev D 07/04
54
Upconverters
Satellites transmit and receive at frequencies above 1GHz because of propagation and other restrictions. Therefore frequency conversion equipment is required between the satellite modem Intermediate Frequency (IF) and the earth station antenna Radio Frequency (RF). The IF is routed through RF frequency upconversion equipment which consists of mixers and local oscillators. These mixers translate the IF into the final RF frequency range for amplification.
On the input side, upconverters accept a modulator IF frequency in the range of 52 MHz 88 MHz (70) 118 MHz 172 MHz (140) 950 MHz 1750 MHz (L)
On the output side, the upconverter sends a small gain RF signal to the Solid State Power Amplifier (SSPA), Traveling Wave Tube (TWT), or Klystron High Power Amplifier (HPA). Typically this frequency range is 5.85 GHz 6.425 GHz (C Band) 5.85 GHz 6.65 GHz (Ext C Band) 6.725 GHz 7.025 GHz (Palapa C Band) 14.0 GHz 14.5 GHz (KU Band) 12.75 GHz 14.5 GHz (Ext KU Band)
Rev D 07/04
55
Upconverters (continued)
Frequency conversion is performed by devices called mixers. Mixers accept an LO (Local Oscillator) signal plus an input signal (IF) and multiply them together to produce the output (RF).
The mixer outputs, which are produced in equal amounts, are the difference and sum of the frequencies of the L.O. and the input signal (IF). The L.O. is either high-side injection (LO = RF + IF) or low-side injection (LO = RF - IF), depending on design trade-offs between the L.O. and filters. If high-side, the mixer inverts the spectrum from input to output. The upconverter input level will typically be 30dBm and the output 0dBm (30 dB of gain). Overdriving the upconverter can lead to compression of the signal, spurious emissions, or in some cases cause severe, even fatal damage to the input stage. The main performance characteristics of upconverters are: 1. Bandwidth the RF bandwidth which defines the capability of the converter to cover the RF band by adjusting the L.O. frequency. The IF bandwidth which defines the capability of the converter to cover all bandwidths of the IF modulator (70 MHz, 140 MHz) 2. Frequency Agility The frequency and channel capacity plan is often altered as traffic through the satellite is changed (increase/decrease). Upconverters which can be adjusted in frequency over the entire RF bandwidth are particularly useful.
Rev D 07/04
56
Upconverters (continued)
3. Linearity In SCPC systems a number of modulators may be frequency converted using only one upconverter. When combining inputs to one upconverter, intermodulation must be taken into consideration and kept to a minimum. This is achieved by using an upconverter with good linearity characteristics. 4. Frequency Stability The frequency tolerance of SCPC carriers is most stringent (+/- 250 Hz) and to realize this specification the L.O. in the upconverter should use a synthesized oscillator to ensure frequency stability.
Rev D 07/04
57
Amplifers
Once the baseband signal has been modulated and upconverted from IF to RF, it is ready for transmission to the satellite.
The signal levels are still relatively low at this point (around 0dB) therefore some form of signal amplification must be employed. The transmitted signal level from an earth station is governed by two factors, antenna gain and the amplifier power. The amplifier power requirement can range from less than 1 watt for a VSAT to several kilowatts for a major earth station. Amplification can be achieved by an intermediate power amplifier (IPA) followed by an HPA (high power amplifier). The HPA is normally housed in an air-conditioned environment to dissipate heat, and co-located with the antenna to reduce losses.
Rev D 07/04
58
Amplifers (continued)
There are two types of HPAs Traveling Wave Tube Amplifier (TWTA) Klystron Amplifier
The TWTA is intrinsically a wideband device covering the entire useable bandwidth of the satellite (i.e. 500 MHz or more), therefore it appears to be the ideal power amplifier for earth stations, since it allows several carriers to be transmitted simultaneously irrespective of the allocated frequencies. The simultaneous transmission of several carriers in the same TWTA produces intermodulation which increases as the operating point of the TWTA approaches saturation. This leads to an output backoff (OPBO) of the TWTA operating point in relation to saturation. TWTAs are the most common type of amplifier operating with power ranging from 100 up to 800 watts. N of carriers 2 4 8 16 32 Output Backoff 5 dB 8 dB 10 dB 12 dB 15 dB
For example, a 100 watt TWTA with 4 carriers will only deliver 25 watts to the antenna due to OPBO requirements. The Klystron is essentially a bandpass device, approximately 40 MHz for 6 GHz and 80 MHz for 14 GHz amplifiers. This narrow bandwidth is sufficient for frequency modulated carriers (example: analogue video) but is inadequate for large numbers of phase modulated carriers. Despite the drawback of narrow bandwidth, Klystrons are generally more economical to operate than TWT amplifiers due to high efficiency, simple power supplies, and long mean time before failure (MTBF). It should be noted that Klystron amplifiers are not available for lower power outputs (i.e. less than 800 watts).
Rev D 07/04
59
Amplifers (continued)
Solid State Power Amplifiers (SSPA) are used for lower power applications, although technology advances now enable SSPAs to replace TWTAs in many applications. SSPAs feature good efficiency (i.e. lower power consumption relative to output power), high reliability, low cost, and are relatively small in size. SSPAs have a sharper saturation curve which means they can be operated with less backoff than HPAs for multi-carrier operation. VSAT SSPAs are normally fitted to the antenna as part of the outdoor unit (ODU) and are encased in a weatherproof housing, usually including the upconverter.
Rev D 07/04
60
Transceivers
Most small earth stations and some older VSAT systems use an integrated package which includes a frequency / power agile transceiver.
Depending on the antenna manufacturer, many transceiver packages can be mounted on the antenna itself. These transceivers typically consist of: Upconverter SSPA LNA Downconverter
Rev D 07/04
61
Transceivers (continued)
Transceivers are matched with 70 or 140 MHz modems and provide a convenient outdoor package with frequency agile converters covering both the uplink and downlink frequencies. A significant advantage of the transceiver system is that the manufacturer can supply a 1+1 redundant package. Power output can range up to 16W for KU and 40W for C.
Block Up Converters
Recently, many VSAT stations and small terminals have begun to take advantage of the recent development in block upconverter (BUC) technology. These small packages are used in conjunction with L-Band modems, such as the Radyne Comstream DMD20 shown below, and use a single stage to translate the output IF from the modulator into RF.
Rev D 07/04
62
Block Up Converters (continued)

BUCs offer a three-fold advantage Cheaper to manufacture since they only perform one stage of frequency conversion and they do not have their own stabilized L.O. Do not require their own power supply (lower power models) Much simpler to install than a traditional transceiver package
DC power is supplied to the BUC via the transmit IFL cable from the modulator (depending on the current sourcing capabilities of the modulator). Stabilized 10 MHz reference is supplied to the BUC via the transmit IFL cable from the modulator. Higher power BUCs (8W, 16W etc). may require an external DC supply to be installed at the indoor modem rack if the DC current requirements exceed the rating of the modem power supply. The DC from this external source would be combined onto the IFL with the L Band signal and 10 MHz reference from the modem.
Rev D 07/04
63
Block Up Converters (continued)

Important Application Note: BUC IFL Cable Unlike traditional tranceivers where the 70/140 MHz modulator signal was the only signal fed to the RF via the interfacility link (IFL), BUCs require the following: L Band modulator signal DC power 10MHz Reference
These characteristics result in a much more stringent requirement for the IFL than was used by traditional transceivers. For uplinks where a BUC is to be installed, Radyne ComStream strongly recommends that customers purchase the 100 Meter ILF kit (RCS P/N 031128-001) for BUC installations. This IFL cable kit includes Radyne ComStream approved IFL cables and connectors. The connectors have been professionally installed and tested at the factory. Use of this kit will eliminate many of the problems that can occur with using customer-supplied (or pre-existing) cable. These problems typically lead to poor or intermittent operation of the system. If the customer would prefer to use an alternate supplier of cable, the IFL cables must meet or exceed the following parameters: Impedance: 50 ohms, +/- 3 ohms Max center conductor DC resistance / 100 meters: 0.45 ohms Max shield DC resistance / 100 meters: 0.54 ohms Max signal attenuation @ 10 MHz / 100 meters: 1.3 dB Max signal attenuation @ 2 GHz / 100 meters: 19.6 dB
Rev D 07/04
64
Intermodulation
Intermodulation distortion occurs when the non-linearity of a device or system with multiple input frequencies causes undesired outputs at other frequencies. In a communications system, this means that signals in one channel can cause interference with adjacent channels. As the spectrum becomes busier and the channels become more tightly spaced, minimizing intermodulation distortion becomes more important. The most significant intermodulation products are third-order intermodulation products because they are closer to the fundamental signals than second-order products, filtering them out is difficult, and they have greater energy than higher order products.
If the intermodulation signal level is too strong, the result will be an increase of the noise floor on the satellite transponder, which may result in interference. Intermodulation products should not be greater than approximately 23dB above the noise floor of the RF (TWT, SSPA) output. Intermodulation is not limited to applications where multiple carriers are combined. Overdriving the input to a transmitter, even when only one input signal is involved, will cause intermodulation that will appear as ears on the fundamental frequency. This occurs as a result of the amplitude to phase modulation conversion.
Rev D 07/04
65
Intermodulation (continued)
In formulating a link budget it is normal to add a certain amount of margin in the calculation to allow for the difference between theoretical and actual equipment performance. As the amplifier enters its saturation stage the effects of intermodulation become noticeable. When two carriers are simultaneously amplified (f, f) intermodulation products are generated as shown below.
Rev D 07/04
66
Due to the effect of intermodulation, power amplifiers must be operated in a sufficiently linear region, well below saturation, when they are used for multiple carrier operation.
SSPAs used in transceiver packages, as well as BUCs, are more linear than TWTA or other types of tube amplifiers. As a consequence they can operate with a smaller OPBO and are more efficient in the digital domain. N of carriers 2 4 8 Tube Amplifier 5 8 10 SSPA 3 6 9
Rev D 07/04
67
In PSK modulation, the AM/PM (input amplitude / output phase) intermodulation is seen as ears which appear each side of a single carrier. If they are transmitted to the satellite, they will increase the noise floor around the carrier and exceed the allocated spectral mask. Intermodulation is always present in a PSK modulated carrier, but only becomes a problem when either the upconverter or the power amplifier are overdriven.
Rev D 07/04
68
Power Amplifier Considerations

Power amplifiers, when used for single carrier operation, can be driven to their 1dB compression point without damage. Their power output is linear up to this point (i.e. a 0.5dB increase on the input will translate to an increase of 0.5dB on the output). The 1 dB compression point of an amplifier is defined as the limit of linear operation. It is the point where increasing the RF input signal results in a gain compression of 1dB. Most amplifiers are rated to operate at the 1dB compression point however the majority of satellite systems are not designed to operate the power amplifier in this region.
Rev D 07/04
69
Frequencies
Of the bands allocated for satellite links between fixed earth stations, those given in the table below are already in use or are likely to be used soon for commercial applications. UPLINK (Tx) 5.85 6.425 5.85 6.65 6.725 7.025 14.0 14.5 Europe 11.7 12.25 12.5 12.75 12.75 14.5 17.34 18.10 27.5 29.5 29.5 30 10.7 12.75 12.5 12.75 17.7 19.7 19.7 20.2 Std. KU Band USA High KU Band Extended KU Band Ka Band Ka Band Ka Band DOWNLINK (Rx) 3.7 4.2 3.4 4.2 4.5 4.8 10.95 11.75 REFERENCE C Band Extended C Band Palapa C Band Std. KU Band
The difference between an uplink frequency and a downlink frequency on a satellite is known as the translation frequency. The 6GHz/4GHz frequency bands were the most heavily used for the first 15 years of satellite communications because they offered the fewest propagation problems and, historically, RF components for these bands have been readily available. It is much more difficult to produce a Ka Band component than it is to produce a C Band component because of the tight tolerance required for the smaller wavelength.
Rev D 07/04
70
Frequencies (continued)
Attenuation of radio waves is a major factor in the selection of an RF frequency. Attenuation is not a significant factor on satellite paths with an elevation of greater than 25 degrees, for frequencies below 10 GHz, because of the short path length through the earths atmosphere, and the low attenuation per kilometer at these frequencies. Sky noise is also lower at 4 GHz therefore it is possible to build receiving systems with lower noise temperatures at 4 GHz then 11 GHz. Satellite systems also share the frequency spectrum with other satellites, and in most frequency bands, with terrestrial radio and microwave. This is one of the most common causes of interference on a satellite, and occurs when an unwanted terrestrial born signal (TI) is inadvertently added to the carrier prior to amplification.
Rev D 07/04
71
Frequency Translation Example
In the example above an L Band modulator feeds the modulated signal to the KU Band BUC, co-located with the antenna. Along with the L Band signal, the following is also supplied to the BUC from the L Band modulator. 10 MHz reference signal 24vdc
All three signals (L Band, 10 MHz, 24vdc) are fed to the BUC using LMR400 IFL cable. For a standard KU Band BUC, the L.O. frequency which is mixed with the L Band signal from the modulator is 13050 MHz. This is a common L.O. frequency used in the KU Band BUC, although other L.O.s may be used. 950 MHz + 13050 MHz = 14000 MHz The RF signal is then fed to the satellite transponder where it is received and translated to a downlink frequency. In this example the L.O. in the satellite transponder is 1250 MHz. 14000 MHz 1250 MHz = 12750 MHz
Rev D 07/04
72
Frequency Translation Example (continued)

The downlink RF signal is received by the LNB and downconverted from RF to L Band. In this example the L.O. in the LNB is 11300 MHz. 12750 MHz 11300 MHz = 1450 MHz This L Band signal is presented to the input of the demodulator.
Rev D 07/04
73
Chapter 3
The Earth Station Antenna
The Antenna Beamwidth Tracking Systems Antenna Types Feed & Polarisation Waveguide EIRP
Rev D 07/04
74
The Antenna
The earth station antenna is available in various sizes, configurations, and materials. Ranging from 0.45 metres to 33 metres, the antenna is the most conspicuous and often the most impressive subsystem of an earth station. The antenna of an earth station is common to transmission and reception and is essentially a passive amplifier. The gain (amplification factor) of the antenna is calculated as a relative function of its size (aperture) and the wavelength of the signal to be amplified. Many factors govern the performance of an antenna, not least being the surface accuracy, rigidity of the main reflector, and the placement of the feed. The features of an earth station antenna are common to transmission and reception and must adhere to the following test related standards: High gain for transmission and reception This requires reflectors which are large in relation to the wavelength and have an accurate reflector contour. C Band antennas are typically larger than KU. Low level of interference (for transmission) and of sensitivity to interference (for reception) This requires a very directional gain envelope with low levels outside the main lobe (low off axis side lobes) Radiation with high polarization purity (xpol)
Rev D 07/04
75
Mechanical Considerations
From a mechanical point of view, the antenna performance calls for great structural integrity. The surface accuracy of the main reflector must be approximately 1/50th of the wavelength (i.e. 1mm for 6GHz). This accuracy must be maintained under all operating conditions (i.e. wind, rain, temperature fluctuations). Typically, a tracking system is not required for any antenna under 4.0m in diameter. This is due to the fact that these smaller antennas have a beamwidth greater than the movement of the satellite.
Rev D 07/04
76
Antenna Beam Width

The larger the antenna, the smaller the main lobe (beamwidth). In the case of an Intelsat Standard-A 30m antenna the accuracy must be approximately 0.015 degrees, which requires an automatic tracking device to control the azimuth and elevation adjustment of the antenna.
Rev D 07/04
77
The Antenna System

The antenna system consists of the following parts: Mechanical system comprising main reflector, back structure, pedestal or mount assembly, and for automatic tracking antenna, the driving gear or servo system. The primary source comprising the illumination horn, the associated reflector sub-assemblies, and non-radiating components (couplers, diplexers etc.).
Rev D 07/04
78
Tracking Systems
Larger antennas require tracking systems to ensure that the narrow beamwidth of the antenna remains aligned to the satellite. The two main methods of automatically tracking satellites are: Step Track this method employs a beacon receiver to sample a satellites signal strength. The antenna is systematically moved to various points around the nominal satellite position, until the position producing the highest signal strength is found. This process is continually repeated to keep the satellite signal strength at its peak. A disadvantage of this method is that tracking accuracy can be directly affected by changes in the incoming signal due to atmospheric changes, beacon instability etc. The signal level will also vary as the antenna peaks the signal. Step track is the most common form of tracking for medium and small earth stations. Some step track systems incorporate a predictive algorithm and other forms of software control, which learn the path of the satellite and can therefore follow it even if the signal is lost for a short time. Monopulse this method has an advantage over the step track method. The monopulse system makes use of special microwave couplers inserted in the antenna feed. As long as the antenna is pointed in the optimal direction there is no need to move the antenna. Several microwave couplers make it possible to receive and compare the beacon frequency simultaneously. As long as the signal strength is equal in each horn, the antenna is locked in its direction towards the satellite. For example, if the tracking system uses a north/south/east/west beacon tracking configuration and the north feed loses 3 dB while the south feed gains 3dB, the system would know how to precisely adjust the antenna back to optimal pointing, whereas the step track system would search around for the best signal.
Rev D 07/04
79
Antenna Types
The most frequently used types of antennas are parabolic antennas and include: Cassegrain Gregorian Prime Focus Offset
Cassegrain and Gregorian antennas make use of a dual reflector system fed by a primary radiator located at the focus of the system. Main earth station antennas are predominantly of the dual reflector axi-symmetric Cassegrain type although there are also Gregorian dual reflector designs. C Band antennas tend to be larger than KU antennas because antenna gain is a function of frequency. Radiated power, or Effective Isotropic Radiated Power (EIRP) is a function of both antenna gain, HPA power, and losses caused by filters and waveguide runs, so designers can achieve the required EIRP by trading HPA power for antenna gain (or vice versa). Advantages of a higher gain antenna are that it benefits both the transmit and receive link budgets, and a single antenna may support many HPAs, each operating to a different transponder.
Classification of Earth Station Antennas

Earth stations that operate in the 6/4 GHz and 14/12 GHz bands are often classified according to the size of the antenna. Large earth stations antenna approx. 15 to 33 metres Medium earth stations antenna approx. 7 to 15 metres Small earth stations antenna approx. 3 to 7 metres Very Small Aperture Terminals (VSAT) antenna approx. 0.6 to 3 metres
There are a number of other classifications used including Standard A and B, Standard F, Standard L and M are examples. These classifications relate to the performance of the station, which is often related to the size and/or purpose of the station. The term VSAT is commonly misinterpreted to denote small antennas. In fact this term is used to denote the relatively small carriers transmitted from these stations.
Rev D 07/04
80
Cassegrain Antenna
The Cassegrain antenna is a rear-fed antenna which provides a convenient location for the complete feed system.
The reflector system consists of a main reflector (which is normally a parabolic) and a secondary reflector called a sub-reflector. A Cassegrain antenna has a hyperbolic sub-reflector.
Rev D 07/04
81
Cassegrain Antenna (continued)

Parabolic antennas that are circularly symmetric about the z-axis are termed axis-symmetric antennas. They may consist of a single reflector or, as in the case of the Cassegrain antenna, multiple reflectors. The ability to modify the sub-reflector gives two advantages: Reduction in spillover Uniform distribution of energy on the reflector resulting in an improvement in antenna efficiency
Cassegrain antennas exhibit lower noise temperatures than front-fed antennas due to the fact that the spill-over radiation from the primary feed is directed towards the sky, whereas with front-fed antennas spill-over is directed to / collected from the ground. Cassegrain antennas suffer from a number of disadvantages: Partial blockage of the main reflector caused by the sub-reflector Direct radiation of the primary feed outside the sub-reflector diameter (spillover radiation) increases the side lobes of the antenna pattern Sub reflector struts are normally placed in the radiation area of the main reflector, causing a scattered radiation (increased side lobes) Blockage by the sub-reflector causes shadowing of the antenna reflector and so decreases antenna gain for transmit and receive
For the reasons stated above, it is very difficult to provide an efficient Cassegrain antenna smaller than 3 meters in diameter.
Rev D 07/04
82
Gregorian Antenna
Gregorian antennas use an ellipsoidal sub-reflector as opposed to the hyperbolic sub-reflector of a Cassegrain antenna. The radiation from the feed hitting the sub-reflector intersects after reflection, but before hitting the main reflector surface. As a result of this design, the structure of the Gregorian antenna cannot be as compact as that of the Cassegrain type. An advantage of using the ellipsoidal sub-reflector is that the feed and subreflector edges are subject to less radiation and consequently less interference is caused. This method of illumination, along with Cassegrain, is commonly found on medium and large earth stations antennas.
Rev D 07/04
83
Antenna Sub-Reflectors
Multiple reflector systems usually use a parabola as the primary reflector and either a hyperbola or an ellipse as the secondary reflector. This is known as a double-reflector system. The secondary reflector is commonly known as the sub-reflector. The feed is positioned at the focus of the sub-reflector nearest to the main reflector, which in turn illuminates the main reflector. Dual reflector systems have the same integral parts as their axis-symmetric counterparts. The feed spillover is directed towards the cold sky.
Rev D 07/04
84
Prime Focus Antenna

Most antenna systems, either parabolic with front feed or Cassegrain / Gregorian types are axisymmetrical. A prime focus antenna (below) gathers the reflected RF in a feedhorn, which is located directly at the focal point.
For axisymmetric antennas, blockage by the feed and associated components, including the feed support structure, causes shadowing of the antenna reflector and so decreases antenna gain.
Rev D 07/04
85
Offset Antenna
Offset antenna are typically used for low power, small and micro station antenna. An offset antenna is a section of a prime focus antenna normally taken 22.5 degrees above the centre line. This type of antenna has significant advantages. The feed is located at the focus of the main parabola but is tilted towards the centre of the main reflector, and is not in the line of sight between the parabolic reflector and the satellite. Therefore there is no blockage of the signal. The offset antenna feed angle also provides better shielding from the hot earth (noise).
Rev D 07/04
86
Antenna Types - Example

Prime Focus and Offset Feed Antenna (all three are looking at the same satellite)
Offset Feed
Cassegrain
Rev D 07/04
87
Antenna Feed
The antenna feed is generally a corrugated, cone-shaped horn, which is placed in the focal point of the antenna (unless a sub-reflector system is used). The feed position is critical to the antenna performance. If the feed is placed too close to the reflector, it will only see part of the reflector. This would cause a reduction in gain of the antenna. If the feed is placed too far from the reflector it will not only see the entire reflector but also the area surrounding the reflector. The result of this would be an increase in system noise. The front of a feed horn is typically covered with a thin film of mylar (a special RF transparent plastic) to keep out moisture and other foreign objects, which would disturb the RF.
Efficiency
If the reflector is in perfect shape and the feed is ideally positioned, then theoretically the antenna would be 100% efficient. In practice an antenna is only approximately 66% efficient, although some more recent dual offset antennas have managed to achieve 70 to 75% efficiency. The larger the antenna, the more difficult it is to achieve efficiency.
Rev D 07/04
88
Polarisation
Most antennas radiate in either linear polarization or circular polarization. Both of these polarization types are elliptical. Linear polarization is obtained by the transmission of one plane of polarization. Perfect cross-polar isolation is achieved if two polarization planes are orthogonal, for example vertical and horizontal. The feed can be rotated to change the plane of polarization.
In the diagram above, if each of these geostationary satellites transmits linear vertical polarization, the earth station must set its polarizer off-vertical to receive from all except the satellite that is due South.
Rev D 07/04
89
Polarisation (continued)
With linear polarization, the earth station polarizer must be capable of being set at different angles in order to receive a vertically or horizontally polarized signal sent from a satellite located on the geostationary arc.
Rev D 07/04
90
The diagram below shows the energy fields in both planes. The red plane represents vertical and the blue plane horizontal. With a vertical polarized receiver, only E1 is received. With a horizontally polarized receiver, only E2 is received.
Unless perfectly aligned, the receiving dipole will receive signals from both E1 and E2. The maximum received energy will come from Er, with the plane of polarization equal to = 45 (if the energy of E1 = E2), and some other angle if E1 E2.
Rev D 07/04
91
In the case of circular polarization, the two states are described as righthand-circular (RHCP) and left-hand-circular (LHCP) relative to the direction of propagation. With circular polarization, the energy is transmitted by both dipoles simultaneously rather than one (as in linear polarity). In addition to the energy fields being equal (E1=E2) a 90 phase shift is introduced to E2. This will force the resultant energy wave (Er) to rotate counter-clockwise in the direction of propagation, producing LHCP. If E1 is phase shifted by 90 to lead E2, then the direction of rotation will be RHCP.
Rev D 07/04
92
Antenna components introduce some cross-polar distortion. Antenna blockage by components such as the feed and support arms, feed misalignment, reflector and sub-reflector imperfections all result in less than perfect cross-pole isolation. An antenna designed to transmit or receive a wave of given polarity can either transmit or receive in the orthogonal polarization, thus enabling two simultaneous links to be established at the same frequency. This concept is known as orthogonal frequency re-use. An important characteristic of the radiation pattern of an antenna is its polarization purity. The RF wave radiated by any antenna contains a co-polarized component (wanted component) and a cross-polarized (unwanted component). The radiation of an antenna, in any direction, can be described by a copolar radiation diagram and a cross-polar radiation diagram. The ratio of cross-polar component relative to the level of co-polar component should be greater than 30dB to minimize the effect of interference on the satellite. Two other names for this ratio are isolation and cross-polarization discrimination.
Rev D 07/04
93
Orthogonal Mode Transducer

An orthogonal mode transducer (OMT) is part of the antenna feed that combines or separates orthogonally polarized signals. An OMT is required if the antenna transmits and receives, or receives or transmits more than one channel on different polarizations. Orthogonally polarized signals are horizontal / vertical (linear) or left-hand / right-hand (circular). OMTs offer approximately 40dB discrimination between polarizations. This is known as cross-polar discrimination (XPD). OMTs operate at RF and may be required to transmit high powers, therefore they are normally constructed using waveguide technology. They have a common port to connect to the antenna feed, and two or four other ports for the individual signals. If the OMT is to be used for transmit and receive purposes the waveguide ports are tuned to match the particular frequency range of use. When performing alignment, very small degrees of rotation must be used, as accurate alignment is critical to obtaining a good XPOL figure.
Rev D 07/04
94
Waveguide
Microwave frequencies present special problems in transmission that are not encountered at lower frequencies. The two-wire transmission line used in conventional circuits is inefficient for transferring electromagnetic energy at microwave frequencies. At these frequencies, energy escapes by radiation because the fields are not confined in all directions. Guided wave tube (waveguide) is the most efficient way to transfer electromagnetic energy. Waveguide is essentially a coaxial line without the center conductor. It is constructed from conductive material and may be rectangular, circular, or elliptical in shape. Advantages of using waveguide include Large surface area reduces copper losses Waveguide can handle more power than coax Much lower attenuation than coax
Disadvantages of using waveguide include Physical size is the primary lower-frequency limitation of waveguide. The width of a waveguide must be approximately a half wavelength at the frequency of the wave to be transported. For example, a waveguide for use at 1 megahertz would be about 500 feet wide. This makes the use of waveguides at frequencies below 1000 megahertz increasingly impractical. Waveguide is difficult to install because of its rigid, hollow-pipe shape. Special couplings at the joints are required to assure proper operation. Also, the inside surface of waveguide is often plated with silver or gold to reduce skin effect losses. These requirements increase the costs and decrease the practicality of waveguide systems at any other than microwave frequencies.
Rev D 07/04
95
Waveguide (continued)
The transfer of RF energy to and from the antenna feed horn has been described using various analogies, but is best described as a gas flowing through a pipe, the waveguide being the pipe. It is very important that the inner surface of the pipe and any joints are smooth so as not to impede ( attenuate) the flow of the signal. Copper is the most common material formed into a tube, which is matched to the frequency of interest. Circular tube (WC) is used to carry both polarities and rectangular (WR) or elliptical (EW) tube for single polarity energy fields.
Rev D 07/04
96
Rev D 07/04
97
Typical KU Band waveguide and their losses @ 12GHz are shown below:
WC (rigid) = 6.0 dB / 100m
WR (rigid) = 12.5 dB / 100
WR flexible = 20.5 dB / 100m
Heliax = 50 dB /100m
Rev D 07/04
98
Waveguide attenuation can be increased dramatically by Poor end fittings Damage (bends, kinks, dents) Foreign objects (nuts, bolts, metal shavings)
Moisture is one of the most common problems experienced in waveguide systems. Water can enter the waveguide via a damaged feed horn (damaged Mylar feed window), poor fittings (loose flange screws, missing o-ring, incorrect flange mate type) or cracked / damaged waveguide. Moisture will corrode the inside of the copper waveguide causing rough spots, which can lead to attenuation. Moisture also concentrates the energy and can lead to voltage arcing inside the waveguide. On medium and high power earth stations an air dehydration and pressurization system is frequently used to maintain a positive pressure of dry air in the waveguide and feed system. This system can also act as an alarm to indicate critical leaks, as any significant drop in pressure can activate an alarm.
Rev D 07/04
99
RF Connection
Given that waveguide is fragile and susceptible to losses, the ideal location to mount the transceiver would be directly to the transmit / receive ports of the OMT, as close as possible to the feed horn. If this mounting configuration is possible, the only losses that will be experienced will be through the feed system and connectors less than 0.5dB. This configuration is widely used in low power VSAT operations where the transceiver uses SSPAs, which are compact and lightweight.
Larger power amplifiers mounted at the feed may impede the performance of the antenna due to the weight and size of the transceiver system. The additional weight of the transceiver system may result in feed misalignment caused by the feed being pulled away from the focal point of the main reflector. Once aspect of mounting the transceiver package at the feed which is frequently overlooked, is the effect of wind forces on the feed support system. In some cases additional wind loading could move the feed away from the focal point, or in extreme cases move the antenna off of the satellite. This can result in a reduction in the system gain (fluctuating power levels both on the satellite and at the receiver) and a degradadation in the XPOL isolation purity, resulting in interference.
Rev D 07/04
100
RF Connection (continued)
Moving the transceiver away from the feed to overcome weight problems creates additional waveguide losses due to the extra distance the RF energy must travel from the transceiver to the feed. Medium and large sized earth stations using TWTAs or Klystron HPAs would normally be used with a rear fed antenna system such as Cassegrain or Gregorian. In this case, the HPA is installed in a cabin, which is located adjacent to the antenna and connected using longer runs of rigid and semi-flexible waveguide. These runs of waveguide can often be in lengths of 15 metres or greater, which contributes significantly to the 2 or 3 dB of loss encountered in such stations.
Rev D 07/04
101
Effective Isotropic Radiated Power (EIRP)

EIRP is the measure of power available from an earth station. It is the product of antenna gain and the input power at the antenna flange. It is expressed as a ratio relative to 1 Watt (dBW). The output power of the amplifier used in a particular system is dictated (primarily) by the EIRP requirement as identified in the link budget calculation. The following equation is used to calculate the maximum EIRP of an earth station: (10 log PA) + GA FL OL To determine this we first need to know 1) 2) 3) 4) 1) GA The gain of the antenna can be calculated as a function of the number of wavelengths captured by the area of the reflector. Using the following formula 10 log ( 9.9 * (D/) ) * eff where: = the wavelength in metres D = the diameter of the antenna in metres Eff = the efficiency of the antenna system Fortunately this is done for us by the antenna manufacturer. A typical 2.4m KU Band antenna has a transmit gain of 49 dBi. The larger the antenna, the greater the gain, therefore gain is directly proportional to the diameter of the antenna. GA the gain of the antenna in dBi PA the HPA output power in Watts FL typical feed losses in dB OL other losses in dB
Rev D 07/04
102
EIRP (continued)
2) 10 Log PA The amplifier power is readily available from manufacturers specifications. It is important to note that many tube amplifiers use a passive device called a circulator, which is fitted to the final output stage and attenuates the active device output power by up to 20%. A 650 watt TWTA, for example, may only provide 520 watts at the output flange. Smaller devices such as SSPAs and BUCs are easier to qualify. For example, the following sizes in KU Band are available. 1, 2, 4, 8, 16, 20, and 40 watts Watts 10 log PA 1 0 2 3 4 6 8 9 16 12 20 13 40 16
The resultant figures are in dBW 3) FL The RF power must be transported from the amplifier to the antenna feed horn through a waveguide system. Losses will occur in the waveguide and will be dependent on the type of waveguide used (rigid, flex etc.). Typically on small to medium earth station antenna this is estimated at 0.5 dB, since the RF is often installed close to the feed and in some cases, directly to it. On larger earth stations there may be longer waveguide runs as well as combiner and redundant switching networks through which the RF must pass. This can result in losses as high as 2 or 3 dB.
Rev D 07/04
103
EIRP (continued)
4) OL Other losses are related to the mechanical state of the antenna and its mount. VSAT and small earth stations may suffer from mis-pointing errors, therefore a typical figure of 0.5dB is given as OL. Some satellite provider link budgets include an EIRP degradation loss figure if the station is declared as non-tracking, in consideration of the effects of satellite movement. On a 3.7m antenna this is usually taken as 1 dB. With our benchmark 2.4m antenna, let us calculate the maximum EIRP when fitted with a 1 watt SSPA. ( 10 log PA ) + GA FL OL Maximum EIRP = 0 + 49.0 0.5 0.5 = 48 dBW With a 2 watt PA the maximum EIRP would be 51 dBW With a 4 watt PA the maximum EIRP would be 54 dBW
Rev D 07/04
104
EIRP (continued)
The satellite downlink beam coverage maps show contour lines where each line refers to a particular power level from the satellite. The lines are marked with EIRP values like 45dBW, 44dBW, 43dBW, 42dBW etc. in descending order from the maximum. The highest number towards the middle of the coverage map shows where the downlink beam is strongest and most easy to receive. In the centre of the beam a smaller receive dish on the ground is required. As you move further away from the beam peak, the beam becomes less powerful and a larger dish is required. In practice, an EIRP contour of 53 dBW could be produced by a satellite using a 200 watt transmitter (200W = 10log(200) = 23dBW) plus a satellite transmit antenna with a gain of 30 dBi (maximum).
Rev D 07/04
105
Block and Level Calculation

Based on what we have discussed, the following is an example of the calculation of maximum input power to the BUC from the modulator EIRP
We will assume that we are using the following 8 watt SSPA 2.4m KU antenna with gain spec of 49 dBi
Step One: First we must convert the SSPA rated power to dBm as follows: Power (dBW) = 10 log ( Pwatts) = 10 log ( 8 ) = 9.0 Step Two: Next we must convert the power in dBW to dBm as follows: Power (dBm) = P(dBW) + 30 = 39 dBm Step Three: We must obtain from the SSPA manufacturer the gain of the SSPA. In this example we will use a gain of 60 dB. Therefore the maximum input to the SSPA will be Maximum SSPA input = SSPA Power dBm - Gain SSPA = 39 60 = -21 dBm Step Four: Simple EIRP without feed or associated losses can be calculated as 10 Log (Pwatts) + Antenna_Gain = 10 Log (8) + 49 = 58 dBW
Rev D 07/04
106
Chapter 4
The Downlink
1101101011010011100 1111101011101101101
LNA / LNB Downconverters Transmit Reject Filter Demodulation Eb/No BER G/T
Rev D 07/04
107
The Downlink
The term downlink covers a complete system where a signal is transmitted from a satellite and then

captured on an earth stations main reflector focused into the feed horn passed through the feed system and OMT fed into an amplifier (LNA) or amplifier/downconverter (LNB/LNC) transported across an inter-facility link (IFL) to the demodulator demodulated and decoded back to a digital signal
LNA
In all downlink systems the signal gathered by the antenna is initially fed into a low noise amplifier (LNA). Signals originating from a satellite reach the earth station at a very low level, the path loss being around 200 dB. The low level signal must be amplified at the antenna while adding as little noise as possible, in order to increase the G/T of the earth station. The LNA is mounted as close as possible to the feed, since any signal loss between the two contributes directly to the noise performance of the receiver. Typically these amplifiers have between 20 and 60 dB of gain. The amount of gain is governed by the downconverter input level specification as well as signal loss due to the IFL length between the antenna and downconverter.
LNB / LNC
In many VSAT applications the amplified signal is sent several metres to the indoor equipment via an interfacility link (IFL) cable. Cable losses are significantly reduced if the signal is first down-converted. Both amplification and down-conversion take place in a combined LNA and block downconverter known as an LNB or LNC, located close to the antenna feed.
Rev D 07/04
108
LNB / LNC (continued)

The LNB/LNC downconverts a block of frequencies, from which the demodulator is free to select for decoding.
The LNB uses an internally or externally referenced local oscillator (LO), which is mixed with the incoming RF signal. RF INPUT GHz 3.7 4.2 10.95 11.7 11.7 12.2 12.2 12.75 LOCAL OSCILLATOR 5.15 GHz 10.00 GHz 10.75 GHz 11.30 GHz
Rev D 07/04
109
LNB / LNC (continued)

There are essentially three common frequency bands for downconversion using an LNB/LNC: LNC 70 MHz +/- 18 MHz (55 88 MHz) 140 MHz +/- 36 MHz (104 176 MHz) LNB L Band +/- 400 MHz (950 1750 MHz) The low-sided oscillator frequency (as used in KU Band signal conversion) is subtracted from the received RF frequency. The result is the L Band signal sent to the demodulator. 12,575,600 L.O. (11,300,000) = 1275.600 MHz A high sided calculation results in an inverted frequency 1,825,000 L.O. (5150,000) = - 1325.000 MHz Power to the LNB is normally supplied by the demodulator via the IFL cable. Frequency stability is important for LNBs used in digital applications. Typically, PSK demodulators have a signal capture range of only +/- 20 KHz. A standard dielectric resonant oscillator (DRO) LNB will drift by +/- 1 MHz or more. PLL LNBs are significantly more stable offering LO drift of +/25 KHz or better. DRO LNBs are commonly used for video or larger carrier reception such as greater than approximately 1.5 Msps when using QPSK.
Rev D 07/04
110
Downconverters
Stand alone downconverters are mainly employed in medium to large earth stations and are typically rack mounted.
The stand alone downconverter is intended for use where multiple carrier downlinks need to be established. Multiple converters in a rack can be protected by a redundancy switch which will allow protection should a failure occur. The main performance characteristics of downconverters are: Bandwidth the RF bandwidth which defines the capability of the converter to cover the RF band by adjusting the L.O. frequency. 2.The IF bandwidth which defines the capability of the converter to cover all bandwidths of the IF modulator (70 MHz, 140 MHz). Frequency Agility The frequency and channel capacity plan is often altered as traffic through the satellite is changed (increase/decrease). Downconverters which can be adjusted in frequency over the entire RF bandwidth are particularly useful.
Rev D 07/04
111
Downconverters (continued)
Linearity In SCPC systems a number of demodulators may be frequency converted using only one downconverter. Good linearity will ensure that intermodulation between carriers is kept to minimum. Frequency Stability The frequency tolerance of SCPC carriers is most stringent (+/- 250 Hz) and to realize this specification the L.O. in the downconverter should use a synthesized oscillator to ensure frequency stability.
Transmit Reject Filter

In all downlink systems the signal gathered by the antenna is initially fed into a low noise amplifier (LNA). Signals originating from a satellite reach the earth station at a very low level, the path loss being around 200 dB. The low level signal must be amplified at the antenna. If a high power signal is present, the LNA can suffer from high levels of intermodulation. On an earth station which transmits as well as receives, the orthogonally opposed port is used to transmit signals to the satellite. This high power transmit energy must be prevented from entering the LNA along with the low power receive signal. To suppress these unwanted and (potentially) damaging frequencies, a band pass filter is used. This filter, commonly called a transmit reject filter or low pass filter, blocks frequencies higher than 13.25 Ghz (in KU for example) by up to 80 dB. If a filter of this type was not installed, a receiving earth station would lose lock as soon as the transmitter was enabled.
Rev D 07/04
112
Demodulation
The purpose of the demodulator is to recover the baseband data from the IF signal which is received from the downconversion equipment. Once the demodulator is configured for the correct receive parameters, the IF signal is processed by demodulation decoding
The carrier recovery circuit detects and regenerates a signal that is both frequency and phase coherent with the original transmit signal. In this state, the demodulator is said to be in sync or locked to the carrier, and the demodulator circuitry is able to recover the binary data from the demodulated spectrum. The recovered symbols are then passed to the decoder.
Rev D 07/04
113
Demodulation (continued)
The decoder or data processor does exactly the opposite as the data processor in the modulator. The demodulated signal is converted to a digital signal and passed to the FEC decoders which recover clocking, remove any scrambling and/or differential coding that was applied at the modulator, then pass the baseband data stream to the data interface on the modem.
Eb/No
For digital receivers, the received signal strength is measured by a value known as Eb/No. Eb/No is a ratio of the energy in the carrier signal devoted to a single bit time interval (Eb) to the noise power density per 1 Hz of bandwith (No). If the energy is too low (i.e. the noise level too high) the decoder will not be able to identify if a 1 or a 0 was transmitted. A digital satellite receiver measures Eb/No the same way regardless of the data rate, coding, or modulation type. Different forms of modulation -- BPSK, QPSK, 8PSK, etc. -- have different curves of theoretical bit error rates versus Eb/No . These curves show the communications engineer the best performance that can be achieved across a digital link with a given amount of RF power.
Rev D 07/04
114
Eb/No (continued)
Rev D 07/04
115
Bit Error Rate (BER)

In telecommunication transmission, the bit error rate (or BER, as it is commonly called) is just one of many statistical measures of a communications link or channel. BER is the percentage of bits that have errors relative to the total number of bits received in a transmission, usually expressed as ten to a negative power. SENT BITS RECEIVED BITS 1101101101 1100101101
error BER = # of Wrong Bits = 1 = 0.1 # of Total Bits 10 BER is normally displayed in Scientific Notation. For example, a transmission might have a BER of 1E-6, meaning that, out of 1,000,000 bits transmitted, one bit was in error. If the BER is 1E-12, we expect that there will be one bit in error, on average, for every 1 trillion bits. The more negative the exponent, the better the BER. Better than 1E-6 is needed after FEC for the system to operate. Errors as a result of noise or from intermittent causes can have the same BER but very different effects. Errors that are spread out are due to noise while errors that are grouped together are due to intermittent problems (ex. loose connector). SPACED ERRORS 1101101011010011100 BURST ERRORS 1111101011101101101
The above example shows the same error rate but the burst errors are more difficult to correct. A BERT (bit error rate test or tester) is a procedure or device that measures the BER for a given transmission.
Rev D 07/04
116
BER (continued)
In order to measure a bit error rate, enough bits must go by and enough errors must occur to allow you to determine the statistics. There must be at least 30 errors to provide any reasonable confidence that the values for the mean and deviation are useful. Let use an example of1E-9, which implies we need to wait until 30 billion bits go by. At 1 Mega bit per second, that amounts to 30,000 seconds, or about 8 hours The basic BER test method is to send a known data pattern along the link, and to count how many errors occur on a bit-by-bit basis. The most common patterns to use are Pseudo Random Bit Sequences (PRBS), which are defined by the length of polynomial used to calculate the pattern. In general a long pattern is preferable because it will be more sensitive to the effects of noise, clock synchronization and pattern specific errors. It is important to note that link errors may manifest themselves in one type of pattern and not in another, therefore BER tests should be performed using more than one pattern.
Typical BERT Patterns

BERT Pattern 0's 1's Description
Test pattern consisting of all 0's that is used to test line coding Test pattern consisting of all 1's that is used to test alternating line volt and repeaters Pseudo-random repeating test pattern that consists of 2,048 bits Pseudo-random repeating test pattern that consists of 32,767 bits Pseudo-random repeating test pattern that consists of 1,048,575 bits Pseudo-random repeating pattern that is 8,388,607 bits long. This pattern is only available for an E3 interface. This is an ASCII text message consisting of all letters and digits. THE QUICK BROWN FOX JUMPS OVER THE LAZY DOG 1234567890
2^11 2^15 2^20 2^23
Fox
Rev D 07/04
117
G/T
The figure of merit of an antenna is an expression of the gain-to-noise temperature ratio. Gain is simply the gain (in dBi) of the satellite receive antenna. The temperature is the system noise temperature of the satellite receive system. This will primarily be the satellite LNA/LNB noise temperature plus the earth surface temperature (approx) which, of course, is fully visible to the beam from the satellite. G/T can be calculated using the following formula G/T = gain (dBi) 10 log (system noise temp K) Calculating the system noise temperature is not something that the average installer needs to determine, in depth, since typically this figure is offered by the manufacturer and is accurate (+/- 0.5 dB) within reason. Using a 2.4m KU Band antenna as an example, with a gain of 49 dBi, fitted with a 0.7 dB LNB, the G/T would be 49 10 log (51 + 30 + 15) = 29.177 dBK where: The LNB = 51K (0.7 dB) (1 dB = 75 Kelvin) The antenna noise @ 30 elevation = 30K The feed and OMT contribution = 15K
Rev D 07/04
118
Appendix A
C/N Measurements G/T Measurements Eb/No vs. C/N G/T Revisited The Decibel
Rev D 07/04
119
C/No Measurements
This picture shows a signal displayed on a spectrum analyser. The signal is at a frequency of 6.5 GHz as it is in the center of the screen and the center frequency of the analyser is 6.5 GHz. Each square of the display is a division, the analyser is set to 5 dB/DIV which means that each square is 5 dB. The resolution bandwidth (RBW) is set to 10 KHz, which is the width of the sweep filter inside the analyser. This filter is swept across the frequency span defined by the SPAN of the analyser, in this case 20 MHz. The RBW defines the amount of power captured by the analyser and displayed on the screen as a signal. Having an RBW too narrow will not capture all of the signal and will result in a less accurate display.
In order to measure the C/No we first 10 Log the RBW. 10 x Log (10,000) = 40 dB Next we add the size of the signal which is 5 squares or 25 dB. 40 dB + 25 dB = 65 dB. Now because of some complicated maths we subtract 2 dB to finish. 65 dB - 2 dB = 63 dBc/Hz. This means a C/No of 63 dB (carrier power) in 1 Hz bandwidth. This is a fast and easy way to measure. The long way and more accurate way involves measuring the noise floor power in 1 Hz and subtracting the power of the signal.
Rev D 07/04
120
G/T Measurements
Simplified G/T Measurement using Spectrum Analyzer
G/T = C+N Boltzmans Constant EIRP Beacon + Path Loss No where: - C+N/No - -228.6 - satellite EIRP - path loss measurement method: Video averaging ON Move marker to peak signal Set marker to NOISE AVERAGING Measure and record C+N/No G/T = 56.16 228.16 9 + 205 = 24 dB/K dB/Hz dBw/K/Hz dBw dB (measured using spectrum analyzer) (Boltzmans Constant) (supplied by space segment provider) (use 205.44 for KU and 196.70 for C)
Rev D 07/04
121
G/T Measurements (continued)

Below is information retrieved from the Eutelsat web site pertaining to W2, from which the beacon measurements were made.
Rev D 07/04
122

The measured value of G/T in the previous example is 24 dB/K, so how does this compare to a theoretical value ? We will use the following values for the theoretical calculations: Channel Master 1.2m KU Rx-Only Gain: 42 dBi (from manufacturer) LNB Noise Temp: 0.7 dB which equates to 51.0 K (from manufacturer) Antenna noise 24 K (at 30 degrees elev.) (from manufacturer) Feed and OMT noise negligible since LNB is bolted directly to feed
The formula for G/T is as follows: G/T = Ant Gain (dBi) 10 log (system noise temp K) G/T = 42 10 log (51 + 24) = 23.2 This is very close to our measured value in the previous page and verifies that the LNB is operating to spec and the antenna is correctly aligned to the signal.
Rev D 07/04
123

Below is the gain and noise temperature information for the receive-only antenna that was used to measure the beacon frequency on W2.
Rev D 07/04
124
C/N = Eb/No
Measure the C/N using a spectrum analyzer and using the table below add or subtract the result to calculate the expected Eb/No. Read the Eb/No as displayed on the demodulator. If there is a significant difference then there may be a phase noise issue. If the difference is common to all receive stations on the network then it is coming from the uplink. If the problem is isolated to one site then check the downlink equipment. Different modulation indices and FEC rates provide differing levels of energy per bit, and using the following table it is possible to approximate the expected Eb/No from a known (measured) C/N. BPSK 3.00 1.25 0.58 QPSK 0.00 -1.75 -2.42 8PSK -1.77 -3.53 -4.19 16QAM -3.00 -4.75 -5.42
R 1/2 R 3/4 R 7/8
The addition of either Turbo or Reed Solomon will provide an advantage however they do not affect these results. Both Turbo and Reed Solomon error correction provide a significantly higher BER at a given Eb/No over standard Viterbi and Sequential coding.
Rev D 07/04
125
The Decibel
The bel was named in honor of the inventor of the telephone, Alexander Graham Bell. The bel B is used to indicate power levels and is defined as the common logarithm of the ratio of two power levels P and P. Therefore, the number of bels B is calculated as B = Log (P / P). A positive value for B would represent a power gain, a negative value a power loss. A bel is a rather large unit of measurement for use in electrical applications so a smaller and more convenient unit is used. The decibel (dB) has a magnitude of 1/10 of a bel. The calculation is dB = 10 log (P / P)
dB/10
or conversely, P / P = 10
By using this equation, we can now determine the power difference between the two levels. If an amplifier received a signal of 10 watts and increased its strength by a value of 100 to 1000 watts, it would have a gain of 20 dB. dB = 10 log (1000 / 10) = 10 log 100 = 20 dB Similarly, if a 1 watt signal was increased to 1 million watts, the amplifier gain would be 60 dB. If the second produces twice as much power than the first, the difference in dB is 10 log (P2/P1) = 10 log 2 = 3 dB.
Rev D 07/04
126
The Decibel (continued)

It becomes obvious that to say a satellite dish or LNB has a gain of 1 million is impractical. For example, in this style of annotation, a 53.7 dB gain for an LNB at 11 GHz would be 233,333.00. In satellite communications, decibels are commonly expressed as a power relative to a referenced value dBW ( 1 watt) or dBm (1 milliwatt). Therefore a 60 dBW signal has a power of 1 million watts. Antenna gain is relative to that of an isotropic antenna (one which radiates equally in all directions) and is expressed as dBi. It is important to note that dB, dBi, and dBm can all be added together but dBW must be converted to dBm before any manipulation. By far the most common absolute and useful unit used in communications is the dBm, which is the number of decibels above or below 1 milliwatt. dBm = 10 log ( P / 1mW ) For example, the power of a 50 watt transmitter, expressed in dBm would be dBm = 10 log (50 / 1mW) = 10 log 50000 = 46.98 dBm (usually rounded up to 47 dBm) A doubling of power is an increase of 3 dB which means that 50 dBm is 100 watts, or 44 dBm is 25 watts. If you increase power by 3 dB, you double the power. If you decrease power by 3 dB you halve the power.
Rev D 07/04
127
G/T Revisited
Noise has a degrading effect on all satellite links. Noise is inherent in all components of an earth station downlink system the antenna feed, LNA, and downconverter being the main contributors. LNB Noise The incoming satellite signal passes through the feedhorn and exits into the receiving system's first stage of electronic amplification (LNA/LNB). A certain amount of noise is generated within any electronic circuit. Any noise created by the LNB circuitry itself will be amplified and passed on to succeeding stages. For best overall system performance, this noise must be kept to a minimum Less noise here means that more signal will actually arrive at the indoor receiver. Today's high performance LNBs use Gallium Arsenide (GaAs) semiconductor and High Electron Mobility Transistor (HEMT) technologies to minimize the noise level of the LNB. The noise performance of C band LNBs is quantified as a noise temperature measured in degrees Kelvin (K), while Ku band LNB noise performance is expressed as a noise figure measured in dB. Today's C band LNBs commonly achieve a noise temperature of 40 K or less, while Ku band noise figures of less than 1 dB are commonly available. In either case, the lower the LNB's noise performance rating, the less noise introduced into the LNB by its own circuitry. The conversion chart presented below shows the relationship between these two commonly used LNB noise measurement systems:
NF(dB) 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
T (K) NF(dB) 7 14 21 28 35 43 51 59 67 75 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2.0
T (K) NF(dB) 84 92 101 110 120 129 139 149 159 170 2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.9 3.0
T (K) NF(dB) 180 191 202 214 226 238 250 263 275 289 3.1 3.2 3.3 3.4 3.5 3.6 3.7 3.8 3.9 4.0
T (K) 302 316 330 344 359 374 390 406 422 438
Rev D 07/04
128
G/T (continued)
Antenna Noise As well as collecting signals, antennas also collect external noise from sources such as the ground, the atmosphere (especially if rain or cloud is present), and extraterrestrial sources. Whereas LNC/LNB noise temperatures are fixed values available from the manufacturer, antenna noise temperature varies with elevation angle, dish size, frequency and weather conditions. Variation with dish size is because a major component of antenna noise is pickup of the ground temperature (say 290K) by the antenna. Below are some examples of antenna noise temperature vs. elevation angle. Dia (m) Ku 10 deg Ku 30 deg Ku 50 deg C 10 deg C 30 deg C 50 deg 1.2 1.8 2.4 3 3.6 58 50 43 35 41 49 42 35 26 32 45 38 31 22 28 42 36 31 25 30 33 29 24 18 23 29 25 20 14 19
Waveguide Noise The transmission line temperature is a measure of the noise power within the receiver bandwidth generated by the resistive losses in the transmission line/ waveguide between the antenna and the receiver. The actual noise temperature is dependent on the OMT/waveguide configuration used in a particular earth station, however 35K can generally be used as a safe rough estimate for the waveguide/OMT contribution to the G/T of a small to medium earth station.
Rev D 07/04
129

Introduction To Satellite 2

Enviado por

Dados do documento

Descrição original:

Título original

Direitos autorais

Formatos disponíveis

Compartilhar este documento

Compartilhar ou incorporar documento

Opções de compartilhamento

Você considera este documento útil?

Este conteúdo é inapropriado?

Direitos autorais:

Formatos disponíveis

Introduction To Satellite 2

Enviado por

Direitos autorais:

Formatos disponíveis

Chapter 1

The Journey Into Space

Examples of Satellite Orbits

Examples of Satellite Orbits (continued)

Antennas, Transponders, Power supplies, Command and telemetry, Thrust Stabilization

Satellite Subsystems (continued)

Satellite Subsystems (continued)

Satellite Subsystems (continued)

Regional Coverage Eutelsat W1

Spotbeam Coverage Eutelsat W1

Modulation Types (continued)

Modulation Types (continued)

M-Phase Modulation (continued)

M-Phase Modulation (continued)

M-Phase Modulation (continued)

Constellation Diagram (continued)

8PSK and 16QAM

8PSK and 16QAM (continued)

Bandwidth vs. Modulation

Modulator Data Interface (DCE)

Synchronous Data (continued)

Synchronous Data (continued)

Synchronous Data (continued)

Synchronous Data (continued)

Synchronous Data (continued)

Synchronous Data (continued)

Synchronous Data (continued)

ASI Data Video Broadcast Example

A Few Words on DVB

Ethernet Interface (continued)

Ethernet Interface (continued)

Buffer Clock SCTE

Buffer Clock RXSAT

Modulator Timing ( continued)

Data Processing (continued)

Forward Error Correction (FEC)

Forward Error Correction (continued)

Forward Error Correction (continued)

Forward Error Correction (continued)

Forward Error Correction (continued)

Turbo and Trellis Coding

Turbo Coding (continued)

Symbol Rate and Bandwidth Revisited

So why doesnt everyone use 16QAM ?

Block Up Converters (continued)

Block Up Converters (continued)

Power Amplifier Considerations

Frequency Translation Example

Frequency Translation Example (continued)

Antenna Beam Width

The Antenna System

Classification of Earth Station Antennas

Cassegrain Antenna (continued)

Prime Focus Antenna

Antenna Types - Example

Orthogonal Mode Transducer

WC (rigid) = 6.0 dB / 100m

WR (rigid) = 12.5 dB / 100

WR flexible = 20.5 dB / 100m

Effective Isotropic Radiated Power (EIRP)

Block and Level Calculation

LNB / LNC (continued)