INTRODUCTION TO MPEG/DVB AND DTT/DTH OPERATION

Ivan Reuter – Regional Director, West Africa

AGENDA

Introduction to Audio and Video interfaces

Introduction to MPEG compression
Introduction to MPEG-Transport Stream (MPEG-TS)
Introduction to DVB
Satellite for combined DTT/DTH operation
Conditional Access System (CAS)
Set Top Boxes (STB)
Monitoring and troubleshooting

3
Introduction to Audio and Video interfaces
INTRODUCTION TO AUDIO AND VIDEO INTERFACES

The video stream generated by a camera can be either

è PAL
è NTSC
è SECAM
è HDTV
è UHDTV
These standards mainly define the video frame rate, the resolution lines
A camera can produce a video stream under 2 formats
à Analog (component or composite)
à Digital (SD-SDI / HD-SDI / 3G-SDI / 6G-SDI)
The bitrate at the output of the camera are
è About 270Mbps for a SD-SDI stream
è About 1Gbps for a HD-SDI stream
è About 3Gbps for a 3G-SDI stream

5
INTRODUCTION TO AUDIO AND VIDEO STANDARDS

The common audio interfaces are

è Mono (1 channel)
è Stereo (2 channels)
è Dolby (several channels)
The Video and Audio streams are being produced at the studio and post-
treated before being delivered to the head-end/broadcaster
Usually the audio and video are being delivered at the head-end under the
following format
è SDI + embedded audio (coaxial cable)
è SDI + separate stereo or Mono (coaxial cable + XLR)
è CVBS + separate stereo or Mono (coaxial + XLR)

6
Introduction to MPEG compression
INTRODUCTION TO MPEG COMPRESSION

MPEG stands for Moving Picture Experts Group, set up by the International
Standardization Organization (ISO)

MPEG is one of the most popular audio/video compression techniques

MPEG constitutes a range of standards suitable for different applications

based on similar principles

Website : [Link]

8
INTRODUCTION TO MPEG COMPRESSION

Compression is needed because the generated video at the source is more

than 200Mbps (for SD) and more than 1Gbps (for HD), thus requiring high
storage capacity and high network bandwidth for real-time transmission

The 2 main advantages of compression are the following :

è Smaller amount of storage needed for a given amount of source material
à Enable a full movie to be recorded and stored on a Tape or DVD or HD-DVD/Bluray
è Smaller bandwidth required on telecommunication network to transport real-
time video
à Enable real-time video broadcasting over satellite networks, terrestrial networks, IPTV
networks where bandwidth /spectrum is a limited and costly resource

9
INTRODUCTION TO MPEG COMPRESSION

Video compression relies on 2 fundamental components of the signal :

those which are novel and unpredictable and those which can be
anticipated

The novel component is called entropy and represent the true information
in the signal. The reminder is called redundancy because it is not essential

Redundancy may be spatial (pixels adjacent to each other having almost

the same value) or temporal (similarities between successive pictures
over the time)

All compression systems work by separating the entropy from the

redundancy in the encoder

10
INTRODUCTION TO MPEG COMPRESSION

There has been over the years many versions of MPEG encoding standard,
that have improved in efficiency
è MPEG-1 : mainly aimed at encoding audio and video to be recorded on a medium
such as Video-CD. The famous MP3 audio encoding was released at that time
è MPEG-2 : mainly aimed at transporting TV over terrestrial and satellite networks
but also to record video on DVD
è MPEG-4 : mainly aimed at transporting in a more efficient way TV over digital
terrestrial and satellite networks, making HDTV transport viable throughout
these networks, but also to record HD video on HD-DVD/Bluray DVD
è MPEG-HEVC : latest standard designed to store and transport UHDTV
Every new standard have provided a significant decrease of throughput
while keeping the same video quality
The switch from MPEG-2 to MPEG-4 allowed to save 50% of bandwidth
while keeping the same subjective video quality

11
INTRODUCTION TO MPEG COMPRESSION

Video and audio elements are encoded into a sequence of Frames

A frame is a single “image” from a video or audio sequence
In a PAL system, frames occur every 40 milliseconds (25 frames per
second)
Each frame is encoded in one of the 3 following ways
è I Frames
à Intra-picture encoding, similar to JPEG encoding (exploiting spatial redundancy)
à Compressed solely on information found in the frame
à Highest amount of data of all other frames type
à GOP (Group Of Pictures) always starts with “I” frames
à B & P frames are calculated from the I Frame
à I stands for Intra-coded

12
INTRODUCTION TO MPEG COMPRESSION

è B Frame
à Bi-directional encoding, using motion adjusted delta from a previous and a future frame
(exploiting temporal redundancy)
à Uses data from preceding and following I and P frame
à Depends on data before and/or after in video sequence
à Cannot calculate from another B frame

è P Frame
à Predictive encoding, using motion adjusted deltas from a previous reference frame
(exploiting temporal redundancy)
à Compressed using data from closest preceding I or P frame
à GOP with back-to back P frames usually indicate fast motion

13
INTRODUCTION TO MPEG COMPRESSION

Encoder transmit sequence of encoded frames which size vary

Encoded frames are packed into Packetized Elementary Stream (PES)

packets

PES packets are packed into MPEG transport packets (all packets for single
video stream have the same PID value)

Compression ratio is typically 50:1 (MPEG-2) or 100:1 (MPEG-4) or even

more

14
INTRODUCTION TO MPEG COMPRESSION

The Group of Pictures (GOP) is series of frames consisting of a single “I”

frame with several B and P frames

GOP begins with an “I” frame and ends with last frame before the next
“I” frame

All frames within the GOP depend directly or indirectly on data from the I
frame

Relationship between GOPs : OPEN or CLOSED

CLOSED GOP is self-contained : no dependence on data outside GOP

OPEN GOP can use data from I frame of the following GOP
15
INTRODUCTION TO MPEG COMPRESSION

The video quality at the output of the encoder mainly depends on

è The Output bitrate
à The higher the bitrate, the higher the quality
à The lower the bitrate, the lower the quality
è The chosen GOP structure vs the type of content

In MPEG-2 SD, the typical bitrate of the encoded video shall be 4Mbps
In MPEG-4 SD, the typical bitrate of the encoded video shall be 2Mbps
In MPEG-4 HD, the typical bitrate of the encoded video shall be 5-6Mbps

The more the content is moving/changing fast, such as sport content, the
higher the bitrate of the encoded video should be in order to avoid video
quality degradation / macro blocking

16
Introduction to MPEG-Transport Stream
INTRODUCTION TO MPEG-TS

The MPEG-TS defines the way the audio, video and data streams are
multiplexed and transported together
The MPEG-TS may contain 1 or more channels
è Video channel(s)
à A video Stream
à 1 or more associated audio stream(s)
à Possibly 1 or more data stream(s)
è Audio channel(s)
à 1 or more audio stream(s)
à Possibly 1 or more data streams(s)
è Data-only channel(s)
à 1 or more data stream(s)

18
INTRODUCTION TO MPEG-TS

The MPEG-TS is made up of 188-bytes transport packets

è 184-bytes of payload
è 4 bytes of header
Each transport packet contains any one kind of information (video, audio,
data, PSI tables)

Transport packets have multiple interleaved elementary streams (video,

audio, data, PSI)

19
INTRODUCTION TO MPEG-TS

Packets belonging to the same elementary stream are identified by Packet

ID (PID) in packet header (same color in illustration below)

20
INTRODUCTION TO MPEG-TS

Structure of an MPEG-TS packet

Sync byte : find packet boundary

PID : Used while demultiplexing stream to identify the content
Continuity counter : identify packet loss
PCR stamp in adaptation field : Clock synchronisation

21
INTRODUCTION TO MPEG-TS

MPEG-TS Header field : Sync byte

è When a decoder first tune, all it sees is a stream of 0 and 1
è The decoder must first identify the beginning of packets before it can interpret
the stream
è The decoder uses the Sync Byte field to do this

The Sync Byte of a packet is always 0x47 (hexadecimal) or 01000111

binary

22
INTRODUCTION TO MPEG-TS

If you don’t have Packet sync, the decoder cannot find packets
boundaries. You will not be able to decode at all

Packet sync problems typically occur at hardware level at packet

boundaries with format converters, edge devices, demodulators
è ASI to Gig-E conversion
è ASI to Microwave
è Satellite to ASI

23
INTRODUCTION TO MPEG-TS

Data about included data is called Metadata

Metadata tells the decoder which kind of information are contained in each
PID, and which PID go together

There are two kind of metadata

è Program Specific Information (PSI) – tables in the stream defined in the MPEG-
Standard
è Program and System Information Protocol (PSIP) – tables defined in ATSC

24
INTRODUCTION TO MPEG-TS

The PSI tables (Program Specific Information) are an essential component

of an MPEG-TS. They are the tables that bind all the elements of a
transport Stream together and contains all the information the decoder
needs to demultiplex the stream and associate the Elementary streams
(video, audio and data) that go together to create a program

25
INTRODUCTION TO MPEG-TS

When the decoder first receives a transport stream, it demultiplexes that

stream based on PID

The SI tables (System information) are also an essential component of the

MPEG-TS. They intend to bind a number of services and TS together in
order to provide a Multi-channels Broadcasting environment
26
INTRODUCTION TO MPEG-TS

The PSI tables are as follows

è PAT (Program Association Table)
à Appears in PID 0x0000
à Identifies MPEG programs in Transport Stream and gives PIDs for their PMTs
à The PAT is on PID 0x00. This is the first PID or “base PID” a MPEG decoder looks for
è PMT (Program Map Table)
à Identifies Elementary Streams in program (video, audio, data) and give their respective
PIDs

27
INTRODUCTION TO MPEG-TS

è CAT (Conditional Access Table)

à Appears in PID 0x01
à Used for Conditional Access management of keys used for decryption of restricted
streams
à The CAT contains privately defined descriptors of the system used and the associated
EMM PID
è NIT (Network Information Table)
à Groups a number of Transport Streams together providing tuning information for the
receiver

28
INTRODUCTION TO MPEG-TS

The SI tables are as follows

è NIT (network Information Table)
à As described before
è BAT (Bouquet Association Table)
à A bouquet is a group of services which are presented to the user as if they were on the
same Transport Stream although they are physically on different Transport Stream. The
BAT allows for associating together all these Transport Streams while transparently re-
tuning the receiver
è SDT (Service Description Table)
à Provides description of a service : name and optionally related information such as
language codes, running status and country availability
è EIT (Event Information Table)
à Carries information coming from an EPG (Electronic Program Guide), providing
information regarding content being currently displayed but also next content available
in few hours

29
INTRODUCTION TO MPEG-TS

MPEG packet header : Continuity Counter

è The continuity counter is a 4 bits field in the header which increments by 1 each
time a packet comes out on a specific PID

When a PID “skips” one value of the continuity counter, it is called a

“continuity error”. This means one or more packets were lost

30
INTRODUCTION TO MPEG-TS

Audio and Video are encoded separately, so they must be synchronized

during play
MPEG has to allow for great distance and network processing between the
encoder and decoder and still allow for correct decode of the transport
stream
In order for the audio and video Elementary streams to remain in sync, the
encoder clock and the decoder clock must remain in sync
When the encoder creates packets, it embeds the current value of its
27MHz clock into the stream
This time reference is called the PCR (Program Clock Reference)
MPEG imposes that one PCR packet appear in the stream every 100ms

31
INTRODUCTION TO MPEG-TS

The MPEG encoder and MPEG decoder use a 27MHz clock to

encode/decode incoming audio and video

The clock is actually a counter which advance every 1/27000000 second

32
INTRODUCTION TO MPEG-TS

What could conceptually cause PCR timestamp issues?

è Encoder possibly time stamped incorrectly
è Decoder possibly failed to consume time stamps
è PCR packet was accidentally lost in transmission
When PCR time stamps go awry, we have “PCR jitter”
The PCR jitter is defined as
è Difference between the actual value of the PCR time stamped by encoder and
the expected value of the PCR as calculated by decoder based on the clock rate
and the time at which the PCR value is received
PCR jitter spec : 500ns
PCR frequency offset is the difference between the clock frequency
calculated at the decoder based on actual PCR values received and an
“ideal” 27MHz clock which is the clock rate dictated by the MPEG-2 TS
standard
PCR frequency offset spec : +/- 810Hz
33
INTRODUCTION TO MPEG-TS

When multiple streams are muxed together, the spacing between the PCR
packets in each stream changes
The physical shift results in a temporal shift as well, throwing the time
stamp off
The temporal shift in PCR values is refered to as “PCR jitter”
The MUX has to restamp all the PCR values to correct for the change in the
packet spacing – this is very hard to do!
è The more services at the output, the harder it is to restamp
è The fewer “null” packets at the output, the harder it is to restamp
PCR jitter might also happen at the source encoding

34
Introduction to DVB
INTRODUCTION TO DVB

The Digital Video Broadcasting Project (DVB) is an industry-led

consortium of over 200 broadcasters, manufacturers, network operators,
software developers, regulators and others from around the world
committed to designing open interoperable technical standards for the
global delivery of digital media and broadcast services

The DVB project develops specifications for Digital television systems,

which are turned into standards by international standards body such as
ETSI. It does so through collaboration of its members in numerous
working groups. Once specifications have become standards, they are
promoted for international adoption and utilisation

36
INTRODUCTION TO DVB

In the framework of satellite television (DTH), the transmission standard

of interest are
è DVB-S
è DVB-S2
The DVB-S and DVB-S2 standards define the technics being used for the
transmission over satellite, that is to say the modulations and coding
schemes, as well as payload framing
On the TV content protection side, the interface between the CAM
(Conditional Access Module) provided by the CAS manufacturer and the
Set Top Box / TV Set is standardized and defined under the following
standards
è DVB-CI (Common-Interface)
è DVB-CI plus

37
Satellite for combined DTT/DTH operation
SATELLITE FOR COMBINED DTT/DTH OPERATION

A typical DTH network infrastructure is composed by

è Contribution links, typically between the studios where the content is being
produced and the head-end facility
è A DVB Head-end where the content processing/encoding/multiplexing/Uplink is
being performed
è Distribution link, which is the Direct-to-Home connection from the head-end to
the consumers
The contribution links can be done by satellite or microwave or IP and are
typically using [Link] encoding in order to preserve the highest quality (but
can be done in [Link])
The distribution link is broadcast in [Link] quality and consist of
transmitting an MPEG-TS containing all the TV channels
Is it recommended to implement Statistical Multiplexing (StatMux) on the
Head-end in order to optimize the quality and the bandwidth

39
SATELLITE FOR COMBINED DTT/DTH OPERATION

A typical DVB head-end infrastructure is composed by the following

elements
è Input router (CVBS or SDI)
è N+m encoders (N main encoders + m redundant encoders)
è 1+1 multiplexers
è A Network Monitoring & Control System
è Optionally
à A Conditional Access System
à A Subscriber Management System
à An Electronic Program Guide
The head-end are more and more full IP (from the input of the encoders to
the output of the MUX, as well as the connections between the encoders
and the MUX)

40
SATELLITE FOR COMBINED DTT/DTH OPERATION

41
TYPICAL DTH NETWORK

From a transmission standpoint, DTH architecture is quite straightforward

è The A/V content are multiplexed into a MPTS
è The MPTS is broadcast over DVB-S or S2 satellite signal in Ku-band
è The consumer DTH Set Top Box (STB) is able to render the A/V content from the
received MPTS

42
DVB-T2 ADAPTATION

The output of the multiplexer (MPTS) must be converted into a DVB-T2

compliant multiplex
This adaptation is done at the head-end by a T2-Gateway
The output of the T2-Gateway is a dedicated protocol called T2-MI
(Modulator Interface)
The T2-MI is not readable by the DTH receivers !

43
TYPICAL DTT NETWORK

The A/V content are multiplexed into a MPTS

The MPTS is encapsulated into T2-MI Protocol (DVB-T2 distribution
Protocol)
The T2-MI stream is broadcast over DVB-S or S2 signal in Ku or C-band
The transmitter sites receive the T2-MI stream and broadcast it on the
DVB-T2 network
The consumer DTH STB are not able to read the T2-MI stream

44
TYPICAL DTH+DVB-T2 ARCHITECTURE

45
DVB-T2 OVER DTH ARCHITECTURE

46
Conditional Access System (CAS)
CONDITIONAL ACCESS SYSTEM (CAS)

A Conditional Access System (CAS) is meant to protect the content of a

multiplex or a TV bouquet over a transport network in order to avoid
unauthorized viewing

There is a substantial number of CAS manufacturers/vendors across the

world whose level of security/reliability depends on the encryption key
algorithm complexity

One must choose its CAS taking into consideration the following aspects
è Security/reliability of the CAS (if the system is being hacked, it is all the pay-TV
revenue model who fails immediately)
è Potential existing bunch of set-top-box on the ground
è Price model of the CAS

48
CONDITIONAL ACCESS SYSTEM (CAS)

A CAS generally consists of 2 main subsystems

è A scrambling subsystem that scrambles the signal to prevent non-subscribers
from receiving it and descrambles the signal at the subscriber’s receivers
è An access control subsystem that process access control messages to determine
whether descrambling is to be performed

The general requirements that scrambling systems must satisfy are as

follows
è It must be difficult for a third party to perform unauthorized reception
è Scrambled signal content must not be understandable
è Quality must not deteriorate on restoring the signal
è Receiver and operation cost must be low

49
CONDITIONAL ACCESS SYSTEM (CAS)

The system should perform

scrambling according to the
properties of the signal in
question
The system should change
the key regularly to maintain
the security of the
scrambling system, and
transmit this key information
to the receiver in a securer
manner using a hierarchical
encryption system
For the purpose of operating
pay-tv broadcasting,
reception should be
controlled according to the
details of each user’s
subscription

50
CONDITIONAL ACCESS SYSTEM (CAS)

Technology for
encrypting digital
signals can be broadly
classified as Stream
ciphers and Block
ciphers
There are 2 methods
for encrypting a
digital signal
è The first method
scrambles the
Transport Stream (TS)
è The second method
scrambles Packetized
Elementary Streams
(PES)

51
CONDITIONAL ACCESS SYSTEM (CAS)

The key configuration and transmission of key information

è In signal scrambling, a scrambling key (ks) determines the scrambling pattern
and this key is typically changed every second in order to maintain security
è This Ks must be continuously transmitted to the subscriber’s receiver and it is
done by encrypting and transmitting it within related information called
Entitlement Control Messages (ECM) together with the scrambled signal
è ECMs include program attribute information for determining whether a
subscriber is entitled to view a program on the basis of his subscription
è To prevent ECM that include the Ks to be understood by a third party, it is
encrypted before transmission by using a work key (Kw) that is also updated
typically on a monthly or yearly basis. This Work key Kw is sent to the receiver
through related information called Entitlement Management Messages (EMM)
together with subscription contents that are sent with subscription updates
è The EMM can be either sent over-the-air or typically under the form of a smart
card and is encrypted by a master key (Km) unique to each receiver

52
CONDITIONAL ACCESS SYSTEM (CAS)

ECMs and EMMs are sent to the receiver in the Transport Stream and the
PID values used to send these ECMs and EMMs sections are included in the
Program Map Table (PMT) and Conditional Access Table (CAT)
In the CAT, specific ECM and EMM contents can be specified according to
business format, billing system, etc… enabling a variety of CAS to be
implemented
The Conditional Access Receiver (STB) consist of a section that
descrambles the scrambled signal and a section that processes related
control information such as ECMs and EMMs as follows

53
CONDITIONAL ACCESS SYSTEM (CAS)

Configuration A integrates the signal-descrambling section and related

information-processing section in the receiver unit
Configuration B includes the signal-descrambling section in the receiver
unit but implements the related information-processing section in a
removable security module
Configuration C makes both the signal-descrambling section and the
related information-processing section removable

54
CONDITIONAL ACCESS SYSTEM (CAS)

To deal with threats to security, changes to the billing system, etc.,

configuration A requires that the entire unit be replaced

Configuration B can deal with such occurrences by simply replacing the

security module, which can be implemented in the form of a smart card
with a built-in processor. This smart card approach is the widest adopted
as of today. This configuration doesn’t allow changes to be made to the
signal descrambling scheme, whereas configuration C allows even the
descrambling algorithm to be updated, enabling system security to be
improved all the more

55
Set Top Boxes
SET TOP BOXES

A set Top Box in a consumer-grade integrated receiver/decoder composed

by
è One or more DVB-S / DVB-S2 tuner(s), usually up to 45MBaud / 8PSK
è A video decoder
è An audio decoder
è A software layer enabling the display of information related to the STB
configuration, but also Electronic Program Guide for instance
è Optionally a PVR (program recorder via an integrated or external HDD)
The STB is usually equipped with a broad range of interfaces
è Video
à Usually CVBS and/or RGB and/or S-Video and/or HDMI
è Audio
à Usually RCA and/or S-PDIF

57
SET TOP BOXES

The STB first scan the L-band spectrum at its input and search for carriers
Once the STB has identified carriers, then it demodulates them and scan
their content in order to display the list of channels
For single-tuner STB, only 1 carrier at a time can be demodulated
The Tuner of the STB controls the universal LNB as follows
è Sub-band control
à 0KHz signal for low sub-band (10,75-11,75GHz)
à 22KHz signal for high sub-band (11,75-12,75GHz)
è Polarization control
à 13V signal for Vertical polarization
à 18V signal for Horizontal polarization

If, for some reason, the STB hasn’t automatically found all the satellite
carriers, there is always a manual mode allowing to enter the desired
frequency, symbol rate, modulation, etc…
58
SET TOP BOXES

In the framework of a Free-to-Air (FTA) network, the choice of STB

available in the market remains fairly large, providing that the
audio/video encoding format and the transmission parameters remain
standard/common

In the framework of a Free-to-View/Pay-TV network, the choice of STB

becomes narrower
è A STB implements the descrambling and ECM/EMM system proper to a particular
CAS provider

59
Monitoring and Troubleshooting
MONITORING AND TROUBLESHOOTING

The most important equipment to monitor and troubleshoot a DVB Head-

end is an MPEG-TS analyzer
The MPEG-TS analyzer perform what we call a TR 101 290 (ETSI standard)
analysis of the MPEG-TS stream whereby all the important parameters of
the streams are being captured and displayed
There are usually 3 priorities of errors
è Priority 1
è Priority 2
è Priority 3
The MPEG-TS analyzer shall be inserted at the output of the multiplexer so
it can analyze what goes out of the Head-end and quickly identify what is
the source of the problem

61
MONITORING AND TROUBLESHOOTING

The 3 main monitoring and troubleshooting lessons are

è Fix continuity problems first
à Continuity problems create alarms in other areas
à Until you resolve continuity problems, it will be very hard to determine if you have other
problems in your stream
è Check for PCR jitter
à PCR jitter can cause packet overflow or underflow
à Lip sync errors may also be result of PCR jitter
è Move backward through your network as you debug
à Isolate the offending device first, then go for detailed problem solving steps
Putting an IRD at the output of the multiplexor allows for rapid channel
scanning and decoding and helps to check the audio/video quality. It
represent a good and straightforward tool to monitor the subjective
quality of the programs before they are being sent over to the satellite

62