DVB-T2
DVB-T2 is an abbreviation for "Digital Video Broadcasting – Second Generation Terrestrial"; it is the extension of the television standard DVB-T, issued by the consortium DVB, devised for the broadcast transmission of digital terrestrial television. DVB has been standardised by ETSI.
This system transmits compressed digital audio, video, and other data in "physical layer pipes", using OFDM modulation with concatenated channel coding and interleaving. The higher offered bit rate, with respect to its predecessor DVB-T, makes it a system suited for carrying HDTV signals on the terrestrial TV channel., it was implemented in broadcasts in the United Kingdom, Italy, Finland, Germany, the Netherlands, Sweden, Thailand, Flanders, Serbia, Ukraine, Croatia, Denmark, Romania, and some other countries.
History
Preliminary investigation
In March 2006, DVB decided to study options for an upgraded DVB-T standard. In June 2006, a formal study group named TM-T2 was established by the DVB Group to develop an advanced modulation scheme that could be adopted by a second generation digital terrestrial television standard, to be named DVB-T2.According to the commercial requirements and call for technologies issued in April 2007, the first phase of DVB-T2 would be devoted to provide optimum reception for stationary and portable receivers using existing aerials, whereas a second and third phase would study methods to deliver higher payloads and the mobile reception issue. The novel system should provide a minimum 30% increase in payload, under similar channel conditions already used for DVB-T.
The BBC, ITV, Channel 4 and Channel 5 agreed with the regulator Ofcom to convert one UK multiplex to DVB-T2 to increase capacity for HDTV via DTT. They expected the first TV region to use the new standard would be Granada in November 2009. It was expected that over time there would be enough DVB-T2 receivers sold to switch all DTT transmissions to DVB-T2, and H.264.
Ofcom published its final decision on 3 April 2008, for HDTV using DVB-T2 and H.264: BBC HD would have one HD slot after digital switchover at Granada. ITV and C4 had, as expected, applied to Ofcom for the 2 additional HD slots available from 2009 to 2012.
Ofcom indicated that it found an unused channel covering 3.7 million households in London, which could be used to broadcast the DVB-T2 HD multiplex from 2010, i.e., before DSO in London. Ofcom indicated that they would look for more unused UHF channels in other parts of the UK, that can be used for the DVB-T2 HD multiplex from 2010 until DSO.
The DVB-T2 specification
The DVB-T2 draft standard was ratified by the DVB Steering Board on 26 June 2008, and published on the DVB homepage as DVB-T2 standard BlueBook. It was handed over to the European Telecommunications Standards Institute by DVB.ORG on 20 June 2008.The ETSI process resulted in the DVB-T2 standard being adopted on 9 September 2009. The ETSI process had several phases, but the only changes were text clarifications. Since the DVB-T2 physical layer specification was complete, and there would be no further technical enhancements, receiver VLSI chip design started with confidence in stability of specification. A draft PSI/SI specification document was also agreed with the DVB-TM-GBS group.
Tests
Prototype receivers were shown in September IBC 2008 and more recent version at the IBC 2009 in Amsterdam. A number of other manufacturers demonstrated DVB-T2 at IBC 2009 including Albis Technologies, Arqiva, DekTec, Enensys Technologies, Harris, Pace, Rohde & Schwarz, Tandberg, Thomson Broadcast and TeamCast. As of 2012, Appear TV also produce DVB-T2 receivers, DVB-T2 modulators and DVB-T2 gateways. Other manufacturers planning DVB-T2 equipment launches include Alitronika, CellMetric, Cisco, Digital TV Labs, Humax, NXP Semiconductors, Panasonic, ProTelevision Technologies, Screen Service, SIDSA, Sony, ST Microelectronics and T-VIPS. The first test from a real TV transmitter was performed by the BBC Research & Development in the last weeks of June 2008 using channel 53 from the Guildford transmitter, southwest of London: BBC had developed and built the modulator/demodulator prototype in parallel with the DVB-T2 standard being drafted. Other companies like ENKOM or IfN develop software based decoding.NORDIG published a DVB-T2 receiver specification and performance requirement on 1 July 2009. In March 2009 the Digital TV Group, the industry association for digital TV in the UK, published the technical specification for high definition services on digital terrestrial television using the new DVB-T2 standard. The DTG's test house: DTG Testing are testing Freeview HD products against this specification.
Many tests broadcast transmission using this standard are being in process in France, with local Gap filler near Rennes CCETT.
DVB-T2 was tested in October 2010, in Geneva region, with Mont Salève's repeater, in UHF band on Channel 36. A mobile van was testing BER, strength, and quality reception, with special PCs used as spectrum analysers, constellation testers. The van was moving in Canton Geneva, and France. However, none were demonstrated in TELECOM 2011 at Palexpo.
The standard
The following characteristics have been devised for the T2 standard:- COFDM modulation with QPSK, 16-QAM, 64-QAM, or 256-QAM constellations.
- OFDM modes are 1k, 2k, 4k, 8k, 16k, and 32k. The symbol length for 32k mode is about 4 ms.
- Guard intervals are 1/128, 1/32, 1/16, 19/256, 1/8, 19/128, and 1/4.
- FEC is concatenated LDPC and BCH codes, with rates 1/2, 3/5, 2/3, 3/4, 4/5, and 5/6.
- There are fewer pilots, in 8 different pilot-patterns, and equalization can be based also on the RAI CD3 system.
- In the 32k mode, a larger part of the standard 8 MHz channel can be used, adding about 2% extra capacity.
- DVB-T2 is specified for 1.7, 5, 6, 7, 8, and 10 MHz channel bandwidth.
- MISO may be used, but MIMO will not be used. Diversity receivers can be used.
- Multiple PLPs to enable service-specific robustness at a particular bit rate.
- Bundling of more channels into a SuperMUX is not in the standard, but may be added later.
System differences with DVB-T
For instance, a UK MFN DVB-T profile and a DVB-T2 equivalent allows for an increase in bit rate from 24.13 Mbit/s to 35.4 Mbit/s. Another example, for an Italian SFN DVB-T profile and a DVB-T2 equivalent, achieves an increase in bit rate from 19.91 Mbit/s to 33.3 Mbit/s.
Recommended maximum bit-rate configurations for 8 MHz bandwidth, 32K FFT, guard interval 1/128, pilot pattern 7:
| Modu- lation | Code rate | Bitrate | Frame length LF | FEC blocks per frame |
| QPSK | 1/2 | 7.4442731 | 60 | 50 |
| QPSK | 3/5 | 8.9457325 | 60 | 50 |
| QPSK | 2/3 | 9.9541201 | 60 | 50 |
| QPSK | 3/4 | 11.197922 | 60 | 50 |
| QPSK | 4/5 | 11.948651 | 60 | 50 |
| QPSK | 5/6 | 12.456553 | 60 | 50 |
| 16-QAM | 1/2 | 15.037432 | 60 | 101 |
| 16-QAM | 3/5 | 18.07038 | 60 | 101 |
| 16-QAM | 2/3 | 20.107323 | 60 | 101 |
| 16-QAM | 3/4 | 22.619802 | 60 | 101 |
| 16-QAM | 4/5 | 24.136276 | 60 | 101 |
| 16-QAM | 5/6 | 25.162236 | 60 | 101 |
| 64-QAM | 1/2 | 22.481705 | 60 | 151 |
| 64-QAM | 3/5 | 27.016112 | 60 | 151 |
| 64-QAM | 2/3 | 30.061443 | 60 | 151 |
| 64-QAM | 3/4 | 33.817724 | 60 | 151 |
| 64-QAM | 4/5 | 36.084927 | 60 | 151 |
| 64-QAM | 5/6 | 37.618789 | 60 | 151 |
| 256-QAM | 1/2 | 30.074863 | 60 | 202 |
| 256-QAM | 3/5 | 36.140759 | 60 | 202 |
| 256-QAM | 2/3 | 40.214645 | 60 | 202 |
| 256-QAM | 3/4 | 45.239604 | 60 | 202 |
| 256-QAM | 4/5 | 48.272552 | 60 | 202 |
| 256-QAM | 5/6 | 50.324472 | 60 | 202 |
Technical details
The processing workflow is as follows:- Input pre-processing
- * Physical Layer Pipe
- ** PLPs, which had already been introduced in DVB-S2, are logical channels carrying one or more services, with a modulation scheme and robustness particular to that individual pipe.
- ** PLP creation: adaptation of Transport Stream, Generic Stream Encapsulation, Generic Continuous Stream, or Generic Fixed-length Packetized Stream
- Input processing
- * Mode adaptation
- ** Single PLP : data are assembled in groups called BaseBand Frames, with lengths of bits, defined by modulation and coding parameters, in a 'normal' length or 'short' length version
- *** Input interface
- *** CRC-8 encoding
- *** BaseBand header insertion
- ** Multiple PLPs
- *** Input interface
- *** Input stream synchronization
- *** Delay compensation
- *** Null packets deletion
- *** CRC-8 encoding
- *** BB header insertion
- * Stream adaptation
- ** Single PLP
- *** Padding insertion
- *** BB scrambling: a Pseudo Random Binary Sequence with generator is used to scramble completely every BBFRAME
- ** Multiple PLPs
- *** PLP scheduling
- *** Frame delay
- *** In-band signaling or padding insertion
- *** BB scrambling
- Bit Interleaved Coding and Modulation
- * Forward Error Correction encoding: each BBFRAME is converted into a FECFRAME of bits, by adding parity data. Normal FECFRAMEs are 64,800 bits long, whereas short FECFRAMEs are 16,200 bits long. The effective code rates are 32,208/64,800, 38,688/64,800, 43,040/64,800, 48,408/64,800, 51,648/64,800, 53,840/64,800
- ** Outer encoding: a BCH code, capable to correct 10 or 12 errors per FECFRAME, is used to compute parity data for the information data field. The BCH generator polynomial is of the 160th, 168th, or 192nd grade
- ** Inner encoding: a Low Density Parity Check code is cascaded to the BCH
- * Bit interleaving
- ** Parity bits block interleaving
- ** Twist column interleaving
- * Bit demultiplexing to cell words
- * Gray mapping of cell words to constellations: either QPSK, 16-QAM, 64-QAM, or 256-QAM maps are used
- * Constellation rotation and cyclic quadrature delay: optionally, the constellations may be tilted counterclockwise by an amount of up to 30 degrees. Furthermore, the quadrature part of the cells is cyclically shifted by one cell
- * Cell interleaving
- * Time interleaving
- Frame building: the transmitted stream is organized in super frames, which are composed by T2 frames and FEF parts
- * Cell mapping: cells are mapped to OFDM symbols. A T2 frame is composed by a P1 symbol, one or more P2 symbols, regular data symbols, and a Frame Closing symbol. The P1 symbol is used for synchronization purposes, the P2 symbols convey L1 parameter configuration signaling, whereas the data symbols carry PLP data, auxiliary streams, and dummy symbols used as space filler
- * Frequency interleaving: random interleaving is done on every OFDM symbol
- OFDM generation
- * Multiple-Input Single-Output processing: Alamouti pre-processing is optionally applied to pairs of OFDM symbol cells. Given the input cells, and transmitter group 1 and 2 cells, the mapping is done as and for group 1, and as and for group 2
- * Pilot insertion and dummy tone reservation: three classes of pilot tones are added. They are either continual, scattered, or edge. There are 8 different configuration for scattered pilots. Moreover, a number of dummy carriers are not modulated and reserved to reduce the dynamic range of the DVB-T2 output signal.
- * Inverse Discrete Fourier Transform : classic IDFT is used to switch from the frequency domain into the time domain, after having adjusted carrier position relevant to the central transmit frequency. 1k to 32k carriers are available. There is also an extended mode, which allows to fill more data in the available bandwidth, using more active carriers and reducing the number of guard band carriers.
- * Peak-to-Average-Power-Ratio reduction, also called crest factor reduction
- * Guard interval insertion: a cyclic prefix is inserted before the IDFT symbol, to recover from transmit channel echoes. Lengths from 1/128 to 1/4 of the IDFT length are allowed.
- * P1 symbol insertion: the P1 symbol is a particularly crafted 1k OFDM symbol, always inserted at the head of a T2 frame. It conveys few bits of information, as it is mainly dedicated to fast synchronization at the receiver side. It is prepended and postpended by frequency shifted repetitions of itself, to ease receiver lock even if the nominal center frequency of the T2 signal is up to 500 kHz off.
- * Digital-to-Analog Conversion : the T2 samples are converted into an analog BB complex signal at a sample rate that depends on the channelization bandwidth. For instance, in 8 MHz wide channels, the complex sample time is 7/64 μs.
- Adaptive coding and modulation ACM
- *Adaptive coding and modulation, allows to adapt the transmission's parameters to the reception conditions required by the terminals, for example changing into a lower coding speed during attenuation.