latest on r&d on wired transmission technologies · breakdown of wired transmission...
TRANSCRIPT
FEATURE
3
In this report, we aim to present the latest updates in our research and developments on the transmission of audio and video signals over wired networks. We will start by review-ing the progress we have made so far, then describe in detail the technologies for delivering video directly to the home via cable television and the internet. We will also look at IP and other new technologies in the field of program produc-tion and contributions to it.
1. NHK STRL’s Research and Development on Wired Transmission Technologies
Terrestrial and satellite broadcasting services use radio waves to deliver programming to each household. On the other hand, cable television (CATV) services distribute pro-gramming over coaxial cables and other wired networks. CATV emerged before the advent of digital broadcast-ing and has seen much growth, especially in areas where over-the-air reception is limited. More recently, watching high-quality video via fiber-optic internet connection has also been gaining ground. On the contribution side, broad-casters are now starting to use optical fiber and the public internet to transmit their video material, in addition to con-ventional transmission over radio waves. See Figure 1 for a breakdown of wired transmission technologies used in and around broadcasting.
Since the 1980s, when tests on satellite and analogue HD broadcasting had just begun, the STRL has been con-ducting research and development on video transmission technologies using fiber-optic cable. In order to promote satellite broadcasting, we established a transmission system whereby CATV facilities receive satellite signals, convert them to intermediate frequency (IF) signals, and distribute them to each household via optical fiber. We have also made
progress in transmitting major sporting events over optical fiber to enable live public HD screenings at remote venues.1) These research studies were mainly focused on finding tech-nical solutions to issues in converting analogue electrical signals into optical signals so that they can be transmitted over fiber-optic cable.
In the 1990s, HD program production began moving to-wards digitalization. We started working on technologies to send uncompressed digital HD signals over long-distance optical fiber at a speed of 1.5 Gbps, which was very fast at the time. We were also looking at how to send digitally modulated radio frequency (RF) signals over CATV, in an-ticipation of the upcoming digitization of satellite broad-casting.
In 1995, we set out to develop a new audiovisual system with higher video resolution than HD, and as we entered the new millennium, our R&D on 8K Super Hi-Vision (8K UHD) was fully implemented. We also started studying 8K CATV broadcasting and technologies for transmitting uncompressed 8K material over optical fiber for program production purposes.
In order to transmit full-specification uncompressed 8K signals, we need a transmission speed of around 144 Gbps. This is roughly a hundred times faster than HD transmis-sion, which raises considerable challenges in developing compatible equipment and enabling long-distance trans-mission. We decided to take advantage of the technological achievements made in the field of communications and to apply Ethernet technology that had already been in wide use for data transfer.1)
All of the above studies relate to improving signal for-mats between transmitters, or enhancing their transmission properties. If we look at it in terms of the Open Systems
Latest on R&D on Wired Transmission TechnologiesYosuke Endo, Akinori Hashimoto, Masao Yamamoto, Takuya Kurakake
Wired Transmission�Technologies
Distribution of Broadcast Programs to the Home�
Transmission of �Video Material�(Contribution)
CATV Transmission・Multiple carrier transmission・Baseband optical distribution
Internet Distribution・MPEG-DASH
Fiber-optic Transmission・Transmission of uncompressed 8K material �
IP Transmission
Figure 1: Wired Transmission Technologies in Broadcasting
FEATUREInterconnection (OSI) reference model, often mentioned in data communication, we may say that these research studies have been centered mainly on the first layer or the physical layer. (Figure 2)
By around the year 2000, wired high-speed broadband services had emerged, serving customers via telephone and CATV networks. Optical fiber services followed, provid-ing broadband access to consumers while boosting internet speed. Mobile broadband services started promoting de-vices that are compatible with 3G (third-generation wire-less mobile telecommunication technology) or later. The STRL kept pace with these trends by embarking on R&D on streaming technologies to offer video over the internet to complement programming over the air.
We also started investigating a new hybrid service that integrates broadcasting and communication using MPEG media transport (MMT), an industry standard to multiplex video, audio, and data for the advanced digital satellite broadcasting system. These studies are based on the prem-ise that the physical layer, which in this case is the transmis-sion circuit, is already available as part of the infrastructure. It means that our research on wired transmission technolo-gies has broadened its scope to cover a wider range of lay-ers (in terms of the OSI reference model referred to above).
In the following sections, we discuss the latest develop-ments in our research on wired transmission technologies. In Section 2, we look at CATV transmission; in Section 3, we discuss streaming video over the internet. In Section 4, we talk about uncompressed 8K transmission as well as IP transmission for video contribution, and in Section 5, we explain more about MMT.
2. CATV TransmissionAccording to figures by the Japanese Ministry of Internal
Affairs and Communications, CATV penetration in Japan is now over 50%. It has become an important part of the communication infrastructure for NHK as well, as a means to deliver its programming to even more households. Be-ginning in the days of analogue broadcasting, the STRL has been doing research on CATV transmission technologies to
distribute programming with the same high quality as over the air.
The current CATV system comprises a “headend,” a transmission network that includes trunklines and ser-vice drops, and a reception device at individual subscriber homes. A headend is a regional facility for cable compa-nies to receive over-the-air television signals and distribute them together with their own programming over the CATV system. (Figure 3.) The conventional standard was to use coaxial cable throughout the entire system. However, sys-tems like the hybrid fiber coaxial (HFC) distribution, which uses optical fiber for trunklines and coaxial cable for ser-vice drops, and fiber-to-the-home (FTTH), where both the trunklines and service drops are fiber, have become more common.2)
A single-carrier 64-QAM modulation scheme was ini-tially adopted as the modulation standard for domestic digi-tal cable television transmission. This scheme was in line with the European standard3) except for the roll-off factors and transmission rates. Later on, the 256-QAM modulation scheme was standardized following further consideration on multi-valuing.4)
In developing a 4K and 8K transmission technology that is compatible with both the HFC and FTTH systems, ideal-ly it will be a method that can operate alongside traditional digital transmission and not require changes to the existing framework. One way to transmit high-capacity signals us-ing the current digital broadcasting modulation scheme is to combine multiple channels. In 2000, the Telecommunica-tions Advancement Organization (TAO, predecessor of the current National Institute of Information and Communica-tions Technology/NICT) proposed a multiplexing method5) using a frame configuration called Transport Stream Mul-tiplexing Frame (TSMF)6), which was adopted for digital cable broadcasting. However, because this method requires all the channels to be running at the same transmission speed, it was not always compatible with the 256-QAM scheme. Thus, we proposed a method that allows multi-plexing channels at different speeds7). We also proposed a multiplexing scheme to enable the transmission of MMT
4
OSI Reference Model Main Functions
7. Application layer
6. Presentation layer
5. Session layer
4. Transport layer
3. Network layer
2. Data link layer
1. Physical layer
Communication services
Presentation of data
Management of sessions, i.e., beginning and end of sessions
Communications management
Routing
Transmission of data between devices
Connection through a physical medium
Figure 2: OSI Reference Model
FEATURE
and other variable-length packets in addition to traditional MPEG-2 Transport Stream (TS) signals.8)
We submitted our proposal to the Working Group on Ad-vancing Cable Transmission (ACT-WG) at the Japan Cable Television Engineering Association (JCTEA). The ACT-WG set up a task force and conducted a series of verification tests to assess the practicality of the method and its consis-tency with various requirements, such as high transmission rates and flexibility in multiplexing services.9) The assess-ment was reflected in discussions by the CATV UHDTV Working Group of the Broadcasting System Subcommittee under the Department on Information and Communications Technology of the Information and Communications Coun-cil. As a result, a report on the UHDTV technical require-
ments for cable television was submitted, which led to the formulation of technical standards.
We are also looking into baseband optical distribution systems (henceforth, “baseband method”) as a potential option for FTTH networks. The baseband method involves combining 8K or HD baseband signals using time-division multiplexing (TDM), which will reach around 10 Gbps, and transmitting it over optical fiber. In conventional subcar-rier multiplexing (SCM) transmission for FTTH networks, light intensity is modulated using an Frequency Division Multiplexing RF signal. The baseband method, on the other hand, modulates light intensity binarily, meaning either ON or OFF. This enables a simple structure and requires only a low signal-to-noise (S/N) ratio, which will help in
5
Headend
Modulator
Subscriber Home
V-ONU*3Wall Outlet
Wall Outlet
ProtectorOptical Node
Trunkline
Optical Fiber
Optical Fiber
Trunkline Service Drop
Hybrid Fiber-Coaxial (HFC)
Fiber-to-the-Home (FTTH)
Coaxial Cable
Feeder Line
Amplifier
Tap-off*1
Closure*2 Transmission
*1 A distributing device connecting the service drop to subscriber home*2 A device used for connecting and distributing optical fiber*3 Video-Optical Network Unit. Optical receiver for video
Service Drop
Opt
ical
Tran
smitt
er
Protector
Figure 3: Example of Cable Television Configuration
*1 Timestamped Transport Stream: Transport stream with a 4-byte timestamp tag added at the head of each packet. Each packet totals 192 bytes.
*2 Real-time Transport Protocol: Network protocol for delivering data streams, such as video, in real time.*3 Pseudo-Random Binary Sequence: A binary sequence with a characteristic that looks like a truly random sequence.*4 Small Form-factor Pluggable+ : An enhanced version of SPF modules that supports 10 Gigabit Ethernet connection.
Transmitter
SFP+*4
Time-stamping�(TTS*1)
MPEG-2 TS
MPEG-2 TS
Dummy Packet Generator(PRBS*3/UDP/IP)
UDP/IP
UDP/IP
IP Packetizing�(RTP*2/UDP/IP) TS Clock
Recovery
10G
Eth
erne
tM
ultip
lexe
r
Receiver
IP PacketSeparator
Linking IP addresses to transport streams Selecting a transport stream based on IP address MPEG-2 TS
Figure 4: Test Device for Baseband FTTH Digital Distribution
FEATUREbringing down the cost of distributing high-capacity signals over optical fiber. We compared SCM and optical baseband transmissions in a modelled scenario, where both systems distribute an equal amount of data to the same number of households in the same area. It revealed that the baseband method can reduce the number of optical amplifiers by about 20% compared with SCM.10) With the recent avail-ability of high-speed 10G Ethernet, we have developed a test device for baseband FTTH digital distribution to evalu-ate the transmission properties of the MPEG-2 TS signal. We now have a good prospect of establishing a system to distribute multiple channels to each home, just as we do with SCM.11) See Figure 4 for the configuration of the test device.
3. Video Streaming over the InternetWith broadcast and broadband coming closer together,
NHK has been working on creating services to deliver more programming and contents to a wider audience using the internet. The STRL is researching technologies for a stable, large-scale distribution of video. In this section, we will first look at the MPEG-Dynamic Adaptive Streaming over HTTP (DASH) player that we have developed at the STRL, followed by the latest progress on our research on stable distribution technology using the player.
Conventional media players for watching Internet video on television, computer, and mobile devices are all based on different technologies. This required the distributors to set up separate facilities for each system, and cost reduc-
tion and streamlining were the major challenges. The STRL developed an MPEG-DASH player,12) a common player that can be used on the web browser and serves as a uni-form video distribution platform for computers and mobile devices, including smartphones (Figs. 5 & 6). This player uses the streaming technique “MPEG-DASH,” which has been jointly adopted as the international standard by the In-ternational Organization for Standardization and the Inter-national Electrotechnical Commission (ISO/IEC), and can be used with browsers supporting HTML5, the latest ver-sion of the markup language for the web. It also complies with the operational regulations on Hybridcast laid out in December 2014 by the IPTV Forum Japan13), the domestic body that formulates the technical specifications regarding IPTV in Japan. It means that users will be able to switch between broadcast signal and streamed video-on-demand (VOD) when using Hybridcast, the integrated broadcast and broadband service in Japan.
We have been researching how we could use this MPEG-DASH player to establish a stable and reliable distribution technology. One potential is to install a function that moni-tors the reception of the video signal on the receptor de-vice14) and reports it back to the distributor. The distributor can analyze this data to dynamically alter individual dis-tribution paths accordingly, in order to avoid heavy traffic and concentrated access to the server.15) We are also study-ing ways to ensure stable and reliable streaming of video on television and other devices that are generally equipped with smaller memories and low-performance CPUs com-
6
Television
Web Browser
Laptops Mobile Devices
VideoDistribution
Media Player
Media Player
Media Player
Media Player
Broadcaster
Internet
The media player is based on the programming language JavaScript and distributed at the beginning of each video streaming session
User Interface for Television
Web Browser
Media Player
User Interface for Laptops
Web Browser
Media Player
User Interface for Mobile Devices
Uniform media player that is compatible with any device
Figure 5: MPEG-DASH Player as a Uniform Video Distribution Platform
FEATURE
pared with computers. To this end, we have conducted labo-ratory testing to identify the problems behind the current browser system. Together with other commercial broad-casters, we have raised the issue16) to the Web and TV Inter-est Group at the World Wide Web Consortium (W3C), the international body for web standardization.
IPTV Forum Japan has been working on establishing a reliable streaming environment by ensuring compatibility between different systems using the MPEG-DASH tech-nique. The Forum regards the MPEG-DASH player as its standard player.
4. Contribution Technology4.1 Uncompressed 8K Contribution Technology
In HD program production today, video signals are un-compressed throughout the production chain, from camera to monitors and editing suites, in order to ensure high qual-
ity and low latency. 8K video has approximately 33 mil-lion pixels, or 16 times more pixels compared with HD. Ideally, 8K production and transmission equipment should offer broadcasters the ability to evolve their facilities later on to handle uncompressed signals, just like they do with the current HD systems. To make this possible, the STRL is working on technologies to transmit uncompressed 8K sig-nals using optical fiber. Potential applications include facili-ties for the exchange of 8K video material between multiple points within a broadcasting station, and long-distance uni-lateral transmission facilities to send material to the station from the crew on site (Figure 7). Transmission may involve so-called “dark fiber” or unused optic fiber no longer than about 300 km, or longer-distance networks provided by communications operators.
As a basic study, we looked into how we may apply net-work communications technology such as the Local Area
7
Hybridcast-enabled Television
Tablet
Laptop
Figure 6: Play-out using MPEG-DASH Player on various devices
BroadcasterTransmission Facility
Dark Fiber
Switching Center
Network Connectivity Technology
Communication Network
Long-Distance Optical Transmission over Dark Fiber
LAN Connectivity TechnologySwitching Center
Less than 300 km
More than 300 km
In-house Facility
Internal Network
Live Field
Broadcaster
Figure 7: Applications of Uncompressed 8K Optical Transmission
FEATURENetwork (LAN) and Storage Area Network (SAN) to the transmission of uncompressed 8K signal. For the years 2007-2011, we were commissioned to undertake the proj-ect “The Next Generation High-Efficiency Network Device Technology Development” by the New Energy and Indus-trial Technology Development Organization (NEDO) to develop an ultrahigh-speed LAN-SAN system for 8K dis-tribution inside broadcasting stations. In this system, 4 × 40 Gbps optical transmission frames called Optical Channel Transport Unit-3 (OTU3)17) are bundled into one 160 Gbps signal using Optical Time-Division Multiplexing (OTDM). We have prototyped an experimental device that can effi-ciently multiplex uncompressed 8K signals into OTU3s. Evaluation tests have given us good results.18)
Then came the 100 Gbps Ethernet, which is much faster than the 40 Gbps OTU3s. It is expected to lower the prices of relevant equipment, so we decided to use this Ethernet technology to develop a new system. In conventional in-house networks within a broadcasting station, high-speed
signals such as uncompressed HD had been transmitted over a dedicated system, independent from the Ethernet network handling file data from servers. However, we are starting to see an increased use of 10 Gbps Ethernet for the trans-mission of uncompressed HD. With 100 Gbps Ethernet,19) we will be able to establish an efficient integrated in-house network for local 8K transmission. Our studies so far in-clude frame conversion technology for the transmission of 72 Gbps uncompressed, full-resolution 8K signals over 100 Gbps Ethernet, methods to control the video clock on the receiving end for a reliable and low-latency transmission,20) and measures against packet loss.21) We are also working on transmitting 144 Gbps uncompressed, full-specification 8K signals over two 100 Gbps Ethernet networks.22)
One of the challenges in using dark fiber for transmis-sion is the amount of time and money needed to install an erbium-doped fiber amplifier (EDFA) every few dozen ki-lometers over the entire network. This has led us to look into the possibility of using distributed Raman amplifica-
8
AIST
Full-resolution 8K/60P�Camera output (72 Gbps)
+22.2 ch Audio
Interface Conversion
Transmission Waveform�(12ch DWDM*)
Error Correction/�Wavelength Mux Transmitter
(Developed by AIST)
* Dense Wavelength Division Multiplexing
NHK STRL
Wavelength Demux/Error Correction Receiver
Interface Reconversion 8K Monitor
Optical PathNetwork
Dark Fiber laid by AIST Distance: 173 km Total Loss: 72 dB
Optical Amplifiers: 4
Receiver Waveform10.692 Gbps per channel
Figure 8: Configuration of the 173 km Long-Distance Transmission Experiment
FEATUREtion (DRA) in which the optical fiber for transmission also serves as the amplifier. We conducted an experiment to send full-resolution 8K signals over a distance of 300 km with-out a relay station, and we have concluded that using DRA enables us to send the signals 100 km farther than without DRA.23) The full-resolution 8K signals in this case were multiplexed, bringing two 43 Gbps optical signals into one using return-to-zero differential quadrature phase-shift key-ing (RZDQPSK) modulation.
Another issue with developing a transmission network using dark fiber is the interface compatibility between equipment. In March 2014, the Association of Radio In-dustries and Businesses (ARIB) in Japan set out the stan-dard ARIB STD-B58 (“Interface for UHDTV Production Systems”). It is an interface (I/F) standard designed for the transmission of ultrahigh-resolution television signals be-tween studio equipment, and as such, only envisions short-distance transmission. We need to establish a system that conforms to this standard but is able to transmit uncom-pressed, full-resolution 8K signals over longer distances. We therefore developed a method of incorporating error-correction codes (ECCs). Through a series of lab testing, we found that the ECC method can extend the transmis-sion distance by more than 20 km compared with the non-ECC method. We conducted further tests with the National Institute of Advanced Industrial Science and Technology (AIST). Using a single-core optical fiber and optical am-plifiers, we were able to successfully transmit a 72 Gbps uncompressed, full-resolution 8K signal between STRL in Tokyo and the AIST facility more than 173 km away.24) See Figure 8 for the configuration of this experiment.
4.2 Using Wired IP Technology for Wireless IP Transmission
Underwater
The STRL has made progress in R&D on wired IP trans-mission technologies that adjust the video compression rate according to how heavy the network traffic is.25) We are now
working on applying this technology to wireless IP trans-mission in order to establish a reliable method of sending video over the types of network that may be affected by external factors, changing the transmission speed dramati-cally. In this section, we will discuss wireless IP transmis-sion underwater. It is a very challenging environment for wireless transmission, but the STRL has succeeded in de-veloping a wireless method to send live video from under-water cameras (Figure 9).
Radio waves are greatly affected in water and is virtually impossible to use for wireless transmission. That is why conventional live transmission from underwater has been carried out using long camera cables. However, camera cables can be pushed around by waves and tides, causing serious safety hazards. Producers need to have several div-ers to secure the cable and ensure safety underwater.
Our newly developed underwater transmission system eliminated camera cables by using blue visible light, which is less affected by water. We have also developed a system to automatically optimize the transmission rate to ensure that the IP transmission is suitable for live transmission, even if it is obstructed by occasional fish or by seaweed in the water. When the transmission light is temporarily dis-rupted and results in loss of data, the receiver automatically sends a request for a prioritized retransmission of missed-out data to ensure stable transmission.
The following is the description of how the adaptive transmission rate control (ATRC) works in wireless under-water IP transmission.1) Error correction code is added to the video from the un-
derwater camera before being sent over to the wireless transmitter (Figure 10(a)). Minor errors such as those caused by a slight deviation of the optical axis can be corrected with the error correction data.
2) When the transmission is temporarily but totally disrupt-ed by an obstacle, the video data is lost and becomes uncorrectable (Figure 10(b)).
9
Underwater Camera
Video Monitor(on boat, etc.)
Video Transmitter
Video Receiver
Underwater
Above water
IP Technology to Stabilize Video Transmission
Blue Visible-Light for Underwater Wireless Transmission
Figure 9: Underwater Wireless IP Transmission
FEATURE
3) When the obstacle moves on and transmission is re-sumed, the transmission rate for the ongoing video and error correction code is lowered, allowing time to resend the lost data. Retransmission of the lost data is given pri-ority to minimize the break in video transmission (Fig-ure 10(c)).We conducted a simulation test using a network simu-
lator to see how well the ATRC works when the wireless transmission is disrupted26) (Figure 11). We used H.264/ Advanced Video Coding (AVC) and Advanced Audio Cod-ing (AAC) for video and audio coding, respectively, and
User Datagram Protocol/Internet Protocol (UDP/IP) for transmission. We allocated a total maximum transmission rate of 48 Mbps for video, error correction, and retransmis-sion data. Video data rate was varied between 3-8 Mbps depending on the condition of the transmission, allowing the rest of the data rate to be allocated for error correction and retransmission.
The result of the simulation test is shown in Table 1. The IP Packet loss rate refers to the rate of packets lost out of all the UDP/IP packets and indicates the loss rate for the transmission circuit. The Media Packet loss rate is the rate
10
(a) Transmission of Video Data and Error Correction Codes
Error Correction Codes
Video Data
RetransmissionRequest for Retransmission of Missed Data
Error Correction
Video Data
(b) Transmission Disrupted by Obstructing Object
(c) Transmission Resumed after Object is Cleared
Minor errors during transmission can be corrected using error correction codes
Data is disrupted, making it difficult to reconstruct video
Priority is given for the retransmission of missed-out data
Video Transmitter
Video Transmitter
Video Transmitter
Video Receiver
Video ReceiverObstruction
Video Receiver
Figure 10: Optimizing Transmission Rates in Underwater Wireless IP Transmission
Video�Player IP Video�Transmitter Network�SimulatorHD-SDI*1 HD-SDIUDP/IP*2 UDP/IP
IP Video�Receiver
*1 SDI : Serial Digital Interface�*2 UDP/IP : User Datagram Protocol/Internet Protocol
Video�Monitor
Figure 11: Configuration to Test Performance of the Transmission Rate Optimization
Transmission Disruption Pattern Packet Loss Rate (ave. per min)
Duration of Disruption
Duration ofTransmission
Media PacketIP Packet
ITU-T standardization
200 ms
400 ms
600 ms
600 ms
400 ms
200 ms
25%
50%
75%
0.0%
0.0%
15%
Table 1: Results of the Transmission Rate Optimization Test
FEATUREof loss after ATRC as well as retransmission of any packet initially lost, so a media packet loss rate of 0.0% indicates there is no disruption in the video.
Table 1 shows that if the IP packet loss is less than 50%, the media packet loss is 0.0%. In other words, we now know that if we can keep the packet loss of the transmission network to less than 50%, we can do live transmission from underwater without video disruption.
5. R&D on Media Transport Methods and their Standardization5.1 R&D on Transmission Technologies using MMT
MMT is a standard for multiplexing broadcast and/or communication streams for the distribution of contents. It specifies the requirements on various components, including the video and audio formats, media transport protocol, and signaling information messages that indicate the structure of the content. Coordinated Universal Time (UTC) is used for the management of presentation time, enabling services that require receivers to synchronize multiple video and audio feeds delivered via different distribution paths.
The Moving Picture Experts Group (MPEG), a working group under ISO/IEC, set out to develop and standardize MMT in 2009. NHK has been participating in its discus-sions from the beginning, while also working on the standardization of 4K and 8K satellite broadcasting, which began test broadcasting in August 2016. We have also been prototyping relevant hardware to conduct various verifica-tion tests.
During the annual STRL Open House in May 2015, we conducted a public 8K satellite broadcasting experiment
(Figure 12). 8K video multiplexed at NHK’s broadcasting center using MMT was transmitted via the BSAT-3b broad-cast satellite and received at the STRL. The demonstration also featured a CATV transmission scheme that receives and retransmits MMT multiplexed 8K programming.
Also during the 2015 Open House, we demonstrated two potential hybrid services, multiview and targeted program promotion, as part of a test transmission using equipment conforming to ARIB standards.27) The Japanese telecom NTT provided the optical fiber network testbed called GEMnet2 while CATV Jupiter Telecommunica-tions (J:COM) contributed its hybrid fiber-coaxial (HFC) network for the test28) (Figure 13).
A multiview service offers additional, synchronized video over the internet to complement the main broadcast programming. The complementary content may be video from an alternative angle. In the demonstration, the main 4K video was transmitted from the MMT “master” server located within the STRL, went through a pair of satellite modulator and demodulator, and was received by the MMT receiver “A”. The 2K complementary video was transmitted from the MMT “slave” server located at a separate J:COM facility. It goes through the distributor modem called cable modem termination system (CMTS), J:COM’s HFC net-work, the receiver’s modem, and finally to the same MMT receiver “A”. (The signal is converted to Data Over Cable Service Interface Specifications (DOCSIS) 3.0 when going through the HFC network). The MMT receiver “A” was able to refer to the UTC timestamp multiplexed in both the main and complementary videos to synchronize the two and provide a multiview experience.
11
Camera System
Transmission
Scrambling
Multiplexing MMT
Coding
Reception
Descrambling
Demultiplexing
DecodingMultichannelmicrophone
Closed-captioning
22.2 ch sound
Cable TV transmission system
Broadcasting Satellite
* U-SDI: Ultrahigh-definition Signal/Data Interface
8K Television
Record
ContributionTransmission
U-SDI*Interface
MMT
Figure 12: Overview of 8K satellite broadcasting experiment
FEATURE
The other example of hybrid services demonstrated at the Open House was the targeted program promotion. It uses user data stored in the receiver, such as gender, age, and program preferences, to show viewers either the main pro-gramming broadcast over the air or the alternative program-ming sent via broadband. For the test (Figure 13), the main programming from the master MMT server at STRL was transmitted via the same path as the main multiview video, then received by the MMT receivers “B” and “C”. The alter-native programming was sent from the same master MMT server but was switched out to go through the GEMnet2 network, finally reaching the MMT receiver “C”. Receiver “C” was able to refer to the UTC timestamp muxed on both the main and alternative programmings, replacing the main programming with the alternative without any glitches.
5.2 Standardization of Media Transport Methods
In Japan, MMT has been adopted as the transport method for the 4K and 8K satellite broadcasting, which started test broadcasting in August 2016. Its standard, the ARIB STD-B60, was set out in July 2014. In December 2015, the Next-Generation Television and Broadcasting Promo-tion Forum (NexTV-F) published a volume of technical data as operational guidelines (NEXTVF TR-0004 ver. 1.0).29) Internationally, MPEG has been working on the standardization of a media transport method suitable for the IP network environment. The work began in 2009 and MMT was designated as an international standard in March 2014.30) It was followed by the publication of the Implementation Guidelines31) by the ISO in February 2015, in which Japan’s 4K and 8K broadcasting system is referred to. In parallel to this move, the International Telecommuni-cation Union Radiocommunication Sector (ITU-R) adopted
a new recommendation32) in June 2015, referring to the Japanese ARIB STD-B60 in detail. In the United States, the nonprofit organization Advanced Television Systems Com-mittee (ATSC) has been working on the development and standardization of the next-generation terrestrial television system, ATSC 3.0, since September 2011. A protocol stack is expected to be adopted in the near future, which would specify MMT and DASH/ROUTE* in its transport layer.
6. SummaryIn this report, we detailed the transmission technologies
that the STRL has been working on to transmit video and audio signals over wired networks. While R&D on ter-restrial and satellite broadcasting takes a holistic approach and aims to enhance everything from transmission to delivery based on the spectrum allocated for broadcast-ing, R&D on wired transmission stands on the basis that relevant infrastructure is already there. In other words, it needs to consider how to implement broadcasting services onto existing infrastructures such as CATV, the internet, and communications networks. Technologies around these infrastructures evolve extremely fast. We intend to keep up with their development and work on our own R&D to deliver even better services to our audiences.
12
* ROUTE: Real-time Object Delivery over Unidirectional Transport
1GHz Band IF
HFC Network
Inter-Server Synchronization Control
Targeted Trailer
MultiviewMain Video(4K) Satellite
ModulatorSatellite
Demodulator MMTReceiver A
MMTReceiver B
MMTReceiver C
Multiview
GEMnet2Switcher
GEMnet2SwitcherRouter
DOCSISCMTS
DOCSISModem
Targeted Trailer(Main Prg.)
MMTServer
(Master)
Targeted Trailer(Alternative Prg.)
MultiviewComplementary
Video (2K)
NHK STRL
J:COM
NTT
MMTServer(Slave)
Figure 13: Configuration of Hybrid Service Experiment
FEATURE
13
References
1) T. Nakatogawa, K. Oyamada: “Research and development history and vision of optical fiber transmission technologies by NHK Science & Technology Research Laboratories,” The papers of Technical Meeting on “History of Electrical Engi-neering”, The Institute of Electrical Engineers of Japan His-tory of Electrical Engineering, History of Electrical Engineer-ing, HEE-14-017, pp.17-22 (2014) (in Japanese)
2) Ministry of Internal Affairs and Communications: “Current status of CATV (in Japanese),” February 2016, http://www.soumu.go.jp/main_sosiki/joho_tsusin/pdf/catv_genjyou.pdf (in Japanese)
3) ETSI EN 300 429, “Digital Video Broadcasting (DVB) : Framing Structure, Channel Coding and Modulation for Cable Systems” (1998)
4) ITU-T Rec. J.83, “Digital Multi-Programme Systems for Television, Sound and Data Services for Cable Distribution” (2007)
5) H. Ohsuka, N. Nakamura, T. Noda, K. Oyamada, S. Itoh: “A New Division Transmission Method for High-speed MPEG-2 TS Signal by Using Multiple Carriers over Cable TV Net-works,” The Journal of The Institute of Image Information and Television Engineers, Vol.59, No.1, pp.58-64 (2005)
6) ITU-T Rec. J.183, “Time-division Multiplexing of Multiple MPEG-2 Transport Streams over Cable Television Systems” (2001) (in Japanese)
7) Y. Hakamada, N. Nakamura, T. Kurakake, T. Kusakabe and K. Oyamada: “UHDTV(8K) Distribution Technology and Field Trial on Cable Television Networks,” ITE Trans. on MTA, Vol.2, No.1, pp.2-7 (2014)
8) Y. Hakamada and T. Kurakake: “An Encapsulation Scheme of Variable-Length Packets for UHDTV Distribution over Existing Cable TV Networks,” IEEE ICCE2016, pp.480-481 (2016)
9) T. Kurakake, Y. Hakamada, T. Konishi, N. Maeda, H. Hira-oka, N. Nakamaru, A. Sugimoto: “JCTEA Evaluation Tests for Advanced Cable TV Transmission System with a Chan-nel Bonding Technology,” The Journal of The Institute of Im-age Information and Television Engineers, Vol.69, No.12, pp. J361-J363 (2015) (in Japanese)
10) T. Kusakabe, T. Kurakake, K. Oyamada and Y. Fujita: “Evalu-ation of Optical Network of Multichannel Baseband Digital Broadcasts over FTTH,” ITE Trans. on MTA, Vol.2, No.3, pp. 267-277 (2014)
11) Y. Hakamada, N. Nakamura, K. Oyamada: “Multi-Channel Digital Broadcasts Distribution System,” IEICE technical report, Vol.114, No.119, CS2014-38, pp.127-130 (2014) (in Japanese)
12) https://www.nhk.or.jp/pr/marukaji/pdf_ver/396.pdf13) IPTV Forum Japan: “Operational Guideline of Hybridcast (in
Japanese),” IPTVFJ STD-0013 2.0 (2014)14) http://www.nhk.or.jp/strl/open2015/img/pdf/tenji_12.pdf15) S. Tanaka, S. Nishimura, M. Yamamoto, Y. Endo: “A Study
of A Delivery Path Control Method for Live Video Stream-ing over Mobile Network,” IEICE General Conference 2016, B-6-128 (2016) (in Japanese)
16) K. Hoya, S. Nishimura and S. Harada: “MSE/EME : Poten-tial Implementation Issues on TV Sets A Case Study of IPTV Forum-Japan Player,” W3C TPAC2015 Web and TV Interest Group, https://www.w3.org/2011/webtv/wiki/images/6/66/ Webtv_mse_eme_iptvfj_player_20151026.pdf
17) ITU-T Rec. G.709, “Interfaces for the Optical Transport Net-work”
18) T. Nakatogawa, K. Oyamada, H. Onaka, S. Namiki, T. Asami: “Fundamental Experiment of Ultrafast Optical LAN-SAN Technology for Full Resolution Super Hi-Vision Distribu-tion,” IEICE Technical Report, Vol.110, No.417, IA2010-80, pp.51-54 (2011) (in Japanese)
19) J. Kawamoto, T. Nakatogawa, K. Oyamada: “Development of a Super Hi-Vision Signal Transmission on 100-Gigabit Ether-net,” ITE Annual Convention, 7-2 (2013) (in Japanese)
20) J. Kawamoto, T. Nakatogawa, K. Oyamada: “A Proposal of Clock Recovery and Control Method for an 8K UHDTV Sig-nal Transmission over 100 Gigabit Ethernet,” The Journal of The Institute of Image Information and Television Engineers, Vol.J98-B, No.10, pp.1127-1136 (2015) (in Japanese)
21) J. Kawamoto, T. Nakatogawa, T. Kurakake: “Study of Error Correction Method to Improve Burst Loss Tolerance of Video Signal Transmission Using IP Packet,” ITE Technical Report, Vol.39, No.38, BCT2015-70, pp.5-8 (2015) (in Japanese)
22) J. Kawamoto, S. Oda, T. Kurakake: “Development of 100 Gigabit Ethernet Transmission Equipment for Full-specifica-tion 8K UHDTV,” IEICE General Conference 2016, B-8-14 (2016) (in Japanese)
23) J. Kawamoto, T. Nakatogawa, K. Oyamada: “Long Haul Op-tical Transmission Experiment of a Full-Resolution Super Hi-Vision Signal Using Distributed Raman Amplifiers,” ITE Winter Annual Convention, 5-12 (2012) (in Japanese)
24) J. Kawamoto, T. Nakatogawa, T. Kurosu, S. Suda, S. Namiki and T. Kurakake: “Field Transmission of an Uncompressed 8K Ultra-High Definition Television Optical Signal with For-ward Error Correction Codes,” OptoElectronics and Commu-nications Conference 2015 (OECC 2015), JMoA.34 (2015)
25) S. Oda, K. Kamimura, H. Hoshino, K. Aoki, M. Yamamoto: “Enhancement of Hi-Vision IP Transmission Device pursu-ing Available Network Bandwidth,” ITE Technical Report, Vol.34, No.14, BCT2010-39, pp.7-12 (2010) (in Japanese)
26) M. Kurozumi: “IP Transmitter/Receiver for Underwater Wire-less Transmission,” ITE Annual Convention, 31D-2 (2015) (in Japanese)
27) “MMT-Based Media Transport Scheme in Digital Broadcast Systems,” ARIB STD-B601.0 (2014) (in Japanese)
28) Y. Kawamura, K. Otsuki, A. Hashimoto and Y. Endo: “Func-tional Evaluation of Hybrid Content Delivery Using MPEG Media Transport,” ICCE2016, pp.267-268 (2016)
29) Next Generation Television and Broadcasting Promotion Forum, Technical Report: “Operational Guidelines for Ad-vanced Digital Satellite Broadcasting,” NEXTVF TR-0004 1.0 (2015) (in Japanese)
30) ISO/IEC 23008-1:2014, “Information Technology - High Ef-ficiency Coding and Media Delivery in Heterogeneous Envi-ronments - Part 1: MPEG Media Transport (MMT)” (2014)
31) ISO/IEC 23008-13:2015, “Information Technology - High Ef-ficiency Coding and Media Delivery in Heterogeneous Envi-ronments - Part 13: MMT Implementation Guidelines” (2015)
32) Rec. ITU-R BT.2074-0, “Service Configuration, Media Trans-port Protocol, and Signaling Information for MMT-based Broadcasting Systems” (2015)