DVB-NGH for Next-Generation Digital Broadcasting
March 6, 2012
Since I wrote last month's article on Next Generation Digital Broadcasting, I've been researching what other countries are doing to meet the need for a new broadcast platform that meets consumers' demands for broadcast content anywhere, anytime on any device.
I don't see how next generation broadcasting can be made compatible with today's ATSC DTV tuners and achieve the desired spectrum efficiency and interoperability with wireless broadband. Ideally the system should be able to cover use cases from fixed reception on larger screens in home (with or without an outdoor antenna) and small screens (smartphones or tablets, for example) in the home, in the office, in cars, on trains and wherever these devices are likely to be used.
ADDING NEW FEATURES TO DVB-T2
I first looked at DVB-T2. This technology incorporates many of the desired features and its "Future Extension Frame" (FEF) provides the opportunity to add new features. However, I found DVB-T2 was not originally designed for use with mobile/handheld devices. That capability can be provided through the FEFs, as has been done with DVB-T2-Lite, a less complex version of DVB-T2 designed for use with smaller devices. The BBC has conducted tests if DVB-T2-Lite transmitted along with two HDTV program streams on a DVB-T2 signal. (See references at the end of this column).
Release 1.3.1 of the DVB-T2 standard, produced in June 2011, extends the DVB-T2 "base" standard by adding a "mobile profile." Additional improvements are planned for DVB-NGH ("Next Generation Handheld"). DVB-NGH specifications are expected to be published by the DVB project early this year. It provides a robust transmission platform based on DVB-T2 and the necessary protocols to allow interoperability with wireless broadband networks using LTE 3GPP technology.
The best document I found describing how DVB-NGH works is from Mobile Multi-Meda (M3), a French collaborative project: Release 1.0 of "D21.1 – Analysis on 3GPP E-MBMS/DVB-NGH Physical Layer Convergence," published July 2011. The document shows why DVB-NGH is better than the LTE/3GPP E-MBMS ("Evolved Multimedia Broadcast Multicast Services') in terms of spectrum efficiency and utilization of existing broadcast antennas. The business model at the start of the document closely matches what's been proposed for the "Broadcast Overlay" model I recently wrote about.
The LTE E-MBMS platform is not a good fit for a wide area broadcast model because the maximum guard interval is only 33.33 µs, leading to a maximum cell radius of 10 km (six miles). DVB-NGH include much large guard intervals, allowing a maximum cell radius of up to 107 km (67 miles), suitable for a high power, high antenna transmitter site and multiple low/medium power transmitters in a single frequency network.
According the M3 analysis, DVB-NGH offers a 50 percent advantage in effective spectrum efficiency over E-MBMS. Because latency is not an issue in a broadcast mode, DVB-NGH permits much longer interleaving—up to 250 ms compared to 1 ms, although in practice it is likely to be limited to 50 to 100 ms to decrease receiver memory requirements.
NO ONE SIZE FITS ALL
Different transmissions complicate developing a convergent Mobile DTV standard. The M3 paper starts with the traditional 10 MHz E-MBMS signal with an FFT size of 1024 and scales it to work in 6, 7 or 8 MHz bandwidths. If 3GPP parameters are extended to 6 MHz, the effective bandwidth is 5.4 MHz, very close to the bandwidth of an ATSC 8-VSB signal. A second mode is proposed which would increase the effective bandwidth to 5.76 MHz.
The M3 paper covers the parameters of the DVB-NGH signal but does not provide examples of a system design including signal-to-noise performance, interference and real-world coverage. MIMO technology is not discussed either, although given the small size of handheld devices compared to VHF and UHF wavelengths, a MISO (multiple-input/single-output) technology is more applicable. DVB-T2 can use modified Alamouti coding, also called "space-time block coding," to improve coverage. This can be accomplished using one site with two transmit antennas with opposite polarity at the same site or two different sites, each with its own antenna.
EBU–TECH 3348, "Frequency and Network Planning Aspects of DVB-T2," provides more detail on system technical parameters, including data rate and C/N performance for various configurations including Gaussian, Ricean, Rayleigh and 0 dB echo channels for VHF and UHF channels.
Could DVB-NGH or a variation of it provide the basis for ATSC 3.0? The DVB-NGH standard is still being finalized and testing will be needed to see how well it performs in the field. There are DVB-T2 transmissions on the air now in a number of countries and work is being conducted on enhancements, including MISO and time and frequency slicing (TFS), which would allow frequency diversity, should provide some guidance.
One aspect of DVB-T2/NGH that would make it easier to introduce here is it can use the same 6 MHz channel bandwidth broadcasters are now using for ATSC and ATSC-MH, allowing stations to transition one-by-one to the technology using the same antenna they use for ATSC 8-VSB and the same transmitter. DVB-T2 includes techniques for reducing the peak-to-average power ratios of the signal, allowing higher average power than DVB-T.
I'll have more on DVB-NGH and other options for next generation digital broadcasting future columns. I welcome your comments. Do you think broadcasters will be able to agree on a new standard that isn't compatible with ATSC? If so, will it be in time to avoid losing mobile/handheld viewers to wireless broadband IP multicast downlink only technology like AT&T is proposing for use in the spectrum it is acquiring from Qualcomm?
I welcome your comments and suggestions! Email me at firstname.lastname@example.org.