If you're in the market for a new smartphone or wireless broadband device, you've probably seen the term "LTE" in conjunction new high speed "4G" offerings from Verizon. The current LTE release does not meet the ITU definition of 4G, but advanced LTE releases will. LTE is being deployed by companies like Verizon on what used to be TV broadcast channels 52-–69 and will also be used at 1.6 GHz in LightSquared's hybrid satellite/terrestrial network. This month I'll provide a brief explanation of LTE and also look at the multicast (broadcast) capability included in the standard. Could it replace ATSC at some point in the future?
WHAT IS LTE?
LTE is an acronym for "Long Term Evolution." It is a 3GPP standard providing, in the current release, uplink speeds of up to 50 Mbps and downlink speeds of up to 100 Mbps in a 20 MHz channel. At the physical layer, it uses Orthogonal Frequency Division Multiplexing (OFDM) and Multiple Input Multiple Output (MIMO) transmission. While OFDM has been used for many years for WiFi and broadcasting (DVB-T and DVB-H plus variations such as FLO), this is the first use, as far as I know, of OFDM for cellular networks. Prior to LTE, cellular standards were based on single carrier transmission, just as ATSC is today.
After building out single carrier cellular networks, why would wireless companies change to LTE? The advantages include downlink spectrum efficiency 3 to 4 times that of HSDPA (one of the most efficient single carrier methods), lower latency (under 10 ms. after the link is established), scalable bandwidths (1.25, 2.5, 5.0, 10.0 and 20 MHz), and support for MIMO configurations up to 4 x 2 in the downlink mode. The standard also supports M-SFN and repeaters. LTE repeaters can operate in an "on-frequency" mode when there is sufficient receive/transmit antenna isolation or use a time offset between the received and transmitted packets when needed to reduce self interference.
Readers who followed the COFDM versus 8-VSB debate will also remember that OFDM is less susceptible to multipath interference as long as it is within the guard interval. MIMO minimizes the impact of frequency selective fading. However, as you may remember from that debate, OFDM has its disadvantages, the main one being a much higher peak-to-average ratio, which reduces RF amplifier efficiency. Indeed, LTE uses Single Carrier Frequency Division Multiple Access (SC-FDMA) for the uplink to reduce amplifier power demands in user handsets and devices.
Verizon and AT&T both recently announced plans to use some of their spectrum for broadband multicasting (broadcasting is the old-fashioned term for it) to reduce the load (one stream per user) video streaming is placing on their unicast IP networks. It isn't surprising that LTE includes provision for broadcasting. LTE Multimedia Broadcast Multicast Services (MBMS) can operate in either a single-cell mode using the downlink shared channel (DL-SCH) or in a multi-cell mode where transmissions from cells are synchronized to form a Multicast/Broadcast—Single Frequency Network (MB-SFN). While the LTE standard specifies full performance at distances up to 5 km per cell, it suffers only slight degradation from 5 km to 30 km and should support operation up to 100 km. Obviously these distances are a bit short for single stick broadcasting. More study is needed, but at first glance it appears to me that fewer transmitters would be needed than with an ATSC-based distributed transmission system.
This workflow chart illustrates the concept of "Universal Broadband Broadcasting." Source: Mark Aitken, director of advanced technology, Sinclair Broadcast Group NEXT-GEN ATSC STANDARD?
It isn't surprising that a rough plan for using LTE for broadcasting by TV stations has already been presented to the FCC. Mark Aitken, Sinclair Broadcast Group's Director of Advanced Technology presented an outline of next generation broadcasting using LTE in a meeting with FCC staff from the Office of Engineering and Technology and the Media Bureau. The presentation, which is available on the FCC Electronic Comment Filing System (ECFS), calls the technology "ATSC-EV." The system would be based on OFDM and incorporate the LTE multicast options discussed earlier but is presented separately from the LTE MBMS format.
Could it work? LTE offers 100 Mbps downlink speeds in a 20 MHz wide channel before subtracting bandwidth for error correction, signaling and other housekeeping functions, The presentation shows ATSC-EV (Next Gen) offering 154 Mbps raw (before subtracting error correction and signaling) data bamdwidth. Since new demodulators would be required for ATSC-EV, it would make sense to switch to a more efficient video codec. However, the presentation shows that 14 channels would be able to provide 330 Mbps (using DVB-T2 parameters from the UK) for fixed service only. Even if the 19.39 Mbps data bandwidth per station is retained, 17 stations would be able to fit into the same spectrum (including guard bands) where 14 stations are now.
Whether or not ATSC-EV is the ultimate solution, it seems clear that TV broadcasting has to be given the regulatory opportunity to modify its transmission technology to keep up with other industries if broadcasting is going to survive. Figuring out how to get there from here without disenfranchising a large number of viewers is the tough part. Sinclair received help from engineers at Rohde and Schwarz and SES World Skies in its FCC presentation.
Comments are welcome! E-mail me at firstname.lastname@example.org.