One is Good, Two are Better
Feb 1, 2005 1:00 AM, By John Battison, technical editor
The road to IBOC development has been long and sometimes rough. The swell of stations considering and implementing the system show favorably on the technology, and the anticipated NRSC announcement to further establish a standard helps to cement the path that has been set. The system itself is still seeing development an innovation. While stereo has been the pinnacle of enhanced radio for many years, surround sound and multichannel sound audio have moved into the spotlight, mainly through home theater and enhanced consumer audio formats.
In the November Insight to IBOC, we looked at the progress of surround sound with IBOC. In this installment we turn our attention to the multichannel technology, such as that developed for the joint project from NPR, Harris and Kenwood called Tomorrow Radio. The official technical designation is supplemental audio channel (SAC). The industry has more generically named the system HD Multicast. Conceived by NPR in 2002 and demonstrated in early 2004, this expansion of audio services is particularly gratifying to NPR stations because it opens new avenues for high-quality programming. On Jan. 9, 2004, the triumvirate announced the completion of development and successful testing of the project.
Figure 1. The original mobile test setup.
Although this new system was fronted by a noncommercial entity, its use is not restricted to noncommercial stations. As a direct result of the tests at the end of 2003, the future looks bright for acceptance of this system. Following the successful development of transmitting and receiving equipment, a series of in-depth field tests were conducted to evaluate reception and coverage not only at fixed locations, but also in mobile receivers. Standard FM reception characteristics in the stationary receivers are well known, but the vagaries of mobile reception needed to be evaluated to ascertain their effect on the SAC signal. It was important to ensure that the signal was sufficiently robust in the real world to provide adequate second program service in the absence of the blend-to-analog feature of the main program channel.
Supplemental audio channels were added to the standard IBOC radio spectrum by splitting the digital FM spectrum into two channels. For the initial tests, the regular 96kb/s was split into a main channel of 64kb/s and a supplemental audio channel of 32kb/s respectively. Since that time, NPR listening tests have shown that two 48kb/s channels will provide satisfactory audio performance.
NPR retained the engineering firm of Hammett and Edison to perform the necessary field testing of the system. Four areas of the country were selected. In 2001/2002 IBOC preliminary testing was carried out by WETA (90.9MHz) in Washington, DC, and WNYC-FM (93.9MHz) New York. The routes selected for the new test series were similar to those used in earlier IBOC signal tests. In San Francisco the test involved KALW (91.7MHz) where earlier IBOC test routes were used and KKJZ (88.1MHz) in Los Angeles/Long Beach where three new test road loops were used.
All four of the test areas were originally measured in August and September of 2003 and rechecked from October through December 2003. The original installation used two receive antennas, a preamplifier and a power splitter. This was not considered satisfactory because of RF noise and uncertainty involving the RF network. The change to a single antenna resulted in an improved, simplified arrangement.
Figure 2. The modified mobile test setup.
The original PAC audio coder used in the earlier tests was replaced by the HDC audio coder for the later tests. This retesting was performed in the Washington and New York markets. Apparently the difference in the results was not considered to be strongly significant and it was confined to the eastern market stations. Figure One shows the mobile test equipment as originally connected. Figure Two shows the test equipment setup as finally modified.
To provide clear and concise test results the report includes maps that show the routes followed, the measured field strengths and signal assessment. The four test areas were chosen to represent four types of terrain. The standard FCC method of predicting coverage obviously would not suffice for these widely differing terrain conditions. The Terrain Integrated Rough Earth Model (TIREM) was used because it evaluates the profile between sites and selects the most probable mode of propagation that results based on terrain profiles. This system uses the USGS three-second terrain database and is considered to be the most accurate method of predicting signal strength where propagation paths are accurately known.
The FCC defines the receiving antenna height as 30ft. above ground. This value was used in the TIREM calculations. Because of this, some of the contours shown represent specified field intensities, which may or may not be the same as FCC service contours. The Hammett and Edison Report includes a large number of maps that show the performance of the received SAC signal while driving around the metro of the test city.
Because of the size of the city of New York several maps were provided in the report. Shown on page 6 are the results of one of the New York City campaigns. The report shows that a 61dBu TIREM service area was obtained for the SAC channel.
Based on the contents of the report, it appears 95 percent certain that Tomorrow Radio will provide a usable service area within an FM station''s 60dBu to 70dBu service area, assuming that the IBOC performance of production line receivers from Kenwood and other manufactures will be the same as the equipment used in the tests. It is possible that short-spaced co-channel or adjacent-channel stations and unusual terrain features could degrade the Tomorrow Radio service area. On the other hand, given ideal conditions the service area could exist below 60dBu.
Results of one of the New York field Tests.
While the Phase Two work was performed many months ago, NPR has not been inactive with the project. During 2004, NPR conducted three tests to better understand and evaluate various aspects of multichannel IBOC. The tests covered listener perception of the quality of low bit-rate codecs, listener perception of the Ibiquity HDC codec at various bit-rates, and interference between the host analog signal and the extended-hybrid operating mode.
The listening tests on coding rates provided two interesting conclusions. First, that listeners cannot find a noticeable difference between the HDC codec at various rates from 48kb/s and above. In an A/B comparison, differences could be heard, but the normal listener will not have the A/B opportunity. The second conclusion finds that the 96kb/s FM mode could be split into two 48kb/s streams with minimal if any disturbance to the listener. This mode would provide an equal level of quality for two streams instead of the original 64kb/s vs. 32kb/s split in the Phase Two Tomorrow Radio tests.
At the 2005 CES in January, NPR presented a request for information for a potential group buy of as many as 50,000 HD Multicast receivers to all licensed HD Radio receiver manufacturers. This news piqued the interest of the radio receiver manufacturers at the convention. In addition, NPR will launch new NPR HD Radio program streams covering music and news/talk, which will be targeted for secondary channel use. The target launch date for the first of these NPR channels is June 2005. In addition, the NRSC has created a new working group to work with this technology.