Calls transmitter separation proposal ‘spectrally inefficient and overly conservative’
The FCC seems intent on squeezing every last bit of spectrum from UHF broadcasters, even if it means adopting different wireless band plans in different parts of the country. A channel that's used for broadcasting in one market could be auctioned for wireless broadband in an adjacent market if it meets the proposed interference limits outlined in a Public Notice released January 29.
The document, titled “Office of Engineering and Technology Seeks to Supplement the Incentive Auction Proceeding Record Regarding Potential Interference Between Broadcast Television and Wireless Services, says that a number of commenters have raised concerns about co-channel and adjacent-channel interference between television and wireless services in response to the FCC's NPRM and 600 MHz Band Plan Supplemental Public Notice.
The most common approach suggested to reduce interference between the services uses predefined separation distances between the services. The separations suggested ranged from 100 to 500 km.
In the Public Notice the FCC dismisses the separation distance approach, noting:
“We are concerned that prescribing a pre-defined separation distance as proposed by some commenters may be spectrally inefficient and overly conservative. Specifically, this approach lumps together all of the above cases and applies separation distances based on a worst case scenario without considering factors such as the actual technical characteristics of the DTV transmitter (e.g., power level, antenna height, and radiation patterns), terrain variability and the density of population in areas predicted to receive interference.”
The document further stated that, in addition, this separation approach fails to take into account “technologies and techniques that wireless licensees might employ to mitigate potential interference, such as antenna characteristics and resource block provisioning.”
While broadcasters will be most concerned about interference to their viewers from cell phone base stations and mobile devices, wireless carriers are concerned about interference from high-power broadcast stations to their base stations and mobile devices. Under the plan outlined in the Public Notice, a modified version of the Longley-Rice based OET Bulletin 69 method of calculating coverage and interference would be used to allow wireless carriers to operate on the same or adjacent channels very close to TV station's coverage areas, even creating some interference inside a station's contour if that interference falls below some allowed percentage.
Interference from wireless operators would be calculated using the same statistical parameters as interference from other TV stations. Calculating interference between two TV stations is simple because in all but a few cases, each station has only one transmitter site.
However, how do you calculate interference from wireless sites that could be spread through the over a large area?
The FCC proposes creating hypothetical base stations spaced uniformly on a grid every 10 km, each with antennas at 30 meters above ground in every wireless license area within 500 km of the DTV facility. The predicted field strength from each hypothetical base station is predicted at each grid point (or cell) within the DTV station's noise limited contour and the D/U (desired to undesired) ratio is determined. The D/U ratios from the hypothetical base stations are used to determine interference to broadcast stations.
From what I can see from the Appendix to the Public Notice, the impact of combined interference from multiple hypothetical base stations would not be considered if none of the individual base stations are predicted to cause interference.
The proposed D/U ratios vary depending on the offset between the 6 MHz broadcast channel and the wireless channel being studied when calculating interference in either direction.
As noted, the interference studies would use Longley-Rice and the parameters outlined in OET-69, with a few differences. The FCC proposes to include clutter loss based on a modified version of NLCD land-use categories. The determination of whether or not a TV station has coverage in a cell is made without taking clutter into account, so it isn't clear how this will impact the D/U interference calculation.
Given the wide range of environments in which both wireless devices and broadcast devices exist, I see problems with this approach. A viewer using a TV antenna on a high mast or a viewer in a high building won't have the same clutter loss as a viewer using an antenna at ground level. The same applies to interference to wireless devices. There may be no calculated interference on the ground, but three floors above ground the cell phone may experience interference from broadcast stations. A footnote says clutter losses for wireless devices will be adjusted to account for the “presumed 1.5 meter user equipment height.”
TV receive antenna patterns were used in calculating interference between TV stations, but I question whether it makes sense to use such patterns when calculating interference from wireless base stations, especially when many viewers now use indoor antennas with patterns much wider than those used in creating the original DTV table of allotments, leading to the possibility that wireless interference sources can hit the TV antenna from different angles. Another complication is that wireless base stations likely won't necessarily be located on 10-km grids, but will likely have non-uniform spacing to provide coverage to dense urban areas and to take advantage of high elevation sites when covering sparsely populated areas.
I'll have more comments on the Public Notice when I've had more time to analyze the proposals and see what comes up at the workshop the FCC will hold on Feb. 21 to explain it.
I don't see the need for such a complicated approach to squeeze the last bit of spectrum from broadcasters and thwart the efforts of small white space providers trying to find sufficient spectrum to provide an alternative to high-priced wireless broadband in areas where major wireless carriers may not even be interested in serving.
Based on what I've learned so far, it seems to me that adopting this approach, instead of a nationwide band plan or even a plan based on the distance from broadcast stations' contours, will add another layer of complexity to the forward auction.
Wireless carriers won't want to see a repeat of the problems they've encountered with spectrum close to TV Channel 51. Even though broadcasters have accepted a statistical model for determining coverage that allows for 50 percent of the locations to have less field strength or more interference than predicted, I wonder if wireless carriers will be willing to take the same risk? Spectrum that's well protected from interference will be more valuable than spectrum with usefulness based on a propagation analysis and statistics which may or may not be usable in the real world environment.
To learn more, see the Public Notice DA 14-98
, “Office of Engineering and Technology Seeks to Supplement the Incentive Auction Proceeding Record Regarding Potential Interference Between Broadcast Television and Wireless Services.”