The past year has been a busy one for opening access to spectrum above 100 GHz.  On March 15, 2019 the US Federal Communications Commission (FCC) voted to allow access to spectrum above 100 GHz for the first time.[1]  Since 2003, its upper limit for other than experimental licenses had been 95 GHz.  On January 17, 2020 the UK’s Office of Communications (Ofcom) requested comments on new proposals to allow use of parts of 100-200 GHz.[2] These new proposals expand on previous Ofcom policies that had permitted since 2012 use of the 122-123 GHz and 244-246 GHz bands at powers up to 20 dBm EIRP.  (In 2014 these same limits were adopted by the whole EU.)

FCC authorized 4 unlicensed bands as shown in Table I:

* “For fixed point-to-point transmitters located outdoors, the average power of any emission shall not exceed 82 dBm and shall be reduced by 2 dB for every dB  that the antenna gain is less than 51 dBi “ -- § 15.258 (b)(2)

In addition, FCC created “Spectrum Horizons Experimental Radio Licenses” which in theory allow use of any band between 95 GHz and 3 THz in a geographic area requested and sale of equipment for a period of up to 10 years.  However, such licenses must “incorporate an interference analysis that explains how the proposed experiment would not cause harmful interference to other services.”[3]  It is not clear if this is practically possible in bands with passive satellites in the Earth Exploration Satellite Service (passive) as there is no clear stated criterion for successful sharing and NASA has been quite strenuous with its objections to even small scale academic experiments in such bands.  It appears that FCC has still not received any applications for this type of license so there is no track record yet for approvals.

Ofcom’s new proposals are summarized in Table II:

The 116-122 GHz is almost the same as FCC’s 116-123 GHz band and the 174.8-182 and 185-190  GHz bands are exactly the same on both sides of the Atlantic.  The two lower bands have a lower EIRP in case of outdoor use than in the FCC case but there is also a provision for a simplified licensed use of these bands with powers up to 55 dBm EIRP subject to antenna pattern restrictions to keep power out of the sky where it might interfere with passive satellites. 

FCC allows higher power without the formality of any license if the operator uses a narrow beam antenna.  Thus EIRP may be 82 dBm if the antenna has a gain of at least 51 dBi.  (This provision is modeled after parallel FCC provisions for point-to-point unlicensed use of 2.4 GHz[4], however, it fails to deal with the issue that while there is a simple general relationship between gain and beamwidth for dish and horn antennas, that does not apply to planar antennas.[5]  Thus an option to comply with the narrow beamwidth goal by direct measurement of actual beamwidth – more expensive than measuring gain - would give designers more flexibility to meet passive satellite protection goals.  It would also allow the use of MIMO antennas that have implementation losses in their internal power distribution)

Why would one want to use spectrum above 100 GHz?  The main reason is potential access to large blocks of contiguous spectrum for fiber optic-like transmission speeds and actually better latency. (The index of refraction in glass slows information group velocity by about 30% compared to air or a vacuum.)  But the FCC policies and the new Ofcom proposal have a maximum bandwidth of 7.2 GHz, not much greater than the 5 GHz which has been available in the US since 2003 at 71-76 GHz where technology is simpler and less expensive.  A rationale for use of FCC’s band at 244-246 GHz is even more difficult since it only offers 2 GHz of bandwidth.

FCC addressed the issue of higher powers for point-to-point communications by requiring a high antenna gain in such cases although high antenna gain or narrow beam size does not by itself result in low sidelobes at high elevation angles – the main threat of interference to EESS.  Ofcom takes an approach based more on the physics of the situation by 1) allowing higher powers for indoor use where building loss generally severely limits illumination of the sky, 2) limiting maximum elevation angle for higher power outdoor uses to 20^o, and 3) limiting sidelobe EIRP for higher power outdoor use based on angle away from main beam.  FCC might wish to learn from the slightly more complicated Ofcom proposal that increases EESS protection without a significant burden on unlicensed spectrum users.

Neither FCC nor Ofcom have addressed the issue of spectrum access for terahertz spectroscopy, a technology that allows detailed sensing of material at close distances and has been used in applications ranging from the NASA Space Shuttle program to verify that thermal tiles were properly glued to the vehicle to checking the thickness and structure of plywood as it  is moving at high speeds while being manufactured.  Terahertz spectroscopy technology is being sold by several firms for such uses and there is real policy confusion.  FCC in its original proposals for above 95 GHz stated that it evaluates ISM devices for terahertz spectroscopy above 95 GHz on a “case-by-case” basis[6].  However, a manufacturer of this equipment filed comments at FCC[7] stating it hat it felt that such equipment complied with present FCC rules, that it had self-certified such equipment under applicable provisions, and had sold such equipment for “nearly 20 years” without any complaints.  Since FCC never addressed this issue in its later decision adopting rules and made no explicit provision for terahertz spectroscopy there is major regulatory uncertainty at present for manufacturers or potential manufacturers of this technology.

In summary, the FCC decision and the Ofcom proposal for spectrum use are similar in many ways.  Both are very influenced by the need to protect EESS. WRC-19 decided that “to the extent practicable, the burden of sharing among active and passive services should be equally distributed among the services to which allocations have been made”.[8] More effective dialogue is needed between proponents of active and passive uses above 100 GHz to review practicable win/win alternatives for sharing the large amounts of spectrum which have passive allocations that permit interference free sharing in bands that have actual passive use.