Why 220MHz?

The 220MHz band may hold great promise for business and industrial users. It certainly has a colorful past.

While existing almost entirely in the hands of amateur radio operators for the past several years (shared with military users for a while), the 220MHz–225MHz segment has had interesting—though limited—use. Amateurs have used the band for a variety of purposes, even moonbounce communications, but most have used it for FM repeaters.

Repeater users enjoy the band’s good coverage, similar to VHF highband, yet with better predictability and less co-channel interference, similar to UHF. Amateur repeaters use 25kHz channel spacing and a unique 1.6MHz offset between transmit and receive frequencies.

Others have been using nearby spectrum. The FCC designated 40 pairs of channels at 217MHz–220MHz for the Automated Maritime Communications System, also known as the Inland Waterways System. Licensees in this band are permitted to offer radio coverage for inland waterways and a substantial portion of a coastline. Radios for this band operate in FM mode on 12.5kHz channel spacing. Four major licensees offer radio services ranging from voice to waterborne GPS-based vehicle location service.

Mobex Communications, Jeffersonville, IN, operates AMTS networks under the Regionet Wireless name in the West Coast, East Coast and Great Lakes regions. The networks carry voice and data, and when combined with a GPS engine, provide automatic vehicle location service. To see a map that shows the possible future coverage area of the Regionet communications system, click here.

The U.S. Navy also uses spectrum at 216MHz–217MHz. The U.S. Naval Space Command uses its space surveillance (SPASUR) system to track earth-orbiting satellites and to monitor over-the-horizon threats from sea and air forces. SPASUR operates a surveillance network of nine field stations across the southern United States. These surveillance stations produce a “fence” of electromagnetic energy that can detect objects to an effective range of 15,000 nautical miles. The fence transmits a narrow beam of RF directly skywards from a two-mile-long, ground-mounted antenna array.

220MHz goes commercial

In 1988, thanks in part to the urging of United Parcel Service, the FCC reallocated the 220MHz–222MHz part of the band for commercial use. Relatively narrow at only 2MHz wide, the band was devoted to ultra-narrowband operations, with channels set for 5kHz spacing.

The reallocation proceeding took so long that UPS moved to other radio communications alternatives, and the new band was chopped up and offered via lottery to other commercial entities (and speculators) in five-channel blocks. Some frequencies were set aside for public safety use. Later, under the direction of then-FCC Chairman Reed Hundt, the FCC decided to auction channels in this band. The first 220MHz auction (No. 18) was held in 1998.

By setting such narrow channel allocations, the FCC hoped to spur manufacturers into developing advanced modulation technologies, including amplitude-companded single-sideband (ACSSB) variants. Furthermore, regulations were changed so that, unlike the end-user loading requirements connected with other SMR licenses at the time, a 220MHz SMR system operator needed only to construct its base stations. Any specific subscriber-use notification to the FCC was waived.

Many hoped that equipment developed for this band would pave the way for the FCC’s refarming initiative. In that initiative, the commission would grant type acceptance in 2005 and thereafter only for new products capable of operating on 6.25kHz channels.

Complications

The way the FCC configured the five-channel blocks changed over time, perhaps complicating operations in the 220MHz–222MHz band. Earlier assignments separated each channel in the five-channel block by 150kHz. This separation was enough to allow the use of cavity-type combiners. But that relatively narrow separation made the combiners more expensive and difficult to design compared to wider channel separations in other bands.

Later assignments placed each channel in the five-channel blocks adjacent to one another, making the use of cavity-type combiners impossible. Instead, expensive, lossy hybrid combiners had to be used. Typical five-channel repeater equipment costs increased to about $70,000 per site.

With the ebb and flow of the telecommunications marketplace, only two companies now offer type-accepted equipment for ultra-narrowband operations: SEA and Microwave Data Systems. SEA offers subscriber-based products, while MDS concentrates on SCADA applications. Other companies have been designing products for this band, but few have entered the marketplace as of yet.

Several original licensees put commercial sites on the air. Some were speculators. Others operated airtime businesses and had minimal success with the equipment available at the time. As a result, few systems built for the originally issued licenses are actually in use today. Many are operational, but without subscribers. In short, the band has become underused.

Unfortunately, some operators overestimated how well a 220MHz ultra-narrowband system would operate. Despite advances in digital signal processing that dramatically helped ACSSB systems operate more reliably, the size of their coverage areas was sometimes disappointing.

Booth & Associates developed a propagation prediction model using Softwright’s Terrain Analysis Package to understand how well an ultra-narrowband system with 5kHz or 6.25kHz channel spacing would operate compared to an FM system on a 12.5kHz channel. This prediction tool has been used to successfully coordinate multisite statewide build-out plans for a licensee with a nationwide 220MHz license. The model works equally well for mountainous or flat terrain.

One licensee expected coverage at least 30% greater area than what the model predicted. With ultra-narrowband technology, the coverage is always less. Several reasons account for the difference.

Relative output power

— Output power on an ACSSB system is considerably less than that for FM systems. Typical ACSSB output power ranges from 60W to 100W PEP (peak envelope power). To convert PEP power to the power equivalent for CW (continuous wave, the appropriate measurement for FM), divide the PEP output power by a factor of 2.47. ACSSB PEP output of 60W to 100W is the equivalent of 25W to 40W for FM.

Moreover, with those contiguous five-channel blocks, the ones with 5kHz channels adjacent to one another, hybrid combiners further reduce the output. In a two-antenna, five-channel configuration, hybrid combiners exhibit as much as 9dB loss. It has not been uncommon to measure post-combiner output power at 7W PEP, equivalent to 2.8W FM.

Even worse, 220MHz licensees have been subjected to the FCC’s height above average terrain rules that sometimes limit output power to as little as 5W PEP, depending on the base antenna elevation.

Terrain factor

— Terrain seems to be a significant factor when working with 5kHz or 6.25kHz channels. As a result, only the Bullington calculation method seems to work accurately for propagation modeling.

The Bullington method computes free-space field strength and adjusts for obstruction attenuation for knife-edge diffraction losses. (See “Radio Propagation for Vehicular Communications” by Kenneth Bullington, IEEE Transactions on Vehicular Technology, November 1977.) Because terrain is the largest calculation variable, high-resolution terrain data must be used to obtain high prediction accuracy. The U.S. Geological Survey makes data available for most of the country at 30-meter resolution. Similar data are available from the Canadian government.

The widely used Longley-Rice method seems to give the most accurate predictions for coverage of FM systems. Great success has been found using a calculation factor of 90% availability for 90% of subscribers, as well as variables including soil conditions, ground conductivity and dielectric constant. But if ultra-narrowband systems are modeled using the Longley-Rice method, predicted coverage areas tend to be exaggerated by 30% or more.

Meanwhile, using Longley-Rice to calculate narrowband FM coverage appears to give accurate results. Empirical experimentation has led to the conclusion that wider bandwidths (+/- 2.5kHz deviation) and the basic characteristics of FM make radio signals at this frequency propagate further, and they are less susceptible to knife-edge diffraction losses.

The accompanying map compares the coverage of a five-channel ultra-narrowband system with a two-channel narrowband FM system. The hilltop radio site has an HAAT of 816 feet.

FM could operate in an ultra-narrowband channel assignment. With at least five contiguous ultra-narrowband channels at a particular site, two narrowband FM signals could fit into the 25kHz-wide channel block. Although converting ultra-narrowband channels in this way reduces the number of usable channels by 60%, using FM can restore the range.

When this article was written, two manufacturers had submitted equipment to the FCC for type acceptance of such use. In addition, equipment already used for AMTS is designed to operate between 216MHz and 225MHz. The only major difference is the trunking format used by AMTS.

How effectively would a two-channel trunking system work? With MPT-1327, one channel is used for control data and the other for voice. The control channel can also be used for vehicle location data services, and the voice channel can be used for longer data messages via the radio’s MAP27 serial data interface. This configuration has been used successfully in systems around the globe for many years.

Through several user studies, Booth & Associates found that many business and industrial users desire mobile data services. Without usable spectrum, those wants generally go unfulfilled. Rudimentary models for automatic vehicle location and status reporting showed that a data system could reduce voice communications from 50% to 90%. It was also calculated that simple data services could increase loading potential by as much as 4:1 over voice services. Therefore, both the market and technical conditions make a two-channel data trunking system practical in many situations.

But any two-channel system would have a practical subscriber limit. Coupled with improved coverage range, the system may not be subscriber-efficient. Because the 800MHz range is limited by comparison, multiple 800MHz sites can be built with each capable of serving a large number of subscribers. In this way, high densities of subscribers in urban areas can be profitably served.

In a two-channel 220MHz system, only so many subscribers can be served in a broader coverage area. That makes it important to consider technologies and services that permit higher loading capabilities—such as mobile data products.

A digitally multiplexed system, perhaps similar to Motorola’s IDEN, may offer another way to offer spectrum- and subscriber-efficient services. Advanced mobile data services that could be fit into a 25kHz window could be offered to even more subscribers.

A digitally multiplexed system should have propagation similar to narrowband FM. Metropolitan areas and portions of interstate highways could fall within a digitally multiplexed site’s coverage.

Services could include AVL, status messaging, paging, email, telematics (a kind of mobile SCADA) and automatic crash notification. It’s possible, however, that more spectrum would be needed to make a digitally multiplexed system with these features practical.

Because the 220MHz band has few active systems in operation, an operator with a reasonable business plan that would use a digitally multiplexed system might find 220MHz spectrum available for purchase or lease at an economical price. Existing license holders, whether speculators or unsuccessful system operators, could be elated to find a buyer or lessee.

Aggregating licenses for such purposes would eventually require the approval of the FCC’s Office of Engineering and Technology, as would a petition to use “dry” geographical locations in the AMTS band and the underused IVDS band.

In December 2001, the FCC issued a Report and Order and Memorandum Opinion and Order (FCC 01-382) that addressed the “dry” or “white area” where AMTS is not used. Although the FCC said that the spectrum soon would be auctioned for land mobile radio use, the agency did not say what channel allocations and bandwidths would be specified prior to the auction. It seems likely that FM systems on 12.5kHz-wide channels would be allowed because such an emission mask would allow current manufacturers of equipment in that band to apply for a permissive change to their products’ type-acceptance documentation. That’s easier and less expensive than a fresh application for type acceptance.

Lastly, the easiest target of spectrum opportunity may be the 222MHz–225MHz amateur radio allocation. Although radio amateurs have been appropriately vocal in defense of their spectrum, their pleas to leave their 222MHz–225MHz band alone may go unheeded if the FCC receives a petition that proposes a compelling commercial use for the spectrum.

Despite the radio amateurs’ contention that increasing numbers of them were using the 220MHz–222MHz band, the FCC still reallocated it. In the controversial decision, the FCC knowingly disregarded extensive use of the band. The commission referred to a directory for amateur radio repeaters in the 222MHz–225MHz band and ignored weak-signal and inter-city relay operations by amateurs in the lower band segment.

The FCC changed its amateur radio rules to allow entry-level licensees who pass a written exam to use 222MHz–225MHz without taking a Morse code test. But the number of amateurs using that band hasn’t risen as few choices of radio products are available for the band.

Overall, the 220MHz band dangles tantalizing possibilities. Technically, the band offers desirable propagation. It has long been the darling in the eye of the FCC’s technical planners. However, without a complement of suitable equipment manufacturers and without a track record of business success by licensees, the spectrum lies fallow.

With the clamor caused by the Nextel white paper that proposes to displace business and industrial users from their present channels at 800MHz, more attention is being paid to their future spectrum needs. Perhaps a re-born, re-planned 220MHz band is the answer.

The author Todd Ellis has more than 15 years of experience in the computing and communications industries. As the operations manager for the Telecommunications Division at Booth & Associates, Raleigh, NC, he oversees a staff that addresses various mobile data and computing system-level projects. Ellis has conducted extensive business in the form of consulting, contracting, system design and project management both domestically and internationally. He has written or contributed to several magazine articles published in communications-related magazines, including Mobile Radio Technology and Cellular and Mobile International magazines.