E-TOLL STANDARDS: Activists vs passivists

from Toll Roads newsletter

E-TOLL STANDARDS:

Activists vs. passivists

by Peter Samuel

Electronic toll developments are driving a vigorous struggle worldwide to shape new standards for vehicle-to-roadside communications (VRC) often known as DSRC (dedicated short-range communications). In North America free market competition has resulted in essentially three distinct electronic tolling systems:

• in the mid-Atlantic US (NY, NJ, PA, DE, MD, VA), the mid-west (IL, OH) and Ontario Canada an active transponder system based on open domain Hughes Corp developed technology as represented by equipment built and marketed by Mark IV, and by Hughes itself for trucks and for the new highway 407 in Canada

• in the south (TX, OK, FL, GA, LA) and the Canadian maritimes a passive backscatter technology developed and marketed by Amtech, lately with contributions from Motorola

• in the west (CA, CO) and Massachusetts passive backscatter tags built by Texas Instruments to a California state standard (called Title 21) and marketed by MFS

Europe: The European scene is more difficult to summarize but virtually all the systems use passive backscatter technologies and mostly at higher frequencies (2.45GHz and 5.8GHz) than the US systems (all 0.9GHz). European regulators are liable to require licensing for an active transponder on the grounds that each device is a radio transmitter so development of active RF systems has been inhibited there and all the toll equipment is passive backscatter — which reflects back to an over-the-road reader a 'modulated' or data-added radio signal transmitted from an overhead transmitter. Though still 'reflective' or passive backscatter, these systems mostly use a battery-powered amplifier to work at sufficient range for high speed operation. There are a handful of different proprietary European systems, and recently efforts have been going onto make these interoperable — in parallel with work by the European standards body CEN. A European draft standard for DSRC known as CEN-278 has majority support in the European Community but not quite the required 71% of votes to become mandatory yet. A group of manufacturers (Saab, Cegelec and Bosch) say they will build to CEN-278 but unless it gets legal force through a 71%+ vote toll authority customers will buy what they want. There is strong resistance to CEN-278 from the Italians who have a large installed system of transponders that is in some ways more capable than CEN-278, and from others in Spain and Portugal with very different systems. At the ITS World Congress in Oct an Italian e-toll woman casually disparaged the standard-setters of CEN as 'amateurs' which provoked red-faced rage among some Germans at the session.

Japan: Despite its electronics prowess and developed toll road network of 6,700km, Japan does not have a single deployed system for electronic (e-) tolling, but they have just completed a very thorough testing and evaluation of a variety of active and passive e-toll systems. Moreover Japan leads the world in other systems that make use of VRC such as traveller information services and in-vehicle navigation. The Japanese are going with a Hitachi/Oki-designed active transponder system that is in essence a more capable (higher data rate) version of the so-called 'Slotted Aloha TDMA' Hughes/Mark IV kind of technology, but at 5.8GHz compared to the US 0.9GHz. The lower US frequency being used for e-toll equipment is fully occupied in Japan by cellular telephones and the higher 5.8GHz frequency that is favored in Japan and Europe for e-tolling is heavily used in the US by the Pentagon for radars and satellite communications, and is only liable to be available after several years of permitting efforts. 5.8GHz is presently an expensive frequency at which to build equipment (gallium vs silicon), and requires more battery power which generates reluctance on the part of US users to move to it. However there is a widespread view in the US e-toll business that eventually parts of DSRC will 'migrate' to the higher frequency because the unlicensed 0.9GHz band will become more heavily congested and interference an increasing problem. Also non-American standardization in the 5.8Mhz band will be driving costs down and present opportunities that US exporters can best exploit by having a home market. The chip design or ASIC need not change for a frequency upgrade.

Double-edged sword: Standards are a double-edged sword. In their favor they can save duplication and help make systems interoperable, generate economies of scale, help drive down prices, create confidence among users that service and parts will continue to be available, and generally help develop markets. Against standards is that they may freeze technology by limiting product innovation, may reduce choice, advantage some manufacturers and users unfairly over others, and diminish competition. A good standards-setting process attempts to maximize the advantages while minimizing the downside or negatives, which suggests the need to keep the business open for innovation. The European tradition tends to be one of setting up expert committees which develop standards that become mandatory. The US tradition is more for so-called stakeholders to try and develop standards which then become conditional in government procurements. The trouble with having present 'stakeholders' decide a standard is that it leaves out of the room yet-to-be-born companies with new solutions and future products. Whichever way is taken standards development is a messy and imperfect process with a lot of uncertain tradeoffs, and may do net good or net harm.

Critics say futile: The planners approach taken in the National Architecture for ITS is to try and anticipate which technologies will sell and to find a place for each within an organized matrix. The idea of reserving spectrum and setting a DSRC standard is one part of that effort. "It is completely nuts," says Ira Brodsky a Chicago consultant on telecommunications, principal of Datacomm Research Company and author of several books. Brodsky says a DSRC standard is "utterly futile" because it presumes that ITS needs a separate infrastructure. "ITS will never go anywhere if it has to build a separate telecom infrastructure. It will be too expensive." Brodsky and other critics say that there is already a huge infrastructure in the US for mobile data communications to handle ITS needs and cites the so-called PCS and other digital data services that exist or are being developed in the marketplace: "We are getting a glut of low-cost communciations capability." He says building equipment specially for ITS purposes at 5.8Hz is "quite insane" because so much more economical equipment is available at many other frequencies including 800MHz and 1.9GHz. Bruce Abernethy at Kimley-Horne in Dallas has characterized the standards setting of the ITS National Architecture as a misguided Pentagon style effort to rigidly specify equipment in advance, though he says it has recently loosened up somewhat in its approach. Brodsky says "standards can only evolve out of commercially scuccessful products" and that the ITS National Architecture is a presumptuous socialist style effort to predict and plan both technology and consumer demand in a top-down fashion. "It just won't work. It can't work. It never has. It never will."

Passivists vs activists: Until the latter half of 1996 the US standard-setting exercise as organized by a commitee organized by the trade group ITS America and Federal Highway Administration centered heavily on the active transponder system of Hughes which has only been applied in truck electronic clearance tests in the mid-west (H-401/I-75) and west (I-5/I-10). This draft standard known as ASTM version 6 (v6) based on slotted Aloha (yes, developed in Hawaii) TDMA was decisively rejected in meetings late summer and fall. Essentially this clash saw the e-toll industry divided between the proponents of active transponder technology ('activists') and those who favor the passive backscatter tag systems ('passivists'). The activists included Hughes and other GM subsidiaries, Mark IV, 3M and system integrators Lockheed Martin and Syntonic. On the other side the passivists were led by Amtech with strong support from MFS the system integrator and sales representative for the Texas Instruments transponders, Combitech and AT/Comm (which has a proprietary active system.) Following the defeat of the ASTM /v6 there were a rapid series of developments.

CEN pledge: Amtech, Combitech, MFS, TI and other 'passivists' signed a joint letter to ITSA suggesting that the US adopt the European CEN-278 standard, as amended to allow use of the 915MHz frequency. Amtech in an effort to force the issue announced that it would build future systems to the CEN standard. Combitech based in Sweden says it is already building equipment close to the CEN standard and together with Bosch (G) and Cegelec-CGA (F) demonstrated an interoperable system at the Orlando ITS World Congress to underline the commitment to CEN.

Federal fiasco loomed: Senior officials of the Federal Highway Administration and ITS America saw the decisive defeat of ASTM/v6 as a serious breakdown in the standard setting process. This represented an immediate crisis and a longterm setback — immediate because the federal government wants to proceed with CVISN, a joint program with several states to develop electronic clearance and one-stop-shopping for permits for heavy trucks and needs to prescribe a DSRC standard, and longterm because a DSRC standard is a crucial component of the so-called National Architecture for ITS, a grand scheme (some say grandiose) for trying to plan the electronic inter-relationships of transportation's future high technology.

Will seven be lucky?: The FHWA/JPO via consultants and a number of key industry figures have been moving to resuscitate the US standard setting process and the outlines of a more promising ASTM draft (#7) are emerging. Simple imposition of the Hughes/Mark IV "slotted Aloha TDMA" (Time Division Multiple Access) active system now looks too contentious and extreme and simply won't fly politically. At the same time the adoption of CEN standard as urged by the 'passivists' creates problems for the US. For a start the largest tolling systems in the country (NY/NJ/PA/IL) are active TDMA by Mark IV, and so are the truck systems of Hughes.

Unarchitectural: CEN doesn't fit the US national architecture for ITS and staff working the issue say "No way" to CEN at 915MHz. All the European testing of CEN is at 5.8GHz and although it does not formally exclude active transponder systems, it is designed around passive backscatter systems. For some non-toll related purposes active systems have clear advantages over passive. Active is well suited to parallel processing and has attractions for open road tolling. It also has longer range (100m vs 50m) and higher data transfer rates that are helpful if not essential for complicated tolling, and use of smart cards. Moreover the technical people say the US national architecture for ITS requires the active TDMA for non-toll DSRC functions such as automated highway operation, border crossing data exchange, in-vehicle signage, intersection collision avoidance, mainline screening, emergency vehicle pre-emption, and electronic license plate systems for trucks. At the same time passive has its own strengths: lower cost, frequency adaptability, long life, simplicity, ruggedness. For many purposes passive works well as is demonstrated by the many large operating e-toll systems in southern California, the US south, France, Italy and elsewhere. The Japanese embrace of active TDMA systems as their e-toll standard (Hitachi/Oki's design)strengthens the case against the US simply following the Europeans to a short range passive standard like CEN-278.

Accomodation via hybridization: An effort at accomodation of 'activists' and 'passivists' is the only way to go for both technical and political reasons. As we go to press there is an effort by the federally funded consultants to devise a US standard (ASTM/v7) that would somehow allow for interoperability of active and passive DSRC systems and their coexistence side by side — much like FM and AM radio. The DSRC committee is now working to a February 6 deadline to report "significant progress" by the federal government or it has said it will mandate ASTM/v6 for federally funded projects. The general approach of v7 is to develop protocols to allow active and passive systems to work on the same facility, the active probably being used for the wider area (100m) and more complex data exchanges and the backscatter being available for more rudimentary vehicle toll systems. At a meeting in New Orleans as we go to press the backscatter interests voted down efforts by Hughes to have v7 incorporate its existing system as they felt this would give it an unfair advantage. It will have to develop new v7 systems like the other manufacturers. The general approach will be to adopt the Hitachi two-way (full duplex) operations for an active system, a kind of Hughes Upgrade, to work alongside a new passive backscatter standard. One approach is that a certain proportion of milli-second time slots in the time-divided system could see the active transponders suppressed to allow the passive system its air time. The actual proportion of active versus passive time could be a matter of programming, depending on the proportions of each type of system in a particular area at a particular time.

Dallas Cowboys: Amtech is by far the largest 'passive' manufacturer and its people have been outspokenly critical of active technologies and the lead advocates of the US moving quickly to the CEN standard. An alternative might be movement toward California Title 21. Amtech has made its more sophisticated read/write 'Intellitag' compatible with the Title 21 standard and on the Georgia 400 pike it says both the older read-only tags (not Title 21 compatible) and the Intellitags are interoperable. Whether a hybrid standard of the kind envisaged for v7 makes economic and technical sense remains to be seen. It envisages new roadside equipment, which has to be specified, analysed, prototyped and tested. The more complicated and compromising a standard becomes the more expensive it is likely to be and the more difficult it will be to sell to tollsters. The New Orleans meeting laid down guidelines that ASTM/v7 "does not require backwards compatibility with existing standards or forwards compatibility with other draft standards" (read CEN-278, Japan's Hitachi/Oki standard), specifications that seem certain to reduce its saleability.

Like trash collection: Electronic toll collection bears some comparison with trash collection in New York City. It is an important public service to the citizenry but it is not always the cleanest of work and it has attracted some colorful characters who can fight rough. Few feel secure. Overall, profits so far are small. Some e-toll shops are for sale, others could simply close, or be merged with former rivals. Law suits run rampant. This is hardly an atmosphere conducive to disinterested and studious consideration of the longterm public interest. The 'standards' play sometimes looks quite Darwinian. However the genius of capitalism is that suppliers have to both compete and also to collaborate to survive and that in the end they get their money from satisfied customers or not at all, so the marketplace may work for the public even though the standards committees fail. (From Toll Roads newsletter December issue, sample copies available on request 301 631 1148 or tollroads@aol.com)