DSRC STANDARDS USDOT slams other standards

DSRC STANDARDS

USDOT slams other

standards

The Joint Program Office (JPO) of the US Dept of Transp (USDOT) has published a scathing criticism of other standards making efforts for vehicle to roadside communications known as DSRC (Dedicated Short Range Communications), the major application of which is currently electronic tolling (ET). The JPO posting is on its website (www.its.dot.gov/dscr.htm) and according to Bill Jones a senior official there, it is an attempt to dispel misinformation about the strengths of some other standards.

The posting claims that the European CEN standard leaves open key operational details which have been decided collaboratively by three European manufacturers, but USDOT says the 3-company interpretation is itself incompatible with the literal wording of CEN, so problems of incompatibility remain. Moreover the Japanese and European approaches are extremely high cost compared to the North American solution, greatly reducing the potential for selling systems to motorists and getting them widely deployed.

The sharpest words in the USDOT posting are about the shortcomings of the European CEN standard.

CEN a ‘cafeteria menu’

“Too much of the information surrounding the CEN standard is based on marketing pitches rather than the cold facts... occasionally, it is said that the CEN standard supports the option of operating in an active mode. There are no parameter definitions in the standard that would allow this, simply some very vague reference to it being an option... there is a possibility of significant problems if this were attempted.

“The standard is very much a ‘cafeteria menu’ standard... there are 6 different bit rates defined. There is no definition of how to arbitrate between any two devices which have implemented any of the options not currently in use. Such devices would be incompatible with each other.

“The (CEN) standard lists a number of applications, including commercial vehicle operations (CVO), for which it is supposed to be applicable. The only consideration during its development, however, was toll collection. It remains to be proven that it will fully support many of the applications, such as CVO. The products available or under development today will only support toll collection. There are several reasons to suspect the CEN standard’s ability to support these other applications, among them the operating range...

“To support other applications with the same transponder would require a relatively powerful microprocessor to perform the ASN.1 encoding/decoding required. To keep the transponder costs down, the existing designs use only ASICs, no general purpose microprocessors, and thus can handle only the one application for which they are designed and with several technical limitations that are not evident from the standard itself. In other words, these existing devices may be considered non-compliant as they do not support the capabilities defined in the standard...

“The (European) VASCO program has drifted away from the standard as a reference and products deemed compatible may not actually be compliant with the (CEN) standard. This started with the need to clarify the documents and the options described in them. The (European grandly misnamed – TRnl) Global Specification for Short Range Communication (GSS) was created to describe how to construct compatible systems based on the CEN standard. The GSS is not a part of the standard, it is a copyrighted product of the companies pushing the standard, but by providing clarification, it became the ‘bible.’ When the standards validation testing was done, the GSS document was used as the reference, not the standards themselves. There are some deviations between the details of the standard and the GSS document, and by using the GSS as the base, the test did not fully test the standard. Then, when performing tests for compatibility between products, the initial prototypes of these three companies were used as the ‘gold standard’ for comparison. The problem is that the engineers developing these prototypes collaborated with each other to resolve incompatibilities without using the (CEN) standard as the basis for arbitrating differences or problems that existed. The resulting prototypes thus were not tested to assure that they did, in fact, represent an accurate manifestation of the standard.

“The result is that if a company designed a new product with total reliance upon only the standards documents, there is a very good chance that it will be deemed non-compliant. Some recent test results (show) this to be the case.”

$250 vs $40

USDOT says that the Japanese standard by contrast has been thoroughly tested and prototyped in a wide range of applications, but its problem is its cost — $250 vs $35 to $50 for North American transponders. It comments: “They are not worried about this price within their home market, but recognize that their system will not be marketable outside of Japan.”

The European approach also will be very high cost for systems with a wide range of capabilities, says USDOT.

Only the North American ASTMv7 standard offers the possibility for supporting a wide range of applications at modest cost, the posting says.

“There are many features of the (N.Am) standard that are designed to facilitate the migration from existing systems to the new standard as it has been found that many of the existing systems and other standards could cause interference with each other during the transition period when multiple system types must coexist. (ASTMv7 also provides) that transponders of one type, such as active, will not interfere with readers of the other types, such as backscatter. Either transponders or readers can be dual mode to provide 100% compatibility whenever this is determined to be worth the additional cost involved. None of the other standards address these issues.

“The Application Layer, defined in a separate IEEE standard (for v7), has a unique characteristic that allows data to be written to and read from the transponder as a simple bit string. This makes the transponder a mobile mailbox from which roadside systems can read and write messages to each other. The result is a very cheap transponder while offering a wide range of capability and high performance. For instance, data stored on the transponder can be fully encrypted for security, and only authorized roadside systems will have the ability to decrypt the data. It also enables any existing and all possible future applications to be supported by this simple transponder. The limitations of the CEN standard, which requires an ASN.1 processor to support multiple applications, is solved by this approach.”

JPO reports that ASTMv7 is “nearing completion.” The physical layer has been balloted, the data link layer is balloting now, and the application layer (an IEEE exercise) will be balloted within weeks. The final standard will be published “within a few months” and there will then be a test validation program of protoypes.

The USDOT posting also gives short shrift to the California standard (Title 21), also recently recommended by Parsons Brinckerhoff for Brazil (TRnl#28 Jun 98 p7).

“(Title 21) is inadequate to assure compatibility between different vendors’ products. To date, each of the manufacturers judged compliant have spent time working with others to resolve compatibility problems uncovered during development... (T)he standard was developed with (only) toll collection in mind and may not be adequate for other applications.”

Background

JPO is the office within USDOT in charge of congressionally mandated efforts to develop standards for North America for DSRC which for the moment concern heavy truck clearance (CVO) at weigh stations and border crossings, though all but about 50k of about 3.5m transponders in use in North America are principally for electronic tolling. There are minor uses of the transponders for parking access, and growing subsidiary use for monitoring urban traffic flow speeds for incident detection. There are plans to use transponders for origin-destination surveys, and suggestions they may be of some help in providing real time traveler information, route guidance and so forth.

Canadian, US and Japanese representatives prevented a German effort at an ISO (Internat Standards Org) panel a year ago to take the European CEN-278 standard international (see TRnl#21 Nov 97 p1). Leading North Americans and Japanese officials believe the European CEN standard is fatally flawed and say that the European officials and company representatives have made false claims about it. Lee Armstrong a Boston consultant leading the North American standards-making exercise has made some of these points personally (see TRnl#25 Mar 98 p1) but they probably carry some more weight coming as a carefully prepared and issued statement of the official US agency.

JPO’s Bill Jones told us that his boss Christine Johnson and others at USDOT have constantly heard misinformation and confusion over the state of standards and the relative strengths of the different standards. The posting is designed to lay out the situation as best the US government can establish it, he says. (Contact Bill Jones USDOT 202 366 2128 www.its.dot.gov)