|
The motivation for the research discussed in this article is to enable better driving through the use of modern communications technology. Therefore, one- and two-way communication systems have been investigated for their specific advantages in the improvement of driver information, safety, and traffic flow. The possibilities of various cellular communications systems are presented.
Traffic Management Improvement by Integrating Modern Communication Systems
Autonomous In-Car Navigation Systems
RDS/TMC
GSM
Overview
GSM Data Services
Short Message Service
General Packet Radio Service
MOBITEX
TETRA
The SOCRATES Concept and Test Site
Conclusion
Acknowledgments
References
Looking for Related References?
Biography
Reaching a destination by car should be done without the burden of traffic jams; also, the
pollution produced while driving should be as low as possible. Without any further help
drivers choose their route based on very individual knowledge which can be improved if
the actual traffic conditions are given to them by voice using FM radio, supported by the
ARI-(Automotive Road Information) system. Nowadays, an advantage is provided by
startup of the RDS-TMC (radio data system-traffic message channel) [1] transmission
because the data rate is much higher and the digital coding allows in-car storage and
processing. Nevertheless, transmission is still based on one-way communication.
Traffic management systems can help free the driver from deciding how to find the best
route because such systems gather as much information from the street network as possible
and, based on that, calculate an optimized traffic flow. The aims are safer driving, lower
pollution, and shorter travel time. The messages given to drivers are results of that
optimization.
Such traffic management systems will be improved by integrating dynamic traffic data;
two-way communication is urgently required [2]. Older systems (e.g., ALI [3] or LISB [4,
5]) employed their own transmission equipment and protocols, but that is inefficient and
cannot be done for large-scale implementations. The aim for future road traffic systems
must be use of an existing communication infrastructure with open architecture, so the cost
can be reduced because traffic message transmission is now one of several offered
services.
Two-way communication enables cars to work as sensors. They measure their own travel
times and locations and send the information on the uplink to the traffic control center. This
allows better traffic modeling, prediction, and route calculation. On the downlink the
vehicle equipment receives new traffic information from the traffic management center.
Most advantageous communication systems are based on cellular radio with good
frequency reuse, low power consumption, interference reduction, and so on. The main
features of the following cellular mobile communications networks which fulfill the
mentioned requirements should be presented in this article from a technical point of view:
There might be the necessity to adapt some attributes of such a system to the special
demands of road traffic characteristics. The possibilities of the above-mentioned
communication systems will be shown in the following chapters, and an overview is also
given on navigation systems and traffic management. Some other cellular communication
systems have been developed, for example, the code division multiple access (CDMA) IS-
95 system by Qualcomm [6], which offer overall advantages for supporting traffic
management; but no investigation of their peculiarities will be made here.
Autonomous in-car navigation systems guide the driver of an equipped vehicle by acoustic
and optical instructions to the desired destination. Such systems consist of a navigation
computer, wheel sensors, compass, global positioning system (GPS) receiver, and CD
player ( Fig. 1). The information (road map with street attributes) required to calculate the
route is stored on a CD-ROM.
At the beginning of a journey the driver inputs his destination. Based on the knowledge of
the current location provided by the GPS receiver and the digital map on CD-ROM, the
navigation computer calculates the best route. According to the instructions which were
presented by the system, the driver starts to drive. During the trip the computer keeps track
by correlating the information gathered by the wheel sensors, the compass, and the GPS
receiver with the calculated route based on the digital map. Early enough it gives acoustic
instructions, for example, to turn left or right or to use another lane. Additionally, a display
( Fig. 2) provides assistance.
Should the driver miss a turn or alter the route, the navigation computer would recognize
this and immediately compute a new route to be presented after a few meters of driving.
Autonomous in-car navigation systems have reached marketability, able to perform static
route guidance for drivers. These systems can be improved by giving them knowledge of
the current traffic situation to enable dynamic route guidance. This information can be
transmitted via different communication systems, to be discussed in the following sections.
In addition to an FM-radio broadcast signal, a subcarrier modulated by digital data is
transmitted. This is called a radio data system (RDS). The raw data rate is ~1200 b/s.
Although RDS has been developed to present additional coded information concerning the
program (name of the radio station, alternative frequencies, etc.) further possibilities have
been identified. One example is the transmission of traffic information to in-car navigation
systems. This has been defined as the traffic message channel (RDS-TMC) [7]. Due to the
capacity of RDS, it is possible to transmit approximately one traffic message per second. A
traffic message gives information about one traffic incident, such as: Traffic jam on
motorway A33, direction north between Paderborn South and Paderborn Centre, travel
time increases to 3.5 minutes. This message will be transmitted in coded form using not
more than 32 bits.
RDS/TMC is an efficient method to transmit traffic information to vehicles and will replace
spoken traffic messages in the future. Because traffic data are transmitted in broadcast
mode, a good information level can be achieved. Nevertheless, concerning the route
guidance application there are some disadvantages of RDS/TMC:
Consequently, the need for a two-way communication system arises. Furthermore, it is
desirable to be able to broadcast only relevant traffic messages to specific areas.
GSM [8] is the pan-European mobile cellular radio network which provides two-way
communication links between mobile stations and base stations. Various voice and data
services are implemented for communications within the mobile network or between mobile
stations and the public telephone network.
GSM seems to overcome the disadvantages outlined above for RDS-TMC. The cells are
significantly smaller than the coverage area of an FM radio station.
Besides speech, GSM offers the very important data communication service. For road
traffic information only this pure data transmission is of interest.
GSM allows data rates from 9.6 kb/s (TCH/F9.6 traffic channel/full rate) to 2.4
kb/s(TCH/H2.4 traffic channel/half rate). Normal connection mode is circuit-switched with
dialup. This kind of transmission requires that a circuit be set up between a subscriber at
one end to a subscriber at the other end by dialing a phone number. The end-to-end circuit
consists of a number of links such as cables, time slots, or frequency bands [9].
In the original concept GSM offers a point-to-point connection especially for handling large
data volumes (e.g. file tranfer ). However, in an RTI (road traffic informatics) application
small data volumes of 10 to 200 bytes usually occur.
So it can be stated that for an individual service, traffic messages can be sent to the in-car
equipment using one of the defined data channels of GSM to update dynamic navigation
systems periodically or on demand; but in this mode every car equipment must be dialed
individually. This works fine if only a few drivers use the service and the cost can be
limited. Furthermore, it is useful to send messages to groups of cars in specific geographic
areas.
In the future telecommunication equipment will become a normal part of modern vehicles as
the demand for voice communication increases. For better exploitation of such equipment a
value-added data service like the transfer of messages concerning the actual status of the
street network should be established. More general traffic management purposes need
additional solutions which will be shown in the following section.
Within the specifications of GSM an additional transfer mode has been defined. It is a
paging service called SMS that supports transmission in point-to-point and broadcast
mode. There are a mobile-originated point-to-point (SMS-MO/PP) and mobile-terminated
point-to-point (SMS-MT/PP) service and cell broadcast (SMS-CB). SMS is of high interest
because data can be transmitted to and from vehicles. The main advantage is that data can
be sent simultaneously to speech because of the integration of SMS into the signaling
channel which normally holds such information as the actual condition, setup, or breaking-
down of the transmission path. The maximum message length of the different services are:
However, the disadvantage of such a transmission is the low data rate because in PP mode
160 bytes will be sent per second, whereas in CB mode 93 bytes will be sent only once per
minute. Because the data rate is not specified by GSM, the different operators have to
define them according to their own capacity.
The throughput of the whole system would decrease if the SMS data rate were increased
because of the embedding of SMS into the signaling channel.
However, for the first time, SMS presents a standardized data service in a cellular system
with full coverage of Europe which offers two-way communication and enables
improvement of traffic management by updating the dynamic data of at least major street
segments by floating car data. Using SMS-CB, the traffic information can be restricted to
relevant data for each cell. The communications costs are quite low (e.g., in Germany less
than 0.50 DM/transmission block). Establishing a connection is done very quickly.
Currently, new systems are planned in Germany for using SMS traffic transmission.
The General Packet Radio Service (GPRS) is a packet-oriented data service for GSM
currently being defined. This service should accomplish the following requirements [10]:
What makes GPRS interesting for the route guidance application? To answer this question,
the main two types of messages have to be considered: first, traffic information sent by the
traffic information center, and second, mobile-originated traffic data such as floating car
data.
For the first message type the multicast mode is highly desirable because all service
subscribers in the cell are updated simultaneously. Since the load of the communications
network can essentially be reduced, GPRS would be the basis for a low-cost service.
Additionally, the feature geographical routing enables informing all subscribers in an area
normally covered by more than one cell. This is very convenient and efficient for
addressing all subscribers affected by a traffic incident.
Since acknowledgment of multicast or broadcast messages is not important, only the
downlink of the communication link is used. For the application it is not a serious problem
when a message is not received because the traffic messages can be repeated ten or more
times per minute. The free capacity on the uplink can be occupied by other services.
The second type of message is transmitted on the uplink of the communications network,
where only single-cast messages are possible. The equipped vehicles send their reports on
the road status to the traffic information center at every junction or at regular intervals every
minute. These messages are not longer than 200 bytes, which would not justify the normal
access procedure of GSM, which loads the network for several seconds. It is much more
efficient to transmit this information via a packet-oriented service. Therefore, GPRS has
advantages
over the existing data services of GSM, especially for short uplink messages and many
subscribers.
The transmitted traffic information can be restricted to relevant data for each cell or a cell
cluster.This allows overall reporting on more traffic incidents. Initiated by the work of the
System of Cellular Radio for Traffic Efficiency and Safety (SOCRATES) working group,
cell-related messaging has been defined and will be part of the next GSM phase, including
GPRS.
In contrast to the idea of a circuit-switched mobile cellular phone network is the cellular
packet-switched data network, in which the data stream is seperated into shorter units called
packets. Every packet searches its own way through the network depending on the traffic
load of the links. The message is rebuild in the destination node. Because the packet size is
not the same for different subsribers, the network performance will be improved. Based on
this idea, the Ericsson Mobitex [11] system has been developed. It comprises all
possibilities for a future system: are point-to-point, point-to-multipoint, and broadcast
mode. It has the potential for transmitting data nation-wide or - as an extra service -
internationally.
The data are transmitted at a maximum rate of 8 kb/s, whereas the bandwidth of a single
channel is 12.5 kHz. The packet size can be adapted according to the application and could
be up to 512 bytes, but is internally split into smaller fragments. If more than 512 bytes are
needed, higher protocols must be used. If the transmission is corrupted by distortion or
noise, retransmission of such fragments is done using automatic repeat request (ARQ)
protocols.
Individual subscription and addressing of closed user groups is provided. A single user can
belong to more than one group.
For subscribers a geographical working area is defined, so they can be active preferably in
that area. If data are transmitted to other areas, there are extra charges. This proceeding can
be very attractive for road management.
Emergency message transfer is an integrated part of the system, so subscribers can transmit
and receive such messages.
Mobitex is not limited to a special frequency band. Therefore, vacant frequencies in
different countries can be used; but a widespread RTI message system requires a unique
frequency band, which is available in most European countries.
The existence of very small terminals for wireless connection to handheld computers with
standardized interfaces is very interesting to European countries. Because they are planned
for additional mass-market applications (e.g., RTI traffic transmission) can lead to more
subscribers and very cheap equipment for vehicles.
At present, a Mobitex system is under construction in Germany and licenses for roaming
have been signed with companies in some European countries. Traffic management system
test sites using Mobitex have been built in Gothenburg, Sweden, and Paris, France.
Another system that should be mentioned here, but without going into much detail, is
TETRA, which is under definition by the European Telecommunications Standards
Institute (ETSI) RES 6 committee [12]. This is not a running system but a standard for
trunked voice service and wide-area packet service. Both will use the same physical layer
and therefore the same transceiver equipment. One special goal is to define an air interface
and lower-layer data protocols so that the system response time is short and the price of the
transceiver low.
For an RTI application TETRA could become interesting because it comprises all the
required functions (e.g., point-to-point mode, cell broadcast mode, closed user groups, 36
kb/s channel rate). Nevertheless, the gap between such a standard and a running system
must be filled in the future.
Some of the above communication systems are used or discussed in the DRIVE (Dedicated
Road Infrastructure for Vehicle Safety in Europe) project SOCRATES [13] which
investigated the application of cellular radio to improve RTI. Besides the main aspect,
support of navigation systems, this project covers some other traffic management
applications:
The principle of SOCRATES for transmitting traffic messages to enable dynamic route
guidance is shown in Fig. 3. To be able to transmit accurate traffic messages, a huge
amount of information gathered from many sources has to be processed.
All information is processed in a regional traffic information center. Together with historic
data, the gathered data are used for traffic modeling. The traffic is predicted for the next
few hours. Based on the prediction, suitable messages are generated to inform the in-car
navigation system of traffic incidents and to control the traffic to avoid the development of
congestion.
The application of the two-way GSM system enables better traffic modeling because
vehicles can act as information sources. This improved traffic model allows more accurate
traffic messages, which results in superior route guidance.
Although this proposal achieves each individual driver receiving the data he needs, the
disadvantage is that the SOCRATES system leads to a significantly increased load of the
cellular network. To overcome this weak point, it was proposed to extend GSM by the
addition of GPRS, described previously.
Two different test sites have been established within SOCRATES: one in Sweden
(Gothenburg) and one in Germany (Frankfurt and the Rhine Hessen area) [14]. For the
communication task Mobitex is used in Sweden and GSM in Germany. Because only a few
test vehicles and only pure data service TCH/F2.4 dial-up mode was available during the
test period, this service was used in the Frankfurt area. The application protocols are
standardized by the SOCRATES consortium and submitted to CEN and the International
Standardization Organization (ISO). The telecommunication infrastructure is used as a
bearer service. For the different tasks the content and format of the application protocols
have been defined, for example, for floating car data, vehicle location (Table 1), and the of
the travel time histogram curve for a specific street segment.
The SOCRATES transmission was put on a well-established system in Sweden, and the
test showed that the data capacity would not be sufficient for a running system with many
users and a repetition rate of 6 times/s for the floating car data.
The situation in Germany was a little bit different. The main problem was that dialing every
single test vehicle and establishing a connection was very time-consuming. There was a
bad relation between the time for setting up a connection and the time for transmission of
100 bytes or less. Therefore, in the future commercial systems serving many motorists
must use other connection modes such as are used in SMS. GPRS in particular will offer
very fast call access for groups and individuals.
However, both test sites showed that the traffic management task had been solved very
well by integrating the floating car data sent by the individual cars on their ways through
the street network.
There are different systems and services which are able to establish communication
between a vehicle and a traffic information center. The differences between these systems
are communication mode, data rate, geographical area (local, nation-wide, Europe-wide),
number of different hardware components inside one car, availability, and others. Another
issue will be the charging policy of the service providers which becomes relevant in the
future when providers establish a commercial RTI service.
RDS-TMC is certainly the cheapest solution to providing dynamic data to a navigation
computer; but because of its restricted capacity, it is necessary to concentrate on the most
important traffic messages. The central point is the definition of the application protocol and
a very efficient coding of the data, especially because this RDS/TMC is focused on Europe.
SOCRATES, a European Commission (EC) RTI research project, tested two of the above-
mentioned data transmission services. In the Gothenburg test site which comprises a small
number of cells for communication the Mobitex system was applied. In the RHAPIT test
site (Rhine Hessen Area Project for Integrating Traffic management) GSM carried the road
traffic data on the TCH traffic channels in circuit-switched mode. Special test scenarios in
1994-1995 showed the possibilities and advantages of those services [15]. The proposed
GPRS will come in the late 1990s after it has been defined in the next GSM
recommendation phase.
Another system that comes up on the horizon is a digital mobile communication system,
TETRA. It has integrated all the ideas of the systems presented above and standards of
speech and packet data transmission. Because the discussion on such a standard is running,
the special demands of RTI applications should be fixed in the new specifications.
To reduce the pieces of equipment inside a car there should be the need for only one radio
box from one service provider for speech and traffic data.
It can be stated that the next generation of RTI systems will be based on two-way
communication services; it will be up to the operators and providers to offer users a system
with advantages while driving.
Because the data rate of the future third-generation mobile communication systems (e.g.,
UMTS - Universal Mobile Telecommunication System) will increase greatly, new
possibilities are coming up. Updated high-quality maps can be transmitted to cars very
quickly. On the uplink additional data on the cars' status can be transmitted to the
management center so that the calculation results are more precise. Thus, there is a benefit
for all drivers.
The author would like to thank Bosch GmbH K7/EFI, Hildesheim, Germany, for their
support.
[1] S-R Ely, "RDS-ALERT: A DRIVE Project to Develop a Proposed Standard for the
Traffic Message Channel Feature of the Radio Data System RDS," IEE Colloquium on
"The Car and its Environment - What DRIVE and PROMETHEUS Have to Offer,"
London, U.K., Jan. 1990.
[2] H. Deuper, H. Pfeiffer, and H. Schulz, "Improvement of Safety in Traffic Using
Autonomous Navigation Systems and Duplex Data Communication between Mobiles and
Traffic Centres," Proc. EUROSAFE Conf., London, U.K., Conf. Rep. S. 108-114,
1992.
[3] P. Br\144gas, "Function, Equipment and Field Testing of a Route Guidance and
Information System for Drivers (ALI)," IEEE Trans. Vehic. Tech., vol. VT-29, no. 2, p.
216.
[4] "LISB Leit-und Informationssystem Berlin," A 1900-N9-11, Siemens AG, Bereich
Stra\137enverkehrstechnik, M\174nchen, Germany.
[5] R. von Tomkewitsch, "Dynamic Route Guidance and Interactive Transport
Management with ALI-SCOUT," IEEE Trans. Vehic. Tech., vol. 40, no. 1, 1991.
[6] W. C. Y. Lee, Mobile Cellular Telecommunications Systems, New York: McGraw-
Hill, 1989.
[7] R. J. Sanchez, "CDMA Digital Cellular Overview and Field Trials Results, Proc. 1995
IEEE MTT-S Symp. on Technologies for Wireless Appls., Vancouver, BC, Canada,
1995.
[8] S. J. Parnall and J. L. Riley, "RDS developments," Proc. Int'l. Broadcasting
Convention, Brighton, U.K., Sept. 21-25, 1990, pp. 234-40.
[9] M. Mouly and M.-B. Pautet, The GSM System for Mobile Communications, published
by the authors: 49 rue Louise Bruneau, F-91120 Palaiseau, France, 1992.
[10] M. Schwartz, Telecommunication Networks, Reading, MA: Addison-Wesley, 1987.
[11] Tdoc SMG4 299/93rev1, "Requirements of General Packet Radio Service," ETSI
STC SMG1/SMG4, Vienna, Austria, 1993.
[12] G. Berntson, "Mobitex. A New Network for Mobile Data Communications," Ericsson
Rev., vol. 66, no. 1, 1989.
[13] J. L. Haine, P. M. Martin, and R. L. A. Goodings, "A European Standard for Packet-
Mode Mobile Data," Proc. 3rd. IEEE PIMRC, Boston, MA, 1992.
[14] I. Catling and F. Zijderhand, "SOCRATES - What Now?" Proc. VNIS '92 Conf.,
Oslo, Norway, Sept. 1992.
[15] P. Gilchrist and B. G\174nther, "The Path from Socrates/Rhapit Pilot to a Commercial
System, Proc. IEEE/VTS 44th Vehic. Tech. Conf., Stockholm, Sweden, 1994.
Wolfgang Schulz was born in 1946 in Bochum, Germany. He received his Diploma in
electrical engineering in 1975 from the RWTH Aachen, Germany, and did research from
1975 to 1978 at the Institute of Communications Equipment and Data Management at
RWTH Aachen. Since 1979 he has worked as a research scientist with the Department of
Communications Engineering at the University of Paderborn, Germany, where he received
his Ph.D. in 1984. Currently he heads a research group in the field of communication
technology and traffic applications.
(c) 1996 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.