CANopen in food-packaging machinery

Would you like to read more? Please subscribe to the CAN Newsletter. It's free!

To top

CAN-Bluetooth on the search for applications

Bluetooth was developed for short distance wireless transmissions in an office environment. Other uses soon became imaginable and the development has extended far into other areas, such as automotive and industrial, where CAN is employed as a major embedded network. Most of the applications are still at an early test phase, however there definitely is a push into the direction of combining the advantages of CAN with those of Bluetooth commercially. 
The Bluetooth wireless specification includes both data link layer and application layer definitions, which support data, voice, and content-centric applications. Radios that comply with the specification operate in the unlicensed, 2.4-GHz radio spectrum. These radios use a spread spectrum, frequency hopping, full-duplex signal at up to 1600 hops per second. The signal hops among 79 frequencies at 1-MHz intervals to give a high degree of interference immunity. Up to seven simultaneous connections can be established and maintained. The specification contains the information necessary to ensure that diverse devices supporting the Bluetooth wireless technology can communicate with each other worldwide.
The communication with a CAN network always entails one or more gateways. Two scenarios are feasible: One, a wireless CAN-to-CAN bridge via two or more CAN-Bluetooth gateways thus connecting independent CAN networks with each other. Two, the wireless access to a CAN network via a laptop or a PDA. The Bluetooth transmission method is a point-to-point method. Each of the up to seven nodes is addressed independently from the master. No broadcast as in CAN is possible. The distance between two gateways or a gateway and a computer can be about 20 m. Theoretically it can be much higher, up to 100 m, but that range depends on several factors, such as electro magnetic interference, transmission medium (air, water, etc.), temperature and signal power, among others.
Some areas in which CAN is used could very well profit from wireless data transmission as provided by Bluetooth. No large-scale application information is available yet but according to Thomas Pohlmann of RM Michaelides (www.rmcan.de) the company’s CAN-Bluetooth gateway was the first one that was available off-the shelf and has been used in several applications. Infotainment, for example, in the automotive industry, may use wireless hands-free speech-controlled phones, audio and navigational equipment. Bluetooth is likely to find its way into comfort and security tasks such as adjustment of seats and mirrors and operating locks also. An even likelier application is automotive diagnostics, whether it be at the production plant or in the service shop. With an off-the-shelf notebook with a Bluetooth interface, the analysis of CAN messages is possible without ever so much as touching a vehicle. The remote access grants methods of surveying all available CAN networks via their Bluetooth gateways, listing them on the notebook and picking the appropriate one for diagnosis via a unique identifier and a pre-defined text string from the gateway, which would be located in the vehicle’s CAN powertrain by default. Diagnosis is then possible from one central point in a production line or a service shop without any cabling, which may be expensive, prone to mechanical wear and tear and bulky. Cable cost, installation and maintenance also plays an important part in industrial applications. Here, as well as in automotive diagnosis, access to the CAN network can be cumbersome. At times the environment is a high-voltage installation, an explosive area or not in direct line-of-sight of staff. Radio transmission of CAN messages may very well amount to savings in working around those problems. Sensor signals can be transmitted to a central point and program information be distributed from such a point to network nodes.
Cabling is also next to impossible in robotic tasks during which the robot turns around several axes multiple times. The option of remote control is of vital importance to the task since without it the robot would be strangled by cable. A possible set-up for remote control tasks might also be in road construction, in smaller construction machines, in remote control of agricultural pumps or watering places, agricultural machinery, cranes, automated guided vehicles, etc. Logistics could benefit from a remote status indication of loading status or weight of a vehicle. Mobile displays at mass events or in road construction or traffic control can be remotely and wirelessly connected. Uses in medical device control are imaginable, etc.

Range and bandwidth pose challenges

There are, however, considerable problems contained in this wireless technology: The radio transmission of CAN messages via Bluetooth is just for short distances. At the moment a realistic range up to 20 m is possible. Developments for a range up to 100 m are under way but as soon as the range is increased, meaning the signal power is increased, the electro magnetic interference and the signal noise increase to a point where the signals are no longer error-free. Also, in most scenarios where a CAN-Bluetooth gateway is imaginable and useful, metal objects are heavily used, which will attenuate radio signals and at worst create a Faraday cage, in which no transmission at all is possible. Well-designed antennas seem to be just a feeble effort to avoid the underlying problems. The design of Bluetooth chips has so far just allowed use in operating temperature ranges from 0 °C to approx. 50 °C, challenging exactly the one area, where the technology is most likely to find quantity customers: the automotive industry. 
Another serious drawback of CAN-Bluetooth is the low data throughput of the gateway due to bandwidth restrictions and thus the lack of real-time capability for such systems. As long as such a set-up cannot guarantee real-time performance, its uses are very limited, indeed. However, currently there are efforts to develop a TTCAN-Bluetooth gateway in order to achieve real-time capability. 
The Special Interest Group Bluetooth (www. bluetooth.org), which includes promoters such as 3Com, Agere, Ericsson, IBM, Intel, Microsoft, Motorola, Nokia and Toshiba, is very well aware of the problems and has tried to clear some of the obstacles with a new version of the Bluetooth specification (Version 1.2), which is available for members already and will be accessible for non-members shortly. 
This version has improved on the previous versions in the following areas: 
w A traffic management has been defined rather than the piconet solution that has prevailed so far. In a piconet a master controls the joining slaves and specifies the hopping sequence. All communication takes place between the master and a single slave at a time since there is no broadcast mechanism for application data. Inter-slave communication is not possible. The simplest form of a piconet is a point-to-point connection between a master and a single slave. Also a topology with overlapping piconets, a scatternet, has been possible but implementations lacked the capability of participating in more than one piconet [1]. Details on the new traffic management are not yet available.
- Bluetooth messages so far were unprotected except by the frequency hopping of the technology. No encryption or anonymity of sender was provided. In the 1.2-version the physical sender address is masked, which is supposed to protect the transmission from unwanted listeners.
- The quality of speech connections has been improved upon, from which speech-recognition in noisy environments will benefit especially.
Many ideas exist about possible applications for Bluetooth, but few of them have found their way into actual projects. Those that have been implemented are mostly still at a testing stage without reliable reports on results. Also, no application requiring large numbers of gateways really necessitates Bluetooth-CAN yet.
“We are actively looking for application areas from which our customers might benefit”, says Matthias Fuchs of Esd (www.esd-electronics.de), “Current applications are simplified and save cost by using wireless technology, but it’s not as if they couldn’t work without it altogether.” Other wireless technologies are available for identical applications and make the choice for customers, which wireless technology to choose even harder. 
Joerg Blumeyer from I+ME Actia (www.ime-actia.de) thinks along those lines, too: “As long as we don’t extend the Bluetooth range to 100 m, the automotive industry will prefer other solutions, such as wireless LAN”, he says, well aware of the problems in interference that come along with raising the signal power. 
New developments will have to work particularly on the signal range and operating temperature range in order to prevail over other wireless technologies, such as wireless LAN (www.wlana.org), ZigBee (www.zigbee.org) or Wi-Fi (www.wifialliance.org).


Would you like to read more? Please subscribe to the CAN Newsletter. It's free!