This is my last "Parity Check" column for Industrial Networking because I'll be starting a new position as this issue goes to press. Therefore, this seems a good time to go out on a limb, and "bravely" predict where industrial network solutions are going, along with the associated hurdles that need to be overcome to make these visions a reality.
Bill Gates says we always over-predict what will happen in two years and under-predict what will be available in five years. Please bear this in mind as we press on.
The ISA's SP-100 committee will be the driving force behind a new series of Wireless in Automation standards that will be adopted by the IEC as the basis for a related group of international documents. Similar to the IEC 61508 and 61511 standards on safety, the resulting standards will be "split" to be suitable for end users and manufacturers of wireless products. The standards themselves will not be prescriptive, defining the characteristics of a wireless network rather than the physical definition and creation of the protocol. Going this route will prevent another round of "fieldbus wars" and the resulting multi-headed standard. Instead, the various existing protocols will, by following the definition, "merge" towards a product or series of products that are interoperable themselves.
Many of us already experience the challenge of bandwidth availability in the ISM bands with our cellular communications products. Of course, adding the bandwidth requirements of video/television-type phones simply compounds the problem. What's a person to do to insure that critical automation signals aren't "jammed" by an important playoff game? Two options are available: assign one or more of the UHF television bands being released later this decade (with the forced migration to high definition) to be for industrial communications and license-free; or, just as we can now do with Ethernet messages, assign a "priority" to certain types of signals over others. This second option is likely more politically charged. Who's to say that my control signal is more important than a call to 911? My hope is that we'll get some dedicated bandwidth from the UHF band. If you agree, be sure to tell your elected representative.
The challenge of power to wireless devices will be addressed through a combination of better power supplies and power management, as well as the ability to recharge the remote device over the wireless signal itself when not in transmit/receive mode.
Consistent with the move to smaller devices with more processing power, and less power consumption, nanotechnology will make motes feasible for wide-area monitoring. Motes will be used for such tasks as environmental monitoring (of specific pollutants, temperature, moisture, wind direction, shear, etc.), and as alternates to deploying people when looking for chemical weapons or in areas suspected of being highly dangerous to personnel.
For less-challenging applications and environments, RFID devices won't only identify what they're attached to and how they got where they are, but will also perform a limited degree of on-board processing to communicate where they should go next. They'll also be able to interact directly with maintenance and operational tools, such as handhelds in their "sphere of influence," to inform passersby that they require attention in the near future. The RFID tag will communicate with internal electronics in the device, so it can update the tag with diagnostic information that could affect its performance.
Despite all the wireless developments, there will still be a need for hubs, access points, and similar devices to be physically connected via cable. However, this cable will be more than the signal carrier. With the Power over Ethernet (PoE) standard available — driven by VoIP and Wireless Access Points — the same technology will migrate to the industrial arena. The first challenge that efforts to use PoE in industrial devices will have to overcome is converting the 48 V in the IEEE standard to more widely used 24 V for automation. In parallel, Intrinsically Safe PoE also will be available to distribute Ethernet to Zone 1 areas. In addition, to meet the needs of extending the Ethernet network beyond the 100 m limit of UTP, fiber-optic cable with appropriate receivers and transmitter also will be developed to transmit the necessary power and signals over longer distances.
Will all this technology mean the demise of "conventional" fieldbus protocols? I think not. Instead, the proliferation of Ethernet to the field will be the catalyst to more fully extend the digital plant. The fieldbus protocols serve a special need — based on the way they're designed — to meet the requirements of specific processes. The needs of the process will not change, but the way in which the devices on the fieldbus connect to the central control host will change.
The end result of all these networking developments will be a seamless-to-the-end-user suite of products that are "just there" when needed with a reliable secure signal available wherever and whenever it is needed.