Flying by wireless - is it safe?
Steve Russell, WRc's senior instrument engineer, looks at the future of wireless devices in the field of telemetry
At the WWEM conference in June this year, John Hother of Proneta UK gave an upbeat presentation on his company's experience of using GSM, the public cellular phone service, for data transfer or telemetry (see page 32). The discussion that followed looked at the potential conflict between the need for security of supply of a telemetry service, and the advantages of the latest mass market communications technology. Over the next couple of pages we shall unpack some of these issues for wireless devices, and illustrate them with reference to a current WRc project on a low-power radio protocol.
Expansion in telemetry
It seems inevitable that more telemetered data will be needed in the future. However, there are some specific drivers:
I believe that the greater part of this expansion will be in wireless systems, largely on the grounds of cost.
The water businesses are not newcomers to wireless telemetry - scanning telemetry links for major fixed sites such as treatment plants and point-to-point data transfer across sites have been used for decades. More recently there has been an expansion in the use of cellular phone networks to provide remote access data logging of pressure and flow in the potable water network.
New wireless applications are emerging from recent research. Installation of wireless monitoring systems is beginning on unmanned WwTWs, where research has shown there to be major cost benefits from using wireless links across treatment plant sites instead of hard-wired connections. Monitoring of CSO spills has been implemented using a mixture of hard-wired and wireless methods.
There are several current projects exploring the use of remote digital cameras as tools to allow operators to check on assets, to determine maintenance needs and to follow up on alarms. These applications are using a range of wireless equipment and services including low-power unlicensed radio, GSM, GPRS (General Packet Radio Service) and HSCSD (High Speed Circuit Switched Data) over the cellular phone network.
So far the installations are at small scale, and the cost consequences of a service being withdrawn are not great. However, if a much greater investment is to be made, security of service becomes very much more important, and is affected by several factors.
Using wireless products
Telemetry is a complex and 'techy' activity. The diagram shows five interlinked factors which can support a good price/performance
To meet basic legal requirements, a suitable portion of radio spectrum must be available over the life of the product. The spectrum is regulated by Ofcom in the UK, where a recent policy review will see a gradual shift to a market-led use of the 70% of spectrum which is licensed. Under the new rules, spectrum trading will be allowed, but Ofcom also promises that it will give users clearly defined rights to plan for the future. It is also to increase the available unlicensed
spectrum to encourage its use.
For data of critical importance, licensed spectrum is preferable, as Ofcom will prosecute illegal use of spectrum to protect licensed users. Unlicensed bands are cheaper to use, but are open to the risk of becoming overused in the vicinity of a site, with the potential for unreliable transmission. Ofcom offers no assistance in this situation; users can either tolerate the service or change their equipment.
For cellular phone services, the service-provider manages the licence as part of the package. When negotiating for these services, it may be worth including clauses about the service life, but in reality the decision to end a service will be driven by its main market and the commercial value to the service-provider. An informed judgement about the state of the market may therefore be more valuable than penalty clauses in judging the life expectancy of a service. The possibility of withdrawal of services applies to satellite-based services as much as to terrestrial ones, but there is also a parallel risk in using proprietary equipment. Single-supplier services pose a serious risk to long-term users, and here standards can be used to create a more open market.
Standards are essential to the establishment of reliable data communications. Traditionally, water utility telemetry has used proprietary systems from a few main suppliers, each with their own private standards. The WITS (Water Industry Telemetry Standards) group is now working towards a common and open protocol for regional telemetry, with benefits in the marketplace and in assurance of supply. If water utilities are prepared to agree on an open standard, it benefits suppliers by creating an open market, but it may also increase the size of the market. This can make the difference between suppliers choosing to invest in a product for the water utility businesses or investing elsewhere.
Critical to obtaining low-cost products is the use of suitable electronic chips. Chips are available which are practically a complete radio, but they are usually targeted at specific high-volume applications and some chip functions will be disabled and other circuitry added for a utility application. This has two main effects:
Chips will be designed for specific bands, and may also be designed to support specific protocols, hence the dots in the diagram.
Many of WRc's collaborative projects include the step of writing a user requirements specification. This has often proved very valuable in spelling out to suppliers what is wanted, making visible a market which they may have missed. It also helps suppliers to understand the application properly, so for example appropriate packaging and battery life are provided.
Low-cost radio protocol initiative
WRc has been working with a group of interested water utilities and suppliers plus the Low Power Radio Association to develop a radio
protocol for applications such as transmitting data across WwTWs. It is intended to be simple and low-cost, and has therefore used an unlicensed band. The specification will provide for interoperability of devices from different suppliers, and in spite of using an unlicensed band, it should offer good robustness against interference. It takes advantage of new developments in European spectrum management which provide for an unlicensed band at 863-870 MHz for systems using listen before talk (LBT) mode of operation.
The new protocol defines ten channels across the band, providing variable robustness in addition to the LBT method and radio quality requirements, such as out-of-band blocking specifications. In remote areas, systems can use a single channel and users can expect reliable operation. In populous areas, higher-cost devices can frequency-hop channels to get data through. The protocol specification development is now quite advanced, and we plan to begin testing devices in early 2006. The choice to develop a new open protocol was made after careful consideration of the current and emerging technologies. This is not an exact science, but our judgement was that other open-system products would be aimed at domestic and other larger industrial markets, and would not meet utility needs.
Future applications and standards
It is likely that additional sensors will be 'hung' on to the existing telemetry infrastructure at sites such as pumping stations, reservoirs and treatment plants. It will extend the monitoring capability to address the needs of energy management, improved management of risk, more sophisticated control, and tracking of medium-life assets such as pumps. The low-cost radio protocol will probably meet most of these needs, though there may be room for cheaper devices of shorter range using RF tagging or similar technology.
The pipeline networks are the highest-value and least monitored assets, and a major expansion in their monitoring seems very likely. A monitoring strategy is needed for these widely distributed assets, a strategy which is low-cost, supports the businesses in locating and operating the assets, and provides data for prioritising spend on maintenance/rehabilitation/renewal.
Current monitoring methods using the cellular phone network are too expensive for intensive use, and it is likely that a fixed network of radios will be used to collect data across a catchment. Such a network could be shared by clean and dirty networks and include AMR where installed. This is an obvious application for 'mesh' radio, where radios at intervals of perhaps 50m could pick up and relay the data. For this application it is particularly important to ensure a long life for the service employed, as the cost of replacing hundreds or thousands of devices below ground would be a major embarrassment.
Remote imaging offers a potential solution for difficult-to-monitor assets such as CSO screens. However, remote eyeballing may become widely used as a means of making a quick check on a site. The high data rates needed have forced this application to use the cellular phone network so far. The best approach to future-proofing is probably to ensure a modular construction so that the modem can be changed if the service is withdrawn.
Obviously, in the choice between mass-market products and specific water industry products, it comes down to risk, and risk grows with scale of deployment and the cost of accessing the devices.
There is a good case for a water industry body responsible for data communications standards to put in the effort to manage these complex issues. Candidate groups include: the Water UK Standards Group; TAUWI (Telecommunications Association of the UK Water Industry), which currently manages the use of radio spectrum for scanning telemetry; or WITS, which could become a permanent body with a wider role. By defining and maintaining standards, it would provide improved procurement and security of supply. Both the WITS initiative and the low-cost radio protocol project have shown that a group of water industry users can be very influential with suppliers.
Such a body could make a huge difference to the future cost of data by representing the water industry at UK and EU level. To my knowledge, there is no effective representation at EU level, where radio spectrum policy for the next decade is currently being decided. Viewed from WRc this seems a very clear need, with large potential benefits and little cost.
© Faversham House Group Ltd 2005. edie news articles may be copied or forwarded for individual use only. No other reproduction or distribution is permitted without prior written consent.