What’s next : A potted history of mobile communications in the water industry

The waste and clean water pipelines in the UK are nearly all underground and as such, the majority of battery powered devices are monitoring underground assets and are often themselves mounted underground. Having an underground antenna presents its own problems for mobile communications with signals normally optimised to arrive a metre or so above ground (i.e. at normal mobile phone height) having to penetrate earth to get to the device’s antenna.  As the carrier frequency in use by the mobile network goes up, the ability of that signal to travel through and around materials is reduced. The materials sometimes used as part of the chamber such as concrete and steel further reduce the amount of signal received.

During my time working around the UK water industry I had the good fortune to be working for a major telemetry equipment vendor as their Engineering Director. Among other things, they made, sold and supported battery-powered outstations. These used mobile communications to send telemetry data back to the master stations run by the water companies. At that early point in time, mobile communications were based on 2G and Circuit Switched Data (CSD), so the signal available to the modem would have to be good enough to allow a voice circuit to be set up over which it was possible to squeeze 9600 bits per second of data.

With the advent of GPRS the situation improved allowing modems to use packet based communications instead of CSD. Instead of maintaining a “circuit” between the mobile modem and the other phone all the time, data was just sent when and as required; this better suited the requirements of telemetry. At the same time Mobile Service Providers (MSPs) started to increase the price of data over CSD making GPRS a more sensible path to take.

Poor coverage at some sites was an inevitability when deploying devices over a wide geographical area. Instead of restricting oneself to a single network, roaming SIMs, which could select the network giving the best signal strength, from all networks visible to the modem, allowed further advantages to be gained. Roaming SIMs also helped where GPRS signals were available in both bands (900MHz and 1800MHz in the UK). The higher frequency didn’t play well with underground modems in concrete containers. However, if the 1800MHz base station was close and well situated with respect to the modem it could outperform less local 900MHz base stations, so it was good to have the flexibility offered by roaming.

As always, technology never stands still and hot on the heels of 2.5G and GPRS came 3G. This introduced higher frequency bands at 2100 MHz to cater for the higher data rates possible, however a higher frequency is yet another problem for underground mobile assets. Luckily 3G began to extended down into the lower frequency bands. This permitted more units to receive the signal and extended coverage from the base station at the expense of data rates, however, in telemetry and likewise with IoT, the data rates required are not normally high and so this did not prove a barrier to adoption.

3G and also now 4G are available in some countries on the 700 MHz band, for instance the USA and Australia. In the rural parts of these countries, which are normally the last to have mobile services rolled out (due to low population density against the cost of deployment), 2G rollout was often replaced by 3G and/or 4G rollout on the lower frequency bands. The frequency coming down to 700 MHz would be an advantage for telemetry in the UK, however that spectrum will not become available until 2022. On the plus side the technology available by that date will be well tested and easily available.

Ofcom’s Connected Nations Report indicates that 4G should allow 98% of premises to receive an indoor signal and a 90% geographical coverage by 2017. 4G is already pushing on the door of 5G technology with techniques such as carrier aggregation already being used to enhance data rate and provide greater connection resilience. So what is next for mobile and 5G?

In mid-November I attended an IET event “Towards 5G Mobile Technology – Vision to Reality” at the Royal Society in London. This event brought together industry leaders, innovators, researchers and regulators presenting current views on the state of 5G as an emerging technology. Professor Andy Sutton, who spoke on the day, has co-authored an interesting paper which gives a good outline of the 5G technologies. Rather than go through the technical details from the day, I have compressed the day into the few points I took away from that meeting. 

  • 5G is not here yet. We are currently in a research and development phase which started around 2011 and should result in standardisation around 2020, with commercial devices available around 2022.
  • 5G is targeting 1ms latency.
  • 5G is targeting peak throughput of 20 GBps, with a normal throughput of 10GBps and cell edge throughput of 100 MBps.
  • Frequencies for 5G are initially likely to be in the sub 6 GHz range with future interest in the above 6 GHz and above 30GHz bands.
  • Discussions are ongoing between national spectrum organisations to attempt to achieve international harmonisation in the frequencies selected.
  • Higher frequencies result in reduced signal coverage, however they also result in smaller antennas, and more opportunities for fast small steerable antennas which could be used to counter some of that loss in coverage.
  • Whereas most peoples’ experience of 3G and 4G may be patchy coverage, sometime working, sometimes not, 5G aims to provide a highly reliable service which does not drop out. It aims to do this through techniques such as carrier aggregation where the user should experience a single reliable connection even though, under the hood, many different connections are made to provide that experience.

One thing is for sure, over the coming years we will be seeing a lot of innovation in the mobile sector and these will slowly feed into the devices which provide telemetry for utilities. If the challenging targets for 5G are met, many new applications which are difficult to imaging at the moment are likely to surface. Arguably we are in for some big changes on what the devices can do and what the utilities choose to do with them.

The author Mark Davison is a director at Terzo Digital, and has 25 years of experience of embedded devices and communications technologies.