Connecting industry with IIoT

October 06, 2016 // By EDN Europe
Nick Robins Technical Director at Alpha Micro Components
A key promise of the Internet of Things that Really Matter is the revolution of industry, interchangeably labelled Industry 4.0 or the Industrial Internet of Things (IIoT)

Commercial viability is easily satisfied by efficiency improvements alone, yet furthered by the creation of fresh revenue streams that can effectively maintain themselves. It’s not all plain sailing though, the establishment of this connected industry ecosystem suffers challenges far greater than any commercial equivalent.

IoT 101 focused on gathering a plethora of raw data, analysing and presenting it generally to a human, yet IIoT demands more. For every piece of raw data, industry needs to know “where and when?” to historically unprecedented levels of accuracy - and the integrity of this “context rich” data must be infallible. As we strive for increasingly autonomous manufacturing and logistics, decisions borne from imprecise data can be catastrophic.

Connected sensors powering smart agriculture, with our ever growing population, are expected to secure the global food supply; whilst that aging population will rely on similar sensors to monitor their own vital statistics. Imprecise, compromised or uncommunicated data in either use case could be disastrous to the individual, or entire population’s very existence. Inaccuracy or high latencies, tolerable in the consumer world as mere inconveniences, simply cannot abide here. However, IIoT cannot permit improvements in these key areas at the expense of others relevant to harsh industrial environments – critically power consumption and environmental conditions.

The drive towards centimetre level accuracy has inspired both, combining existing global positioning entities and the development of fresh techniques for measuring signal angles, employing methodologies similar to those historically used to triangulate cellular signals from mobile phones.

A shift in emphasis in cellular from high bandwidth to unprecedented coverage, low power and cost has seen the emergence of Narrowband IoT enabling end-point devices such as sensors to connect directly to the cloud, negating often costly gateway devices. Simultaneously, the industry has realised the smartphones we all carry already combine both sensors, multiple wireless protocols and cloud connectivity and can suffice as IoT gateways or even end-to-end solutions all on their own.

Facilitating contextual data farming

A plethora of wireless communication interfaces serves to boggle IoT devices, with new protocols continuously emerging and the evolution of those long ubiquitous blurring past lines. Revised specifications serve to address historical shortfalls against competing protocols and drive new functionality such as mesh networking; yet as an industry, so many protocols are counterproductive in a time where we need rationalisation to ensure a cross-compatible IoT ecosphere. For example, traditional Bluetooth is intended to increase range to a kilometre! Bluetooth Low Energy will transform to double its 1 Mbits/s data rate, whilst slashing latency and adding mesh networking capability; sub 10ms latencies are vital for fail-safe reactions within IIoT.

With received signal strength indicator (RSSI) derived location determination, accuracy is restricted to the nearest metre and worsens as distance increases. IIoT itself demands a higher level of accuracy, yet in vehicle-to-everything (V2X) applications accuracy down to the centimetre is critical to make valid decisions and thus drive vehicular autonomy. Deriving location via AoA and AoD (angle of arrival/departure) is well studied and proven, whilst practises such as ToF measure packet times from transmitter to receiver and can achieve