Wireless measurement accuracy improvement
NPL has worked with wireless sensor network developers, Senceive, to help them understand their measurement challenges, and improve their products.
Wireless sensor networks (WSNs) are groups of sensors that collect, process, and send information about the environment around them. Depending on their design, they can collect information on temperature, pollution levels, sound, vibration, pressure, location and many other quantities. They are used in many areas of measurement, security, structural health monitoring, and environmental monitoring.
One of the main barriers preventing their use in new, safety-critical, applications is the difficulty of proving the accuracy of the sensors' results. One way to improve confidence in a sensor's output is to check it against another sensor. By combining the data gathered with a model of the system you can predict what an individual sensor should be reading and hence check how reliable the sensors' measurements are.
Senceive's main application area is long-term infrastructure monitoring, particularly in transport and heritage. Their WSNs are highly complex embedded systems, meaning it can be a challenge to quantify the uncertainties involved.
Senceive needed NPL's help to improve their tilt sensing system, and verify its accuracy. This system was designed for structural health monitoring-type applications, so generating reliable data is critical.
Building upon work in self-validation and data fusion, NPL characterised Senceive's tilt sensors' output, in terms of linearity, jitter and overall uncertainty. NPL also advised Senceive on system improvements and on future deployments.
This has given Senceive a greater understanding of the measurement issues they face, and prepared them for an upcoming large-scale deployment.
About the Customer
Senceive are a Fulham-based supplier of wireless asset monitoring systems. Their systems use wireless sensor networks to protect infrastructure spread across wide areas on industrial machines, data centres, railways, utilities and many other applications. By retrieving data from almost any sensor type either on-site or over secure internet, our customers monitor their sites from anywhere in the world.
Representatives from Senceive became aware of NPL's expertise after attending a knowledge transfer event.
Senceive needed to overcome a significant measurement issue which was preventing their system being used in a number of applications. To do this they needed to understand how different environments could affect their sensors. In particular, Senceive wanted to understand how temperature affected their sensors, and whether the sensors could be made to self-validate, or correct, for temperature-induced sensor response.
Wireless sensor networks combine sensing, computation and communication within a single, often very small, device. They are quickly becoming a significant enabling technology in many areas of measurement, security and environmental monitoring. In the next decade, industry will begin to benefit from autonomous sensor networks that can determine their position and adapt to their environment. But, this development raises a number of measurement challenges relating to reliability, system confidence, calibration and validation.
Self-validation is a valuable tool for extending sensing systems' operating range as well as making them more robust and efficient.
To solve this measurement problem, Senceive and NPL worked together to put the sensors through a series of tests to see how they performed in laboratory conditions. The results of the tests allowed Senceive to make some significant improvements to the sensors, and NPL advised them on measurement principles which would help them with further improvements of their products.
By working closely with NPL, Senceive were able to cut many weeks of development time and save money, as they were able to tap into NPL's expertise and make use of their facilities.
The results from the work have given Senceive the confidence of knowing how well their sensors perform with real data to back up their datasheets.
The overall knowledge gained from working with NPL has also given them a better understanding of how the sensors are affected by temperature and will help them further when the time comes to upgrade their product offerings.
"Working closely with (NPL scientist) Michael Collett and tapping into the great depth of expertise at NPL has been invaluable in helping us understand the measurement challenges we are faced with, and enabled us to incorporate some significant improvements into our system design."
Michael Gois, CTO Senceive Ltd
Do you have a measurement challenge that you'd like NPL's help with? If so, why not apply for NPL's Technology Innovation Fund ?
Please note that the information will not be divulged to third parties, or used without your permission