Q&A with Qoitech founder: optimising energy consumption for IoT devices
05 September 2018
After working on field testing with many partners, Qoitech CEO Vanja Samuelsson (VS) and her team identified a common need for a compact, flexible, everyday measurement tool. Reluctantly, Samuelsson founded Qoitech, a Sony Group start-up, to develop and bring to market their Otii solution: a comprehensive toolkit for energy optimisation of IoT devices. EPDT Editor Mark Gradwell (MG) interviewed VS to learn more about Otii…
For the digital issue of this piece, please visit this link – or click here to register for EPDT's magazine.
MG: Other forms of power measurement are available, so what is ‘disruptive’ about this one?
VS: Our focus when developing Otii was to elevate the user experience when it comes to measurements, visualisation and the analysis of energy consumption. It is really important for us at Qoitech that we offer a wide range of developers a much more user-friendly alternative for their power measurements.
Energy optimisation is important throughout the development process, regardless of whether it is done in hardware, firmware or software. We believe that every engineer can contribute to it, to develop the best and fastest end result. The concept of energy use should be a fundamental design parameter that is at the core of every development team, with tools available on every developer’s desktop to monitor and evaluate it.
This is why Otii was created to be a cost-effective plug-and-play, compact solution, but comprehensive in terms of its features. With its high dynamic range, scripting and battery simulation capabilities, Otii breaks down barriers in more ways than one.
MG: What insights does the Otii Arc provide, beyond those offered by a digital storage oscilloscope or digital multimeter (DMM)?
VS: Well, what our customers appreciate about the Otii Arc is that it measures both voltage and current simultaneously, which enables the software environment to properly calculate energy usage, and it can do this over long periods of time, such as days and weeks. Both digital storage oscilloscopes (DSO) and multimeters are usually limited per measurement. A DSO will typically capture only short sequences and a regular multimeter will typically only give an average value.
MG: OK, how would engineers integrate these insights into their established approach to design?
VS: Since Otii Arc is a power source as well as a measurement unit, a developer can use it in their normal setup, to power their device under development and measure the power consumption in the time domain, while also debugging the code. For those who practise continuous integration, Otii is very easily integrated into the design flow.
For software developers, it is about making energy consumption measurements part of their workflow for continuous integration. This means verifying this additional parameter before the code is merged into the main branch, to give insights on how their work will affect the end result – and how they can improve it to avoid firefighting and delays later. For the product development projects, this is crucial but is rarely utilised, due to a lack of affordable and easy-to-set-up tools – until today.
MG: This seems to be an empirical approach to development, rather than theoretical. What evidence do you have to show this produces better results?
VS: Battery-driven devices exist in a very fast-paced market, which is dictated by consumer demand for regularly updated, good quality products. Yet, despite this drive, they are particularly complex hardware designs, with layers of software on top. Otii fits well with the agile approach required, where designers can develop, review and adjust – and then repeat the process, to gain insights into the energy consumption of a device early on in the process. It also gives scope to continuously improve with every new change and enhancement that is applied.
This shortens the time spent developing – and ultimately results in better product quality. These sorts of results are ones we often see from our customers, from within the mobile industry right through to IoT devices. Our customer, Sensative, used Otii to help produce its smart home Strips, and managed to achieve a battery life of ten years; in comparison, the typical battery life of similar IoT products still lies around two years.
MG: Would you agree that the semiconductor manufacturing process results in a distribution curve with respect to power and performance, which is why manufacturers cite ‘typical’ values in their datasheets? If so, do you need to take this spread into account when using Otii? How do you advise this is achieved?
VS: Sure, there is definitely a distribution curve that needs to be considered by making multiple measurements, and one way to accomplish this would be to take several different measurements over a wide temperature range. However, the developer will most likely associate the curve with battery performance, rather than the semiconductors’ power consumption, due to the larger spread in capacity.
But in general, I would say that one of the most important design aspects for developers to really focus on when using Otii would be relative power consumption; these are improvements that can be more directly attributed to changes to the code.
MG: To what extent does the ability to write LUA scripts to the Arc contribute to its usefulness? Do you expect that most customers will use this feature or is it just for those who need to automate multiple tests?
VS: The LUA scripts allow developers to implement long-running tests, as well as automate testing. Battery discharge profiling is a great example of this longer testing.
This type of profiling is very interesting from the perspective of your question about a distribution curve, as there is a need to understand the performance of the battery in a realistic setting, and this is something not typically found in a datasheet.
Contact Details and Archive...