Low data accuracy helps save energy - University of Tampere in Finland is coordinating a project to train young scientists from around the world to develop solutions

Computers use more and more energy. APROPOS – an international project funded by the European Union and coordinated by the University of Tampere in Finland – aims to improve the energy efficiency of global data use by up to 50 times, thus reducing data accuracy. Young researchers participating in the project come from all over the world to work in Finland, Sweden, the Netherlands, Austria, Italy, Spain, Switzerland, France and the United Kingdom.

Tampere, Finland 11/11/2021 –

It is estimated that by 2040, computers will need more electricity than the world’s energy resources can generate. The Approximate Computing for Energy and Energy Improvement Project (APROPOS) will train 15 junior researchers across Europe to meet the energy efficiency challenges of future embedded computing and high performance using disruptive methodologies.

This press release includes multimedia. Watch the full version here: https://www.businesswire.com/news/home/20211111006122/pt/

Professor Gary Nurmi is coordinating the APROPOS project and co-supervising four of its researchers. His current research interests include approximate and reconfigurable computing, software-defined radio and networking, and wireless positioning devices. Photo: Sari Labote

The power consumption of mobile broadband networks can be compared to that of data centers. The Internet of Things model will soon connect up to 50 billion devices over wireless networks. APROPOS project It aims to reduce power consumption in distributed computing and communications for cloud-electronic physical systems.

“Fortunately, in many parts of the world’s data acquisition, transmission, computing and storage systems, it is possible to reduce the accuracy that will allow to reduce energy consumption and time. With the introduction of precision in design optimization, energy efficiency can be improved by up to 50 times,” said the professor. Gary Normi From the Electrical Engineering Unit at the University of Tampere.

Normi points out that, for example, many sensors measure noisy or inaccurate inputs, and algorithms for processing the acquired signals can be random. Many sensors that measure natural environments produce noisy or imprecise data that can be transmitted and processed with less accuracy, without losing the fundamental trends of the observed phenomenon.

“Applications that use data may not need completely correct results; acceptable accuracy may be sufficient. This means that the system can be resilient against random errors, and for example, an approximate classification can be sufficient for a data mining system.”

New solutions are needed to deal with rising energy consumption

The overall energy consumption of computing and communication systems is growing rapidly despite recent advances in semiconductor technology and energy sensing system design. The APROPOS project will train fifteen research fellows in an international, multi-stakeholder environment to form the basis for enhanced resources in energy-conscious product and systems designs.

“Researchers in the early stage of APROPOS will be trained in both commercial and academic directions. Thus, they will be able to develop the commercial potential of their research and provide ideas for innovative products and services,” Nurmi added.

APROPOS is a four-year project funded by the Marie Sklodowska-Curie Innovative Training Networks of the European Union’s Horizon 2020 programme. It is formatted by University of Tampere Finland. Many young researchers come from outside the European Union to work in Finland, Sweden, the Netherlands, Austria, Italy, Spain, Switzerland, France and the United Kingdom.

Read more on http://www.apropos-itn.eu

University of Tampere

The Polydisciplinary University of Tampere is the second largest university in Finland. The driving forces for our research and learning are technology, health, and society. The university is committed to overcoming the biggest challenges facing our society and creating new opportunities. Almost all internationally recognized fields of study are represented at the university. The University of Tampere and Tampere University of Applied Sciences together comprise the University of Tampere community of more than 30,000 students and nearly 5,000 employees.www.tuni.fi/en

The original text of this announcement is the official approved version. Translations are provided as an annex only and must refer to the text in the original language, which is the only copy of the text with legal effect.