Our guest post is from Positium, a data analytics company from Tartu, Estonia. They are the only company in the world specialising in mobile positioning data for official statistics. Positium has over 10 years of experience working with global clients and scientists alike, always striving to improve modern statistics with timely and accurate data. They provide a technological platform and methodology for processing mobile phone data for human mobility monitoring, analyses and statistical indicators.

There are over 5 billion people subscribed to mobile services around the world [1]. We generate vast amounts of data through mobile devices every day. Yet mobile phone data is still virtually unharnessed today despite its potential to modernise policymaking everywhere. According to Claire Melamed, CEO of the Global Partnership for Sustainable Development Data, we are using outdated information to make our policies today [2]. Mobile positioning data (MPD) can offer a timely and accurate big data solution for official statistics.

This type of data is collected passively and therefore does not place any additional burden on respondents. MPD is readily available, waiting to be harnessed for the benefit of societies worldwide. At Positium, we are sometimes asked questions like “So you can see where my phone is at any point in time?” or “Are you guys like Big Brother, always watching everyone?”. The simple answer is “No” and, quite frankly, the mobility of individuals is not what interests us – Positium wants to know where, when and how masses of population move to improve planning decisions on local and national levels. However, these questions bring to the forefront people’s main concern over using mobile phone data for statistics – data privacy.

Working with clients from different countries, we, first and foremost, have to ensure the data privacy and confidentiality of the data we use. We cannot emphasise enough that statistics should not come at the cost of people’s privacy. Indonesia is one of the first countries in the world to tap into this unique opportunity. For the past three years, we have been working with the Indonesian Ministry of Tourism to improve the country’s statistics on cross-border tourism.

Indonesia is a land of thousands of islands and not every border crossing is registered in immigration statistics. “Roaming MPD-based statistics can be used to measure cross-border tourism, so that the decision-makers of the state would have reliable statistics,” says Erki Saluveer, CEO of Positium. When MPD reaches Positium, the first step is to clean and process the data to distinguish tourists, as they are defined by the national statistical offices, from the other users in the network. When we started the pilot project in Indonesia, we soon realised that to get the most accurate estimate of cross-border tourism, we needed to focus on analysing data from two of the major mobile network operators in Indonesia. With so many entry points to the country and tourists’ phones constantly switching between network providers, it would have been difficult to pinpoint how many tourists actually enter the country from the neighbouring areas. Making this proposal, however, had its challenges: How to compare the two data sets without neither compromising data privacy nor revealing each operator’s business secrets?

The pilot project, which took place in late 2018, featured privacy-preserving technologies for mobile positioning data, courtesy of Sharemind HI – a new solution by Cybernetica which uses the Intel® Software Guard Extensions (Intel® SGX) technology. Sharemind HI employs the Intel® SGX technology to help create Trusted Execution Environments (TEE’s). Sharemind HI fully utilises the three key concepts that Intel® SGX provides to help protect data – enclaves, attestation and data sealing. The Indonesian pilot project included mobile positioning data analytics from Positium and calculations on encrypted data using Sharemind HI.

The project provided new information on roaming counts and roamer overlap between the two big telecom providers, enabling us to find out the real roaming market share. This is an important part in the accurate calculation of tourism indicators from mobile positioning data. According to Baldur Kubo, account manager at Cybernetica, these technologies allowed us to provide better quality in data as well as overcome barriers in data security and privacy. Sharemind HI is useful in situations where sensitive datasets from multiple sources need to be combined for analysis.

Collaboration is highly important on an international scale – so much so that it is noted in the Sustainable Development Goals (Goal 17) as something we as humans need to strive towards [3]. Positium believes in providing a holistic solution to our clients’ statistical questions, which is why the partnership with Cybernetica was like a match made in heaven. “We were honoured to participate in this initiative together with Positium, to be able to tackle real-life problems that have nation-wide impacts,” Kubo said. “The success of this pilot project can open new avenues for mobile location data.”

The results of this collaboration between Positium and Cybernetica will be presented on 8–10 July at the poster session of the NetMob 2019 conference in Oxford, UK.

[1] https://www.gsmaintelligence.com/
[2] http://www.data4sdgs.org/news/were-constantly-creating-data-lets-use-it-accelerating-sdg-progress-through-timely-data
[3] https://www.un.org/sustainabledevelopment/sustainable-development-goals/