Vodafone is to use machine-learning (ML) and artificial intelligence (AI) to monitor its pan-European cloud-native network in a partnership with Cardinality.io and Google Cloud.

The aim is to give Vodafone’s customers a consistently faster and reliable mobile experience across Europe.

The platform is being implemented across eleven European countries in which Vodafone operates and should transform the way it plans, builds and manages its network over the next five years.

Vodafone’s European operating companies will be able to draw billions of network performance datapoints from a secure cloud data pool using the platform. It will replace over 100 separate network performance applications in use today, providing a single source of clean data in the cloud that can be optimised and analysed using AI.

Vodafone will use the data to monitor, manage and enhance its entire pan-European mobile network, including critical national infrastructure (CNI). The platform will enable Vodafone to focus 5G investment with pinpoint precision in the areas where the demand is greatest such as in cities, travel corridors, ports, industrial hubs and in specialist areas such as mobile private networks, as well as target rural areas with patchy coverage.

Capable of processing and analysing up to eight billion mobile network performance events per day, Vodafone data scientists can have deep insights into network data such as traffic patterns to make informed and instant network-changing decisions. This means Vodafone can add capacity and integrate new network connectivity in response to major incidents, public and political events, manage energy efficiency and restore services after severe weather storms — within minutes and hours, not days — on either a highly localised level or internationally across borders.

“As the needs of our 300 million plus mobile customers evolve so will our network using this new platform,” said Johan Wibergh, chief technology officer of Vodafone. “It is a global data hub that gives us a real-time view of what is happening anywhere on our network, uses our global scale to manage traffic growth cheaper and more efficiently as customer data consumption grows by around 40% per year, and supports the full automation of our network by 2025.”

Performance management tools are used for most network processes including capacity planning, optimisation, troubleshooting, enhancing the customer experience, and introducing 5G and edge computing services. By simplifying data operations at scale and generating real-time insights, the platform should make all these processes faster, smarter and more personalised.

Vodafone has already seen major network and IT incidents fall by nearly 70% this year by simplifying and digitalising core operations and, within a few years, it expects to resolve 80% of all incidents end-to-end without human intervention. In turn, this is freeing up Vodafone’s expanding technology workforce to develop products and services faster, cheaper and at scale.

Also, using these ML and AI monitoring tools, Vodafone can detect and block fraudulent or suspicious activity on the network across Europe often before its customers are impacted. It will underpin the operator’s work in smart planning across its entire European footprint, a method of using key network metrics to prioritise upgrades while significantly reducing the cost of maintaining and extending its networks.

Called Vodafone Unified Performance Management (UPM), the platform combines Google Cloud’s smart analytics portfolio and ML and AI tools with Cardinality.io’s telecoms industry expertise and cloud-native data-ops and analytics platform to deliver on a hybrid cloud architecture. All data are stored within Vodafone’s on-premise data lake on servers in Europe. Data will be protected with appropriate security and privacy protections such as encryption, anonymisation, pseudonymisation and aggregation and will be accessible only by authorised users.

“Moving data to the cloud while providing governance across multiple on-premises sites is an immense challenge,” said Vivek Gupta, director at Google Cloud. “Cardinality.io’s close integration with Google Cloud unlocks immense value for Vodafone as it seeks to provide an unparalleled customer experience to its European subscriber base.”

Prashant Kumar, chief technology officer at Cardinality.io, added: “With Vodafone UPM, Cardinality.io provides the bridge that makes it possible to efficiently move data from multiple far edge locations to a centralised Google Cloud data repository. Thanks to the Cardinality.io platform’s data virtualisation capabilities, open API interfaces and microservices-based architecture, Vodafone can choose the third-party tools it prefers for creating dashboards, reports and advanced analytics.”

Vodafone UPM centralises and stores data from across Vodafone operating companies in 11 European countries and across technology domains, including radio and mobile (2G, 3G, 4G and 5G), voice, data, transport and network virtual infrastructure. A unified web user interface provides access to dashboards, reports and monitoring tools, as well as analytic capabilities across a set of common key performance indicators (KPIs).

Vodafone’s hybrid cloud architecture from Cardinality.io and Google Cloud provides a scalable and real-time mechanism for on-premises data to be ingested, parsed, formatted and enriched in memory before being streamed to a Google Cloud HDFS cluster.

The cloud-native Cardinality.io runs both on-premises and in a centralised Google GKE cloud cluster, providing a data fabric that allows data to be moved from multiple edge locations to core data centres where it is enriched before being moved to Google Cloud storage for monitoring, visualisation, analytics and dashboarding using the Cardinality.io platform.