Continuous Intelligence for 4G/5G Mobile Edge Computing


Combining compute, connectivity, and low latency communication to endpoints lets MEC-hosted applications analyze streaming data on-the-fly and respond before storing it.

Next-gen mobile networks can deliver powerful services that take advantage of low-latency, high bandwidth, and edge computing. This article describes a software architecture for continuous intelligence applications that operate in 4G/5G Mobile Edge Computing (MEC) environments.

MEC offers low latency computing co-located with wireless base-stations and close to fixed networks. It interfaces directly to the provider Radio Access Network (RAN), which manages connectivity to wireless devices. MEC environments are also richly connected – to the provider’s core network, the Internet, and public and private clouds. As a result, MEC offers a unique opportunity to mobile providers to deliver valuable services that use continuous intelligence to analyze, learn, and predict from streaming data, on-the-fly, and respond in real-time.

See also: Why Businesses Are Implementing Edge Analytics in Their Line of Work

Continuous intelligence enables applications to analyze, react (and then store). Valuable, low latency responses are computed on-the-fly in the MEC context, but data storage occurs later, often in a public cloud or on-prem. This simplifies application architecture and compliance because data is analyzed on-the-fly but not stored in the MEC. Continuous intelligence capabilities should be thought of as supplemental to cloud computing services jointly offered by operators and cloud service providers. To deliver continuous intelligence capabilities to applications, providers need to add a data processing layer that is stateful, and that allows applications to continuously compute, driven by data using a stateful paradigm that is best thought of as a “stateful functions as a service” paradigm.

Mobile + Edge Computing = Opportunity

Continuous Intelligence offers mobile providers the ability to turn MEC into a powerful revenue generation platform. It exploits proximity to devices to compute on high-volume data with low latency, and can deliver insights a million times faster, and at a tenth of the cost of on-prem or cloud-based applications.

MEC is optimized for mobile use cases for edge computing and is documented in an ETSI standard. Applications in the MEC environment can take advantage of its rich connectivity and access to the Radio Access Network to discover powerful insights that are impossible to find in a general-purpose edge computing environment. The RAN manages services for all devices and using a continuous intelligence platform enables applications to gather a dynamic understanding of the real-world context of each device in real-time, and is a foundation for powerful new services:

  • Carriers can enhance their network services on the fly, dynamically maximizing connection quality, assigning resources to devices or network slices, identifying faults, delivering quality of service guarantees, ensuring privacy, and enhancing security.
  • Customer-facing MEC applications serve both enterprise and consumer needs:
    • Industrial automation: MEC applications can help predict equipment failures, detect problems, optimize supply chains, manage inventory, and customize production.
    • Augmented reality (AR) and virtual reality (VR) applications help workers understand the environment around them and repair or undertake work that relies on the dynamic fusion of visual and digital information.
    • Retail: MEC is required to deliver immersive in-store environments that require low latency, proximity, and personalization and offer new forms of payment.
    • Security and safety: mobile device location of one or more users can give emergency teams new tools to help in their response to emergencies.
    • Smart cities: Applications can dynamically predict future traffic load and gauge public transit needs for citizens, tailoring each user’s experience to their own travel plans. Information from sensors in utility infrastructure, fused with information from mobile devices, can help predict energy demands and control user environments based on personal preferences.

MEC has a powerful advantage over general-purpose edge computing, namely its proximity to the RAN, which offers low latency access to information about mobile devices. Dynamically building a picture of the real-world allows these applications to react in real-time to complex edge situations.

Exploiting MEC Advantages

The juxtaposition of compute, rich connectivity, and low latency communication to endpoints makes it possible for MEC hosted applications to analyze streaming data on-the-fly and respond before storing, so responses can be delivered a million times faster than with a traditional “store then analyze” approach that is hampered by database latency. Using this approach, insights are computed immediately in-memory, in context of a rich contextual awareness that helps to identify relationships – such as proximity, correlation, containment, and more – that enables deep insights. Continuous intelligence applications

  • Continuously analyze, learn, and predict, driven by data: Each update is statefully processed immediately in-memory. This improves performance.
  • Securely analyze behavior over time: This helps applications discover deep insights.
  • Analyze in context: Applications can discover hidden meaning in data, including real-world relationships – like containment, proximity, and correlation between different sources
  • Always have the answer: Algorithms analyze, estimate, and predict continuously from boundless data streams, as data arrives, rather than relying on batch-based analysis

To deliver these benefits to developers, the MEC needs to support an application platform that continuously updates a live model of the edge environment and its endpoints that can be securely used to deliver powerful new services and to enhance carrier networks.

MEC is not just a “Closer Cloud”

The stateless, RESTful microservice-and-database architecture that has been so successful in the cloud needs to be enhanced with a stateful processing model. Accessing a database to gather state for analysis is a million times slower than stateful in-memory analysis at CPU speed. A stateful computing architecture is key. Although there is a trend toward faster in-memory databases, they don’t run applications and can’t continuously find relationships between entities. Other approaches, such as event streaming, can only act as a buffer between the real world and applications. Reasoning about the meaning of a set of events requires a stateful system model that captures the states of all entities that the application needs to analyze.

Building a “LinkedIn for Things” in the MEC

It’s important to recognize that it is the continuous, concurrent state changes of data sources that are critical for situational analysis. Moreover, continuous intelligence needs to express fluid relationships, unlike traditional databases that capture fixed relationships (trucks have engines). Dynamically computed relationships (a truck “with bad braking behavior” “is approaching” an inspector) are typical in a fluid environment with mobile devices. Insights such as “bad braking behavior” and “is approaching” require continuous, stateful analysis to dynamically determine how the application should respond, moment by moment. The analysis must fuse real-time GPS information with static data – the real street map. The application must respond immediately, in context (an inspector on the same street and ahead of the truck should be alerted, but not others). Since there’s no point telling an inspector to stop a truck that has already passed, responses must be real-time.

The flow of continuous intelligence and automatic responses must be driven concurrently for every truck and inspector. Tens of millions of evaluations may need to be executed concurrently – for each “thing” and its current relationships.

The dynamic nature of the relationships between data sources suggests that we should represent the environment as a graph in the digital domain. Graph databases are used in social networking apps, but their graphs change relatively slowly. A continuous intelligence graph needs to be fluid – relationships are inherently dynamic – so computation must occur in the graph, in memory, driven by the arrival of data. The analysis must occur in the context of the current state of and relationships between sources in (the ever-changing) graph of relationships rather than “over (a pre-built) graph.”

Continuous intelligence demands stateful in-memory processing to optimize performance and to enable continuous computation for real-time responses. It embraces event streaming and other infrastructure patterns, focusing instead on the application layer capabilities needed to develop and operate stateful applications that consume streaming events at scale. Although modern databases can store streaming data for later analysis, and update relational tables or modify graphs, continuous intelligence drives analysis from the arrival of data – using an “analyze, react, then store” architecture that builds and executes a live computational model from streaming data. Whereas streaming analytics applications use a top-down visualization/user-driven control loop, continuous intelligence applications continuously compute and stream insights, deliver truly real-time user experiences, and facilitate real-time automatic responses at massive scale.


MEC offers an opportunity for mobile providers to deliver continuous intelligence services to their customers and to optimize their own network operations. Rather than focus on the delivery of the same kinds of applications that customers can deliver from IaaS clouds, providers should take advantage of key attributes of MEC: low latency, data volume reduction, and rich interconnection to networks and services.

Most importantly, MEC is uniquely situated to identify dynamic relationships between data sources, including proximity and correlation, to surface insights and respond in real-time to real-world needs. Cloud applications that provide value by analyzing and deriving value from streaming or real-time data can adopt continuous intelligence alongside MEC to offer valuable new services to customers. These services can take advantage of the benefits provided by 5G and stateful computing to compete in today’s fast-moving markets and enable new opportunities for network resilience and business continuity.

Simon Crosby

About Simon Crosby

Simon Crosby is CTO of Swim.AI.

Leave a Reply

Your email address will not be published. Required fields are marked *