Building the Future of IoMT: Our Catalyst for Real-Time Orchestration and AI at DTW25

Collaboration Brings Scalable, Real-Time Management to the Internet of Moving Things (IoMT)

The telecom industry is entering a new era—one defined by autonomous systems, hyper-connectivity, and a relentless push toward sustainability. At DTW25 – Ignite, TM Forum’s flagship event, this transformation took center stage. Among the event’s standout innovations was Catalyst 805: .

This groundbreaking Catalyst initiative tackled one of the industry’s most complex emerging challenges: how to orchestrate and manage millions of mobile, connected devices that operate in unpredictable environments. These aren’t traditional, stationary IoT devices—they’re drones, autonomous vehicles, and robotic systems constantly on the move, generating vast streams of dynamic data.

Catalyst 805 brought together a diverse alliance of telecom operators, AI specialists, infrastructure providers, and research institutions to build a composable, event-driven solution capable of scaling and adapting in real time. It was a masterclass in how collaboration fuels innovation—and how that innovation becomes reality.

Why DTW25 – Ignite Was the Perfect Stage

DTW25 – Ignite was designed to accelerate meaningful change in the telecom ecosystem. It emphasized collaborative innovation, cross-industry partnerships, and practical outcomes for digital transformation. The event was a call to action for operators, vendors, and disruptors to co-create solutions that simplify telco models, scale automation, and modernize infrastructure.

Catalyst 805 met these ambitions head-on. It brought together multiple partners from different sectors to build and demonstrate a fully integrated, scalable, and sustainable IoT management platform—proof that cross-disciplinary collaboration can yield production-ready results.

The Challenge: Managing the Internet of Moving Things (IoMT)

IoT ecosystems are evolving rapidly. Devices are no longer stationary sensors mounted on fixed infrastructure—they’re mobile, autonomous systems operating across vast spaces. From delivery drones and self-driving cars to automated port surveillance, these assets demand seamless connectivity, intelligent orchestration, and low-latency decision-making.

Managing these assets presents a multidimensional challenge:

• Dynamic Movement – Devices move unpredictably and interact with changing network conditions.
  
  
• Real-Time Coordination – Orchestration must be instantaneous to maintain operational integrity.
  
  
• Edge Resource Allocation – Compute and network resources must follow the device.
  
  
• Sustainability Requirements – Power consumption must be optimized to reduce environmental impact.
  
  
• Monetization – As 5G and edge deployments scale, telcos must find ways to monetize usage-based services in real time.
  
  

Catalyst 805 tackled these challenges through a cloud-native, standards-aligned solution built on real-time responsiveness, intelligent automation, and composable architecture.

From Concept to Reality: The Catalyst Architecture

The technical foundation of Catalyst 805 leveraged TM Forum’s ODA Canvas and event-driven architecture (EDA). This was critical to enabling:

• Real-time responsiveness: Systems reacted instantly to changes in device state, mission progress, or network conditions
  
  
• Decoupled, scalable architecture: Event streams enabled asynchronous communication, making it easy to adapt and scale
  
  
• Autonomous edge service orchestration: As drones or vehicles moved across locations, services followed them automatically
  
  
• Optimized API interaction: Event-driven orchestration complemented synchronous TMF APIs, offering dynamic service activation
  
  

The architecture connected event producers—such as drone portals and telemetry systems—to Kafka topics that streamed real-time data for AI/ML analysis and intelligent service orchestration. This enabled responsive control, feedback loops, and dynamic coordination across the system, supporting features like sustainability dashboards, generative AI interfaces, and adaptive drone operations—all within a standards-aligned, cloud-native ecosystem.

The Architectural Core: Event-Driven Architecture & Intent-Based Design

At the heart of the Catalyst was an event-driven architecture (EDA) aligned with the principles of the Open Digital Architecture (ODA). This architecture allowed the system to react immediately to incoming events—such as changes in device state, network conditions, or mission status—without relying on static workflows or human intervention.

Key architectural strengths included:

• Asynchronous Communication – Components communicated via event streams, not synchronous API calls, reducing latency and enabling loose coupling between services.
  
  
• Scalability – Kafka-based pipelines handled high-throughput telemetry and service data from thousands of moving devices simultaneously.
  
  
• Intent-Based Control – Business goals and service-level expectations were translated into “intents” that orchestrated dynamic behaviors automatically.
  
  
• Resilience and Agility – The architecture could reroute workflows and reassign resources in milliseconds to maintain service quality even as network or device conditions change.
  
  

This flexible, reactive architecture was essential for orchestrating services across unpredictable environments and ensuring that real-time responsiveness was baked into every layer of the system.

The Intelligence Layer: AI-Driven Autonomy and Sustainability

Catalyst 805 extended beyond orchestration with a robust layer of artificial intelligence (AI) that drives autonomy and sustainability across the ecosystem.

The solution employed a modular multi-agent AI system, each agent with a distinct responsibility—coverage optimization, energy efficiency, anomaly detection, service recommendation, and natural language interface.

What set the AI system apart:

• Continuous Monitoring – AI agents analyzed time-series data from drones, sensors, and network elements to identify inefficiencies or service degradations.
  
  
• Autonomous Decision-Making – Upon detecting anomalies, agents initiated real-time corrective actions, such as rerouting traffic or triggering device reallocation.
  
  
• Sustainability Insights – Energy usage was continuously tracked and optimized, allowing systems to reduce consumption and carbon footprint in real time.
  
  
• Generative AI Interfaces – Operators could interact with the orchestration system using natural language, making it easier to deploy complex configurations and understand operational performance.
  
  

By integrating AI at multiple layers—monitoring, orchestration, and interaction—the Catalyst ensured that intelligence is not centralized, but distributed across the ecosystem.

Deep Dive: The Architecture and Partner Roles

What made Catalyst 805 truly powerful was the breadth and depth of collaboration among its partners. Each one brought essential capabilities to the table, creating a solution that was greater than the sum of its parts.  The functional architecture levered is provided following, with roles of each partner.

‍

Wavelo – The Orchestrator

Wavelo provided the event-driven orchestration platform that served as the system’s central nervous system. It transformed static APIs into asynchronous events, enabling real-time coordination of services such as edge computing, drone telemetry, and billing. Wavelo’s architecture supported usage-based charging by integrating live service consumption data with orchestration logic—unlocking 5G monetization through precision.

Wavelo’s technical solution played a foundational role in the Catalyst by delivering an event-driven architecture that enabled real-time, scalable orchestration across the Internet of Moving Things. At its core, Wavelo transformed traditional synchronous TMF Open APIs into asynchronous event streams, allowing the orchestration layer to react dynamically to events such as device mobility, mission progress, and network state changes. This event-driven platform decoupled service logic from static workflows, enabling flexible and responsive coordination of IoT services as they shifted across network zones and edge environments. Whether managing drone telemetry, orchestrating edge compute allocation, or automating service activation, Wavelo’s platform ensured low-latency, intent-based orchestration in highly dynamic settings.

Beyond orchestration, Wavelo also powered intelligent ordering flows and supported real-time billing capabilities. By integrating real-time network telemetry and service state updates into the orchestration logic, Wavelo’s platform could trigger usage-based billing events and enforce business rules at the moment services are consumed or altered. This enabled telecom operators to monetize 5G-enabled IoT use cases with precision, ensuring that billing reflected actual service utilization across a constantly changing network environment. Wavelo’s adherence to TM Forum standards further ensured compatibility with existing BSS/OSS stacks, while its modular, event-driven architecture positioned it as a strategic enabler for autonomous, responsive, and revenue-generating IoT ecosystems.

CGI – The Brain Behind the Bots

CGI delivered a sophisticated AI system composed of modular agents, each focused on a specific performance or sustainability goal. These agents ingested telemetry, evaluated operational objectives, and provided actionable recommendations. A generative AI chatbot acted as the user interface, enabling operators to interact with the system using natural language commands. CGI’s architecture made AI understandable, accessible, and autonomous.

As an incumbent partner from phase 1, CGI had collaborated with the catalyst champions to engage and to onboard new partners in this phase 2 catalyst. From a technical perspective, CGI worked with the other partners to build an end-to-end solution which demonstrated the moving IoT management and sustainability insights using AI capabilities; from a business perspective, CGI also worked closely with the team to articulate business use cases and values streams, paving the foundation to commercialize the solution in subsequent phases.

The CGI solution comprised a modular, multi-agent AI system built on AWS, designed to

monitor and optimize drone deployments for operational effectiveness and sustainability. Functionally, the application ingested time-series and static data (e.g., telemetry, drone specs, zone configurations) from a Kafka data stream and stored it in Amazon TimestreamDB and DynamoDB. Four AI agents operated in this architecture: the Coverage Monitor Agent evaluated aerial zone coverage; the Energy Sustainability Agent analyzed energy consumption against optimal baselines; the Service Order Agent formatted actionable recommendations (e.g., drone removal); and the Coordinator Agent interfaced with the operator through a chatbot to answer queries and orchestrate interactions among the other agents. 

Technically, the agents were implemented using Amazon Bedrock, enhanced with Retrieval-Augmented Generation (RAG) by indexing operational documents, SOPs, and historical logs stored in S3 and embedded via Amazon OpenSearch. Each agent could either run in scheduled mode (triggered by Amazon EventBridge to publish insights to DynamoDB for dashboard use) or in coordinated mode (triggered by the Coordinator Agent to provide just-in-time responses without persisting to DB). The architecture included Action Groups to allow LLMs to invoke Lambda functions or API endpoints for dynamic data retrieval. Feedback loops and post-deployment summarization allowed agents to critique or refine their insights over time, improving system intelligence. Future enhancements include exploring agent orchestration frameworks like Agent-Squad for scalable multi-agent reasoning.

OPT/NET – The Eye on the Network

OPT/NET’s platform delivered advanced, real-time anomaly detection by ingesting massive volumes of structured and unstructured IoT data. Their AI engine flagged performance irregularities before they caused disruptions and recommended remediation strategies. The platform is highly extensible, offering plug-and-play knowledge packs for various telco domains, making it ideal for smart city and industrial deployments.

OptOSS/MONITORED AI is a novel combination of products from OPT/NET BV, which delivers the market-first Digital Reality platform for IoMT applications in Telecom. This solution was powered with an advanced and awards winning Artificial Intelligence engine which detected anomalies in streaming data from thousands of sensors and moving robotic vehicles. This product was developed by OPT/NET over the last decade, and it is already used for real-time monitoring and analysis of geospatial and telecom related Big Data streams.

Any time-series data (structured and unstructured) could be streamed into the OptOSS/MONITORED AI engine, where irregular patterns were detected autonomously by AI and intelligent insights were produced automatically guiding and advising human operators in difficult and stressful environments. The OptOSS/MONITORED AI software served as a multifunctional and modular launch platform for the purpose-built innovative AI Knowledge Packs (AIKPs) covering many critical industry domains with one coherent solution. This novel approach to a well-known problem has been recognised by several prestigious industry awards.

This modular platform was provided by OPT/NET BV for the IoMT Catalyst for: (I) real-time monitoring and analysing of Big Data streams originating from thousands of mobile robotic vehicles enabled with the Telecom Networks and Edge Compute resources. (II) Any time-series data (structured and unstructured) were streamed into OptOSS/MONITORED AI, where irregular patterns were detected autonomously by AI first, and only then presented to humans and/or dispatched to the intelligent orchestration systems from our partners. (III) Unlike other solutions, the OptOSS/MONITORED AI tandem acted as a base plate of the multifunctional and versatile cross domain platform. It is organised along the TMF eTOM and NGOSS principles and is purposely developed for large scale Telecom carrier-grade applications enabling Smart Cities and Industry 4.0 use cases.

This solution can be expanded with purpose-built AI Knowledge Packs covering many telecom domains End-2-End with one distributed solution. This novel approach proved to be an effective solution for our customers and solves the hardest problems in the telecom industry. OPT/NET is a market leader in application of its advanced AI products in the Telecom sector of the Netherlands, with two out of three incumbent telecoms being publicly referable customers. At the moment OptOSS AI assures quality of digital services for the majority of the residents and businesses in the Netherlands.

Ulster University – The Test Lab

Ulster University provided the simulation environment used to test real-world behaviors under controlled conditions. Their drone simulator introduced environmental variables and stressors such as battery drain, signal loss, and trajectory violations. The team also built a service-level agreement (SLA) engine that monitored and adjusted system behavior to maintain performance targets. These components were critical for validating the solution under realistic operational constraints.

Contributions from Ulster University for the Catalyst solution included a drone simulator and a Service Level Agreement (SLA) co-ordinator, both hosted on AWS. The drone simulator was responsible for flying drones around the area of Belfast Harbour with the goal of monitoring operations on the ground and supporting recovery actions being taken. Simulated anomalies were integrated into the simulator to influence drone operation, which were ultimately detected by Catalyst partners once the relevant metadata was distributed. Anomalies included flying outside the acceptable perimeter of the port and unexpected situations of battery degradation. In response to metadata returned to Ulster University from Catalyst partners, drone flying decisions were manipulated in a manner which prioritised operational sustainability. This included returning a drone to base for recharging or replacement, or advising that the drone repositions.

The contribution from Ulster University also involved a SLA component, which ensured that the way in which drones and environmental sensors collected metadata was prioritised for performance objectives. Metadata collected includes the residual bandwidth availability across the ecosystem, in addition to residual memory stores. Operational reconfigurations could subsequently be made in response to local processing of the metadata to adapt the metadata collection rate to allow sufficient bandwidth and memory to remain so that the drone flight objectives and SLAs could be fulfilled in parallel.

BT Group – The Connectivity Enabler

As the continued Catalyst champion from Phase 1. BT Group bought extensive experience in network connectivity, ensuring reliable and high-performance IoT communications. They provided the necessary infrastructure and expertise to maintain continuous connectivity for moving IoT devices.

BT delivered the backbone network infrastructure, ensuring that autonomous devices maintained continuous, high-quality connections. Their experience with enterprise-grade IoT networks ensured reliability, low latency, and seamless integration across the Catalyst’s components. BT’s ongoing involvement across both Catalyst phases provided consistency and maturity to the project.

AWS – The Scalable Cloud Platform

AWS powered the entire backend infrastructure. Their services supported event streaming, AI training, data storage, and edge orchestration. The Catalyst leveraged AWS’s scalable, serverless infrastructure to ingest and process real-time telemetry, run AI models, and support high availability across all components. This partnership ensured that the solution is deployable at global scale with cloud-native resilience.

From Smart Ports to Smart Cities: Use Cases That Matter

While the flagship demonstration centered around a Smart Port scenario with autonomous drones, the underlying architecture is extensible to a wide variety of industries:

• Urban Mobility – Support for self-driving cars and autonomous public transit with dynamic route optimization.
  
  
• Logistics & Warehousing – Real-time orchestration of robotic fleets for package delivery and inventory management.
  
  
• Smart Cities – Integrated environmental monitoring, waste collection, and traffic control with energy-efficient edge processing.
  
  
• Manufacturing – Orchestration of automated assembly lines and predictive maintenance for connected industrial machines.
  
  

Each use case benefits from the same foundational features: real-time decision-making, intent-driven control, AI-assisted orchestration, and sustainable operations.

A Model for the Future of Telecom Innovation

Catalyst 805 was more than a successful demonstration—it was a strategic template for how telecoms can become platforms for ecosystem-level innovation. It showed how event-driven integration, AI, and collaboration could break down silos and enable new business models in IoT and 5G.

The solution offered a compelling answer and has the potential to have a transformational impact on the Industry.  The solution tackled the industry's growing need for advanced IoT management amid rising autonomous device demand.  It boosted sustainability and efficiency by enabling devices to autonomously optimize energy use, cutting costs and supporting global goals.  And, it enabled new business opportunities with a flexible, scalable IoT platform for monetization

It also exemplified what DTW25 – Ignite is all about: turning bold ideas into operational blueprints, building solutions that are not only visionary but also viable, and creating partnerships that push the industry forward.

As telcos evolve into digital service providers and platforms, Catalyst 805 offered a replicable path—composable, intelligent, and built for movement.

Thanks for joining us at DTW25 – Ignite to witness Catalyst 805 in action and discovering how the telecom industry is engineering a smarter, greener, and more responsive future.

‍