End-to-end Network Lifecycle Automation in Telecom Networks

5G, IoT, and edge cloud offer huge potential for operators to increase their revenue multifold, by transforming the way they offer services to their customers. At the same time, there are concerns about costs related to new spectrum, new radio, 5G core and control plane elements, increased capacity and low-latency transport networks, virtualization/containerization, and slicing and orchestration requirements. Many top tier-1 operators have already taken up network optimization and end-to-end network lifecycle automation to optimize these costs.

Industry Expectations on End-to-End Network Automation

Many top operators have started their first steps in unified IT and network operations. In our view, the automation has to be seamless, modular, and end-to-end for the entire network lifecycle, spanning planning, engineering, deployment/rollout/transformation, operations, and optimization. Telecom operators are looking for cost optimization, less manual intervention, improved quality of service, scalability, and enhanced security. In order to cater all these requirements, capabilities like end-to-end network automation, robotic operations, and zero-touch provisioning must be deployed along with a modernized network.

A modernized network and accurate inventory management systems are basic building blocks to enable E2E network automation. Network modernization can be driven by technology transformation initiatives like NFV/SDN, disaggregated networks, segment routing, and 5G deployment.

Integrated End-to-End Network Automation Architecture

End-to-end network automation requires an integrated automation architecture (see Figure 2) that leverages an analytical data lake provisioned with real-time data, historical/snapshot data, fixed/static data, planned information data, and predicted/projected data sourced from the various network and IT data sources. This data can them be analyzed using cognitive analytics and AI/ML techniques, performing the following end-to-end network lifecycle automation functions:

1. AI-based automated network planning: AI-based network planning addresses the automation of the planning process involving integrated RAN, front-haul, mid-haul, backhaul transport, and core network planning covering the physical, virtual, and containerized network functions as well as multiple layers of the network utilizing AI/ML techniques. The planning should utilize the current processes of the operator, as well as all relevant inputs for such integrated planning, including crowd sourcing data inputs. Cognitive analytics and AI/ML techniques can then produce quarterly/annual network plans to roll out.
2. Flow-through network build: The plans created above seamlessly flow into an automated network build management system that can create a project, then start, run, and track until its closure. The system is an umbrella tool covering all internal actions including the various planning, engineering, testing, operations, procurement, sales, and finance teams of the operator and all external stakeholders (tower companies, OEMs, systems integrators, field partners, data center partners, MS partners, cloud partners, transportation and logistics partners, etc.) The system also produces contextual intuitive reports and incorporates role-based access, integrating the mobile application. The automation also provides an option to chat with/speak to the stakeholder responsible for the task delay/failures.
3. Automated network configuration/build/turn-up and automated network testing/validation. The end-to-end build automation covers device onboarding as part of the network build process including xNF onboarding, xNF configurations, slice provisioning, MACD, and required onboarding into OSS and inventory updates. The system we implement includes automated testing/validation of the built/transformed network using Wipro’s Network Testing Automation Framework (NTAF), leveraging Wipro IP and accelerators like NetAssure and VeVaTo (a verification and validation tool). With decades of experience in the network testing and device accreditation areas of telecom networks, Wipro has built 4,000+ test cases across the device, RAN, fixed access, IP/optical transport networks, mobile/fixed core networks, hybrid cloud, data center use cases, and more. A significant portion of those are ready for regression test automation using Wipro’s NTAF.
4. AIOps (predictive & preventive operations): Once a network build phase is over, the operations part of the solution incorporates significant automation in use cases such as: 
   • AI-based cross-domain correlation of events: Automated alarm/event inputs to the correlation engine from the analytical data lake or fault management system(s) can be correlated using AI algorithms. Raising tickets automatically for the actual event only.
   • Self-healing: An automation engine can orchestrate and execute relevant procedures automatically for each of the ticket journeys or scenarios, updating the ticketing system with relevant inputs, and also raising and invoking the right-priority tickets in the field ticketing tool for tickets that will require field intervention.
     
   • Zero-touch provisioning and bulk configuration: OSS and/or service/domain orchestration systems across the network abstraction layer boundaries enable automated provisioning of the network and services as well as bulk parameter configurations, and also enable configuration audits for automated and secure network management.
     
   • Self-optimizing networks: RAN, transport, and core optimizing systems and related automation logic enable prioritizing network availability for the critical traffic during network incidents.
     
   • Cognitive field operations: Smart field operations may apply to service delivery and/or service/network assurance.The automation and intelligence dispatches the right engineer to the site based on AI-computed parameters like proximity of the current location of the engineer to the fault location, earlier experiences doing a similar task, and spares availability. Operators can also provide field engineer with a mobile app, AR/VR technologies, and required technical support (such as relevant and contextual documentation getting pushed to the engineer), allowing collaboration with the NOC engineers for advanced technical support.
   • Integrated inventory management: Smart integrated inventory is a basic building block for any sustainable automation initiative.
     
5. Unified network and IT operations driving all possible synergies: 

   Our network automation solution also leverages all possible synergies including processes, tools, and automation across IT and network lifecycle activities.
   

6. Continuous monitoring and improvement of automation through agile delivery processes and tools: Using a real-time contextual console, the solution monitors and documents the impacts of the implemented automation releases as well as feedback on the feasibility for further automation that can be executed through agile sprint.

Conclusion

Automation not only improves efficiency in end-to-end processes for new 5G networks but also saves significant costs. Our automation solution addresses the end-to-end network lifecycle instead of just targeting the operations piece. Our approach also introduces novel synergies to how operators operate their IT and networks, which is one of the key areas to address in this new era where the boundaries between IT and networks are shrinking and blurring.

About the Authors

Ajit Garg
Senior Partner, Wipro Consulting

Ajit is a Core Telecom professional with more than 30+ years of experience. He has a proven track record across various leading multinational companies in the network and sourcing domains. Ajit brings expertise in network managed services, network transformation, green field network launch, project roll-out, RFP processes, and partner performance management. 

Amol Sudhakar Telavane
Consulting Partner, Wipro Consulting

Amol has more than 22 years of experience in telecom networks. He is focused on creating network solutions for telcos around cutting-edge network technologies such as 5G, orchestration, virtualization, IP and Optical Transport Networks, cognitive analytics-based hyper-automation, and network optimization.