SliceNet’s smart city pilot
Marius Iordache
Orange Romania
Christian Patachia
Orange Romania
The new 5G network system promises numerous advantages for a large number of vertical sector applications. Horizon 2020 project SliceNet has designed, implemented and executed a smart city pilot to showcase the advantages of network slicing, a new concept that has been introduced in the 5G system. Network slicing enables the operation of independent logical networks over the same physical network infrastructure.
Smart city applications include, among others, metering solutions for gas energy and water consumption, remote monitoring of city infrastructure like pollution, temperature, humidity and noise, real-time traffic information and control, city or building lights management and public safety alerts for improved emergency response times. The smart city pilot in SliceNet has defined a 5G network slicing architecture for the vertical’s smart lighting application as a solution for one of the many smart city applications.
Architecture of the pilot
The smart lighting pilot is based on an end-to-end sliced 5G architecture, vertical-oriented, technically supported through innovative network resources management, control and orchestration, enabled to support the service requirements by providing Quality of Service (QoS) and Quality of Experience (QoE) service assurance through cognitive machine learning algorithms.
The pilot provides the transition from existing concepts of Internet of Things (IoT) implementations, such as those based on the low-power wide-area network protocol LoRa (Long Range) to standards based LTE-M/Nb-IoT and further to the new 5G network system for smart city use cases.
The pilot system is based on different ICT components and capabilities such as IoT applications and services, virtualised network infrastructure, virtualised network functions implementation, dynamic resources, slices and services orchestration, concurrent slices resources allocation, cognitive modules for QoS/QoE assurance and interfaces facilitating the interaction with the vertical application.
Using existing 4G network components, the pilot provides the possibility to instantiate end-to-end slices stretching from the users’ equipment, i.e. lighting poles, up to the IoT application running in a data centre, including all necessary network components that support the use case.
The key achievements, performance and business indicators (KPIs) are demonstrated in a real environment, integrating the technologies and demonstrating the capabilities and key innovative aspects of the project implementation, with a focus towards commercialization. The key aspects addressed by the smart lighting pilot are:
- Adopted model of 5G stakeholder roles and responsibilities, with the vertical stakeholder in the loop
- Time-to-market for business solution delivery
- Customer friendly usage
- Innovative programmable infrastructure deployment, resource, slice and services monitoring and exposure of collected data to the vertical application
- Service QoS/QoE KPIs assurance, metrics (bandwidth, delay, jitter, packet loss)
- Service provision in a massive Machine Type Communication (mMTC) slice
- New operational model for deployment, operation and system service assurance
Deployment and measurements
The smart lighting pilot has been deployed and demonstrated in a real live scenario in Bucharest, Romania. A series of measurements have been conducted to obtain results about the potential capabilities of the system. The measured end-to-end latency is in the range of 27ms on average and well below the target key performance indicator for this application. The measured packet loss observed was 0{b28ae05319d94bff0b4d65c5a9f4524dd588360f05c61ef440e1608e0a1c4144}. Longer-term measurements under a larger variety of configurations have to be conducted, in order to add more realism to the estimation of service reliability.
Impacts
The pilot has important societal and business impacts. The system integration and use case piloting in a real environment provided a basis for wide communication of the potential through Orange’s Romania ties into industry, academic institutes, media and last but not least the city hall. The Orange Fab programme provides the co-innovation environment to channel the results to start-ups and other partners.
Commercialisation opportunities emerge for the whole service, i.e. the smart lighting application or parts thereof such as the programmable infrastructure the service is based upon, the vertical application orchestrator, or the vertical API.
A video summarises the smart lighting application (https://youtu.be/zmxJbxpSTYQ).
Collaboration in the 5G PPP
The pilot provided a unique 5G PPP collaboration opportunity between the projects SliceNet and Matilda. Together the two projects provided a holistic, innovative framework for the design, development, life-cycle-management and orchestration of 5G-ready applications and the related network services. The whole system is based on a programmable infrastructure, virtual network functions and network services, the 5G-ready applications and application components all available through a 5G marketplace. The collaboration showcased how this innovative smart lighting application is deployed from the 5G marketplace into the infrastructure.
Standardisation
It has been clear from the start that the solution must be standards based. Where standards do not exist or are not mature enough yet, the project actively engaged with the relevant bodies and contributed in shaping standards as follows:
- ETSI Work Item for “PoC proposal for Predictive Fault management of E2E Multi-domain Network Slices” (using a smart grid use case as an example);
- ITU-T Work Item for “Vertical-assisted Network Slicing Based on a Cognitive Framework” (using an eHealth use case as an example);
- ITU-T Work Item for “Anomaly prediction and integration for eHealth use case based on vertical feedback” (using an eHealth use case as an example);
- ITU-T Work Item for “Noisy neighbour detection and integration in a virtualized infrastructure” (using a smart city use case as an example);
- ITU-T Work Item for “Machine learning based end-to-end network orchestration and network slice management” (across use cases).
Outlook
Smart cities will make significant use of 5G capabilities to improve their services for citizens. SliceNet piloted and validated business opportunities relying on 5G features like network slicing, flexible network management and provision of mMTC and enhanced Mobile Broadband (eMBB) services at scale. It implemented a closed-loop cognition-based autonomous network slice control, management and orchestration framework.
Further information
SliceNet project website – https://slicenet.eu