The essence of any smart city initiative is to improve efficiency, performance and speed of operations so it is possible to make fast, accurate, automatic decisions and ‘right first time’ problem-solving interventions in high priority situations, such as crime, emergencies, popular events, etc.
Smart cities also need to manage the expansion and evolution of their ‘smart’ infrastructure, not only for more efficient city operations but to also cater for more and more third parties that are requesting to install their own sensor networks.
These third parties need the flexibility to deploy their own intelligence in smart city infrastructure and push collected data to the cloud IoT platforms of their preference, e.g. Amazon Web Services, Microsoft Azure, etc., on top of or completely bypassing the city’s own data hub.
Applying edge and fog computing principles to smart city data hubs will also improve the efficiency of city operations through situational, decentralized decision-making and pushing intelligence from the central data hub to a local loop of programmable edge devices, e.g. IoT gateways and sensors.
Running on the edge has been proven to give faster performance, removes network latency of connecting to cloud and enhances the quality of data collected by carrying out analytics locally where the raw data is collected. All this helps to proactively handle incidents/situations as they unfold.
Members of the TM Forum community have joined together to help formulate the best practices to achieve this in a project aptly named the Smart City on the Edge Catalyst. It is being led by the city of Milton Keynes, Dublin City and Agile Fractal Grid as champions and Cloudsoft, Huawei, BP/Infonova, BT and Exfo as active participants.
The Catalyst will focus on several scenarios including, but not limited to:
- Deploying logic on edge gateways that enable local actuation in reaction to a stimulus e.g. when unusual, potentially suspicious activity is sensed in dark areas, deploy an app on a gateway with logic to switch on street lights;
- Deploying on-demand special purpose apps and analytical capabilities on edge devices, enabling collection of more detailed data as the situation unfolds, e.g. deploy a car plate recognition app on a gateway to help police locate a suspect’s car through cameras in a car park.
- Orchestrating high speed network links to communicate back to a control room for better quality data about the situation, such as high-resolution video streams, CCTV, etc.
The main objectives will be to:
- Increase the speed of decision-making and actions to handle incidents or prevent situations developing in the city;
- Release council human resources from handling situations that may now be handled by apps enabling situational actuation on the edge;
- Effectively expand and evolve a flexible smart city infrastructure;
- Flexibly deploy sensor networks in the city to collect data of interest, analyze data fast in real-time on the edge and feed collected data to the IoT platform of preference;
- Ensure data is secure.
This in turn opens new charging models for city services such as:
- Context-based charging for applications deployed on the edge at locations where more action is observed/sensed e.g. increased presence of people or traffic.
- Charging for apps selected from an application catalog for deployment on the IoT gateways (as-a-service).
- Charging more for deploying a catalogued application to the edge than deploying it to the cloud.
- Charging for committing premium computational resources on gateways to run third-party apps as priority.
The Smart City on the Edge Catalyst will, no doubt, generate a large amount of interest and will be demonstrated at TM Forum Live! In Nice next month.
Watch this video for more information:
If you’d like to discuss the Catalyst further, contact Nektarios Georgalas, BT, directly.