Use Cases in Living Labs
The Living Labs in XGain are defined as a human-centric, open innovation ecosystem of relevant stakeholders, focusing on user support, services provision and wide adaption. Through LLs, researchers/innovators can observe and understand user behaviour patterns, even those that are not immediately obvious.
XGain is aiming to demonstrate the developed Knowledge Facilitation Tool and each application results in series of heterogeneous use cases in terms of location, connectivity needs, local needs, edge and connectivity potential solutions and operational business models. The initial business models provided for the use cases will also be evaluated as per the project methodology in parallel with the assessment of the proposed ecosystem of technologies.
Level of Assessment: Rural Community
Services: a) Drones operation in rural areas, and b) Additional services with high data rate
Leader: I2CAT
A set of drones will be deployed at the pilot site location, connected via 5G to the operations centre that will host various services such monitoring (i.e., live video streaming clients) assisted by video/image processing tasks. High bandwidth, ultra-low latency and reliable 5G communications will be exploited to serve the fleet of 5G connected drone operations. The drone centre can be either indoors or outdoors. To enable the envisioned use case, an end-to-end 5G network will be deployed (5G UE, i.e., drone, 5G base station operating at 3.5 GHz and potentially at 700 MHz, and a virtualized 5G Core) and edge-computing capabilities.
Level of Assessment: Region
Services: a) eHealth robots for elderly, b) High data rate services, and c) Tourism
Leader: CAP
The scenario aims at providing health and wellbeing guidance and interventions for people living in rural areas. 5G will be exploited to deliver natural language processing (NLP) services and for real time interactions/interventions with/to the respective individual. CAP will support the social interaction of people in rural areas by providing means of interaction through the Robot device like videoconferencing, social gaming, and interaction with other users exploiting the full potential of the enhanced mobile broadband service and low latency transmissions of 5G. The usage of existing infrastructure, especially at the touristic parts of the region will be investigated as part of the needed network expansion. In parallel, services requiring high data rate like e-Learning, as well as backhaul connectivity options will be explored and potential tested in lab environment.
Level of Assessment: Community/Island
Service: a) Precision Agriculture and b) Tourism
Leader: CAFA
A series of sensors will be deployed in the farm, forming the input data sources for a digital twin that can be used for the optimisation of production such as optimised water management, and monitoring of the production quality. For the scenario demonstration soil, water, meteorological sensors, and cameras to monitor the field may be used. All sensors will be connected to an edge device where pre-processing of the collected data will take place. The connectivity between the edge and the core network will be expanded by the usage of aerial radio relay balloons ensuring the connectivity till the next connected area of the island.
Level of Assessment: Community/Region
Service: a) Precision Agriculture, and b) Forest Management
Leader: ART
Depending on the strength of the 5G connection (e.g., based on the distance from the cellular station), two main scenarios are considered: direct transmission (when in proximity to the cell centre) or partially direct transmission (when the end device resides at the cell edge). Provided a stable cellular (5G) connection, the collected data could be transmitted to the server (cloud) directly from the drone after the data is pre-processed using on-board edge technologies. In the second case, UAV collects data in fields, forests, or other environmental locations and then returns to the original base (receiving station). Data can be pre-processed (using edge technologies) and then sent to the server (cloud) after the 5G signal strength is improved. It is estimated that at around 40 Mbps data rate should be enough for efficient data transmission directly from the drone. In that case, depending on the server-side algorithm’s performance, the user should be able to get the results and plan his actions after a timeframe of 5 to 10 minutes. The planned edge equipment is suitable for complex nonlinear multidimensional optimization and data analysis models, such as various neural networks, enabling on-situ heavy computation tasks that would provide efficient transmission policies (tailored to the use case needs), e.g., by reducing the volume of data for transmission.
Level of Assessment: Farm/Community
Service: a) Water quality monitoring, and b) Remote oyster farming
Leader: BENCO
The use case will demonstrate how an extended network infrastructure (supplemented by LoRaWAN and satellite comms) can be used to transmit data from offshore sensors to the central monitoring system. Particularly the sensors are connected to the receiver (stations floating on water, float/buoy aid) via underwater cables. Multiple nearby stations can be connected to a LoRaWAN network, where an edge device aggregates data, pre-process it and transmits (either raw or processed) data to other existing platforms and monitoring services. To retrieve data from the edge device the satellite IoT network of Astrocast SA will be employed.
Level of Assessment: Farm/Community
Service: a) Livestock health and b) Farm management
Leader: EVILVO
The use case will have two distinct scenarios showcasing the application for farms with strong network connection (e.g., with fixed cameras on the field) but also cases of poor network conditions where a remote (drone) shepherd will be deployed for data acquisition and monitoring. The aim is to devise a service for the livestock farming industry that can be tailored and adapted to diverse farming ecosystems, that irrespective of the technology and means of connectivity used, provides the same output for the farmer (e.g., modelling accuracy of livestock behaviour, accuracy of analytics, etc.). Particularly, for farm cases where 5G network coverage exist in-situ, direct voluminous data streams (of video or other data) can be analysed, to provide data analytics services tailored to the respective application. On the other hand, in areas with absent or poor network conditions, digital shepherd patrols can compensate by aggregating data from remote site locations, and either process the input data locally (as a far-edge computing service), or offload data when within range of an access point, e.g., 5G gateway, for further processing and analytics at an external server. In more detail, cameras at the border of the field could be used to monitor the health of the sheep/cows at every moment, able to adapt its field-of-view based on perceived conditions by following the animals in the field. A “drone-in-a-box” solution could further automate this monitoring workflow by automatic take-off and allow for a mobile camera solution instead of fixed cameras. The focus will be on counting the animals, registering any isolated animals and detecting abnormal behaviour in their daily patterns, for example by analysing their drinking habits. Other indicator may include the detection of other characteristics of animals within the herd such as lame animals, sick animals, animals that are calving, or are lost, but also lack of drinking water, disturbances of the herd by predators, etc. The drone station will be equipped with XGain technology, and given the available network capacity (e.g., bandwidth) the envisioned service will generate alerts, initiate live video streams with analytics indicators, etc., enabling both limited processing capabilities on the edge, but target application that require full cloud support for data intensive analytics services.