Watch the video of the available demos on 5G TRANSFORMER YouTube Channel clicking here or clicking directly on the title of the demo interest.
In a factory Cloud Robotics service robots and production processes are remotely monitored and controlled in the cloud, exploiting wireless connectivity (LTE/5G) to minimize infrastructure, optimize processes, and implement lean manufacturing.
In the presented video, a mobile robot uses indoor navigation algorithms to shuttle materials between work cells in a plant. Robotic arms are used to load and unload goods from the mobile robot. An automated warehouse is simulated by a rotating platform, and an automated door is placed along the navigation tracks to show a flexible and optimized shuttling of materials between the work cells. The entire sequence is monitored and controlled by a remote server through radio communication using the Ericsson Xhaul optical network infrastructure designed in the 5G Crosshaul project.
The objective of the demonstrator is to verify that the system is able to allocate the suitable resources based on the specific service request to allow the interaction and coordination of multiple (fixed and mobile) robots controlled by remote distributed services, satisfying tight latency requirements.
5TONIC LAB internal demonstration to PSA group:
The purpose of this demo is to show the value chain generated by an Edge Robotics vertical providing its services to a Manufacturer vertical. This is accomplished using a slice of the network provided by the operator that includes Edge devices. By means of virtualization technologies, the Operator provides networking and computing capabilities at the edge of the network and everything is remote and secure.
We had a joint booth with 5G-CORAL and 5G-EX projects, and following demos were shown from 5G-TRANSFORMER project side:
The introduction of Internet and Cloud technologies had a great impact in the last few years over the technological solutions used to provide Media and Entertainment services in sport venues. Nowadays almost every video streaming service relies on these technologies, but the growing demand of media rich content and the low latency requirements still represent a technological challenge. In our demonstration, we show how it is possible to encompass a media streaming service with the current trends in virtualization technologies and provide an on-demand high-definition streaming service.
This demonstration shows media service deployment on the top of heterogeneous infrastructure combined from public and private clouds, While high definition streaming service with very low latency is deployed, the network underneath is abstracted, inter-cloud connectivity and all required configurations are provisioned automatically. To this end, the streaming service provider can easily make requests that are mapped to network services with different constraints, which are instantiated over arbitrary infrastructure, which might be combined of both public and/or private clouds. In this case, the virtual appliances that contain a global repository of videos to be accessed from any venue are hosted at the public cloud,. The virtual appliances to be accessed by the users which are geographically close to the venue are hosted at the edge private cloud. This smart placement and the logic of the VAs allow to stream high definition videos to the users in the venue avoiding the generation of bottlenecks in the network. We show this by requesting different high definition video streams, with the use of a laptop connected to the network of the venue, and analyzing the data rates in different parts of the network. The latency perceived by the user in the laptop is also analyzed.
The demonstration also works as a Proof of Concept (PoC) for two technological challenges. First, provides a PoC of the seamless integration of the video streaming service with orchestrator, based on the open source Cloudify platform (which is the main component of the Service Orchestrator in the 5G-TRANSFORMER architecture). The second PoC is about seamless interconnection of the private and public cloud using a mix of the custom and community Cloudify plugins.
For the last years, the entertainment industry has been working on improving fan engagement solutions on sports venues, and this is especially true for the services related to Ultra-high definition streaming. In the past, all the entertainment services in the venues relied on a physical infrastructure to provide the service. This situation changed in the last few years with the embracement of Cloud and Internet technologies, and nowadays almost all the video streaming services rely on these technologies. The increasing demand of low latency and high definition services however still represents a technological challenge that is far from being solved. Furthermore, it usually takes several days or weeks to deploy ultra-high definition service in one venue, and it usually demands a great expertise in the network infrastructure supporting the service. In our demonstration, we show how it is possible to provision in just some minutes an ultra-high definition streaming service with very strict latency constraints, and without requiring any knowledge of the network underneath.
Our demonstration shows how a streaming service provider can easily request the deployment of the service for a different number of users. These requests are in turn transparently translated to services with different resource requirements which are the ones finally instantiated. The logic behind the resource allocation is smart enough to place virtual appliances (VAs) containing caches of the video streams close to the final users, and at the same time place all the virtual appliances (VAs) which contain the global repository of data (i.e. global repository of videos to be accessed from any venue) somewhere else in the network (in order not to compete for the scarce resources in the edge of the network). With this smart placement and the logic of the VAs themselves, we show how it is feasible to stream high definition videos to the users in the venue without generating bottlenecks in the network. We do this by requesting different high-definition video streams using a laptop connected to the network of the venue and analyzing the data rates in different parts of the network. We also analyze the latency as perceived by the user in the laptop in order to prove that we also fulfill this requirement.
In terms of technological challenges, this demonstration provides two different Proofs of Concept (PoCs): the 5G-TRANSFORMER Vertical Slicer and the seamless integration of the video streaming service with the 5G-TRANSFORMER virtualized environment. In particular, we demonstrate how the media provider can request a service for a Content Delivery Network (CDN) providing different service-level constraints, which are automatically translated in NFV network services dynamically instantiated in the 5G-TRANSFORMER virtual environment. A key feature is the cooperation between the 5G-TRANSFORMER Vertical Slicer and Service Orchestrator. The slicer handles the service business logic and builds network slices customized to meet the application requirements, while the orchestrator manages the NFV network services implementing the slices with the objective of optimizing the resource allocation across the cloud, MEC and transport domains. The result is an operational streaming service dynamically adapted to the business requirements and instantiated on-demand in a few minutes without any overhead for the media provider.
3) Robotic Control Leveraging a Radio Network Information Service (RNIS)
We are demonstrating an LTE network where a robot equipped with an LTE interface is attached to an OpenAirInterface eNodeB and EPC. The RNIS is exposed and applications can consume it over a REST API to retrieve, among others, run-time Channel Quality Indications (CQI) per UE. The RNIS collects this information from agents built into eNodeBs using the FlexRAN southbound protocol. The purpose of this demonstration is to show how applications can make the best of the RAN-level awareness offered by the RNIS for real-time adaptations, particularly focusing on cloud robotics scenarios. The following application scenarios will be shown:
- Channel quality visualization and simple robot control (Required): A robot connected to an LTE cell is slowly moving within its coverage, while a control application hosted remotely (e.g., at the mobile edge) is receiving CQI information for the UE attached to the robot by accessing the RNIS API. In the meantime, the application plots in near real time this CQI information. As soon as the CQI drops below a predefined threshold, the application remotely instructs the robot to invert its direction towards better- covered areas of the cell.
- Adaptive media streaming for remote surveillance (optional): A mobile robot is hosting a streaming server, which captures and transmits live video over LTE from a camera mounted on it. A remote application is monitoring the channel quality at the robot end using the RNIS. As soon as it detects that signal quality drops (i.e., when the robot moves away from the eNodeB), the application remotely commands the streaming server on-boarded to the robot to switch to lower quality (and thus bitrate) video to match the signal conditions.
IEEE INFOCOM 2018
1) OVNES: Demonstrating 5G Network Slicing Overbooking on Real Deployments
Network slicing allows mobile operators to offer, via proper abstractions, mobile infrastructure resources (radio, networking, computing) to vertical sectors traditionally alien to the telco industry (e.g., automotive, health, construction). Owning to similar business nature, in this service orchestration technology we adopt revenue management models successful in other industries (e.g. airlines, hotels, etc.) and so we explore the concept of end-to-end slice overbooking to maximize the revenue of mobile operators. The main innovation of our platform is threefold. First, we design a hierarchical control plane to manage the orchestration of end-to-end-slices. Second, we cast the orchestration problem as a stochastic yield management problem and propose two algorithms to solve it: an optimal Benders decomposition method and a suboptimal heuristic that expedites solutions. Third, we implement a human-friendly slice manager to introduce slice requests and a monitoring dashboard collecting data from all domains of a mobile system. Our demonstration consists of two slice-capable LTE base stations, an Ethernet-based backhaul network and two distributed computing clouds: a centralized pool of servers emulating a datacenter and a commodity server located at the edge. Our results show that slice overbooking can provide up to 3x revenue gains in many realistic scenarios, as compared to employing no overbooking schemes.
MWC 2018 (Mobile World Congress) :
1) Heterogeneous network component of the MTP
The goal of this demo is to illustrate the capabilities to orchestrate end-to-end the network resources in a multi-domain, multi-technology transport network (featuring optical, mmWave, and WiFi links) able to handle both fronthaul and backhaul traffic in an integrated way. This is one of the pillars of the 5G-TRANSFORMER concept, which aims to transform today’s rigid mobile transport networks into a dynamic SDN/NFV-based Mobile Transport and Computing Platform (MTP) covering the specific needs of vertical industries. As such, it benefits from the work carried out in 5G-CROSSHAUL (http://5g-crosshaul.eu/).
FUTURE ACCEPTED DEMOS:
1) Experimental Demonstration of a 5G Network Slice Deployment through the 5G-Transformer Architecture
1) Overbooking Network Slices End-to-End: Implementation and Demonstration