Search for a command to run...
The 5G-TOURS project aims at deploying full end-to-end trials involving real end-users (volunteers who consent to participate) and vertical operational services in three different European cities (Turin, Rennes, and Athens). In the 5G-TOURS “ecosystem” realised in the three cities, 13 use cases related with the themes of the touristic city (5 use cases), the safe city (4 use cases) and the mobility-efficient city (4 use cases) have been deployed. The ultimate goal of this approach was to trial the use cases in real environments to continuously collecting network, service and vertical KPIs and then to evaluate them against a set of predefined verticaloriented criteria. Towards this direction, WP7 - System integration and evaluation, focuses on delivering the integrated 5GTOURS ecosystem that would allow for the realisation of the pilots in all three sites and drive the evaluation of the results of the trials. This deliverable is the final document produced by WP7 and presents: a) the final version of the 5G-TOURS integrated ecosystem; b) the final validation results from the execution of the trials; c) the results from the application of 5G-TOURS QoS/QoE model. Regarding the 5G-TOURS ecosystem, a first version was introduced in the previous deliverable D7.2 [6]. In this deliverable, its final version of the ecosystem is presented, which includes: a) the site infrastructure, b) the 5G-TOURS platform and 5G-TOURS innovations, c) the hardware and software components of the verticals and, in general, d) all the required functionalities which were required for the smooth and successful execution and evaluation of the 5G-TOURS trials. This deliverable acts complementary to deliverables D4.4 [3], D5.4 [4] and D6.4 [5], in which the final sites infrastructure, the hardware and software components deployment and the preparation and execution of the trials were presented for each of the three sites respectively. Regarding the validation results, in D7.2 [6] a feasibility study was realised for each use case to identify which of the KPIs could be technically validated. For the KPIs that were characterised as feasible, a technical validation plan was generated. In this deliverable, the realisation of the validation plan is described for each use case, as well as the process of collecting the metrics, analysing the metrics, calculating the KPIs and, finally, validating the KPIs against the latest use case requirements presented in D2.3 [1]. In this deliverable, in addition to the selected KPIs for each use case, the details behind the validation process are presented including: a) the scenario details using a scenario template; b) the probe positions in the network; b) the probe positions in the protocol layers; c) the trial details (e.g., duration, sampling period, collection method); d) the methodology used during the analysis and validation and; e) any assumptions made during the collection/analysis/validation process. Regarding the 5G-TOURS QoS/QoE model, in the previous deliverable D7.2 [6], the final 5G-TOURS evaluation methodology, which covers QoS, QoE and vertical satisfaction aspects, was introduced. In this deliverable, the results from the application of such QoS/QoE methodology on the use cases are presented. By following the 5G-TOURS QoS/QoE methodology, the level of satisfaction of end-users and verticals’ players in the use cases were measured and evaluated. This evaluation included users' QoE as well as the feedback from the vertical players on how the technology provided can improve their business operations. The final 5G-TOURS evaluation methodology was followed in the trial execution of all use cases. Initially, during the actual trial execution, both the QoS metrics, automatically obtained from the infrastructure, and the QoE metrics (and vertical satisfaction), were collected using appropriate questionnaires, Then, all the collected metrics were analysed and the KPIs calculated and validated against the predefined targets. In addition, insights were provided in the case of successful validation results, while a justification in case of not fulfilled KPIs. Besides, in selected use cases, models for QoS/QoE correlation were created by using correlation-regression analysis