NovaGenesis – Inatel (BR)
NovaGenesis means “new beginning” in Latin, and it is a clean slate architecture aimed at integrating networking, cloud computing, Internet of Things, service-oriented architectures and self-organizing technologies. Two demands are still open:
- (i) how to run NovaGenesis over ZigBee?
- (ii) what is the performance of the solution in a bigger scale?
This is exactly the aim of this project: to investigate NovaGenesis over ZigBee in scale using FUTEBOL facilities. Inatel will use FUTEBOL resources to extend the support of the NovaGenesis tool to Arduino nodes running XBee. To that end, they will exploit the IoT resources available at the UFRGS testbed, together with resources on the cloud, which will require connectivity from UFRGS to the Inatel ICT labs.
AAHCCaVI: Adaptation Aware Hybrid Client-Cache; a smart pre-fetching scheme for DASH for Regular Videos and Immersive Contents – University of Missouri-Kansas City (USA)
Although the Internet gets faster, with the increasing trend of video streaming traffic, users usually face difficulties to watch their favorite streams. A couple of solutions are applied by content providers are caching and pre-fetching. UMKC has previously proposed the AAHCC framework that predicts the pre-fetch bit-rate using an adaptation algorithm. The algorithm uses the throughput from the client and the forecasted throughput at the cache. In previous work, UMKC implemented their idea on the wired GENI testbed, showing that AAHCC improves the cache hit- rate and consequently Quality-of-Experience (QoE).
Using FUTEBOL, the experimenters will extend AAHCC and implement AAHCCaVI, which will perform pre-fetching for both regular videos and immersive contents. This experiment will leverage FUTEBOL in order to test the proposal over a range of networks, including WiFi, 4G and 5G, with content servers at different distances from the final users. Their objective is to use the inter-testbed connectivity of FUTEBOL to create experimental slices containing nodes in both Brazil and in Europe. These slices will be used to evaluate the proposed solution with the cache and clients located in different relative distances.
Testing Adaptive Scaling for Evolved Packet Core in the FUTEBOL Research Infrastructure – Universidad del Cauca (CO)
The 5G mobile networks will offer a wide range of applications for people and things. However, these applications impose new requirements to Network Services (NSs) regarding high throughput and low latency. Network Functions Virtualization (NFV) provides to network services the scaling capability aiming at ensuring performance targets by, for instance, allocating dynamically resources (e.g., processing, networking, and storage).
Since network services are formed by Network Functions (NFs), scaling can be performed by increasing or reducing resources to such functions (i.e., scaling up/down) or creating and removing instances of those network functions (i.e., scaling out/in). This proposal will conduct tests of a novel scaling approach for 5G network services, in which an NFV-based Evolved Packet Core (EPC) can scale adaptively by Reinforcement Learning (RL) agents.
Open-source EPC components will be deployed in the FUTEBOL testbeds using containers, and those can be scaled using the existing control framework tools, such as COPA, ABNO and O2CMF. The testbed that is supporting this experiment is the Iris testbed. Iris -the reconfigurable radio testbed at Trinity College Dublin- provides virtualized radio hardware to support the experimental investigation of the interplay between radio capabilities and future networks. In the previous figure, UE (User Equipment) and eNodeB (base station) are implemented as software defined radios (SDR), particularly the srsLTE library is used. MME (Mobility Management Entity), HSS (Home Subscriber Server) and S/P-GW (Serving and Packet Gateways) are part of the Evolved Packet Core (EPC), which is the core of the system. And finally, PDN is the external packet data network. The blue discontinuous line represents signaling messages and the red continuous line represents data flows.
Evaluating Wireless Body Area Network MAC Protocols Power Consumption – Universidade Federal de Viçosa (UFV)
Currently, wireless body area networks (WBANs) are widely used in medical environment in order to monitor and process vital signs of patients and then send this data for analysis. This research proposes a power consumption evaluation of several WBAN MAC protocols. The ZigBee protocol stack developed by Schmid et al., 2016 is used as a starting point for the implementation our protocols, and later it will be compared with DCAA-MAC protocol [Lee et al. 2011]. The evaluation will be based on the results obtained from simulations in real devices through the use of testbeds and the GNURadio platform.
Experiments in Cognitive Radios – Universidade Federal de Lavras (BR)
The intensive use of ISM (Industrial Scientific and Medical) free frequencies in wireless device communication has caused spectrum pollution. Several solutions have been proposed for the optimization of the spectrum, among them the cognitive radios (CR). The opportunistic use of primary frequencies can improve spectrum usage, bandwidth, and latency. This project proposes the development and adaptation of an architecture for cognitive radios. A proof of concept of this architecture will be made using software-defined radios, in which tests and experiments will be conducted at FUTEBOL testbeds. Several of the FUTEBOL testbeds have software-defined radios: TCD, UFMG, UFRGS, UFES.
Selected Experiments Leaflet: Download