xG Cell Insight
xG Cell Insight, is a fully flexible 4G-LTE radio access monitoring system, utilizing low-cost SDR hardware and software code running on general purpose computing equipment. It is based on the SDR (SOFTWARE DEFINED RADIO) TECHNOLOGY, which enables the development of end-to-end telecommunications platforms (i.e. comprising both the S/W application and the radio transmitting and receiving system), built entirely on S/W and general purpose H/W. At the core of the system lies:
- a generic purpose SDR, such as Ettus Research B210, which captures the air-interface, either uplink or downlink;
- a software code implementing full analysis of 4G L1 and extracts various cell-level and user-level information, such as the number of active users, the utilization of time-frequency resources, the configuration profiles, etc.
Use cases: xG Cell Insight scope is quite broad. It was primarily designed as a cost-efficient solution to monitor 4G mobile networks, especially at challenging conditions (e.g. crowded events). Besides that, it could be also used by academia for education purposes and by research groups as a means to accelerate their research.
Current Status: A product prototype has been recently tested and validated, within the context of a large European R&D project (http://www.flex-project.eu/). In particular, three different private LTE networks have been tested, sopen-source OpenAirInterface Software, another one based on a commercial small-cell provided by ipAccess and Sirran, and thirdly another SDR-based solution provided by AMARISOFT eNB/EPC software. Local live 4G cells have been also decoded. In FLEX-D, the probe functionalities have been enhanced in order to use it as a tool for identifying non-occupied LTE resources, which could be potentially used for underlying D2D communications within a network supporting both cellular and direct access. Two approaches for blindly detecting used/unused LTE resources have been applied. The first is based on a “protocol-level” LTE DL L1 signalling sniffer which recovers UL scheduling information from broadcast L1 messages sent by the LTE eNB, whereas the second is based on a “signal-level” LTE UL filter-bank based sensor/energy-detector, which operates completely blindly. Indicative results for both the “protocol”-based and “signal-processing”-based probe components are shown in associated figures:
Indicative results for LTE DL L1 signalling sniffer experiments
Indicative results for LTE Filterbank-based UL Sensor/Detector experiments
lte-sidelink is an open software library developed in MATLAB by Feron Technologies P.C., that implements the most important functionalities of the 3GPP LTE sidelink interface.
Sidelink is a new LTE feature introduced in 3GPP Release 12 aiming at enabling device-to-device (D2D) communications within legacy cellular-based LTE radio access networks. Sidelink has been enriched in Releases 13 and 14 with various features. D2D is applicable to public safety and commercial communication use-cases, and recently (Rel.14) to vehicle-to-vehicle (V2V) scenarios. In legacy uplink/downlink, two UEs communicate through the Uu interface and data are always traversing the LTE eNB. Differently, sidelink enables the direct communication between proximal UEs using the newly defined PC5 interface, and data does not need to traverse the eNB. Services provided in this way are often called “Proximity Services” (or ProSe) and the UEs suppporting this feature “ProSe”-enabled UEs.
The library provides an (almost) complete implementation of the sidelink physical signals, physical channels and transport layer functionalities described in the 3GPP standard. In addition it provides the neccessary receiver processing functionalities for generating and/or recovering a real sidelink signal which is either simulated/emulated or sent over the air and captured from an SDR board. The code is highly-modular and documented in order to be easily understood and further extended.
The library has many usages. Typical use-case examples are the following:
- LTE sidelink waveform generator.
- End-to-end sidelink link-level simulator.
- Core component of a sidelink system-level simulator.
- Platform for testing new resource allocation/scheduling algorithms for D2D/V2V communications
- Tool to experiment with live standard-compliant sidelink signals with the help of SDR boards
Current Status: The library is freely and openly available in Github (https://github.com/feron-tech/lte-sidelink). New features and functionalities will be added in the next few months. A real-time software modem based on the above library and running in typical GPP hosts is also under development. The development of the library has been partially supported by EU Project FLEX-D.
Open Source Software Tools for Measuring the Quality of 4G Networks
Feron Technologies along with the leading National MNO, COSMOTE S.A., work together in the EU MONROE H2020 Project towards developing a set of automated software tools targeting the benchmarking of commercial networks. The tools can be executed either at remote MONROE nodes or at any general-purpose linux-based host equipped with mobile broadband devices (e.g. Wi-Fi/4G USB dongle, Mi-Fi, etc.).
The specific software tools enable the automated execution of a set of widely used mobile data network performance tools. The applications allow to characterize latency and throughput. Currently the following standard applications are supported:
- ping, for measuring RTT latency.
- iperf3 forward and reverse TCP tests, for measuring throughput.
- speedtest, for measuring latency, download and upload speed.
- HTTP transfer file download (GET) and upload (POST) using the curl package, for measuring throughput in both directions.
- Video Streaming, i.e. playing a video from a remote server (youtube or standalone server), for measuring various quality of service and experience indicators. The VLC client has been selected to be the core of our video testing tool, thanks to the availability of an extensive open API/library which exposes various video streaming parameters and performance indicators. Currently, the tool is able to automatically stream both YouTube videos (using a parser) and videos stored to dedicated servers. Various video quality levels (including 1080p) and network caching configurations are being supported. The tool enables the capture of a set of application-level quality of service/experience indicators, such as:
- video playback initiation latency,
- number of (potential) video stalls,
- progress of video playing time vs. the actual time,
- application bit-rate.
Instructions for deploying and configuring the necessary server side components are also provided.
Current Status: The tools are freely and openly available in Github (https://github.com/feron-tech/movement). The repository will be continuously upgraded with new features and functionalities, e.g. the measurement of OTT VoIP applications, the investigation of 4G-WiFi offloading scenarios, etc.