CCRI: SpecScape: Enabling a Global Spectrum Observatory through Mobile, Wide-band Spectrum Sensing Kits and a Software Ecosystem

Description

SpecScape is a geographical wireless data collection system that integrates Software Defined Radio(SDR) with Android mobile devices. It uses cross-compiled version of GNU Radio running on the phone to collect wireless signals and perform basic analyses on the phone. It is designed to support researchers doing wireless measurement in the field by offering a portable choice. The system utilizes tripartie relationship consisting of Controllers, a Broker and Endpoints. Controllers are researchers who design wireless measurment program. The Broker serves as an intermediary, distributing programs among Endpoints. Endpoints execute tasks and backhual collected data to the Broker. By leveraging a domain-specific language, SpecScape enables flexibility for controllers design their own program while securely executed on Endpoints.



Background

SpecScape is a platform designed for researchers to collect wireless signals in the field and backhaul IQ samples to cloud. It also allows user to run some light analysis on SDR and Android devices while put heavy computation analyses on cloud. It utilizes Android device's ecosystem with computation, Internet, power and storage resources to collect data and backhaul processed data back to the Broker. This platform needs to be flexible, secure and portable to different wireless measurement programs. This platform separates designers and executors mission where researchers can sit in the lab and design programs while there are endpoints in the field to collect data. It is a signal data collector and analyzer that can be used to collect and analyze RF signals in real-time. SpecScape is a tripartie system consisting of Controllers, a Broker and Endpoints. Controllers design flowgraph and upload to the Broker, then the Broker will allocates tasks to Endpoints according to the configure file required by controllers. These three components together, they form a robust platform for comprehensive RF signal collection and analysis, enabling users to gain a deeper understanding of the signal environment in the field.


Background



Running SpecScape

To run SpecScape, first design a flowgraph using GNU Radio 3.8 with the approved blocks, available at this link {\url}. Ensure that the flowgraph is runnable and set to "no_gui" mode in the top_module. Save the flowgraph in .grc format. Next, create an experiment configuration file tailored to your specific research needs. This file should include geolocation details for signal collection and specify the duration of data collection. After preparing both the configuration and flowgraph .grc files, log onto the platform, select an endpoint, and upload both files. The platform will automatically assign the flowgraph to an endpoint when it is online and meets the criteria specified in the configuration file. The endpoint will then execute the flowgraph and transmit the data back to the platform. During the experiment, endpoints will continuously relay log information in real time, allowing researchers to monitor the experiment and identify any emerging errors. All results will be published on the platform, and researchers can download the experiment results directly. Additionally, our platform offers a cloud service. For instance, researchers can visualize the spectrogram of the collected IQ samples by selecting a historical result and clicking the "View on IQ-Engine" button. We also provide a default data collection flowgraph for cellular networks and run a Matlab decoder on the platform, enabling researchers to view the decoded MIB/SIB results directly.


Running SpecScape

User Interaction with the SpecScape Interface




Contributors

    Dinesh Bharadia profile picture
    Dinesh Bharadia

    Principal Investigator, Associate Prof. ECE

    Aaron Schulman profile picture
    Aaron Schulman

    Principal Investigator, Associate Prof. CSE

    Zesen Zhang profile picture
    Zesen Zhang

    PhD

    Yunxiang Chi profile picture
    Yunxiang Chi

    PhD

    Rohith Reddy Vennam profile picture
    Rohith Reddy Vennam

    PhD

    Maiyun Zhang profile picture
    Maiyun Zhang

    Undergraduate