RCSE WS19/20 Group Study: Performance analysis framework for base station placement using IEEE 802.11

Links

• A LaTex project documentation.
• An Experiment report.
• Group Study Presentation.

Some tips

Access Points deploying

1. Go to directory docs/experiments/ap_provisioning
2. Change parameters, especially check if AP's stick adapter name is changed properly as it's seen in the OS
3. Run ansible-playbook (it will asks for password to escalate privileges):

// Deploy and provision of CnC
ansible-playbook -i inventory.yml cnc_run.yml --ask-become-pass
// Deploy and provision of APs
ansible-playbook -i inventory.yml ap_run.yml --ask-become-pass

Early access

Now it is possible to run a Proof-of-Concept of the system. The PoC is taking into considering:

• Visualization done by offline Jupyter Python Notebook (here)
• Instead of a real ClientApp, there is a mock client class - RealisticWIFIPayloadGenerator, which is running as a part of PoC.
• Data producing by RealisticWIFIPayloadGenerator is not truly realisitic, especially in GPS location, but produced RSS taken by a formula visualized here.

To start up the PoC, you should check if you have Docker and docker-compose installed:

docker version
docker-compose --version

If they are installed, first you need get required base images or build them from Dockerfiles:

docker-compose pull
docker-compose build

Then type:

docker-compose up -d

Docker will run:

• MongoDB NoSQL database - to store records.
• Eclipse Mosquitto MQTT Message Broker - to be a bus for components.
• DataBroker - to receive messages from MQTT bus and save it in the Storage.
• RealisticWIFIPayloadGenerator - to generate testing data.
• DataVisual - a simple HTTP API backend, if you need integration.

After docker is done, you can havigate to the notebook and run the cells.

Available pages

• Information on the project
• Project Documentation
• Extra information to keep in mind while working