We suggest Wi-Fi positioning based on Fingerprinting

• Figerprinting - A unique characteristic of the radio signal (e.g. Identify a position based on a set of RSSIs associated with APs

Develop Android application for indoor positioning (AI 공학관 4, 5층) using Wi-Fi and sensors

• Set the role for each member
• Create Admin App
• Implement Wi-Fi sensing
• Create User App

• Collect data using another sensors (e.g. GPS, Barometer)
• Implement back-end server using firebase realtime database
• Implement python environment for analysis sensor data

• Make an Admin app that can measure the RSSI value of Wi-Fi
• After uploading the measurement value to the database (FIREBASE), convert it using Python

Use the spinner to select the number of floors and and press the START button to find the RSSI(Received Signal Strength Indication Value) value of nearby Wi-Fi and inform the WAP(Wireless Access Point) and distance.

Contains the RSSI of the WAP at the measured location. When RSSI values are expressed in negative format, the closer the value is to 0, the stronger the received signal.

Only the reception strength of the measured value is measured in the (FIREBASE) data set. Convert the measured value to a data frame using Python

#원하는 호 수 => 504호 => (5층/4호)
ref = db.reference('?층/??호')
p = ref.get()
get_list = []

df = pd.DataFrame(p)
df.index = np.arange(len(df))

#호 수로 입력
df["target"] = ???

df

• Dataset collection
• Dataset Reorganization
• Implement restAPI using google AI platform
• Associate information with user apps

• Using Google AI Platform:
  AI Platform is a service created by Google Cloud Platform that helps you implement and deploy machine learning models for production
  
• Tools to help implement, deploy
  • Serving, TFX, Explainable AI, TPU
  • AI Hub, Kubeflow
  • Various tools are available and are increasingly being added

After deep learning the dataset through the Google AI Platform and completing modeling, it is released using RestAPI
It uses Google's Identity Platform Rest API to perform common user tasks, such as user token tasks.

• Android - Mobile development platform
• Firebase - Back-end design
• Python - Sensor data analysis

• minSdk:21
• targetSdk:31

• 1st week - MinHyung Lee - MinHyung
• 2nd week - Kyungho Choi - Kyungho
• 3rd week - Hyegyun Yim - hgyim00
• 4th week - Yunho Kim - YunHo
• 5th week - MinHyung Lee - MinHyung

This project is licensed under the MIT License - see the LICENSE.md file for details

• Knowledge about 802.11 protocol
• Basic android programming skill with java
• Basic Python programming skill
• Data analysis technique