Beacon sniffer

Test script for debug the beacons.

Installation

You must to install node v9 for ARMv7 devices. Follow this tutorial. Also, follow noble instruction for install noble.

install node v9

cd ~/

get https://nodejs.org/dist/latest-v9.x/node-v9.11.2-linux-armv7l.tar.xz

tar -xf node-v9.11.2-linux-armv7l.tar.xz

cd node-v9.11.2-linux-armv7l/
sudo cp -R * /usr/local/

node -v
npm -v

How to run the main app.

Run the app with

sudo node app.js

This will log the telemegry data for the debug device into the *.sqlite3 , and will output in the console:

pi@raspberrypi:~/bluetooth-rpi-beacon $ cat logs_app 
[INFO] Database created!
Starting scan...
Scanning started.
TABLE CREATED with response: [object Object]
{
	"addrs": "c1:93:da:19:eb:cd",
	"rssiDB": -87,
	"manufacturer": "Kontakt",
	"uuid": "fe6a",
	"telemetry": {
		"timeStamp": "2019-03-26 03:33:48",
		"x": 16,
		"y": 4,
		"z": 60,
		"temp": 28
	}
}
[INFO] Inserting data
Data INSERTED with log [object Object]
{
	"addrs": "c1:93:da:19:eb:cd",
	"rssiDB": -89,
	"manufacturer": "Kontakt",
	"uuid": "fe6a",
	"telemetry": {
		"timeStamp": "2019-03-26 03:33:49",
		"x": 229,
		"y": 5,
		"z": 54,
		"temp": 27
	}
}
[INFO] Inserting data
Data INSERTED with log [object Object]
{
	"addrs": "c1:93:da:19:eb:cd",
	"rssiDB": -84,
	"manufacturer": "Kontakt",
	"uuid": "fe6a",
	"telemetry": {
		"timeStamp": "2019-03-26 03:33:54",
		"x": 224,
		"y": 4,
		"z": 53,
		"temp": 28
	}
}

How to plot

From the root of the proyect, cd into timeseries/ folder and run the script exploredata.py

$ cd timeseries/
$ python exploredata.py

This script exploredata.py expects the sqlite3 db telemetry_data.py in the root of the project for run the plots. if __init__ is :

if __name__ == '__main__':
    model = DataManager()

    database = "../telemetry_data.sqlite3"

    # Clean dataframe
    df = model.get_df(database, use_all=False)

    # Add magnitude
    df = model.add_magnitude(df)
    
    # Convert to series wtih 60 seg window
    series = model.df_to_series(df, '60s')

    # Plot
    model.plot_df(series, save=False)

Update 28.mar.2019

Updated exploredata.py script for plot the corrent fields in telemetry_data.sqlite3 Update README.md with instructions of how to run the plot script.

Update 25.mar.2019

Refactor of insertData methods so the beacon can log the gyro, temp signal data into the SQL in the format:

The process of data collection for the algorithm creation has begun...

Update 06.mar.2019

Added temperature address from 22 lenght vector

Added main frontend scripts for draw time series data into the folder frontend/index.hmtl

TODO

Add Temperature value to the JSON Log
Implement REST API for get data values and plot live charts with the data using the frontend scripts.
Implement main classification algorithm for move/no-move.

Update 27.feb.2019

Thanks to the plots, we identify the x y z componets of the telemetry in the array with 22 length. The next plot shows the behavior of moving the beacon in a table in X , later in Y and the Z componets.

a13 seems to be some kind of Force Sensor , accidentally I drop the device to the floor and this down in the graph for a13 reflect this event.

Refactor

A refactor process was made in the repo, now the principal modules of the script are in separate folders and sub classes. You can run the application with:

pi@raspberrypi:~/beacons $ sudo node app.js

with the following log:

pi@raspberrypi:~/bluetooth-rpi-beacon $ sudo node app.js 
Database created!
Starting scan...
Scanning started.
{
	"addrs": "c1:93:da:19:eb:cd",
	"rssiDB": -75,
	"manufacturer": "Kontakt",
	"uuid": "fe6a",
	"telemetry": {
		"timeStamp": "2019-02-27 05:14:26",
		"x": 232,
		"y": 205,
		"z": 246
	}
}
{
	"addrs": "c1:93:da:19:eb:cd",
	"rssiDB": -73,
	"manufacturer": "Kontakt",
	"uuid": "fe6a",
	"telemetry": {
		"timeStamp": "2019-02-27 05:14:39",
		"x": 36,
		"y": 42,
		"z": 25
	}
}
{
	"addrs": "c1:93:da:19:eb:cd",
	"rssiDB": -74,
	"manufacturer": "Kontakt",
	"uuid": "fe6a",
	"telemetry": {
		"timeStamp": "2019-02-27 05:14:40",
		"x": 63,
		"y": 231,
		"z": 241
	}
}
{
	"addrs": "c1:93:da:19:eb:cd",
	"rssiDB": -60,
	"manufacturer": "Kontakt",
	"uuid": "fe6a",
	"telemetry": {
		"timeStamp": "2019-02-27 05:14:41",
		"x": 13,
		"y": 244,
		"z": 194
	}
}
{
	"addrs": "c1:93:da:19:eb:cd",
	"rssiDB": -67,
	"manufacturer": "Kontakt",
	"uuid": "fe6a",
	"telemetry": {
		"timeStamp": "2019-02-27 05:14:42",
		"x": 12,
		"y": 244,
		"z": 195
	}
}
...

TODO

Identify the Temperature Service Address (wth the service discovery script)

Update

Once the telemetry is activated in the beacon, we can access to the new peripheral.address c1:93:da:19:eb:cd, this peripheral address emit 3 vector arrays types, from which we are just interested in the array with length of 22 .

[3,9,2,16,0,2,62,255,255,224,12,6,1,185,135,108,92,100,3,5,255,27]

The example array has the properties for detect the beacon movements. We log this information into a my_db*.sqlite3 database. Here is the plot for the values of the array in a time series fashion, the most important arrays are the ones which are ploted below.

Feature imporance

The feature importance was checked using a histrogram for the timeseries data. The feature importance is showed below:

For Movement case

For Static case

There are 22 features, from which it seems that just 8 participate in the movement-sensors.

For Movement case

For this scenario, the beacon was carried by a pedestrian for around 30 min, the movement of the pedestrian is displayed below.

With y-logaritmic scale

With normal scale:

For the Static case

In this state the beacon is left on a table for around 20 minutes.

With y-logaritmic scale

With normal scale

How to debug devices

For find beacon devices:

pi@raspberrypi:~/beacons $ sudo node advertisement-discovery.js

The result is:

peripheral discovered (cd1244573708 with address <cd:12:44:57:37:08, random>, connectable false, RSSI -20:
	hello my local name is:
		undefined
	can I interest you in any of the following advertised services:
		["feaa"]
	here is my service data:
		"feaa": "00dcf7826da6bc5b71e0893e65666a6e4e64"

peripheral discovered (f40dc8467fc6 with address <f4:0d:c8:46:7f:c6, random>, connectable true, RSSI -44:
	hello my local name is:
		Kontakt
	can I interest you in any of the following advertised services:
		[]
	here is my service data:
		"fe6a": "0208010364f452737669485a"

peripheral discovered (e7dadf9ced70 with address <e7:da:df:9c:ed:70, random>, connectable false, RSSI -21:
	hello my local name is:
		undefined
	can I interest you in any of the following advertised services:
		[]
	here is my manufacturer data:
		"4c000215f7826da64fa24e988024bc5b71e0893ee92392f3b3"

For check the peripherals info:

pi@raspberrypi:~/beacons $ sudo node peripheral-discover.js cd1244573708

peripheral with ID cd1244573708 found
  Service Data      = [
  {
    "uuid": "feaa",
    "data": {
      "type": "Buffer",
      "data": [
        16,
        220,
        2,
        107,
        110,
        116,
        107,
        46,
        105,
        111,
        47,
        101,
        100,
        100,
        121,
        115,
        116,
        111,
        110,
        101
      ]
    }
  }
]
  Service UUIDs     = feaa

services and characteristics:

Fase de escaneo BLE finalizada

PERIPHERAL Peripheral {
  _noble: 
   Noble {
     initialized: true,
     address: 'b8:27:eb:91:8b:c4',
     _state: 'poweredOn',
     _bindings: 
      NobleBindings {
        _state: 'poweredOn',
        _addresses: [Object],
        _addresseTypes: [Object],
        _connectable: [Object],
        _pendingConnectionUuid: null,
        _connectionQueue: [],
        _handles: {},
        _gatts: {},
        _aclStreams: {},
        _signalings: {},
        _hci: [Hci],
        _gap: [Gap],
        _events: [Object],
        _eventsCount: 21,
        onSigIntBinded: [Function: bound ],
        _scanServiceUuids: [] },
     _peripherals: { e7dadf9ced70: [Peripheral], cd1244573708: [Circular] },
     _services: { e7dadf9ced70: {}, cd1244573708: {} },
     _characteristics: { e7dadf9ced70: {}, cd1244573708: {} },
     _descriptors: { e7dadf9ced70: {}, cd1244573708: {} },
     _discoveredPeripheralUUids: [ 'e7dadf9ced70', 'cd1244573708' ],
     _events: 
      { warning: [Function: bound ],
        newListener: [Function: bound ],
        stateChange: [Function],
        discover: [Function] },
     _eventsCount: 4,
     _allowDuplicates: undefined },
  id: 'cd1244573708',
  uuid: 'cd1244573708',
  address: 'cd:12:44:57:37:08',
  addressType: 'random',
  connectable: false,
  advertisement: 
   { localName: undefined,
     txPowerLevel: undefined,
     manufacturerData: undefined,
     serviceData: [ [Object] ],
     serviceUuids: [ 'feaa' ],
     solicitationServiceUuids: [],
     serviceSolicitationUuids: [] },
  rssi: -47,
  services: null,
  state: 'disconnected' }

For check the traffic:

pi@raspberrypi:~/beacons $ sudo node puck.js

The result for MAC cd:12:44:57:37:08 is

cd:12:44:57:37:08 -72 undefined undefined [ 32, 0, 11, 167, 31, 64, 6, 152, 167, 215, 6, 28, 216, 90 ]
cd:12:44:57:37:08 -71 undefined undefined [ 32, 0, 11, 167, 31, 64, 6, 152, 167, 230, 6, 28, 216, 100 ]
cd:12:44:57:37:08 -73 undefined undefined [ 32, 0, 11, 167, 31, 0, 6, 152, 167, 245, 6, 28, 216, 120 ]
cd:12:44:57:37:08 -74 undefined undefined [ 32, 0, 11, 167, 31, 64, 6, 152, 168, 7, 6, 28, 216, 130 ]
cd:12:44:57:37:08 -59 undefined undefined [ 32, 0, 11, 167, 31, 0, 6, 152, 168, 22, 6, 28, 216, 140 ]
cd:12:44:57:37:08 -73 undefined undefined [ 32, 0, 11, 167, 31, 0, 6, 152, 168, 88, 6, 28, 216, 200 ]

cd:12:44:57:37:08 -31 undefined undefined [ 32, 0, 11, 167, 34, 0, 6, 152, 178, 222, 6, 28, 226, 0 ]
cd:12:44:57:37:08 -17 undefined undefined [ 32, 0, 11, 167, 34, 0, 6, 152, 178, 240, 6, 28, 226, 10 ]
cd:12:44:57:37:08 -45 undefined undefined [ 32, 0, 11, 167, 33, 192, 6, 152, 178, 255, 6, 28, 226, 30 ]
cd:12:44:57:37:08 -44 undefined undefined [ 32, 0, 11, 167, 34, 0, 6, 152, 179, 17, 6, 28, 226, 40 ]
cd:12:44:57:37:08 -31 undefined undefined [ 32, 0, 11, 167, 34, 64, 6, 152, 179, 32, 6, 28, 226, 60 ]
cd:12:44:57:37:08 -46 undefined undefined [ 32, 0, 11, 167, 34, 0, 6, 152, 179, 47, 6, 28, 226, 70 ]
cd:12:44:57:37:08 -45 undefined undefined [ 32, 0, 11, 167, 33, 192, 6, 152, 179, 80, 6, 28, 226, 100 ]
cd:12:44:57:37:08 -45 undefined undefined [ 32, 0, 11, 167, 33, 192, 6, 152, 179, 98, 6, 28, 226, 110 ]
cd:12:44:57:37:08 -16 undefined undefined [ 32, 0, 11, 167, 33, 192, 6, 152, 179, 131, 6, 28, 226, 140 ]
cd:12:44:57:37:08 -18 undefined undefined [ 32, 0, 11, 167, 34, 0, 6, 152, 179, 146, 6, 28, 226, 160 ]
cd:12:44:57:37:08 -18 undefined undefined [ 32, 0, 11, 167, 34, 0, 6, 152, 179, 179, 6, 28, 226, 180 ]
cd:12:44:57:37:08 -30 undefined undefined [ 32, 0, 11, 167, 34, 0, 6, 152, 179, 194, 6, 28, 226, 200 ]

Uncomment the line 12 in puck.js for full log for all the aviable services from beacons.

TODOS

Log rssi signal strength in the sqlite db.
Refactor important functions in the code.
Implement a timeseries plot in html and javascript, for real-time movement display in the browser.
Implement algorithm for classify movement/no-movement from data
Log this classification in other sqlite db.

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Repository files navigation

Beacon sniffer

Installation

install node v9

How to run the main app.

How to plot

Update 28.mar.2019

Update 25.mar.2019

Update 06.mar.2019

TODO

Update 27.feb.2019

Refactor

TODO

Update

Feature imporance

For Movement case

For Static case

For Movement case

For the Static case

How to debug devices

TODOS

References

About

Releases

Packages

Languages

Name		Name	Last commit message	Last commit date
Latest commit History 57 Commits
db		db
devices		devices
frontend		frontend
scripts		scripts
timeseries		timeseries
.gitignore		.gitignore
README.md		README.md
app.js		app.js
links		links
package-lock.json		package-lock.json
package.json		package.json
telemetry_data.sqlite3		telemetry_data.sqlite3

stanlee321/bluetooth-rpi-beacon

Folders and files

Latest commit

History

Repository files navigation

Beacon sniffer

Installation

install node v9

How to run the main app.

How to plot

Update 28.mar.2019

Update 25.mar.2019

Update 06.mar.2019

TODO

Update 27.feb.2019

Refactor

TODO

Update

Feature imporance

For Movement case

For Static case

For Movement case

For the Static case

How to debug devices

TODOS

References

About

Resources

Stars

Watchers

Forks

Releases

Packages 0

Languages

Packages