多个蓝牙加速度计的实时同步实验

undefined February 21, 2020 View/edit this page on Colab

Creative Commons License This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.

Setup and run streams

Import packages Input code

import arus
import pandas as pd
import numpy as np
import logging
import time
arus.dev.set_default_logging()

Scan nearby devices Input code

# Get device addr
scanner = arus.plugins.metawear.MetaWearScanner()
addrs = scanner.get_nearby_devices(max_devices=3)
addrs = addrs[0:3]

Output results

[INFO]2020-02-05 19:31:41,626 <P13484-MainThread> Scanning metawear devices nearby...
[INFO]2020-02-05 19:31:42,631 <P13484-MainThread> Scanning metawear devices nearby...
[INFO]2020-02-05 19:31:43,632 <P13484-MainThread> Scanning metawear devices nearby...

Setup streams Input code

# prepare streams
streams = []
ref_st = pd.Timestamp.now()
for addr in addrs:
    generator = arus.plugins.metawear.MetaWearAccelDataGenerator(
        addr, sr=50, grange=8, buffer_size=100)
    segmentor = arus.segmentor.SlidingWindowSegmentor(
        window_size=2)
    stream = arus.Stream(generator, segmentor,
                         name='metawear-stream-{}'.format(addr))
    streams.append(stream)

Start streams Input code

# start stream
results = {}
for stream in streams:
    stream.start(start_time=ref_st)
    results[stream._name] = []

Input code

done = [False] * len(streams)
while True:
    for stream in streams:
        for data, st, et, _, _, name in stream.get_iterator():
            data['ST'] = st
            data['ET'] = et
            i = streams.index(stream)
            logging.info('data from: {}, {}'.format(
                name, i))
            results[name].append(data)
            if len(results[name]) == 20:
                done[i] = True
            break
    if np.all(done):
        break

for stream in streams:
    stream.stop()

combined = []
for name, result in results.items():
    name = name.replace(':', '')
    result = pd.concat(result, sort=False)
    combined.append(result)

combined = pd.concat(combined, sort=False)

Output results

[INFO]2020-02-05 19:31:57,744 <P13484-metawear-stream-D5:D2:01:2B:E7:6D-loading> New metawear connected: MetaWearClient, D5:D2:01:2B:E7:6D: {'model': '5', 'hardware': '0.3', 'manufacturer': 'MbientLab Inc', 'serial': '031371', 'firmware': '1.5.0'}
[INFO]2020-02-05 19:31:57,773 <P13484-metawear-stream-D5:D2:01:2B:E7:6D-loading> starting accelerometer module...
[INFO]2020-02-05 19:31:57,774 <P13484-metawear-stream-D5:D2:01:2B:E7:6D-loading> Waiting for accelerometer callback to start...
[INFO]2020-02-05 19:31:57,782 <P13484-Dummy-8> Accelerometer callback starts running...
[INFO]2020-02-05 19:31:57,790 <P13484-metawear-stream-D5:D2:01:2B:E7:6D-loading> Accelerometer callback started successfully.
[INFO]2020-02-05 19:31:57,833 <P13484-MainThread> data from: metawear-stream-D5:D2:01:2B:E7:6D, 0
[INFO]2020-02-05 19:31:57,905 <P13484-MainThread> data from: metawear-stream-E6:F8:37:54:03:87, 1
[INFO]2020-02-05 19:31:57,977 <P13484-MainThread> data from: metawear-stream-FF:EE:B8:99:0C:64, 2
[INFO]2020-02-05 19:31:58,036 <P13484-MainThread> data from: metawear-stream-D5:D2:01:2B:E7:6D, 0
[INFO]2020-02-05 19:31:58,056 <P13484-MainThread> data from: metawear-stream-E6:F8:37:54:03:87, 1
[INFO]2020-02-05 19:31:58,100 <P13484-MainThread> data from: metawear-stream-FF:EE:B8:99:0C:64, 2
[INFO]2020-02-05 19:31:58,116 <P13484-MainThread> data from: metawear-stream-D5:D2:01:2B:E7:6D, 0
[INFO]2020-02-05 19:31:58,153 <P13484-MainThread> data from: metawear-stream-E6:F8:37:54:03:87, 1
[INFO]2020-02-05 19:31:58,304 <P13484-MainThread> data from: metawear-stream-FF:EE:B8:99:0C:64, 2
[INFO]2020-02-05 19:32:00,033 <P13484-MainThread> data from: metawear-stream-D5:D2:01:2B:E7:6D, 0
[INFO]2020-02-05 19:32:00,265 <P13484-MainThread> data from: metawear-stream-E6:F8:37:54:03:87, 1
[INFO]2020-02-05 19:32:00,281 <P13484-MainThread> data from: metawear-stream-FF:EE:B8:99:0C:64, 2
[INFO]2020-02-05 19:32:01,896 <P13484-MainThread> data from: metawear-stream-D5:D2:01:2B:E7:6D, 0
[INFO]2020-02-05 19:32:02,157 <P13484-MainThread> data from: metawear-stream-E6:F8:37:54:03:87, 1
[INFO]2020-02-05 19:32:02,181 <P13484-MainThread> data from: metawear-stream-FF:EE:B8:99:0C:64, 2
[INFO]2020-02-05 19:32:03,993 <P13484-MainThread> data from: metawear-stream-D5:D2:01:2B:E7:6D, 0
[INFO]2020-02-05 19:32:04,277 <P13484-MainThread> data from: metawear-stream-E6:F8:37:54:03:87, 1
[INFO]2020-02-05 19:32:04,350 <P13484-MainThread> data from: metawear-stream-FF:EE:B8:99:0C:64, 2
[INFO]2020-02-05 19:32:06,126 <P13484-MainThread> data from: metawear-stream-D5:D2:01:2B:E7:6D, 0
[INFO]2020-02-05 19:32:06,234 <P13484-MainThread> data from: metawear-stream-E6:F8:37:54:03:87, 1
[INFO]2020-02-05 19:32:06,268 <P13484-MainThread> data from: metawear-stream-FF:EE:B8:99:0C:64, 2
[INFO]2020-02-05 19:32:08,002 <P13484-MainThread> data from: metawear-stream-D5:D2:01:2B:E7:6D, 0
[INFO]2020-02-05 19:32:08,081 <P13484-MainThread> data from: metawear-stream-E6:F8:37:54:03:87, 1
[INFO]2020-02-05 19:32:08,174 <P13484-MainThread> data from: metawear-stream-FF:EE:B8:99:0C:64, 2
[INFO]2020-02-05 19:32:10,122 <P13484-MainThread> data from: metawear-stream-D5:D2:01:2B:E7:6D, 0
[INFO]2020-02-05 19:32:10,327 <P13484-MainThread> data from: metawear-stream-E6:F8:37:54:03:87, 1
[INFO]2020-02-05 19:32:10,352 <P13484-MainThread> data from: metawear-stream-FF:EE:B8:99:0C:64, 2
[INFO]2020-02-05 19:32:11,932 <P13484-MainThread> data from: metawear-stream-D5:D2:01:2B:E7:6D, 0
[INFO]2020-02-05 19:32:12,252 <P13484-MainThread> data from: metawear-stream-E6:F8:37:54:03:87, 1
[INFO]2020-02-05 19:32:12,315 <P13484-MainThread> data from: metawear-stream-FF:EE:B8:99:0C:64, 2
[INFO]2020-02-05 19:32:13,867 <P13484-MainThread> data from: metawear-stream-D5:D2:01:2B:E7:6D, 0
[INFO]2020-02-05 19:32:14,215 <P13484-MainThread> data from: metawear-stream-E6:F8:37:54:03:87, 1
[INFO]2020-02-05 19:32:14,255 <P13484-MainThread> data from: metawear-stream-FF:EE:B8:99:0C:64, 2
[INFO]2020-02-05 19:32:15,875 <P13484-MainThread> data from: metawear-stream-D5:D2:01:2B:E7:6D, 0
[INFO]2020-02-05 19:32:16,263 <P13484-MainThread> data from: metawear-stream-E6:F8:37:54:03:87, 1
[INFO]2020-02-05 19:32:16,312 <P13484-MainThread> data from: metawear-stream-FF:EE:B8:99:0C:64, 2
[INFO]2020-02-05 19:32:17,709 <P13484-MainThread> data from: metawear-stream-D5:D2:01:2B:E7:6D, 0
[INFO]2020-02-05 19:32:18,021 <P13484-MainThread> data from: metawear-stream-E6:F8:37:54:03:87, 1
[INFO]2020-02-05 19:32:18,048 <P13484-MainThread> data from: metawear-stream-FF:EE:B8:99:0C:64, 2
[INFO]2020-02-05 19:32:19,691 <P13484-MainThread> data from: metawear-stream-D5:D2:01:2B:E7:6D, 0
[INFO]2020-02-05 19:32:20,260 <P13484-MainThread> data from: metawear-stream-E6:F8:37:54:03:87, 1
[INFO]2020-02-05 19:32:20,286 <P13484-MainThread> data from: metawear-stream-FF:EE:B8:99:0C:64, 2
[INFO]2020-02-05 19:32:21,739 <P13484-MainThread> data from: metawear-stream-D5:D2:01:2B:E7:6D, 0
[INFO]2020-02-05 19:32:22,210 <P13484-MainThread> data from: metawear-stream-E6:F8:37:54:03:87, 1
[INFO]2020-02-05 19:32:22,236 <P13484-MainThread> data from: metawear-stream-FF:EE:B8:99:0C:64, 2
[INFO]2020-02-05 19:32:23,648 <P13484-MainThread> data from: metawear-stream-D5:D2:01:2B:E7:6D, 0
[INFO]2020-02-05 19:32:24,105 <P13484-MainThread> data from: metawear-stream-E6:F8:37:54:03:87, 1
[INFO]2020-02-05 19:32:24,176 <P13484-MainThread> data from: metawear-stream-FF:EE:B8:99:0C:64, 2
[INFO]2020-02-05 19:32:25,650 <P13484-MainThread> data from: metawear-stream-D5:D2:01:2B:E7:6D, 0
[INFO]2020-02-05 19:32:26,225 <P13484-MainThread> data from: metawear-stream-E6:F8:37:54:03:87, 1
[INFO]2020-02-05 19:32:26,307 <P13484-MainThread> data from: metawear-stream-FF:EE:B8:99:0C:64, 2
[INFO]2020-02-05 19:32:27,625 <P13484-MainThread> data from: metawear-stream-D5:D2:01:2B:E7:6D, 0
[INFO]2020-02-05 19:32:28,095 <P13484-MainThread> data from: metawear-stream-E6:F8:37:54:03:87, 1
[INFO]2020-02-05 19:32:28,120 <P13484-MainThread> data from: metawear-stream-FF:EE:B8:99:0C:64, 2
[INFO]2020-02-05 19:32:29,599 <P13484-MainThread> data from: metawear-stream-D5:D2:01:2B:E7:6D, 0
[INFO]2020-02-05 19:32:30,013 <P13484-MainThread> data from: metawear-stream-E6:F8:37:54:03:87, 1
[INFO]2020-02-05 19:32:30,043 <P13484-MainThread> data from: metawear-stream-FF:EE:B8:99:0C:64, 2
[INFO]2020-02-05 19:32:31,543 <P13484-MainThread> data from: metawear-stream-D5:D2:01:2B:E7:6D, 0
[INFO]2020-02-05 19:32:32,020 <P13484-MainThread> data from: metawear-stream-E6:F8:37:54:03:87, 1
[INFO]2020-02-05 19:32:32,075 <P13484-MainThread> data from: metawear-stream-FF:EE:B8:99:0C:64, 2
[INFO]2020-02-05 19:32:33,238 <P13484-MainThread> Disconnected.
[INFO]2020-02-05 19:32:34,510 <P13484-MainThread> Disconnected.
[INFO]2020-02-05 19:32:35,743 <P13484-MainThread> Disconnected.

Analyze results

Check the first sample timestamp

  1. referenced start time: The start time used as a reference to segment the incoming data. This value is set before setting up and start the streams.
  2. first sample timestamp: The timestamp of the first sample received from each device. This timestamp has been synced with the computer clock when receiving the data.
  3. first sample device timestamp: The timestamp of the first sample received from each device. This timestamp has not been synced with the computer clock when receiving the data. Input code
first_sample = combined.groupby(by=['MAC_ADDRESS']).apply(
    lambda df: df[['HEADER_TIME_STAMP', 'NO_FIX']].iloc[0, :])
first_sample = first_sample.rename(
    columns={'HEADER_TIME_STAMP': 'first sample timestamp', 'NO_FIX': 'first sample device timestamp'})
first_sample = first_sample.sort_values(by=['first sample timestamp'])
first_sample['referenced start time'] = ref_st
first_sample['diff_ref (s)'] = arus.Moment.get_durations(
    first_sample['referenced start time'], first_sample['first sample timestamp'])
first_sample['diff_device (s)'] = arus.Moment.get_durations(
    first_sample['first sample device timestamp'],
    first_sample['first sample timestamp'])
first_sample.transpose()

Output results

MAC_ADDRESS FF:EE:B8:99:0C:64 E6:F8:37:54:03:87 D5:D2:01:2B:E7:6D
0
first sample timestamp 2020-02-05 19:31:54.511123 2020-02-05 19:31:54.950348 2020-02-05 19:31:57.779814
first sample device timestamp 2020-02-05 19:31:54.511000 2020-02-05 19:31:54.950000 2020-02-05 19:31:57.780000
referenced start time 2020-02-05 19:31:51.878749 2020-02-05 19:31:51.878749 2020-02-05 19:31:51.878749
diff_ref (s) 2.63237 3.0716 5.90107
diff_device (s) 0.000123 0.000348 -0.000186

Output results

MAC_ADDRESS                             FF:EE:B8:99:0C:64  \
0                                                           
first sample timestamp         2020-02-05 19:31:54.511123   
first sample device timestamp  2020-02-05 19:31:54.511000   
referenced start time          2020-02-05 19:31:51.878749   
diff_ref (s)                                      2.63237   
diff_device (s)                                  0.000123   

MAC_ADDRESS                             E6:F8:37:54:03:87  \
0                                                           
first sample timestamp         2020-02-05 19:31:54.950348   
first sample device timestamp  2020-02-05 19:31:54.950000   
referenced start time          2020-02-05 19:31:51.878749   
diff_ref (s)                                       3.0716   
diff_device (s)                                  0.000348   

MAC_ADDRESS                             D5:D2:01:2B:E7:6D  
0                                                          
first sample timestamp         2020-02-05 19:31:57.779814  
first sample device timestamp  2020-02-05 19:31:57.780000  
referenced start time          2020-02-05 19:31:51.878749  
diff_ref (s)                                      5.90107  
diff_device (s)                                 -0.000186

Check number of samples in each segments

Input code

count_segment = combined.groupby(
    by=['MAC_ADDRESS', 'ST']).size().to_frame(name='# of samples').reset_index(drop=False)
count_segment = count_segment.pivot(
    columns='MAC_ADDRESS', index='ST', values=['# of samples'])
count_segment.columns = sorted(addrs)
count_segment.reset_index(drop=False, inplace=True)
count_segment.plot(x='ST', kind='line', ylim=(0, 210))

Output results

<matplotlib.axes._subplots.AxesSubplot at 0x192224ffd30>

<Figure size 432x288 with 1 Axes>