STPbeacon/plotting_tools.py at main · Yphrea/STPbeacon · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
import matplotlib.pyplot as plt
from math import sqrt
from scipy.interpolate import make_smoothing_spline, sproot
import numpy as np


"""plotting of test data for quality evaluation"""

text_box = dict(boxstyle='round', facecolor='whitesmoke', ec='black')

def read_data_from_file(filename):
    """read data from file (collect with 'scan.py')"""


    RSSI=[]; time=[]
    start_time = None; true_time = None

    with open(filename,'r') as fil:
        for line in fil:
            try:
                line = line.split()
                try:
                    int(line[1])
                    assert int(line[1]) < 0 #manually marked "true time" indicated by RSSI=0 in test data file
                    RSSI.append(int(line[1]))
                except ValueError:
                    start_time = [float(i) for i in line[-1].split(':')]
                    start_time = start_time[0]*3600+start_time[1]*60+start_time[2]
                    continue
                except AssertionError:
                    true_time = [float(i) for i in line[-1].split(':')]
                    true_time = true_time[0]*3600+true_time[1]*60+true_time[2]
                    continue
                aux_time = line[-1].split(':')
                time.append(float(aux_time[-2])+float(aux_time[-1]))
            except IndexError:
                break

        try:
            true_time = true_time - start_time
        except TypeError:
            pass
        time_consecutive = [sum(time[:i+1]) for i in range(len(time))]
        return (time_consecutive, RSSI, true_time)


def panelplot_test_data(list_of_data_files, show=False, save=None, bin_time=False, N=10):

    if type(list_of_data_files) != list:
        list_of_data_files = [list_of_data_files]

    # Panel layout
    m = len(list_of_data_files)
    sq_m = sqrt(m)
    width = int(sq_m); height = int(sq_m)

    if m > int(sq_m)**2:
        width += 1
    if m > int(sq_m)*(int(sq_m)+1):
        height += 1

    fig, axes = plt.subplots(height, width, sharey=True)
    try:
        axes = axes.flatten()
    except AttributeError:
        axes = [axes]

    handles = [None, None]
    labels = ['Evaluated', 'Facit']
    for i, data_file in enumerate(list_of_data_files):
        time, RSSI, true_time = read_data_from_file(data_file)
        if bin_time: # histogram over time between advertisements in sample data
            time = [0]+[time[i+1]-time[i] for i in range(len(time)-1)] # time difference between data points
            axes[i].hist(time, density=True, stacked=True, bins=N)
        else: # plain plot RSSI(time) from start of data sampling
            axes[i].plot(time,RSSI,'*')
            ylim = axes[i].get_ylim()
            time_aux = np.arange(0, max(time), 0.05)
            spl = make_smoothing_spline(time, RSSI, lam=10)
            spl_RSSI = spl(time_aux)
            i_spl_max = spl_RSSI.argmax(axis=0)
            axes[i].plot(time_aux, spl_RSSI, '-.')
            handles[0], = axes[i].plot([time_aux[i_spl_max]]*2, ylim,'--k')
            axes[i].text(0.02,0.02,'diff in s: %5.2f'%(true_time - time_aux[i_spl_max]), transform=axes[i].transAxes, bbox=text_box)
            axes[i].set_ylim(ylim)
            axes[i].grid(True)
            try:
                handles[1], = axes[i].plot([true_time]*2, ylim,'-k')
            except ValueError:
                pass # If true_time is None, i.e. not registered
        if i%width == 0:
            axes[i].set_ylabel('RSSI')
        if i >= width * (height-1):
            axes[i].set_xlabel('time')
        if i == 0:
            axes[i].legend(handles, labels)
    fig.set_size_inches(18.5, 10.5)
    plt.subplots_adjust(left=0.05, right=0.98, bottom=0.05, top=0.98, wspace=0.05, hspace=0.05)

    if save != None:
        plt.savefig(save, dpi=300)
    if show:
        plt.show()


#panelplot_test_data('test_data/noise/noise_Nano_32', show=True, save='noise_Nano_32.2.png')
#panelplot_test_data(['test_data/noise/noise_Nano_32', 'test_data/noise/noise_Nano_40'], N=30, bin_time=True, show=True)

# Data from desktop integrated BLE shipset (didn't try to check which) in home environment (less bluetooth disturbance)
#panelplot_test_data(['test_data/walking_it2/juniper_a',
#                     'test_data/walking_it2/juniper_b',
#                     'test_data/walking_it2/juniper_c',
#                     'test_data/walking_it2/juniper_d',
#                     'test_data/walking_it2/juniper_e',
#                     'test_data/walking_it2/juniper_f'], show=True)

# Data from Arduino Nano 33 BLE  in office environment
#panelplot_test_data(['test_data/walking_it2/Nano_a',#],show=True)
#                     'test_data/walking_it2/Nano_b',
#                     'test_data/walking_it2/Nano_c',
#                     'test_data/walking_it2/Nano_d',
#                     'test_data/walking_it2/Nano_e',
#                     'test_data/walking_it2/Nano_f'], show=True, save='test_data_Nano.png')

# Data from Nano taken whilst biking in parkinglot outside HiQ office, arduino was carried in backpack.
# Dataset a-e are with backpack on back, f-k are with backpack on steering bar (front of bike)
panelplot_test_data(['test_data/walking_it3/parkering_f',
                     'test_data/walking_it3/parkering_g',
                     'test_data/walking_it3/parkering_h',
                     'test_data/walking_it3/parkering_l',
                     'test_data/walking_it3/parkering_j',
                     'test_data/walking_it3/parkering_k'], show=True, save='test_data_parkering.png')