File size: 7,391 Bytes
8abe2d5
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
import math
from jddb.file_repo import FileRepo
from jddb.processor import Signal, Shot, ShotSet, BaseProcessor
from typing import Optional, List
from scipy.fftpack import fft
from copy import deepcopy
from scipy.interpolate import interp1d
import pandas as pd
import numpy as np
from scipy import signal as sig
from copy import deepcopy


class SliceProcessor(BaseProcessor):
    """
            input the point number of the window  and overlap rate of the given window ,
        then the sample rate is recalculated,  return a signal of time window sequence
    """

    def __init__(self, window_length: int, overlap: float):
        super().__init__()
        assert (0 <= overlap <= 1), "Overlap is not between 0 and 1."
        self.params.update({"WindowLength": window_length,
                            "Overlap": overlap})

    def transform(self, signal: Signal) -> Signal:
        window_length = self.params["WindowLength"]
        overlap = self.params["Overlap"]
        new_signal = deepcopy(signal)
        raw_sample_rate = new_signal.attributes["SampleRate"]
        step = round(window_length * (1 - overlap))

        down_time = new_signal.time[-1]

        down_time = round(down_time, 3)

        idx = len(signal.data)
        window = list()
        while (idx - window_length) >= 0:
            window.append(new_signal.data[idx - window_length:idx])
            idx -= step
        window.reverse()
        new_signal.attributes['SampleRate'] = raw_sample_rate * len(window) / (len(new_signal.data) - window_length + 1)
        new_signal.data = np.array(window)
        new_start_time = down_time - len(window) / new_signal.attributes['SampleRate']
        new_signal.attributes['StartTime'] = round(new_start_time, 3)
        new_signal.attributes['OriginalSampleRate'] = raw_sample_rate
        return new_signal


class FFTProcessor(BaseProcessor):
    """
        processing signal by Fast Fourier Transform , return the maximum amplitude and the corresponding frequency
    """

    def __init__(self):
        super().__init__()

        self.amp_signal = None
        self.signal_rate = None
        self.fre_signal = None

    def transform(self, signal: Signal):

        self.amp_signal = deepcopy(signal)
        self.signal_rate = signal.attributes['OriginalSampleRate']
        self.fre_signal = deepcopy(signal)
        self.fft()
        self.amp_max()

        return self.amp_signal, self.fre_signal

    def fft(self):
        if self.amp_signal.data.ndim == 1:
            N = len(self.amp_signal.data)
            fft_y = fft(self.amp_signal.data)
            abs_y = np.abs(fft_y)
            normed_abs_y = abs_y / (N / 2)
            self.amp_signal.data = normed_abs_y[:int(N / 2)]
        elif self.amp_signal.data.ndim == 2:
            N = self.amp_signal.data.shape[1]
            R = self.amp_signal.data.shape[0]
            raw_cover = np.empty(shape=[0, int(N / 2)], dtype=float)
            for i in range(R):
                fft_y = fft(self.amp_signal.data[i])
                abs_y = np.abs(fft_y)
                normed_abs_y = abs_y / (N / 2)
                raw_cover = np.append(raw_cover, [normed_abs_y[:int(N / 2)]], axis=0)
            self.amp_signal.data = raw_cover

    def amp_max(self):
        fs = self.signal_rate
        raw = self.amp_signal.data
        amp_cover = np.empty(shape=0, dtype=float)
        fre_cover = np.empty(shape=0, dtype=float)
        N = (raw.shape[1]) * 2
        f = (np.linspace(start=0, stop=N - 1, num=N) / N) * fs
        f = f[:int(N / 2)]
        for j in range(raw.shape[0]):
            list_max = (raw[j, :]).tolist()
            raw_max = max(list_max)
            max_index = list_max.index(max(list_max))
            f_rawmax = f[max_index]
            amp_cover = np.append(amp_cover, raw_max)
            fre_cover = np.append(fre_cover, f_rawmax)
        self.amp_signal.data = amp_cover
        self.fre_signal.data = fre_cover


class Mean(BaseProcessor):
    """
         Given a set of input signals, average each instant
    """

    def __init__(self):
        super().__init__()

    def transform(self, *signal: Signal) -> Signal:
        new_signal = Signal(np.row_stack([sign.data for sign in signal.__iter__()]).T, signal.__getitem__(0).attributes)
        new_signal.data = np.mean(np.array(new_signal.data, dtype=np.float32), axis=1)
        return new_signal


class Concatenate(BaseProcessor):
    """
        calculate the mean and standard deviation of the given signal
    """

    def __init__(self):
        super().__init__()

    def transform(self, *signal: Signal) -> Signal:
        new_signal = Signal(np.concatenate([sign.data for sign in signal.__iter__()], axis=0),
                            signal.__getitem__(0).attributes)

        return new_signal


class AlarmTag(BaseProcessor):
    """
            Give arbitrary signals, extract downtime, timeline,
        and generate actual warning time labels

    """

    def __init__(self, lead_time, disruption_label: str, downtime_label: str):
        super().__init__()
        self.lead_time = lead_time
        self._disruption_label = disruption_label
        self._downtime_label = downtime_label

    def transform(self, signal: Signal):
        copy_signal = deepcopy(signal)
        fs = copy_signal.attributes['SampleRate']
        start_time = copy_signal.attributes['StartTime']

        if self.params[self._disruption_label] == 1:
            undisrupt_number = int(fs * (self.params[self._downtime_label] - self.lead_time - start_time))
        else:
            undisrupt_number = len(copy_signal.data)
        if undisrupt_number < len(copy_signal.data):
            # new_data = np.zeros(shape=undisrupt_number, dtype=int)
            new_data = np.zeros(shape=1, dtype=int)
            for i in range(len(copy_signal.data) - 1):
                if i <= undisrupt_number - 1:
                    new_data = np.append(new_data, np.array(0))
                else:
                    new_data = np.append(new_data, np.array(1))
        else:
            new_data = np.zeros(shape=len(copy_signal.data), dtype=int)

        new_signal = Signal(data=new_data, attributes=dict())
        new_signal.attributes['SampleRate'] = fs
        new_signal.attributes['StartTime'] = start_time

        return new_signal


class RadiatedFraction(BaseProcessor):
    """
            Given the radiated power signal and input power
            signal to calculate the radiated fraction.

    """
    def __init__(self, ):
        super().__init__()

    def transform(self, radiated_power_signal: Signal, input_power_signal: Signal) -> Signal:
        """

        :param radiated_power_signal:
        :param input_power_signal:
        :return:
        """
        resampled_attributes = deepcopy(input_power_signal.attributes)
        new_data = radiated_power_signal.data / input_power_signal.data

        return Signal(data=new_data, attributes=resampled_attributes)


def find_tags(prefix, all_tags):
    """
        find tags that start with the prefix
    param:
        prefix: The first few strings of the tags users need to look for
        all_tags: a list of all the tags that needed to be filtered
    :return: matching tags as a list[sting]
    """
    return list(filter(lambda tag: tag.encode("utf-8").decode("utf-8", "ignore")[0:len(prefix)] == prefix, all_tags))