-
Notifications
You must be signed in to change notification settings - Fork 0
/
filterbank.py
204 lines (163 loc) · 6.29 KB
/
filterbank.py
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
"""
Utilities for reading data from filterbank file
"""
import os
import numpy as np
import matplotlib.pyplot as plt
from header import read_header, len_header
class Filterbank():
"""
Processing .fil files
"""
# pylint: disable=too-many-instance-attributes
def __init__(self, filename, freq_range=None, time_range=None):
"""
Initialize Filterbank object
Args:
freq_range, tuple of freq_start and freq_stop in MHz
time_range, tuple of time_start and time_stop
"""
self.freqs = None
self.timestamps = None
if os.path.isfile(filename):
self.filename = filename
self.header = read_header(filename)
self.idx_data = len_header(filename)
self.n_bytes = int(self.header[b'nbits']/8)
self.n_chans = self.header[b'nchans']
self.n_ifs = self.header[b'nifs']
self.read_filterbank(freq_range, time_range)
else:
raise FileNotFoundError(filename)
def read_filterbank(self, freq_range=None, time_range=None):
"""
Read filterbank file to 3d numpy array
"""
fil = open(self.filename, 'rb')
fil.seek(self.idx_data)
ii_start, n_ints = self.setup_time(time_range)
# search for start of data
fil.seek(int(ii_start * self.n_bytes * self.n_ifs * self.n_chans), 1)
i_0, i_1 = self.setup_chans(freq_range)
n_chans_selected = self.freqs.shape[0]
# decide appropriate datatype
if self.n_bytes == 4:
dd_type = b'float32'
elif self.n_bytes == 2:
dd_type = b'uint16'
elif self.n_bytes == 1:
dd_type = b'uint8'
# create numpy array with 3d shape
self.data = np.zeros((n_ints, self.n_ifs, n_chans_selected), dtype=dd_type)
for i_i in range(n_ints):
for j_j in range(self.n_ifs):
fil.seek(self.n_bytes * i_0, 1)
# add to matrix
self.data[i_i, j_j] = np.fromfile(fil, count=n_chans_selected, dtype=dd_type)
# search for start of next chunk
fil.seek(self.n_bytes * (self.n_chans - i_1), 1)
def setup_freqs(self, freq_range=None):
"""
Calculate the frequency range
"""
f_delt = self.header[b'foff']
f_0 = self.header[b'fch1']
i_start, i_stop = 0, self.n_chans
if freq_range:
if freq_range[0]:
i_start = int((freq_range[0] - f_0) / f_delt)
if freq_range[1]:
i_stop = int((freq_range[1] - f_0) / f_delt)
chan_start_idx = np.int(i_start)
chan_stop_idx = np.int(i_stop)
if i_start < i_stop:
i_vals = np.arange(chan_start_idx, chan_stop_idx)
else:
i_vals = np.arange(chan_stop_idx, chan_start_idx)
self.freqs = f_delt * i_vals + f_0
if chan_stop_idx < chan_start_idx:
chan_stop_idx, chan_start_idx = chan_start_idx, chan_stop_idx
return chan_start_idx, chan_stop_idx
def setup_time(self, time_range=None):
"""
Calculate the time range
"""
t_delt = self.header[b'tsamp']
t_0 = self.header[b'tstart']
n_bytes_data = os.path.getsize(self.filename) - self.idx_data
n_ints_data = int(n_bytes_data / (self.n_bytes * self.n_chans * self.n_ifs))
ii_start, ii_stop = 0, n_ints_data
if time_range:
if time_range[0]:
ii_start = time_range[0]
if time_range[1]:
ii_stop = time_range[1]
n_ints = ii_stop - ii_start
self.timestamps = np.arange(0, n_ints) * t_delt / 24./60./60. + t_0
return ii_start, n_ints
def setup_chans(self, freq_range=None):
"""
Calculate the channel range
"""
chan_start_idx, chan_stop_idx = self.setup_freqs(freq_range)
i_0 = np.min((chan_start_idx, chan_stop_idx))
i_1 = np.max((chan_start_idx, chan_stop_idx))
return i_0, i_1
def select_data(self, freq_start=None, freq_stop=None, time_start=None, time_stop=None):
"""
Select a range of data from the filterbank file
"""
# if no frequency range is specified, select all frequencies
if freq_start is None:
freq_start = self.freqs[0]
if freq_stop is None:
freq_stop = self.freqs[-1]
i_0 = np.argmin(np.abs(self.freqs - freq_start))
i_1 = np.argmin(np.abs(self.freqs - freq_stop))
if i_0 < i_1:
freq_data = self.freqs[i_0:i_1 + 1]
fil_data = np.squeeze(self.data[time_start:time_stop, ..., i_0:i_1 + 1])
else:
freq_data = self.freqs[i_1:i_0 + 1]
fil_data = np.squeeze(self.data[time_start:time_stop, ..., i_1:i_0 + 1])
return freq_data, fil_data
def plot_spectrum(self, i=0, freq_start=None, freq_stop=None):
"""
Method for plotting the data and its shape
TO BE REPLACED
"""
plot_freq, plot_data = self.select_data(freq_start, freq_stop)
# arrange frequency ascending
if self.header[b'foff'] < 0:
plot_data = plot_data[..., ::-1]
plot_freq = plot_freq[::-1]
# select iteration i
plot_data = plot_data[i]
dec_fac_x = 1
plot_data = rebin(plot_data, dec_fac_x, 1)
plot_freq = rebin(plot_freq, dec_fac_x, 1)
plt.plot(plot_freq, plot_data)
plt.xlim(plot_freq[0], plot_freq[-1])
plt.ylabel('Power')
plt.xlabel('Frequency')
plt.show()
def rebin(data, n_x, n_y=None):
"""
Put data in bins
"""
if data.ndim == 2:
if n_y is None:
n_y = 1
if n_x is None:
n_x = 1
data = data[:int(data.shape[0] // n_x) * n_x, :int(data.shape[1] // n_y) * n_y]
data = data.reshape((data.shape[0] // n_x, n_x, data.shape[1] // n_y, n_y))
data = data.mean(axis=3)
data = data.mean(axis=1)
elif data.ndim == 1:
data = data[:int(data.shape[0] // n_x) * n_x]
data = data.reshape((data.shape[0] // n_x, n_x))
data = data.mean(axis=1)
else:
raise RuntimeError('Only 2 dimensions supported')
return data