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
|
import pmt
import numpy as np
from gnuradio import gr
class blk(gr.sync_block):
"""
Apply phase and frequency correction where there is a correlation peak tag.
The correlation peak tags are NOT propagated, and instead replaced with a
frame_start tag.
"""
def __init__(self):
gr.sync_block.__init__(
self,
name='Phase and Frequency Correction',
in_sig=[np.complex64],
out_sig=[np.complex64]
)
# tags should not be propagated, we then output our own tags
self.set_tag_propagation_policy(gr.TPP_DONT)
# because we do block processing, we need to keep track of the last tag
# of the previous block to correct the first values of the next block
self.last = None
self.lastfreq = 0
def block_phase(self, start, end):
"""
Compute a vector for the phase and frequency correction for the samples
between two tags (start and end).
@param start Tag where the samples should start to be corrected
@param end Tag where to stop correcting
@return A vector of phase values for each sample. To correct the samples
the data should be multiplied with np.exp(-1j * phase)
"""
# compute number of samples between tags
nsamples = end.offset - start.offset
# unpack pmt values into start and end phase
sphase = pmt.to_python(start.value)
ephase = pmt.to_python(end.value)
# compute frequency offset between start and end
phasediff = (ephase - sphase) % (2 * np.pi)
freq = phasediff / nsamples
# save this one for the last block (see variable `end' in self.work)
self.lastfreq = freq
# debugging
print(f"Correction for chunk of {nsamples:2d} samples is " \
f"sphase={sphase: .4f} rad and freq={freq*1e3: .4f} milli rad / sample")
# compute chunk values
return sphase * np.ones(nsamples) + freq * np.arange(0, nsamples)
def work(self, input_items, output_items):
counter = self.nitems_written(0)
# nicer aliases
inp = input_items[0]
out = output_items[0]
# read phase tags
is_phase = lambda tag: pmt.to_python(tag.key) == "phase_est"
tags = list(filter(is_phase, self.get_tags_in_window(0, 0, len(inp))))
if not tags:
print(f"There were no tags in {len(inp)} samples!")
out[:] = inp
return len(out)
# debugging
print(f"Processing {len(tags)} tags = {tags[-1].offset - tags[0].offset} " \
f"samples out of {len(inp)} input samples")
# compute "the middle"
enough_samples = lambda pair: ((pair[1].offset - pair[0].offset) > 0)
pairs = list(filter(enough_samples, zip(tags, tags[1:])))
chunks = [ self.block_phase(start, end) for (start, end) in pairs ]
middle = np.concatenate(chunks) if chunks else []
# compute values at the end, we do not have informations about the future
# but we can use the frequency of the last tag to approximate
nback = len(inp) - (tags[-1].offset - counter)
print(f"Processing {nback} samples at the back of the buffer")
end = np.ones(nback) * pmt.to_python(tags[-1].value) \
+ self.lastfreq * np.arange(0, nback)
# compute the "start", using the last tag from the previous call
nfront = tags[0].offset - counter
print(f"Processing {nfront} samples at the front of the buffer")
start = self.block_phase(self.last, tags[0])[-nfront:] \
if self.last and nfront else np.zeros(nfront)
# compute correction
correction = np.exp(-1j * np.concatenate([start, middle, end]))
length = len(correction)
# write outputs
out[:length] = inp[:length] * correction
# save last tag for next call
self.last = tags[-1]
# add tags
for tag in tags:
self.add_item_tag(0, tag.offset, pmt.intern("frame_start"), pmt.PMT_T)
# FIXME: should return `length' but then the last sample is not
# included and self.last does something weird
return len(out)
|
