-
Notifications
You must be signed in to change notification settings - Fork 2
/
Copy pathNEWS
143 lines (84 loc) · 4.87 KB
/
NEWS
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
# NEWS
# news information for LuMP
#_HIST DATE NAME PLACE INFO
# 2013 Sep 22 James M Anderson --- MPIfR start
# 2014 Sep 23 JMA --- GFZ update for LuMP 2.3.0
################################################################################
# 2014 Sep 23 JMA --- LuMP v2.3.0
This version fixes some bugs from the previous version (including the
bug that Stefan reported about a week ago --- fixed in my version for
several months now). A full ChangeLog is still in progress. However,
here is a list of major subsystems:
*** Pulsar recording: There are no significant changes here, except
for minor bug fixes that have not affected the data actually going
into dspsr. The system still supports outputting subbands to disk
separately (LuMP0), groups of subbands to disk (LuMP1), and piping
the data directly into dspsr with no storage to disk (LuMP2).
*** VDIF: a new writing mode has been enabled for writing out the
LOFAR beamformed data in the VLBI Data Interchange Format (VDIF),
so that correlating GLOW data with something like DiFX should be
made a bit easier. (DiFX will still need to be modified to
understand the exact form of VDIF that I have implemented, as
LOFAR beamformed data break some of the rules of the VDIF format
requirements, but...)
*** Channelization: Multiple modes for performing channelization and
optionally to convert the voltage data to power and integrate over
arbitrary lengths of time have been enabled and tested (as far as
I can without access to a real datastream). Both FFT and PFB
implementations are available (you probably want the PFB
implementation to deal with RFI leakage), with a huge assortment
of weighting functions to choose from for both. On my 4 core
i5-2500K system at home (computer from 2011), I can generate 16
channels per subband for all subbands in a LOFAR station lane (1/4
of the whole station) in faster than real time for both the 16 and
8 bit modes for the PFB integrating to power mode. Processing 244
subbands for the 4 bit mode takes about 15% longer than the length
of time of the input data on my home computer, so it is also close
to being realtime. For the lofarBN and lofarCN machines at the
MPIfR, there should be no problem in being able to keep up with
realtime for the 4 bit mode with 16 channels per subband. As the
lofarAN machines are a bit older but have more cores, I cannot
predict whether or not they will be able to keep up. The
channelization output modes of LuMP might be interesting for the
Solar group to get the resolution they want for single station
observations, as my version of LuMP can easily do << 100 kHz
frequency resolution and 1 ms integration times. Other options
(looking for carbon lines, ...) might be possible too.
# 2013 Sep 22 JMA --- LuMP v2.0 alpha
LuMP is now in its v2.0 alpha release stage.
New features in v2.0:
New recording formats:
LuMP1: pulsar recording format for processing with dspsr that can
output multiple beamlets per file
FFT0: discrete Fourier transform channelization of LOFAR beamlet data
PFB0: polyphase filterbank channelization of LOFAR beamlet data
P_FFT0: power (as opposed to raw voltage) recording
format using Fourier transform channelization
P_PFB0: power (as opposed to raw voltage) recording
format using polyphase filterbank channelization
The logging system has been extensively modified
Each beamlet may be assigned a separate RCUMODE value, as opposed
to LuMP v1.0 which assumed that all beamlets had the same
RCUMODE. This allows multiple RCUMODEs to be recorded
simultaneously (such as the LOFAR 357 mode).
Pointing directions and epoch information for each beamlet are now
provided to the program, and recorded for output formats that
support them.
Bugfixes in v2.0:
The reference epoch was off by 1.0 seconds (exactly) in LuMP v1.0
when LuMP was started on an odd second using the 200 MHz sampling
mode of the LOFAR station.
################################################################################
# Copyright (c) 2014, James M. Anderson <[email protected]>
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
################################################################################