% an updated historical narrative for HOPS

At the heart of the VLBI technique is the correlation of the raw
station data using either dedicated hardware or software to find
the correlated signal from the cosmic source.  The correlation is
manifest as an interference fringe that changes in an expected way
as the Earth rotates.  This is a simple, but (computationally)
expensive process that requires good, but nevertheless approximate,
models in order to obtain useful a useful fringe.  Thus some
post-correlation processing software is required to analyze the
fringes to obtain scientifically useful results.

The current Haystack Observatory Postprocessing system (HOPS) was
born from the efforts of Alan Rogers in the late 70's with a program
called FRNGE which was written in Fortran and designed to be efficient
on an HP-21MX (later renamed HP-1000) minicomputer.  With improvements
in hardware and software, a rewrite of the toolset was launched in
the early 90's by Colin Lonsdale, Roger Cappallo and Cris Niell as
driven by the needs of the geodetic community.  The basic algorithms
were adopted from FRNGE; but there was a complete rewrite of the code
into (K&R) C and substantial revisions of the i/o, control and file
structures resulting in the framework of the current HOPS system.
This was followed by a substantial effort in the early-mid
00's to develop tools for optimizing SNR and deriving correction factors
for data with imperfect coherence, based on analysis of amplitude with
coherent averaging time.  Further evolution in the late 00's was provoked
by the re-emergence of software correlation (DiFX), and in the 10's by the
needs of EHT-scale mm-VLBI which brings us to HOPS in its current form.

Acknowledging its geodetic heritage, HOPS was optimized for precision
on per-baseline delay and delay-rate measurements which are the raw
material for the geodetic analysis programs.  Consequently, it is
somewhat light on support for some routine calibration processes found
in some other astronomical software packages (e.g. AIPS or CASA).
Nevertheless, it provides a good framework for the reduction and
analysis of mm-VLBI data, where the vagaries of atmospheric effects
require ever more specialized processing to harvest significant
astronomical results.

For the needs of the EHT Campaigns of 2017 and subsequent years it
was decided to augment the basic HOPS framework with python-based
packages as described in the next sections.....

Whitney, AR & Cappallo, R & Aldrich, W & Anderson, B & Bos, A & Casse, J & Goodman, J & Parsley, S & Pogrebenko, S & Schilizzi, R & Smythe, D. (2004). Mark 4 VLBI correlator: Architecture and algorithms. Radio Science - RADIO SCI. 39. 10.1029/2002RS002820. 


% eof