BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Chicago
X-LIC-LOCATION:America/Chicago
BEGIN:DAYLIGHT
TZOFFSETFROM:-0600
TZOFFSETTO:-0500
TZNAME:CDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0500
TZOFFSETTO:-0600
TZNAME:CST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260522T150117Z
LOCATION:D220
DTSTART;TZID=America/Chicago:20181112T140000
DTEND;TZID=America/Chicago:20181112T143000
UID:submissions.supercomputing.org_SC18_sess172_ws_phpsc102@linklings.com
SUMMARY:Accelerating the Signal Alignment Process in Time-Evolving Geometr
 ies Using Python
DESCRIPTION:Vinay B. Ramakrishnaiah and Zachary K. Baker (Los Alamos Natio
 nal Laboratory)\n\nThis paper addresses the computational challenges invol
 ved in postprocessing of signals received using multiple collectors (satel
 lites). Multiple low cost, small sized satellites can be used as dynamic b
 eamforming arrays (DBA) in remote sensing satellites. This usually require
 s precise metrology and synchronized clocks. In order to mitigate this req
 uirement, correlation searches can be performed across time and frequency 
 offset values to align the signal. However, this process can take consider
 able time on traditional CPUs. We explore the use of heterogeneous paralle
 l architectures to expedite the computation process, while trying to maint
 ain the flexibility and ease of development using Python.\n\nThe Cross-Amb
 iguity Function (CAF) is used to perform correlation searches across a ran
 ge of all possible frequency differences of arrival and time differences o
 f arrival for a given emitter-collector geometry, followed by a phase alig
 nment search. For evolving geometries, maintaining the signal alignment ov
 er long time periods require time evolving CAF searches, which is computat
 ionally expensive. Consequently, we explore the use of massively parallel 
 architectures using both distributed and shared memory parallelism, and sh
 ow performance results. We also propose a simple load balancing scheme for
  efficient use of heterogenous architectures.\n\nWe show that the NumPy im
 plementation provides the same performance as the compiled Armadillo C++ c
 ode. Using different optimization techniques, the results show a performan
 ce improvement of 150x on a GPU compared to the naive implementation on a 
 CPU.\n\nTag: Parallel Application Frameworks, Reproducibility, Scientific 
 Computing\n\nRegistration Category: Workshop Reg Pass\n\n
END:VEVENT
END:VCALENDAR