eprintid: 716
rev_number: 8
eprint_status: archive
userid: 17
dir: disk0/00/00/07/16
datestamp: 2018-05-27 16:44:17
lastmod: 2018-05-27 16:44:17
status_changed: 2018-05-27 16:44:17
type: report
metadata_visibility: show
creators_name: Armour, Tristram
creators_name: Andrianjafinandrasana, Misaina
creators_name: Barton, David
creators_name: Billingham, John
creators_name: Sibley, David
creators_name: Theil, Florian
creators_name: Peremezhney, Nicholai
creators_name: Przemyslaw, Switalski
creators_name: Ranner, Tom
corp_creators: Neil Cade
corp_creators: Neil Brearley
title: Emitter-Platform Association
ispublished: pub
subjects: aerodef
studygroups: esgi73
companyname: Selex Galileo Ltd
full_text_status: public
abstract: Intercepted RF electromagnetic signals provide a good long-ranged source of information on the motions and activities of people, vehicles, instal- lations and organisations. For those emissions that are detected, tra- ditional tracking methods are used to associate the separate low level interceptions and average their characteristics to obtain tracks of the source location and characteristic patterns of the emissions. The Study Group was asked to provide a prediction of the number of underlying source platforms and the association between the emissions and plat- forms.
problem_statement: Intercepted RF electromagnetic signals provide a good long-ranged source of information on the motions and activities of people, vehicles, installa- tions and organisations. Such sources range from mobile phones at the low frequency end, though surveillance radar (air traffic control) to millimetre guidance radar (car collision avoidance), all of which produce intermittent pulse signals of varying frequency, inter-pulse timing and pulse shape. This is a very broad spectrum and typically, broadband RF receivers will potentially be able to detect several hundreds of sources at any time in- stant. For those emissions that are detected, traditional tracking methods are used to associate the separate low level interceptions and average their characteristics to obtain tracks of the source location and characteristic patterns of the emissions.
(1.1.2) This traditional tracking problem is complicated for several reasons:
  The emissions are sporadic, consisting of short bursts of emission interspersed with long quiescent intervals.
  The individual sources have multiple modes of operation (e.g. mobile phones may be transmitting voice or data station-polling signals).
  Location information is available in terms of noisy measurements of azimuth, elevation angle and range.
  Platforms can have multiple sources of emission (e.g. a ship may have several different types of Radar, or a bus may have several people using their mobile phones at the same time).
  The interceptor sensors are also likely to be on moving platforms and will not necessarily have a consistent visibility of the sources (occlusion, multi-path, etc).
  The emission is dense enough that the emission patterns from dif- ferent sources are bound to overlap with at least one other source.
Therefore it is inevitable that the traditional tracking will have introduced some additional track errors from mis-associations that in turn result in incorrect classifications and location distortions. Given intercepted Radio Frequency (RF) emissions, the Study Group was asked to provide a pre- diction of the number of underlying source platforms and the association between the emissions and platforms.
(1.1.3) The proposed study took this emitter track data from the intercepted RF emissions as given. The multi-target track data is a sequence of time- stamped state vectors comprising continuous (angle) components, esti- mates of the maximum possible ranges of the data sources, an identity that associates the individual emission belonging to the separate tracks
1
Emitter-Platform Association ESGI73
 and the associated uncertainties of these characteristics. A simulation was written to create this data.
(1.1.4) The task was to provide a best many to many match, supported by some measure of the quality of match, between the emissions and potential platforms at all times, on the basis of previously seen data. The issues that needed to be addressed include:
  The emission sequences associated with a single identity may be wrong. For example the same type of emission might come from multiple platforms of the same type and may therefore have been incorrectly associated with a single track.
  Platforms of the same type can have very different emissions.
  Platforms can have emissions overlapping in time.
  Ambiguities may become resolved as the targets approach the sensor system or as different platforms move relative to each other.
  The accuracy of the track data can vary greatly between different tracks and over the evolution history of a single track.
  The need to avoid discontinuous jumps in the mappings as time evolves. Ultimately, the primary interest is in the underlying plat- forms and it is particularly disconcerting if the solution chatters between almost equally likely alternatives.
date: 2010
citation:   Armour, Tristram and Andrianjafinandrasana, Misaina and Barton, David and Billingham, John and Sibley, David and Theil, Florian and Peremezhney, Nicholai and Przemyslaw, Switalski and Ranner, Tom  (2010) Emitter-Platform Association.  [Study Group Report]     
document_url: http://miis.maths.ox.ac.uk/miis/716/1/ESGI73-Selex_CaseStudy.pdf