Оцінювання ефективності багатопозиційної системи спостереження на основі додаткового оброблення даних від сенсорів зі змінюваним місцеположенням
The efficiency of a multi-position system depends on the realization of its struc-ture — how many elements it includes, where they are located, and how the envi-ronment and terrain influence its operation. The paper is dedicated to data pro-cessing in a multi-position surveillance system as an addit...
Saved in:
| Date: | 2025 |
|---|---|
| Main Authors: | , , , , |
| Format: | Article |
| Language: | English |
| Published: |
The National Technical University of Ukraine "Igor Sikorsky Kyiv Polytechnic Institute"
2025
|
| Subjects: | |
| Online Access: | https://journal.iasa.kpi.ua/article/view/351408 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Journal Title: | System research and information technologies |
| Download file: |  |
Institution
System research and information technologies| _version_ | 1866303057651826688 |
|---|---|
| author | Tymchuk, Volodymyr Mediakov, Oleksandr Popov, Oleksandr Trysnyuk, Taras Tsybulia, Serhii |
| author_facet | Tymchuk, Volodymyr Mediakov, Oleksandr Popov, Oleksandr Trysnyuk, Taras Tsybulia, Serhii |
| author_sort | Tymchuk, Volodymyr |
| baseUrl_str | http://journal.iasa.kpi.ua/oai |
| collection | OJS |
| datestamp_date | 2026-02-02T20:49:24Z |
| description | The efficiency of a multi-position system depends on the realization of its struc-ture — how many elements it includes, where they are located, and how the envi-ronment and terrain influence its operation. The paper is dedicated to data pro-cessing in a multi-position surveillance system as an additional option, leverag-ing the in-between big data from the system’s elements. A sufficient number of numerical data generated by the multi-position system and its elements–sensors–allows the use of statistical methods and models from machine learning or deep learning. The ontology for quality estimation of the multi-position sys-tem, depending on its configurations, is proposed. The results of the distribu-tions of detected events are presented in graphical forms that allow statistical evaluation of the distributed data. Our findings allow us to ensure the efficiency of a multi-position system in an unpredictable, variable environment by recon-figuring it when it offers better capabilities. |
| doi_str_mv | 10.20535/SRIT.2308-8893.2025.4.02 |
| first_indexed | 2026-02-08T08:06:11Z |
| format | Article |
| fulltext |
V.Yu. Tymchuk, O.O. Mediakov, O.O. Popov, T.V. Trysnyuk, S.A. Tsybulia, 2025
20 ISSN 1681–6048 System Research & Information Technologies, 2025, № 4
TIДC
МАТЕМАТИЧНІ МЕТОДИ, МОДЕЛІ,
ПРОБЛЕМИ І ТЕХНОЛОГІЇ ДОСЛІДЖЕННЯ
СКЛАДНИХ СИСТЕМ
UDC 623.4/519.2+004.75
DOI: 10.20535/SRIT.2308-8893.2025.4.02
THE RESULTS OF THE MULTI-POSITION SURVEILLANCE
SYSTEM’S EFFICIENCY, DEPENDING ON THE LOCATIONS OF
ITS SENSORS, USING ADDITIONAL DATA PROCESSING
V.Yu. TYMCHUK, O.O. MEDIAKOV, O.O. POPOV,
T.V. TRYSNYUK, S.A. TSYBULIA
Abstract. The efficiency of a multi-position system depends on the realization of its
structure—how many elements it includes, where they are located, and how the en-
vironment and terrain influence its operation. The paper is dedicated to data process-
ing in a multi-position surveillance system as an additional option, leveraging the in-
between big data from the system’s elements. A sufficient number of numerical data
generated by the multi-position system and its elements–sensors–allows the use of
statistical methods and models from machine learning or deep learning. The ontology
for quality estimation of the multi-position system, depending on its configurations,
is proposed. The results of the distributions of detected events are presented in
graphical forms that allow statistical evaluation of the distributed data. Our findings
allow us to ensure the efficiency of a multi-position system in an unpredictable,
variable environment by reconfiguring it when it offers better capabilities.
Keywords: data processing, surveillance, detection, multi-position system, system-
of-systems, efficiency.
INTRODUCTION
Being multidisciplinary, science is able to cover different areas simultaneously.
Our research corresponds to this feature — there are several independent areas
from System Engineering, Estimation theory, Geospatial Intelligence, Big Data
Processing, Machine Learning (ML), Deep Learning (DL). For example, there is a
well-known problem of accuracy in Detection theory that has limits when using
its methods and algorithms, but due to collaborating with methods and algorithms
from other scientific fields such limits would be overcome. Another example is a
problem of detection system’s design to be optimized for unpredictable and vari-
able environment. The mentioned problem takes place for multi-positional detec-
tion system. It is a very difficult mission to define the effective structure of multi-
positional detection system especially during warfare. This research investigates a
method for evaluating a detection system's structure based on its operational re-
sults, — the results for in-between data (not the results of direct detection) from
elements of multi-positional detection system are presented. As usual, any system
The results of the multi-position surveillance system’s efficiency, depending on the locations …
Системні дослідження та інформаційні технології, 2025, № 4 21
produces a great part of in-between data that could be additional information
when it is necessary to improve the efficiency of the system. The volume of in-
between data is enormous, so it makes possible to use Big Data Processing, ML
or DL. So, the data preprocessing and statistical analysis were used to present the
detection results on additional grid coordinate system and to make a statistical
evaluation for the data.
PROBLEM STATEMENT
The theory and practice of developing multi-position systems have been well-
studied for a long time [1]. The problem to be solved is formulated as follows.
Having some specific uncertain environment, the scientific and technical task is to
register certain changes in this environment using technical solutions imple-
mented in the systems. There are many areas of application for such a problem —
studying nature and space, medicine, monitoring the security situation at infra-
structure facilities, military monitoring systems [2]. Obviously, there are different
well-known kinds of passive multi-position surveillance system (MPSS) for
acoustic location, GPS tracking, imagery intelligence, seismic reflection, signals
intelligence, thermal imaging, underwater acoustics, video surveillance, wireless
tracking etc. The physical nature for each method is unique, but there are some
common similarities: 1) waves (signals); 2) a set of distributed synchronous sen-
sors; 3) a great volume of measurement results — the datasets.
Therefore, if a certain system has already been developed and is functioning,
then the main next task is to ensure the best (optimal) way to process information
(data, signals).
In modern systems, whether specialized functional systems [3; 4], complex
multi-profile systems [5], or systems of systems [6] — signal and information
processing occurs across multiple stages. These processes are diverse and often
complementary, but the specific way used to identify them is less critical than
their integration. Regardless of the system's architecture, an "integrated" processing
pipeline inevitably incorporates Big Data analytics or the processing of large volumes
of homogeneous data alongside other methods, algorithms, and software solutions [7].
The purpose of the research. Our goal is to make a qualitive analysis of the
efficiency of multi-position system using data preprocessing and statistical data
analysis to change its configuration improving the detection possibilities.
The general aspects of a configuration of a multi-position surveillance system
A passive MPSS with a set of typical sensor posts (SPs) for terrain monitoring
and/or control is done. The configuration of MPSS is the combination of all SPs
on the terrain in some grid-like projection. The example of configuration is on the
Fig. 1, where from 1 to 9 are the typical SPs being synchronous in a passive
MPSS. The SPs are distributed on the terrain in some way, on the picture it is
used a distance of 1.5 km between two sensors in the first (upper) line of the con-
figuration and a distance of 1 km between first and second (below) lines. The
MPSS base line is the distance between two utmost SPs (here are SP1 and SP6).
The sensitivity for a MPSS means the possibility to detect some signals (on the
potentially attainable distance for the signal with some defined level (power)).
Here the range area is a circle of 25 km in radius (for the optimal conditions).
V.Yu. Tymchuk, O.O. Mediakov, O.O. Popov, T.V. Trysnyuk, S.A. Tsybulia
ISSN 1681–6048 System Research & Information Technologies, 2025, № 4 22
As it is clear from the Fig. 1, the potential area of MPSS’s sensitivity may
have some form with 25 km in depth for any side direction from the utmost sen-
sors. The two lines in MPSS’s configuration allow for the determination of the
Area of Interest (AoI) by excluding, for example, signals from back (useless) di-
rections. Such a task is typical in security or military applications where both
friendly (allied) and adversary forces are present. The optimal AoI for some gen-
eral configuration and conditions should have some right form (Fig. 2). The or-
thogonal line (OL) to base line shows a main direction of AoI (main surveillance
angle (MSA) is an angle between OL and main danger direction (MDD)).
A note: 1 — The typical sensors 2–5 from the upper line (Fig. 1), sensors 7 and 9
from the below line are missing here to have a simple picture. MSA here is 0 de-
gree that is the best for AoI
In real-world conditions, an ideal AoI does not exist. The first reason for it is
the terrain features where the MPSS’s configuration is set — there are no possi-
bilities to establish typical sensors on the straight line (or two lines) with even
distance between sensors. Further, the real terrain is characterizing with relief and
natural or artificial coverings — the such reality effects on the MPSS’s sensitiv-
ity. The second reason is some secure restrictions which are inevitable for war-
fare. The technical challenges (such as the problems with a sensor’s operating or
with some kind of destroying in warfare) are the next factor that makes an effect
on the MPSS. So, the real MPSS’s configuration has a non-optimal decision with
a smaller number of sensors in a usage (Fig. 3) and it results in a less achievable AoI
(Fig. 4).
Fig. 1. The example of possible configuration for a typical MPSS (the sketch is not at
scale)
Fig. 2. The form of the MPSS’s sensitivity (dotted area)
The results of the multi-position surveillance system’s efficiency, depending on the locations …
Системні дослідження та інформаційні технології, 2025, № 4 23
So, for such occasional conditions of systems’ utilization there is not just a
technical optimization problem (which defends only from specifications of a sys-
tem), but a procedure one too, especially how to get the best approaches for next
processing of ongoing datasets during the time of MPSS’s functionating.
The acoustic location multi-position surveillance system
The stages of data preprocessing and data processing are presented on the exam-
ple of an acoustic location MPSS.
The principles of acoustic MPSS are well-known: due to automatized detec-
tion and classification procedures, sensitive synchronous sensors and wireless
communications the determination of the sound source from the muzzle wave
Times of Arrival is a simple task for such kind of MPSS [8]. But the problems of
the sound source location accuracy and even the recognition in typical circum-
stances when the various waves emitted by and during a shot (from many sound
sources) are still actual. The reasons have the nature character — the initial pro-
jectile characteristics (the whole variety of its aerodynamic and ballistic coef-
ficients), the range (Fig. 5), the atmosphere (with atmospheric wind (Fig. 6) and
Fig. 4. The variant of the real AoI for s MPSS’s (MPSS’s sensitivity)
Fig. 3. The example of a real MPSS’s configuration for some terrain and secure conditions
(the sketch is not at scale)
Fig. 5. The attenuation of the sound wave during its propagation in the atmosphere [11]
V.Yu. Tymchuk, O.O. Mediakov, O.O. Popov, T.V. Trysnyuk, S.A. Tsybulia
ISSN 1681–6048 System Research & Information Technologies, 2025, № 4 24
sound speed gradients), the ground, possible obstacles (woods, buildings, hills…,
(Fig. 7)), refraction or air absorption, wave alterations, multipath arrival of the
ballistic shock wave and so on. So, the outcome that the localization performance
is affected, sometimes critically [9], is firm. Even more, the practice with highly
intensive shot conditions that took place in Ukraine since 2022-24-02 [10], ap-
proved this statement.
Therefore, many modern researchers are dedicated to improving the accu-
racy of acoustic MPSS in sound source localization. There are two main
directions of the efforts: 1) to improve the established approaches (due to
technical and different tools’ progress, and processing method with some more
efficiency [8; 9; 12; 13]); 2) to search new approaches for problem solving. The
example of the second direction is an application of artificial intelligence —
convolutional recurrent neural networks in [14] or other neural networks on the
proposed software-mathematical models [15], depending on the shape of the
location of the sensors (MPSS’s configuration), the distance between SP, their
number, the parameters of the neural network (the number of hidden layers and
neurons), and the volume of the dataset for training. The authors proclaimed that
their results for the neural network training algorithm ensure the average value of
the absolute error in determining the grid coordinate are not exceed about 1 m and
maximum absolute error value are not exceed 16 m for the Y grid coordinate (it
corresponds the determination in depth) and 4.5 m for the X grid coordinate (it
corresponds the determination in front) for the range 1800 m between sensors and
a source of acoustic signals. Even if it is a fact for testing conditions (optimal
Fig. 7. The probable role of obstacles on the sound wave path in poor operation of a
MPSS’ sensor [11]
Fig. 6. The wind influence on the sound wave and source’s localization [11]
The results of the multi-position surveillance system’s efficiency, depending on the locations …
Системні дослідження та інформаційні технології, 2025, № 4 25
terrain specification and for an ideal configuration (Fig. 1) it is not useful for real
circumstances, because the practical range between SP and a sound source is in
5…15 times much more, that means the average or maximum absolute error
values are increasing properly — up to 50…300 m.
To approve the mentioned statement the field data are presented below.
Some field data from the acoustic MPSS and their explanation
The type of acoustic MPSS for the gathering data in field conditions is HALO
[11]. According to the specifications of the HALO system the average value of
the absolute error should be 100 m on the range up to 100 m and 0.7% of the
range when the distance to the sound source exceeds 15 km.
The configuration for the HALO was occasional (Fig. 3) and included or 4,
or 5, or 6 sensors with distance between them from 1.5 km till 4 km.
The field results for initial utilization of HALO are shown in the Table 1.
T a b l e 1 . The initial data about HALO’s accuracy
Number of sensors with the base line distance
5 SPs for 13.8 km or 6 SPs for 16.5 km 4 SPs for 11.2 km
MSA (see Fig. 4)
Range between
MPSS and
sound source
0…±30 º ±30…±60 º 0…±30 º
The average value of the absolute error
7…10 km 200…300 m 500…700 m 500…800 m
10…15 km 400…500 m 1000 m 1000…1500 m
15…20 km 600…800 m 1500…2000 m -
20…30 km 1000 m - -
A note: 1 — The time period of observation is a month (since 2023-05-04 till
2023-08-05).
To estimate the value of the error it was possible to use the radar system for the same
sound sources. The type of the radar is AN/TPQ-36 with accuracy up to 50 m [16].
T a b l e 2 . The comparison data between Radar’s and HALO’s target grid co-
ordinates
Detection time Nr.
HALO Radar
Location
difference, m
Range (between SPs of MPSS
and sound source), m
1 23:24 23:30 818 22240 26920
2 23:25 23:30 877 - // - - // -
3 10:15 10:20 391 15310 18830
4 10:18 10:20 151 - // - - // -
5 10:42 10:55 108 15480 19020
6 10:46 10:55 46 - // - - // -
7 10:38 10:30 98 13320 16720
8 10:39 10:30 198 - // - - // -
9 16:21 16:20 513 21790 26400
10 08:36 08:38 145 15480 18990
11 08:38 08:38 419 - // - - // -
12 08:19 08:20 153 15390 18940
13 08:21 08:20 226 - // - - // -
14 08:23 08:20 446 - // - - // -
V.Yu. Tymchuk, O.O. Mediakov, O.O. Popov, T.V. Trysnyuk, S.A. Tsybulia
ISSN 1681–6048 System Research & Information Technologies, 2025, № 4 26
It is obviously that the statistical data for reliable conclusions are too little.
But the main value of the data is their real (utilization) nature, not experimental.
It should be noted that the influence of meteorological data (which is re-
quired for HALO) was not considered. The average data are presented in the Ta-
ble 3.
T a b l e 3 . The average data of HALO’s accuracy during the field utilization
Range between SP
and sound source
Manual’s meaning of
HALO’s error
The deviation diapason of sound
source’s coordinates from actual ones
15…20 km 105…140 m up to 100 m
20…30 km 140…210 m 500…900 m
A note: 1 — The MPSS consists of six SPs, the MPSS’s base line is 16.5 km, the MSA is
0…±30º.
So, the average error values may correspond to manual’s ones (according to
HALO’s specifications) on the ranges up to 20 km. But it is not enough for direct
application the surveillance data on next stages of decision-making. So, it is nec-
essary to find approaches how to decrease average and maximum absolute error
values.
The input conditions and restrictions
The objective of monitoring by means of MPSS is to determine the location of the
targets after at least two systems’ sensors produced information about detection of
the target.
The MPSS consists of 6 sensors [10].
The fact of generating some kind of signal that was received/detected by a
sensor(s) will be called an event.
The detected signal is the information about event transformed in some digi-
tal form including some specific features about the event, like time and potential
coordinates.
The digital event set (DES) is the ordered combination of numbers that is
generated by a sensor detecting the signal. Normally the DES includes time of
detection, coordinates of a sensor that detects a signal and coordinates of an event,
derived from some signal and information processing.
The proposed dataset is the collection of DES collected during some period
of MPSS’s operating. Several dataset sources can be used, so the corresponding
indication of those sources is included where needed.
The size of dataset is more than 17 000 rows. The example of the raw dataset
is shown on Fig. 8 (its description is given in the note under the figure’s title). So,
the presented digital data for the data preprocessing and EDA were used similar to
[17]. Gaining the ability to evaluate the performance of the MPSS depending on
the configuration is easily (to a certain degree) doable.
Time of occurrence of the event isn’t taken into account for the statistical
analysis.
The common operational picture (COP) is a presentation of all detected
events on some grid system.
The results of the multi-position surveillance system’s efficiency, depending on the locations …
Системні дослідження та інформаційні технології, 2025, № 4 27
SOLVING THE PROBLEM OF INEFFICIENCY OF MULTI-POSITION
SYSTEM IN UNPREDICTABLE AND VARIABLE ENVIRONMENT
Experiment Design/Data Collection
To represent results of dataset processing including corresponding diagrams and
graphic elements in the best form some abbreviations and acronyms for the terms
and processes taking place in our research are proposed (Table 4).
T a b l e 4 . A physical essence of some processes in MPSS with designations
and symbols
The terms and their designation or symbolization
N
Term or characteristics Acronym Symbol
1 Configuration Conf. jConf
2 Number of sensors in a configuration – K
3 Sensor Post from a configuration with a sequence number1 k SP kSP
4 Configuration Duration2 CD Т
5 Configuration Sencitivity Terrain3 CST ), ( YX
6 Number of Events being Detected4 NED N
7 Event sequence number i i
8 Total Events Space Distribution TESD iW}{
9 Event Space-Gradient Map5 ESGM KW}
~
{
10 Locating Posts’ Number Portion6 LPNP nPˆ
11 Single Post Operating SOP kP
Notes: 1 — It is generated in MPSS automatically; 2 — Time period of MPSS’s operating
with some configuration; 3 — Some terrain (with defined size) that is being achievable for
MPSS to detect events; 4 — Using upper and lower indexes it is differed a NED for one
or another configuration and/or CDs (f.e., per a day, per a week, …); 5 — It shows the
terrain (as a gradient surface) in which events were detected by all SPs; 6 — The ratio of
number of SPs that detected each event to all SPs from a configuration
The methodology aspects for analysis of MPPS
Specialized geoinformatics system (GIS) software is usually integrated into
MPSS to display the current event situation [18]. Usually, such GIS is a 3D model
of terrain with a set of coordinate systems. In the research it was used a relative
simulated Cartesian coordinate system (CCS) that “covers” a CST and displays
the configuration in the center of it. So, all events detected by SPs are reproduced
on this simulated CCS.
The ontology of a configuration is depicted as
j
k
j KConf )}λ {,( , (1)
where j is a relative sequence number for possible configurations (there were 6
ones); К is the number of sensors in a configuration; λ & are the coordinates
of SP (a longitude and a latitude for kSP ). Sometimes the UTM-like coordinates
V.Yu. Tymchuk, O.O. Mediakov, O.O. Popov, T.V. Trysnyuk, S.A. Tsybulia
ISSN 1681–6048 System Research & Information Technologies, 2025, № 4 28
are used instead of latitude-longitude, but the conversion between those is a sim-
ple computational task:
.}{ }λ { YX (2)
ML system has a typical design for such systems [9].
The ontology for configuration quality estimation
A collection of several statistical characteristics that are used for demonstration
and estimation of how a specific configuration operates in a CST is called the
configuration quality (CQ) within the context of the paper.
TESD shows the density of detected events:
i
N
i
i tYXW
total
}, , {}{
ˆ
, (3)
where totalN̂ is a total NED for the j-configuration.
ESGM shows the terrain area where the events were detected by all SPs:
ki
K
kK WW /1 }{}
~
{ , (4)
where kiW /}{ is an event distribution detected by kSP .
Apparently, each event may be detected by a single SP or by a particular
combination of SPs. Consequently, LPNP is defined as:
totalnn NNP ˆ/ˆˆ , (5)
where Kn ˆ1 , meaning that nN ˆ is a sum of NED for the cases detected by a
combination of n̂ SPs (neglecting the actual composition of SPs in the combina-
tion, only the number of SPs is considered).
SOP shows how frequently each single SP detected events registered in the
dataset:
totalkk NNP ˆ/ , )1( Kk . (6)
Configuration quality CQ presentation
So, there were 6 configurations with different parameters according to (1). Each
configuration had a NED that is enough for ML (see depiction for figures). To
analyze CQs it was proposed to calculate and present five graph materials for
each configuration.
TESD presents the distribution of events that were detected in MPSS during
time period corresponding to configuration duration T. Each point for presentation
it was taken from geographic coordinates of corresponding row of dataset (Fig. 8)
according to (3). The geographic coordinates were transformed into grid one ac-
cording to (2). The grid system is conditional and it has no connection with any
known grid systems such as MGRS. It was got just for local terrain. The central
point for this grid is approximately in the geometric center of MPPS’s base that
includes all sensors. As common the form for MPSS’s base is similar to line. The
limits for terrain area are not far than the biggest distance (from central point to
event’s location) projected on grids. In the presented case there is a 50-km area
The results of the multi-position surveillance system’s efficiency, depending on the locations …
Системні дослідження та інформаційні технології, 2025, № 4 29
from the geometric center (in each of four grid directions). Due to terrain scale
restrictions, it was used a gradient scale to present the level of concentration de-
tected events in limited location areas. For example, gradient scale is from 0 to 80
on a Figs. 9, 11, 12, 13, 17, 18.
A note: 1 — MPSS generates automatically unique identification numbers for each sensor
that detects a target and for a whole system when at least two sensors detect the same tar-
get (the number of sensors that detect a joint target is ina column that is headed as nDet).
The corresponding identification numbers (ID from one of six sensors (they are defined as
SP_117, SP_118, etc.) and Loc ID from a MPSS are presented on the example —
1958699491, 1958633570, etc.). The time detection for a target and its coordinates —
geographic longitude and latitude — are also presented in the proper columns
Fig. 8. The example1 of real data from a MPSS
Fig. 9. The Total Events Space Distribution iW}{ (half-left), Event Space-Gradient Map
KW}
~
{ (top-right), Locating Posts’ Number Partion nP (middle-right) and Single Post
Operating kP (bottom-right) for Configuration Nr. 1 of MPSS ( dT , NED (corre-
sponds to the size of dataset) 5575N
V.Yu. Tymchuk, O.O. Mediakov, O.O. Popov, T.V. Trysnyuk, S.A. Tsybulia
ISSN 1681–6048 System Research & Information Technologies, 2025, № 4 30
The next graph of CQs (top-right) shows the gradient distribution of detected
events that were detected by all 6 sensors of MPSS simultaneously (or with un-
derstandable time delay) according to (4). The graph allows to estimate both some
aspects of MPSS’s configuration and terrain peculiarities. In first case the opera-
tor of MPSS may define ineffective sensors in case when the following result is
happened: the event is localized in the whole-area but there is no detected infor-
mation from one (or more) sensors of 6. In second case the operator could analyze
how the terrain influenced on MPSS’s possibilities and what variants of changing
the configuration would be useful for better detection.
The third illustration of CQs (middle-right) is a diagram showing what is a
portion of numbers of sensors in event detection according to (5). As it was men-
tioned in the Fig. 8’s note, there are unique identification numbers in the MPSS —
location ones and numbers from each of 6 sensors. The quantity of unique
identification numbers from all sensors corresponds to number of events. Some of
these events “becomes” targets — it happens when the event has two or more
identification numbers from sensors that makes possible event’s location with
simultaneous generating the location identification number. So, the diagram
shows the distribution of identification numbers depending on how many sensors
define events one-by-one. It is clear that a portion that is corresponded to a single
sensor high-probably identifies the occasional (low, bad, poor, week, strange)
signal. It is impossible to locate an event due to “detection” from a single sensor.
So, such information could inform the operator that configuration has some week
points and it will be good to reduce the portion of identification numbers from
one any sensor. The “ID-portions” for two or more sensors are normal for MPSS
because of variety of signals on sensors’ inputs — the range, terrain features,
source of the signal, etc. make probable possibilities for the sensors to detect
them. It is obviously that what a concrete combination for two (for three and part-
ly for four) sensors for one-by-one events is could guess the operator some other
configuration week points. But the decision needs more high computation for
such analysis — the distribution of identification numbers is multiplied
(5!+4!+3!=150), so it wasn’t executed for this paper.
The fourth diagram of CQs (bottom-right) shows what is a portion of num-
bers of sensors in event detection. In general, the diagram expresses (6).
The fifth graph material of CQs is a normalized contingency (crosstab) table
(see Figs. 10, 14, 15, 16, 19, 20) that shows the relative frequency of detections
made by a specific SP depending on SPs’ combination.
Fig. 10. The crosstab of kSP -detection for Configuration Nr. 1
The results of the multi-position surveillance system’s efficiency, depending on the locations …
Системні дослідження та інформаційні технології, 2025, № 4 31
Let’s show how to make the qualitive analysis of the illustrated data for this
configuration of MPSS (it is the Conf. Nr 1 above).
Each point on the graph for TESD iW}{ shows the sound event that took
place during the observation time. It doesn’t differ what is a source of the detected
sound event — gun fire moment or any kind of explosion caused by a shell, or by
a missile, or by a mine, even engine switch-on is possible to detect (it is clear be-
cause the SP has a stable determined sensitivity (Fig. 5)). For such circumstances
the COP depends on terrain features (relief forms and artificial and natural cover-
ings effect on sound wave propagation). So, making the COP’s observation dur-
ing the determined time period it is possible to get some qualitive analysis —
what concrete terrain areas have much more “signals” (they correspond to sound
events that were detected in MPSS) and what other terrain areas are “problem-
Fig. 11. CQs for 2Conf ( dT , dataset size 4589N ): iW}{ , KW}
~
{ , nP , kP
Fig. 12. CQs for 3Conf ( dT , dataset size 2492N ): iW}{ , KW}
~
{ , nP , kP
V.Yu. Tymchuk, O.O. Mediakov, O.O. Popov, T.V. Trysnyuk, S.A. Tsybulia
ISSN 1681–6048 System Research & Information Technologies, 2025, № 4 32
atic” (there are a smaller number of “signals” then it is expected for typical situa-
tion of the initial preposition that there are continuous uniform distributions for
signal events — in fact it is necessary to combine the intelligence data with other
sources to approve the preposition as it is shown in [17] but for this research it
doesn’t matter). For the “ideal” efficiency it is demanded to have the COP with no
blank terrain areas for the whole attention sector (it corresponds roughly to sur-
veillance area).
So, for the Fig. 9 it is possible to define some problematic areas (and the di-
rections from the MPSS’ configuration center) — they are in the south-east part
of the COP (the top-right figure for the ESGM KW}
~
{ shows the general depiction
for the COP during the time period observation).
Using the diagram for the SOP kP (see bottom-right) helps to see the “prob-
lematic” SP (SPs). Here for the Conf. Nr 1 the “problematic” SP is the “second”
— SP2 with 16,0kP (16.2%-participation in MPSS’ total detection although
other SPs give 50% or more results). So, the qualitive analysis for this case shows
that it is necessary to reset the SP2 on the terrain — to change its location.
Using the diagram for the LPNP nPˆ (see middle-right) allows to identify an-
other “problem” — how often single SP operates. The such qualitive analysis is a
base for next statistical and/or technical analysis because all SPs are identical
ones and sound wave for typical conditions (for all SPs of MPSS they are similar
at least) should be detected by two or more SPs (it depends on distance from the
source and angle of the direction of sound wave propagation). So, the qualitive
analysis for this case allows operator to define some “blind” directions or other
possible reasons that cause less more efficiency for MPSS. The outcome of such
analysis should be adjusting of a “problematic” SP (its sensitivity or other
organizing measures to stop the detection (to blank) of the unwilling or other
parasitic sound wave signals). For this task the normalized contingency table (Fig.
10) should be useful. Here it is presented that SP4 (with 2.2%), SP5 (with 3.1%)
and SP6 (with 2.1%) have higher portion of only theirs SOP (without co-
Fig. 13. CQs for 4Conf ( dT , dataset size 2535N ): iW}{ , KW}
~
{ , nPˆ , kP
The results of the multi-position surveillance system’s efficiency, depending on the locations …
Системні дослідження та інформаційні технології, 2025, № 4 33
detection). Naturally that the qualitive and statistical analysis should be made for
all combinations of different co-detections that took place in the MPSS — such
wide analysis of crosstab’s data from Fig. 3 will allow to define weak points of
the analyzed configuration and to search the ways of MPSS’s adjusting or new
variant of the configuration. The last approach is presented in the paper further-
more — there are five more variants of MPSS’s configuration with CQ’s calcu-
lated data for each configuration as presented on the Fig. 9 and 10.
Using the crosstab allows to find the SPs with the best “collaborative” fea-
tures (optimal co-detection). It means that for the detecting of the sound wave and
next determination of the sound source’s coordinates it is demanded to have the
detection from two or more sensors. So, what SPs have the high level of co-
detection for two sensors or for three sensors in different combinations it will give
the optimal configuration of the MPSS. It is a classic optimization problem for
non-optimal terrain accessibility.
So, following CQs are presented for other MPSS’s configurations — the fig-
ures are grouped correspondingly (see Fig. 11 & Fig. 18 for Conf. Nr. 2, Fig. 12
& Fig. 15 for Conf. Nr. 3, Fig. 13 & Fig. 16 for Conf. Nr. 4, Fig. 17 & Fig. 19 for
Conf. Nr. 5 and Fig. 18 & Fig. 20 for Conf. Nr. 6).
Fig. 14. The crosstab of
kSP -detection for 2Conf
Fig. 15. The crosstab of
kSP -detection for 3Conf
Fig. 16. The crosstab of
kSP -detection for 4Conf
Fig. 17. CQs for 5Conf ( dT 48 , dataset size N=124542): iW}{ , KW}
~
{ , nPˆ , kP
V.Yu. Tymchuk, O.O. Mediakov, O.O. Popov, T.V. Trysnyuk, S.A. Tsybulia
ISSN 1681–6048 System Research & Information Technologies, 2025, № 4 34
Having some field configurations (it means they were actual on the terrain)
itis possible using some qualitive and statistical analysis of the detection effi-
ciency depending on MPSS’s configuration and each or some SP(s) operation to
find the best (the optimal for real terrain or other circumstances) configuration
for the MPSS. The transforming of the configuration means that some SP(s) are
being reset on the terrain (on a new geographic location of the position).
Another way of increasing the MPSS efficiency is just the adjusting some features
of ‘problematic’ SP (sensitivity or blank directions (sectors) establishing).
The mode for MPSS’s transiting to achieve better detection possibilities
The obtained statistics allows to choose the configuration for MPSS for a particu-
lar terrain. For instance, using the computations for (5) and (6) the ‘weak’ points
for each configuration — ineffective SP (SPs) — are obtained. The results al-
lowed to build a priorities scheme (an ontology) to determine where the configu-
ration should be moved towards to — see Table 5.
Fig. 18. CQs for 6Conf ( dT 6 , dataset size N=4589): iW}{ , KW}
~
{ , nPˆ , kP
Fig. 19. The crosstab of kSP -detection
for 5Conf
Fig. 20. The crosstab of kSP -detection
for 6Conf
The results of the multi-position surveillance system’s efficiency, depending on the locations …
Системні дослідження та інформаційні технології, 2025, № 4 35
T a b l e 5 . CQs’ parameters for best configuration
LPNP jConf
for 2ˆ n for 3ˆ n for 1ˆn for 6ˆ n
SOP min
1Conf 31.71 18.53 10.12 0.39 SP2
2Conf 38.29 18.48 12.36 0 SP2, SP3
3Conf 31.85 18.42 18.74 0 SP5
4Conf 38.66 15.53 20.16 12.25 SP4
5Conf 45.67 17.87 0 0 -
6Conf 43.24 14.93 0 13.14 -
Prioritize min max none
The optimal configuration should be one with more portion of two and/or
three SPs, with “maximum LPNP” and absent zero or near-zero SOP.
For presented data the optimal configurations were 5Conf and 6Conf —
approximately the efficiency exceeds on 10…15% (it corresponds to valueless
part of single post operating).
CONCLUSIONS
In the research the variant of additional option in multi-position surveillance sys-
tem’s operating using datasets that are generated in sensors of the system is
showed.
One of the key characteristics is a system’s configuration which is possible
to transit to ensure better efficiency of detection by means of system’s sensors.
Both the methodology and the ontology for qualitive estimation of different sys-
tem’s configurations are proposed in the paper. The corresponding results are pre-
sented using graphic distribution of detected events and diagram for efficiency of
system’s elements including 6 sensors. The efficiency of MPSS exceeds on
10…15% due to defining the weakest point of multi-position surveillance system
and its following transition of system’s configuration.
Although the representations with event distributions and SPs’ frequencies
allow to find the best configuration for MPSS operating, the possibilities of the
results are wider. Next step of statistical analysis can be held with respect to the
time durations, like estimating average event distributions depending per period,
building periodograms and some event dispersions. The ontology and methodol-
ogy also will be explored in the next researches. In general, the gained results
would be useful for system engineering when it’s necessary to design both con-
crete system and system-of-systems including multi-position surveillance system,
communication system, GIS, machine learning or deep learning system, etc.
REFERENCES
1. “Theory and Practice of Control and Systems,” Proc. of the 6th IEEE Mediterranean
conf., Alghero, Sardinia, Italy, 9–11 June 1998, A. Tornambe, G. Conte and
A.M. Perdon, Eds, 1999, 864 p.
V.Yu. Tymchuk, O.O. Mediakov, O.O. Popov, T.V. Trysnyuk, S.A. Tsybulia
ISSN 1681–6048 System Research & Information Technologies, 2025, № 4 36
2. V. Lytvyn, R. Peleshchak, “Directions of using technologies of machine training in
the military sphere,” Artificial Intelligence, 27(1), pp. 161–164, 2022. doi:
https://doi.org/10.15407/jai2022.01.161
3. L.Y. Yurchuk, O.V. Rodinin, “Specialized system of collection and treatment of in-
formation for the meteorological monitoring,” Visnyk of Vinnytsia Politechnical In-
stitute, no. 1, pp. 18–22, 2013
4. S. Balovsyak, V. Vasiliev, I. Fodchuk, “Expert System for Supporting the Construction
of Three-Dimensional Models of Objects by the Photogrammetry Method,” SISIOT, vol.
1, no. 2, 02005, 2023. doi: https://doi.org/10.31861/sisiot2023.2.02005
5. C.-E. Lee, J. Baek, J. Son, Y.G. Ha, “Deep AI military staff: cooperative battlefield
situation awareness for commander’s decision making,” The Journal of Supercom-
puting, 79, pp. 6040–6069, 2023. doi: https://doi.org/10.1007/s11227-022-04882-w
6. T. Li, H. Chen, L. Kong, J. Li, Y. Yang, “The development of the combined weapon
of light and high maneuvering artillery from the view of the warfare object,” in S.
Long, B.S. Dhillon (Eds.) “Man-machine-environment system engineering,” Proc. of
the 23rd Int. Conf. on MMESE, 2023, Lecture Notes in Electrical Engineering, 1069,
pp. 303–308, 2023. doi: https://doi.org/10.1007/978-981-99-4882-6_43
7. S. Faraj, S. Pachidi, K. Sayegh, “Working and organizing in the age of the learning
algorithm,” Information and Organization, vol. 28(1), pp. 62–70, 2018. doi:
https://doi.org/10.1016/j.infoandorg.2018.02.005
8. A. Dagallier et al.,“Long-range acoustic localization of artillery shots using distrib-
uted synchronous acoustic sensors,” The Journal of the Acoustical Society of Amer-
ica, pp. 4860–4872, 2019. doi: https://doi.org/10.1121/1.5138927
9. S. Cheinet, T. Broglin, “Sensitivity of shot detection and localization to environ-
mental propagation,” Applied Acoustics, 93, pp. 97–105, 2015. doi: https://doi.org/
10.1016/j.apacoust.2015.01.021
10. V. Tymchuk, O. Mediakov, A. Poliakov, O. Popov, “The research of the configura-
tions in some locating acoustic system for geospatial modeling in GIS to increase the
coordinate accuracy,” Int. Conf. of Young Professionals “GeoTerrace-2023”, Lviv
Polytecnic National University, 2–4 October 2023. doi: https://doi.org/10.3997/
2214-4609.2023510064
11. “HALO Hostile Artillery Locating System,” Leonardo Ltd., 2022. Available:
https://electronics.leonardo.com/en/products/halo
12. R. Kochan, B.Trembach, O. Kochan, “Methodical error of targets bearing by sound
artillery intelligence system,” Measuring Equipment and Metrology, 80(3), pp. 10–
14, 2019. doi: https://doi.org/10.23939/istcmtm2019.03.010
13. A. Kowalska-Styczen, R. Peleshchak, V. Lytvyn, I. Peleshchak, A. Dyriv, V.
Danylyk, “Automatic Identification of Sound Source Position Coordinates Using a
Sound Metric System of Sensors Linked with an Internet Connection,” Symmetry,
14(11), 2338, 2022. doi: https://doi.org/10.3390/sym14112338
14. S. Kapka, M. Lewandowski, “Sound source detection, localization and classification
using consecutive ensemble of CRNN models,” Proceedings of the Detection and
Classification of Acoustic Scenes and Events 2019 Workshop (DCASE2019), pp.
119–123, 2019. doi: https://doi.org/10.33682/9f2t-ab23
15. S. Artemuk, I. Mykytyn, “System for determining the coordinates of the acoustic
signal source based on the difference-time method and machine learning methods,”
Measuring and Computing Devices in Technological Processes, no. 3, 2023. doi:
https://doi.org/10.31891/2219-9365-2023-75-1
16. V. Tymchuk, “Artillery radars of TPQ series: some aspects of design and operation,
lessons of modification,” Artillery and Rifle Armaments, 1(38), pp. 12–18, 2011.
Available: http://www.irbis-nbuv.gov.ua/publ/REF-0000534126
17. J. Safari Bazargani, A. Sadeghi-Niaraki, S.-M. Choi, “A Survey of GIS and IoT In-
tegration: Applications and Architecture,” Appl. Sci. 2021, 11, 10365. doi: https://
doi.org/10.3390/app112110365
The results of the multi-position surveillance system’s efficiency, depending on the locations …
Системні дослідження та інформаційні технології, 2025, № 4 37
18. T.-Y. Ho et al., “The dawn of AI-native EDA: promises and challenges of large cir-
cuit models,” arxiv.org, 12 March 2024. doi: https://doi.org/10.48550/arXiv.2403.07257
Received 02.05.2024
INFORMATION ON THE ARTICLE
Volodymyr Yu. Tymchuk, ORCID: 0000-0002-3549-2813, Hetman Petro Sahaidachnyi
National Army Academy, Ukraine, e-mail: v_tymchuk@yahoo.co.uk
Oleksandr O. Mediakov, ORCID: 0000-0002-2580-3155, Lviv Polytechnic National
University, Ukraine, e-mail: oleksandr.mediakov@gmail.com
Oleksandr O. Popov, ORCID: 0009-0002-7548-6175, Armed Forces of Ukraine, A1108
military unit, Ukraine, e-mail: vparrvia@gmail.com
Taras V. Trysnyuk, ORCID: 0000-0002-3672-8242, Institute of Telecommunications
and Global Information Space National Academy of Science of Ukraine, Ukraine, e-mail:
taras24t@gmail.com
Serhii A. Tsybulia, ORCID: 0000-0003-0323-1771, National Defense University of
Ukraine, Ukraine, e-mail: kibtor@gmail.com
ОЦІНЮВАННЯ ЕФЕКТИВНОСТІ БАГАТОПОЗИЦІЙНОЇ СИСТЕМИ
СПОСТЕРЕЖЕННЯ НА ОСНОВІ ДОДАТКОВОГО ОБРОБЛЕННЯ ДАНИХ
ВІД СЕНСОРІВ ЗІ ЗМІНЮВАНИМ МІСЦЕПОЛОЖЕННЯМ / В.Ю. Тимчук,
О.О. Медяков, О.О. Попов, Т.В. Триснюк, С.А. Цибуля
Анотація. Ефективність багатопозиційної системи залежить від її структури,
а саме від кількості елементів (сенсорів) у складі системи та від місцеполо-
ження самих сенсорів, які визначають конфігурацію системи, а також від того,
як на роботу системи впливають середовище і рельєф місцевості. Розглянуто
оброблення даних у багатопозиційній системі спостереження як додаткової
операції шляхом збирання та обчислювальних маніпуляцій із великими дани-
ми (набором датасетів) з елементів системи. Значний обсяг цифрових даних,
що циркулюють у багатопозиційній системі (отримуються від сенсорів систе-
ми), уможливлює використання статистичних методів оброблення даних. За-
пропоновано онтологію для оцінювання якості багатопозиційної системи за-
лежно від її конфігурацій. Результати розподілів виявлених подій подано
у графічних формах, що дало змогу виконати статистичну оцінку розподілених
даних. Зроблено висновки про спосіб поліпшувати ефективність багатопози-
ційної системи в непередбачуваному змінному середовищі завдяки зміні кон-
фігурації системи, за якої забезпечують кращі показники якості виявлення.
Ключові слова: оброблення даних, спостереження, виявлення, багатопозицій-
на система, система систем, ефективність.
|
| id | journaliasakpiua-article-351408 |
| institution | System research and information technologies |
| keywords_txt_mv | keywords |
| language | English |
| last_indexed | 2026-02-08T08:06:11Z |
| publishDate | 2025 |
| publisher | The National Technical University of Ukraine "Igor Sikorsky Kyiv Polytechnic Institute" |
| record_format | ojs |
| resource_txt_mv | journaliasakpiua/16/0c01fd25e8d22426813f8fdb101bd316.pdf |
| spelling | journaliasakpiua-article-3514082026-02-02T20:49:24Z The results of the multi-position surveillance system’s efficiency, depending on the locations of its sensors, using additional data processing Оцінювання ефективності багатопозиційної системи спостереження на основі додаткового оброблення даних від сенсорів зі змінюваним місцеположенням Tymchuk, Volodymyr Mediakov, Oleksandr Popov, Oleksandr Trysnyuk, Taras Tsybulia, Serhii оброблення даних спостереження виявлення багатопозиційна система система систем ефективність data processing surveillance detection multi-position system system-of-systems efficiency The efficiency of a multi-position system depends on the realization of its struc-ture — how many elements it includes, where they are located, and how the envi-ronment and terrain influence its operation. The paper is dedicated to data pro-cessing in a multi-position surveillance system as an additional option, leverag-ing the in-between big data from the system’s elements. A sufficient number of numerical data generated by the multi-position system and its elements–sensors–allows the use of statistical methods and models from machine learning or deep learning. The ontology for quality estimation of the multi-position sys-tem, depending on its configurations, is proposed. The results of the distribu-tions of detected events are presented in graphical forms that allow statistical evaluation of the distributed data. Our findings allow us to ensure the efficiency of a multi-position system in an unpredictable, variable environment by recon-figuring it when it offers better capabilities. Ефективність багатопозиційної системи залежить від її структури, а саме від кількості елементів (сенсорів) у складі системи та від місцеположення самих сенсорів, які визначають конфігурацію системи, а також від того, як на роботу системи впливають середовище і рельєф місцевості. Розглянуто оброблення даних у багатопозиційній системі спостереження як додаткової операції шляхом збирання та обчислювальних маніпуляцій із великими даними (набором датасетів) з елементів системи. Значний обсяг цифрових даних, що циркулюють у багатопозиційній системі (отримуються від сенсорів системи), уможливлює використання статистичних методів оброблення даних. Запропоновано онтологію для оцінювання якості багатопозиційної системи залежно від її конфігурацій. Результати розподілів виявлених подій подано у графічних формах, що дало змогу виконати статистичну оцінку розподілених даних. Зроблено висновки про спосіб поліпшувати ефективність багатопозиційної системи в непередбачуваному змінному середовищі завдяки зміні конфігурації системи, за якої забезпечують кращі показники якості виявлення. The National Technical University of Ukraine "Igor Sikorsky Kyiv Polytechnic Institute" 2025-12-29 Article Article application/pdf https://journal.iasa.kpi.ua/article/view/351408 10.20535/SRIT.2308-8893.2025.4.02 System research and information technologies; No. 4 (2025); 20-37 Системные исследования и информационные технологии; № 4 (2025); 20-37 Системні дослідження та інформаційні технології; № 4 (2025); 20-37 2308-8893 1681-6048 en https://journal.iasa.kpi.ua/article/view/351408/338426 |
| spellingShingle | оброблення даних спостереження виявлення багатопозиційна система система систем ефективність Tymchuk, Volodymyr Mediakov, Oleksandr Popov, Oleksandr Trysnyuk, Taras Tsybulia, Serhii Оцінювання ефективності багатопозиційної системи спостереження на основі додаткового оброблення даних від сенсорів зі змінюваним місцеположенням |
| title | Оцінювання ефективності багатопозиційної системи спостереження на основі додаткового оброблення даних від сенсорів зі змінюваним місцеположенням |
| title_alt | The results of the multi-position surveillance system’s efficiency, depending on the locations of its sensors, using additional data processing |
| title_full | Оцінювання ефективності багатопозиційної системи спостереження на основі додаткового оброблення даних від сенсорів зі змінюваним місцеположенням |
| title_fullStr | Оцінювання ефективності багатопозиційної системи спостереження на основі додаткового оброблення даних від сенсорів зі змінюваним місцеположенням |
| title_full_unstemmed | Оцінювання ефективності багатопозиційної системи спостереження на основі додаткового оброблення даних від сенсорів зі змінюваним місцеположенням |
| title_short | Оцінювання ефективності багатопозиційної системи спостереження на основі додаткового оброблення даних від сенсорів зі змінюваним місцеположенням |
| title_sort | оцінювання ефективності багатопозиційної системи спостереження на основі додаткового оброблення даних від сенсорів зі змінюваним місцеположенням |
| topic | оброблення даних спостереження виявлення багатопозиційна система система систем ефективність |
| topic_facet | оброблення даних спостереження виявлення багатопозиційна система система систем ефективність data processing surveillance detection multi-position system system-of-systems efficiency |
| url | https://journal.iasa.kpi.ua/article/view/351408 |
| work_keys_str_mv | AT tymchukvolodymyr theresultsofthemultipositionsurveillancesystemsefficiencydependingonthelocationsofitssensorsusingadditionaldataprocessing AT mediakovoleksandr theresultsofthemultipositionsurveillancesystemsefficiencydependingonthelocationsofitssensorsusingadditionaldataprocessing AT popovoleksandr theresultsofthemultipositionsurveillancesystemsefficiencydependingonthelocationsofitssensorsusingadditionaldataprocessing AT trysnyuktaras theresultsofthemultipositionsurveillancesystemsefficiencydependingonthelocationsofitssensorsusingadditionaldataprocessing AT tsybuliaserhii theresultsofthemultipositionsurveillancesystemsefficiencydependingonthelocationsofitssensorsusingadditionaldataprocessing AT tymchukvolodymyr ocínûvannâefektivnostíbagatopozicíjnoísistemisposterežennânaosnovídodatkovogoobroblennâdanihvídsensorívzízmínûvanimmíscepoložennâm AT mediakovoleksandr ocínûvannâefektivnostíbagatopozicíjnoísistemisposterežennânaosnovídodatkovogoobroblennâdanihvídsensorívzízmínûvanimmíscepoložennâm AT popovoleksandr ocínûvannâefektivnostíbagatopozicíjnoísistemisposterežennânaosnovídodatkovogoobroblennâdanihvídsensorívzízmínûvanimmíscepoložennâm AT trysnyuktaras ocínûvannâefektivnostíbagatopozicíjnoísistemisposterežennânaosnovídodatkovogoobroblennâdanihvídsensorívzízmínûvanimmíscepoložennâm AT tsybuliaserhii ocínûvannâefektivnostíbagatopozicíjnoísistemisposterežennânaosnovídodatkovogoobroblennâdanihvídsensorívzízmínûvanimmíscepoložennâm AT tymchukvolodymyr resultsofthemultipositionsurveillancesystemsefficiencydependingonthelocationsofitssensorsusingadditionaldataprocessing AT mediakovoleksandr resultsofthemultipositionsurveillancesystemsefficiencydependingonthelocationsofitssensorsusingadditionaldataprocessing AT popovoleksandr resultsofthemultipositionsurveillancesystemsefficiencydependingonthelocationsofitssensorsusingadditionaldataprocessing AT trysnyuktaras resultsofthemultipositionsurveillancesystemsefficiencydependingonthelocationsofitssensorsusingadditionaldataprocessing AT tsybuliaserhii resultsofthemultipositionsurveillancesystemsefficiencydependingonthelocationsofitssensorsusingadditionaldataprocessing |