Notice: Undefined index: linkPowrot in C:\wwwroot\wwwroot\publikacje\publikacje.php on line 1275
Strona główna > Publikacje
Publikacje
Pomoc (F2)
[93510] Artykuł:

An indoor tracking system and pattern recognition algorithms as key components of IoT-based entertainment industry

Czasopismo: Photonics Applications In Astronomy, Communications, Industry, And High-energy Physics Experiments 2019   Tom: 11176, Strony: 1-10
ISSN:  1996-756X
ISBN:  978-1-5106-3066-6
Wydawca:  SPIE-INT SOC OPTICAL ENGINEERING, 1000 20TH ST, PO BOX 10, BELLINGHAM, WA 98227-0010 USA
Opublikowano: Listopad 2019
Seria wydawnicza:  Proceedings of SPIE
Liczba arkuszy wydawniczych:  0.80
 
  Autorzy / Redaktorzy / Twórcy
Imię i nazwisko Wydział Katedra Do oświadczenia
nr 3		 Grupa
przynależności		 Dyscyplina
naukowa		 Procent
udziału		 Liczba
punktów
do oceny pracownika		 Liczba
punktów wg
kryteriów ewaluacji
Radosław Belka orcid logo WEAiIKatedra Systemów Informatycznych *Takzaliczony do "N"Automatyka, elektronika, elektrotechnika i technologie kosmiczne100.00.00  

Grupa MNiSW:  Materiały z konferencji międzynarodowej (zarejestrowane w Web of Science)
Punkty MNiSW: 0
Klasyfikacja Web of Science: Proceedings Paper

DOI LogoDOI     Web of Science Logo Web of Science    
Keywords:

Internet of Things  Entertainment Industry  indoor tracking  RTLS  BLE Beacon  pattern recognition  Linear Discriminant Analysis  Weibull distribution 


Abstract:

The paper presents the key components of an innovative Integrated Visitor Support System (IVSS) for the distributed entertainment industry, based on Internet of Things (IoT) technology and classification analysis. Regarding the modern IoT layered models, including indoor tracking technologies and pattern recognition algorithm, the project of IVSS was
presented. Simple tracking concepts based on chaining of Login Records to form a Path Vector (PV) by so-called poly-LR-isation process was proposed. The data generation model was also developed, where a Weibull and Poison distributions were adopted to generating Login Period and Entry Time parameters, respectively. The data in the form of PVs set was generated for a virtual theme park containing twenty monitored Points of Interest (POIs). The 5 different Visitors’ profiles and 4 different behavior types were defined in the model. The efficiency of pattern recognition was calculated according to the Linear Discriminant Analysis concept. Influence of the set of independent variables on the efficiency of correctly classified labels was reported. It has been shown that high efficiency in recognizing the types of behavior can be obtained including the total number of logins in the POIs area and first login times, whereas the proper separation of profiles needs additional data as Login Period.


POWRÓT
Strona główna > Publikacje