Dictionary learning based fingerprinting for indoor localization

Kumar, Chirag; Rajawat, Ketan

doi:10.1109/ncc.2018.8600195

Cited by 2 publications

(1 citation statement)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…From several other methods that can be used in indoor positioning based on received signal strength indication (RSSI) such as trilateration and triangulation, fingerprinting is a method that is widely adopted ISSN: 2502-4752  Improved fingerprinting performance in indoor positioning by reducing duration of the… (Andika Muharam) 237 because of its high level of accuracy [11]. The fingerprinting method contains two main work phases, namely training (offline-phase) [2] and positioning (online-phase) [12], [13]. In the first phases (training) is the process of collecting data s (example: RSSI) from each sensor Node that is collected in a database which can later be used as reference data to help infer the position of an object carried out in the second phase (positioning).…”

Section: Introductionmentioning

confidence: 99%

Improved fingerprinting performance in indoor positioning by reducing duration of the training phase process

Muharam

Wahab

Alaydrus

2021

IJEECS

View full text Add to dashboard Cite

<span>Wireless <span>sensor network (WSN) can be used as a solution to find out the position of an object that cannot be reached by global positioning system (GPS), for example to find out the position of objects in a room known as Indoor Positioning. One method in indoor positioning that can be used is fingerprinting. Inside there are two main work phases, namely training and positioning. The training phase is the process of collecting received signal strength indication (RSSI) data levels from each sensor Node reference that will be used as a reference value for the positioning phase. The more sensor Nodes used, the longer the processing time needed in the training phase. This research focussed on the duration of the training phase, the implementation of which are used 4 sensor Nodes, namely Zigbee (IEEE 802.15.4 protocol) arranged according to mesh network topology, one as Node X (positioning target) and 3 as reference Nodes. There are two methods used in the training phase, namely fixed target parameter (FTP) and moving target parameter (MTP). MTP took 5 seconds faster than FTP in terms of the duration of RSSI data collection from each reference Node. </span></span>

show abstract