Study of localization scheme base on neural network for wireless sensor networks

Tian, Junfang; Shi, Huichang

doi:10.1049/cp:20070084

Cited by 6 publications

(2 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Theoretical results for ANN assert that one single hidden layer network with sufficient numbers of sigmoidal nodes (in the hidden layer) are capable of approximating any continuous function arbitrarily well. This approach has been used in some of the reported works [3,4,5].…”

Section: Introductionmentioning

confidence: 99%

International Journal of Electronics and Telecommunications

Krupanek¹,

Bogacz²

2020

View full text Add to dashboard Cite

One of the ways to improve calculations related to determining the position of a node in the IoT measurement system is to use artificial neural networks (ANN) to calculate coordinates. The method described in the article is based on the measurement of the RSSI (Received Signal Strength Indicator), which value is then processed by the neural network. Hence, the proposed system works in two stages. In the first stage, RSSI coefficient samples are taken, and then the node location is determined on an ongoing basis. Coordinates anchor nodes (i.e. sensors with fixed and previously known positions) and the matrix of RSSI coefficients are used in the learning process of the neural network. Then the RSSI matrix determined for the system in which the nodes with unknown positions are located is fed into the neural network inputs. The result of the work is a system and algorithm that allows determining the location of the object without processing data separately in nodes with low computational performance.

show abstract

Section: Introductionmentioning

confidence: 99%

International Journal of Electronics and Telecommunications

Krupanek¹,

Bogacz²

2020

View full text Add to dashboard Cite

show abstract

“…Simulation results show that neural networks can achieve high level of accuracy in estimating the location, and requires fewer anchor nodes. This technique is neither affected by the NLOS (Non-Line-of-Sight) environment nor by the irregularity of the power transmitted by the anchors (Tian & Shi, 2007). The neural network is employed to predict the NLOS error (L. Povescu, 2001).…”

Section: How Nn's Were Selected For This Thesismentioning

confidence: 99%