RSSI localization with DB-Assisted Least Error algorithm

Jongtack Jung, Kangho Kim, Seungho Yoo, Mungyu Bae, Suk Kyu Lee, Hwangnam Kim

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

RSSI (Received Signal Strength Indication) localization techniques using Wi-Fi presents substantial advantages compared to others. They are light weight both in terms of computation and energy consumption. RSSI localization techniques are mostly used indoors, where many APs (Access Points) are present and no GPS is available. Recently, APs are getting deployed outdoors as well, and urban canyon phenomenon degrades the capability of GPS localization even in outdoor environments, which makes RSSI localization techniques attractive as an outdoor localization solution as well. The downside of RSSI localization is that it is polarized, which means it has either high performance and economic cost or low cost and poor accuracy. Both cases are inadequate for a general deployment; high-cost algorithms can only be deployed in heavily populated area for cost feasibility and the accuracy of low-cost algorithms is nowhere near credible. In this paper, we propose a range-based RSSI localization algorithm that has reasonable accuracy yet has very low cost. The proposed algorithm consists of DB-assistance, ration base algorithm, and an elementary machine learning algorithm. This helps achieving the qualities that can provide a feasible RSSI localization solution that can be employed in a much wider area.

Original languageEnglish
Title of host publicationInternational Conference on Ubiquitous and Future Networks, ICUFN
PublisherIEEE Computer Society
Pages338-343
Number of pages6
Volume2015-August
ISBN (Print)9781479989935
DOIs
Publication statusPublished - 2015 Aug 7
Event7th International Conference on Ubiquitous and Future Networks, ICUFN 2015 - Sapporo, Japan
Duration: 2015 Jul 72015 Jul 10

Other

Other7th International Conference on Ubiquitous and Future Networks, ICUFN 2015
CountryJapan
CitySapporo
Period15/7/715/7/10

Fingerprint

Costs
Global positioning system
Wi-Fi
Learning algorithms
Learning systems
Energy utilization
Economics

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Computer Science Applications
  • Hardware and Architecture

Cite this

Jung, J., Kim, K., Yoo, S., Bae, M., Lee, S. K., & Kim, H. (2015). RSSI localization with DB-Assisted Least Error algorithm. In International Conference on Ubiquitous and Future Networks, ICUFN (Vol. 2015-August, pp. 338-343). [7182561] IEEE Computer Society. https://doi.org/10.1109/ICUFN.2015.7182561

RSSI localization with DB-Assisted Least Error algorithm. / Jung, Jongtack; Kim, Kangho; Yoo, Seungho; Bae, Mungyu; Lee, Suk Kyu; Kim, Hwangnam.

International Conference on Ubiquitous and Future Networks, ICUFN. Vol. 2015-August IEEE Computer Society, 2015. p. 338-343 7182561.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Jung, J, Kim, K, Yoo, S, Bae, M, Lee, SK & Kim, H 2015, RSSI localization with DB-Assisted Least Error algorithm. in International Conference on Ubiquitous and Future Networks, ICUFN. vol. 2015-August, 7182561, IEEE Computer Society, pp. 338-343, 7th International Conference on Ubiquitous and Future Networks, ICUFN 2015, Sapporo, Japan, 15/7/7. https://doi.org/10.1109/ICUFN.2015.7182561
Jung J, Kim K, Yoo S, Bae M, Lee SK, Kim H. RSSI localization with DB-Assisted Least Error algorithm. In International Conference on Ubiquitous and Future Networks, ICUFN. Vol. 2015-August. IEEE Computer Society. 2015. p. 338-343. 7182561 https://doi.org/10.1109/ICUFN.2015.7182561
Jung, Jongtack ; Kim, Kangho ; Yoo, Seungho ; Bae, Mungyu ; Lee, Suk Kyu ; Kim, Hwangnam. / RSSI localization with DB-Assisted Least Error algorithm. International Conference on Ubiquitous and Future Networks, ICUFN. Vol. 2015-August IEEE Computer Society, 2015. pp. 338-343
@inproceedings{147f42df86ca4d2bad82e84ba1988a07,
title = "RSSI localization with DB-Assisted Least Error algorithm",
abstract = "RSSI (Received Signal Strength Indication) localization techniques using Wi-Fi presents substantial advantages compared to others. They are light weight both in terms of computation and energy consumption. RSSI localization techniques are mostly used indoors, where many APs (Access Points) are present and no GPS is available. Recently, APs are getting deployed outdoors as well, and urban canyon phenomenon degrades the capability of GPS localization even in outdoor environments, which makes RSSI localization techniques attractive as an outdoor localization solution as well. The downside of RSSI localization is that it is polarized, which means it has either high performance and economic cost or low cost and poor accuracy. Both cases are inadequate for a general deployment; high-cost algorithms can only be deployed in heavily populated area for cost feasibility and the accuracy of low-cost algorithms is nowhere near credible. In this paper, we propose a range-based RSSI localization algorithm that has reasonable accuracy yet has very low cost. The proposed algorithm consists of DB-assistance, ration base algorithm, and an elementary machine learning algorithm. This helps achieving the qualities that can provide a feasible RSSI localization solution that can be employed in a much wider area.",
author = "Jongtack Jung and Kangho Kim and Seungho Yoo and Mungyu Bae and Lee, {Suk Kyu} and Hwangnam Kim",
year = "2015",
month = "8",
day = "7",
doi = "10.1109/ICUFN.2015.7182561",
language = "English",
isbn = "9781479989935",
volume = "2015-August",
pages = "338--343",
booktitle = "International Conference on Ubiquitous and Future Networks, ICUFN",
publisher = "IEEE Computer Society",

}

TY - GEN

T1 - RSSI localization with DB-Assisted Least Error algorithm

AU - Jung, Jongtack

AU - Kim, Kangho

AU - Yoo, Seungho

AU - Bae, Mungyu

AU - Lee, Suk Kyu

AU - Kim, Hwangnam

PY - 2015/8/7

Y1 - 2015/8/7

N2 - RSSI (Received Signal Strength Indication) localization techniques using Wi-Fi presents substantial advantages compared to others. They are light weight both in terms of computation and energy consumption. RSSI localization techniques are mostly used indoors, where many APs (Access Points) are present and no GPS is available. Recently, APs are getting deployed outdoors as well, and urban canyon phenomenon degrades the capability of GPS localization even in outdoor environments, which makes RSSI localization techniques attractive as an outdoor localization solution as well. The downside of RSSI localization is that it is polarized, which means it has either high performance and economic cost or low cost and poor accuracy. Both cases are inadequate for a general deployment; high-cost algorithms can only be deployed in heavily populated area for cost feasibility and the accuracy of low-cost algorithms is nowhere near credible. In this paper, we propose a range-based RSSI localization algorithm that has reasonable accuracy yet has very low cost. The proposed algorithm consists of DB-assistance, ration base algorithm, and an elementary machine learning algorithm. This helps achieving the qualities that can provide a feasible RSSI localization solution that can be employed in a much wider area.

AB - RSSI (Received Signal Strength Indication) localization techniques using Wi-Fi presents substantial advantages compared to others. They are light weight both in terms of computation and energy consumption. RSSI localization techniques are mostly used indoors, where many APs (Access Points) are present and no GPS is available. Recently, APs are getting deployed outdoors as well, and urban canyon phenomenon degrades the capability of GPS localization even in outdoor environments, which makes RSSI localization techniques attractive as an outdoor localization solution as well. The downside of RSSI localization is that it is polarized, which means it has either high performance and economic cost or low cost and poor accuracy. Both cases are inadequate for a general deployment; high-cost algorithms can only be deployed in heavily populated area for cost feasibility and the accuracy of low-cost algorithms is nowhere near credible. In this paper, we propose a range-based RSSI localization algorithm that has reasonable accuracy yet has very low cost. The proposed algorithm consists of DB-assistance, ration base algorithm, and an elementary machine learning algorithm. This helps achieving the qualities that can provide a feasible RSSI localization solution that can be employed in a much wider area.

UR - http://www.scopus.com/inward/record.url?scp=84944675520&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84944675520&partnerID=8YFLogxK

U2 - 10.1109/ICUFN.2015.7182561

DO - 10.1109/ICUFN.2015.7182561

M3 - Conference contribution

SN - 9781479989935

VL - 2015-August

SP - 338

EP - 343

BT - International Conference on Ubiquitous and Future Networks, ICUFN

PB - IEEE Computer Society

ER -