Abstract | ||
---|---|---|
Utilizing machine learning methods for radio localization is gaining popularity in recent years. This is because of the current technology trends in better connectivity, cloud database, and cheaper processing power. Received signal strength (RSS) fingerprinting is one of the common localization methods because of its relative simplicity and ability to produce well-distinct patterns at different locations. In this paper, we compare the performance of different machine learning algorithms in terms of their mean localization error using RSS fingerprinting. The comparison is based on two key parameters, namely; (i) the correlation distance of the radio shadowing, and (ii) the standard deviation of the shadowing. The studied machine learning methods are the linear regression (LR), k-nearest neighbour regression (kNR), decision tree regression (DTR) and random forest regression (RFR), where extensive simulation demonstrates the performance of these methods under the correlated shadowing scenarios. |
Year | DOI | Venue |
---|---|---|
2020 | 10.1109/ICSPCS50536.2020.9310009 | 2020 14th International Conference on Signal Processing and Communication Systems (ICSPCS) |
Keywords | DocType | ISBN |
localization,machine learning,correlated shad-owing,Internet-or-Things,linear regression,k-nearest neighbour regression,decision tree regression,random forest regression | Conference | 978-1-7281-9973-3 |
Citations | PageRank | References |
0 | 0.34 | 0 |
Authors | ||
3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Iza S. Mohamad Hashim | 1 | 0 | 0.68 |
Akram Al-Hourani | 2 | 468 | 33.31 |
Wayne S T Rowe | 3 | 1 | 2.40 |