Bluetooth positioning based on weighted K-nearest neighbors and adaptive bandwidth mean shift

Abstract

Bluetooth positioning is an important and challenging topic in indoor positioning. Although a lot of algorithms have been proposed for this problem, it is still not solved perfectly because of the instable signal strengths of Bluetooth. To improve the performance of Bluetooth positioning, this article proposes a coarse-to-fine positioning method based on weighted K-nearest neighbors and adaptive bandwidth mean shift. The method first employs weighted K-nearest neighbors to generate multi-candidate locations. Then, the testing position is obtained by applying adaptive bandwidth mean shift to the multi-candidate locations, which is used to search for the maximum density of the candidate locations. Experimental result indicates that the proposed method improves the performance of Bluetooth positioning.

Developing a BLE Beacon-Based Location System Using Location Fingerprint Positioning for Smart Home Power Management

“…Bluetooth Low Energy (BLE) Beacon is a low-power, low-cost signal transmitter that can be detected by nearby Bluetooth 4.0 devices [5]. It is compatible with a smartphone or mobile device, and can be used to execute commands within the sensing range of a BLE Beacon site [16,17]. In connection state, a BLE Beacon frequency changes pseudo-randomly among 40 channels.…”

“…If users want to accurately estimate the distance, the TX power of the BLE Beacon must be measured first to improve the positioning accuracy. Even if users use RSSI for indoor positioning, it may not yield the desired results, because the indoor environment is quite complicated, contains a lot of interference, and a multipath effect is generated, which may cause the signal strength to be a non-ideal attenuation [17]. In the system development of this work, a filter was used to remove noise, and hardware configuration was used to correct the signal to improve positioning accuracy.…”

“…This section introduces the positioning algorithm based on location fingerprinting. Referring to Wang et al [17] formalizing the Weighted K-Nearest Neighbors (WKNN) positioning method, this study developed the proposed positioning algorithm for this work. Figure 6 illustrates the case of location p positioning by three BLE Beacons in the area A, which the location is sampling by the maximum and minimum RSSI values.…”

“…In a practical environment, a perfect linear correlation between distance and RSSI did not exist due to various influence factors, e.g., obstacles or walls of the building. We refer to the method proposed by Wang et al [17] to design the mean filter for this study. The method in [17] used the standard deviation correction to normalize the RSSI deviation value of different the measurements for every sampling experiment.…”

“…We refer to the method proposed by Wang et al [17] to design the mean filter for this study. The method in [17] used the standard deviation correction to normalize the RSSI deviation value of different the measurements for every sampling experiment. Therefore, this study used the mean filter to execute the experiments.…”