The device-free indoor localization (DFIL) research is gaining attention due to the popularity of location-based service (LBS)-based advertisement. In DFIL, a user or an object does not need to bring any device to be localized. In this paper, we propose the Wi-Fi-based DFIL and the random forest algorithm for the fingerprint-based technique. The simple parameter commonly used in indoor localization is the Received Signal Strength Indicator (RSSI). We apply the fingerprint technique because of its reliability to handle the RSSI fluctuation and time-varying effect in a static indoor environment. We conducted an actual measurement campaign to observe the DFIL's implementation visibility. The DFIL system works by comparing the database fingerprint in an empty open office with the database in which a person is inside the measurement area without bringing any devices. Thus, we have the device-free RSSI database for fingerprint technique from both empty rooms and RSSI affected by a person inside the room. We validated the random forest algorithm results by comparing them with the k-nearest neighbor (kNN) and artificial neural network (ANN). The results show that our proposed system's accuracy is better than kNN and ANN with a mean error of 0.63 m than kNN with 0.80 m and ANN with 1.01 m. Meanwhile, the precision of the random forest is 0.63 m, whereas kNN and ANN are 0.67 m and 0.80 m, showing that the random forest performed better. We concluded that our simple DFIL system is visible to apply with acceptable accuracy performance.

The real-life indoor localization implementation in a multi-story building is reasonably necessary. Multi-floor shopping centers, airports, residential areas, especially in the big cities, apply positioning schemes to ease visitors or inhabitants. However, most indoor localization researches still emphasize 2D-indoor localization, and the multi-story indoor localization implementations are still limited. One of the challenges of 3D-indoor localization implementation is the shadowing effect caused by signal propagation obstructed by objects in the room, the walls, and floors between rooms. Some researchers conducted the 3D-indoor localization to consider the elevation property of the position estimation scenario. However, there are still very few experimental results in an actual multi-story building as the authors' concerns. This paper proposes the measurement campaign of a 3D-indoor localization system in the actual multi-story building by applying the range-based and range-free method based on the Wireless-Fidelity (Wi-Fi). This research is essential since Wi-Fi is available in almost all smart devices and is installed almost in every corner globally. Compared to other approaches, we propose a relatively simple Wi-Fi-based indoor 3D localization utilizing the specific parameter, received signal strength indicator (RSSI), in a static indoor lobbies environment. Despite some of its advantages, the RSSI parameter has a disadvantage in signal fluctuation over time.  In our approach, we tried to solve this issue by applying the min-max algorithm to improve the known trilateration method as the range-based method. We implemented the min-max to observe how far the range-based can still give acceptable positioning results in an actual multi-story building. On the other hand, we used the RSSI values for the range-free method to construct the fingerprint database and employed the machine-learning (ML)-based pattern matching algorithm, the random forest algorithm. We expect to solve the shadowing problem with this radio fingerprint method and to achieve minimal errors. We conducted the measurement campaign using the low-cost Wi-Fi module, the ESP-8266, to generate the RSSI. We placed three ESP-8266 nodes on each floor of a two-floor building as the access points (APs) and an ESP-8266 as a target node or a station (STA). We emphasized two performance metrics to evaluate our proposed system performance: the location estimation accuracy observed as the mean square error (MSE) and the precision shown as the standard deviation (Std Dev). The results show that the fingerprint technique yielded the MSE of 0.9m and Std Dev of 0.69 m, while the min-max method resulted in the performance of MSE of 1.79 m and Std Dev of 0.89 m. These results show that the fingerprint technique still gave better accuracy and precision in the same measurement campaign than the min-max. However, the min-max results are also acceptable since the whole multi-floor building has more than 4 m in elevation. The indoor localization system for multi-story buildings can be applied using both the fingerprint and the min-max in a relatively static environment by observing our system performance metric

No abstract

No abstract