A genetic based indoor positioning algorithm using Wi-Fi received signal strength and motion data

Abstrak

The recent trend in location-based services has led to a proliferation of studies in indoor positioning technology. Wi-Fi received signal strength indicator (RSSI) Fingerprinting and pedestrian dead reckoning (PDR) are the two best representatives from both approaches. This research proposed a genetic algorithm to combine Wi-Fi Fingerprinting and PDR. By taking advantage of PDR and genetic algorithm, we only need to collect a limited number of points for the fingerprint dataset with known coordinates, then target trajectories' position can be estimated with high accuracy. Results from our experiments and simulations have shown that even in the scenario of noisy inertial measurement unit (IMU) sensors data, using RSSI measurements and the coordinate of 8 points, our proposed method was able to achieve 1.589 meters of average distance error which is 34.4 percent lower than the conventional Fingerprinting method.