SpotFi is an indoor localization system that achieves high accuracy using standard WiFi infrastructure without requiring hardware or firmware changes. It uses super-resolution algorithms to accurately estimate the angle of arrival (AoA) of multipath components, even with only three antennas. SpotFi also employs novel filtering and estimation techniques to identify the direct path between the target and access point (AP). Experiments in a multipath-rich environment show SpotFi achieves a median accuracy of 40 cm, robust to obstacles and multipath. SpotFi's key contributions include super-resolution AoA estimation, robust direct path identification, and a localization algorithm that combines direct path AoA estimates with RSSI data from APs. SpotFi uses Intel 5300 WiFi cards and achieves a median error of 40 cm, with an 80th percentile error of 1.8 m. It works robustly in challenging scenarios and is lightweight, requiring only ten packets for accurate localization. SpotFi satisfies three requirements: deployability, universality, and accuracy. It outperforms existing systems with fewer antennas and does not require calibration or fingerprinting. SpotFi's design includes a super-resolution algorithm for AoA and ToF estimation, a method to identify the direct path, and a localization algorithm that combines AoA and RSSI data. SpotFi's super-resolution algorithm leverages CSI data from multiple subcarriers and antennas to overcome antenna limitations. It also addresses time synchronization issues by sanitizing ToF estimates. SpotFi's direct path identification uses clustering to determine the most likely direct path based on AoA and ToF variations. SpotFi's localization algorithm combines direct path AoA estimates and RSSI data to determine the target's location. SpotFi is lightweight, requires no calibration, and works with commodity WiFi infrastructure. It achieves high accuracy with minimal hardware requirements, making it a promising solution for indoor localization.SpotFi is an indoor localization system that achieves high accuracy using standard WiFi infrastructure without requiring hardware or firmware changes. It uses super-resolution algorithms to accurately estimate the angle of arrival (AoA) of multipath components, even with only three antennas. SpotFi also employs novel filtering and estimation techniques to identify the direct path between the target and access point (AP). Experiments in a multipath-rich environment show SpotFi achieves a median accuracy of 40 cm, robust to obstacles and multipath. SpotFi's key contributions include super-resolution AoA estimation, robust direct path identification, and a localization algorithm that combines direct path AoA estimates with RSSI data from APs. SpotFi uses Intel 5300 WiFi cards and achieves a median error of 40 cm, with an 80th percentile error of 1.8 m. It works robustly in challenging scenarios and is lightweight, requiring only ten packets for accurate localization. SpotFi satisfies three requirements: deployability, universality, and accuracy. It outperforms existing systems with fewer antennas and does not require calibration or fingerprinting. SpotFi's design includes a super-resolution algorithm for AoA and ToF estimation, a method to identify the direct path, and a localization algorithm that combines AoA and RSSI data. SpotFi's super-resolution algorithm leverages CSI data from multiple subcarriers and antennas to overcome antenna limitations. It also addresses time synchronization issues by sanitizing ToF estimates. SpotFi's direct path identification uses clustering to determine the most likely direct path based on AoA and ToF variations. SpotFi's localization algorithm combines direct path AoA estimates and RSSI data to determine the target's location. SpotFi is lightweight, requires no calibration, and works with commodity WiFi infrastructure. It achieves high accuracy with minimal hardware requirements, making it a promising solution for indoor localization.