Examples disclosed herein relate to an autonomous driving system in a vehicle having a radar system with a Non-Line-of-Sight (“NLOS”) correction module to correct for NLOS reflections prior to the radar system identifying targets in a path and a surrounding environment of the vehicle, and a sensor fusion module to receive information from the radar system on the identified targets and compare the information received from the radar system to information received from at least one sensor in the vehicle.

To perform positioning by a computing device, a position request transmitted by a terminal is received. The positioning request includes target access point information obtained by the terminal. Then a plurality of first geographic grids related to a location of the terminal is determined based on the target access point information. A positioning sequence of the plurality of first geographic grids is determined according to access point information corresponding to the plurality of first geographic grids and the target access point information. The positioning sequence indicates distances between the plurality of first geographic grids and the terminal. The location of the terminal is then determined based on the positioning sequence of the plurality of first geographic grids and location information of the plurality of first geographic grids.

A robot and a method for localizing a robot are disclosed. The method for localizing a robot may include acquiring communication environment information including identifiers of access points and received signal strengths from the access points, generating an environmental profile for a current position of the robot based on the acquired communication environment information, comparing the generated environmental profile with a plurality of learning profiles associated with a plurality of regions, respectively, determining a learning profile corresponding to the environmental profile, based on the comparison, and determining a region associated with the determined learning profile as a current position of the robot. In a 5G environment connected for the Internet of Things, embodiments of the present disclosure may be implemented by executing an artificial intelligence algorithm and/or machine learning algorithm.

Disclosed is an approach for improving positioning quality via a feedback loop between a radio-based positioning system and a client device, such as a tracking system. In particular, the tracking system or other client device may request and then receive a position estimate from the positioning system. The tracking system could then make a determination that the position estimate is incorrect, such as by determining that it is an outlier relative to a location trace, for instance. Responsive to this determination, the tracking system may transmit, to the position system, an indication of radio node(s) associated with the incorrect position estimate. Based on this indication, the positioning system could then exclude one or more of those radio node(s) from a radio map, thereby improving quality of the radio map and in turn quality of future position estimates, among other advantages.

Apparatuses and methods in a communication system are disclosed. Data on error between the location of more than one user terminal and the estimated location of the more than one user terminal is obtained, the estimated location obtained utilising radio frequency fingerprinting. It is evaluated whether the error is greater than a given threshold, and if so it is determined whether the error is due to radio frequency fingerprinting or not. Based on the determination decision is made on initialising update of radio frequency fingerprinting map.

Inter-alia, a method is disclosed comprising: determining at least one fingerprint information indicative of at least one identifier of a cellular and/or a non-cellular radio network, wherein based on the identifier at least one cell of the cellular radio network and/or at least one node of the non-cellular radio network is identifiable; sending the at least one fingerprint information, wherein the at least one fingerprint information comprises or is accompanied by a first set of tuning parameters in case such a set of tuning parameters is stored in a memory; receiving a second set of tuning parameters in response to the sending of the at least one fingerprint information, wherein the first and/or the second set of tuning parameters respectively comprise a correction value for compensating a bias for the at least one fingerprint information; and storing the received second set of tuning parameters, wherein the first set of tuning parameters is updated with the second set of tuning parameters enabling to iteratively converge the set of tuning parameters to the actual bias. It is further disclosed an according apparatus, computer program and system.

Disclosed are various embodiments for estimating missing wireless attributes of a wireless fingerprint generated for a given location and confirming that the given location is a correct location. A collection of wireless fingerprint samples for the given location can be used to train an attribute prediction model that can estimate missing attributes in wireless fingerprints. For example, a wireless fingerprint may be incomplete as the fingerprint may fail to include attribute data for some AP devices associated with the given location. The attribute prediction model can be used to reconstruct wireless fingerprints to include the missing attributes. The reconstructed wireless fingerprints can then be used to train a location confirmation model which can be used for accurate location confirmation.

Disclosed is an approach for improving performance of a radio-based positioning system by detecting and excluding mobile radio nodes. The disclosed approach involves processors (e.g., of positioning server(s) and/or of a computing device) making a determination (i) that a speed of a computing device is at or above a threshold speed, and (ii) that at least one condition is met, the at least one condition indicating that a radio node is moving substantially along with the computing device. In response to making the determination, the processors may deem the radio node to be a mobile node. Given this, the processors may in turn exclude the radio node for positioning purposes, which may ultimately improve performance of the radio-based positioning system.

Embodiments described herein provide for an electronic device comprising a set of classifiers that can determine whether the electronic device is likely on an airplane. Upon a determination that the electronic device is likely on an airplane, one or more wireless radios on the electronic device (e.g., an ultra-wideband ranging radio, a cellular radio, etc.) may be disabled or a prompt can be displayed to enable a user to place the device into airplane mode.

A mobile device collects data, which enables a generation of a radio map for a site. The collection comprises detecting at least one indication of a floor, at which a user is located, in a user input; and performing measurements on radio signals to obtain for each of a plurality of locations of the site a set of characteristics of radio signals. The mobile device receives, during the collection of data, data from a barometer, verifies a most recent indication of a floor in a user input based on changes in the data from the barometer, and outputs a warning to the user in the case of a discrepancy. The mobile device associates each set of characteristics with a floor indicated by a user. The mobile device provides the sets of characteristics and their association with a floor for a generation a radio map supporting a positioning at the site.