Energy Efficient indoor LOcalization leveraging UWB radio for Crowded environment

Scientific context

Knowing the position of a device in its environment becomes an important issue for applications providing services based on the position. Many solutions have been proposed going from satellite-based techniques (i.e. GPS…), video tracking or radio-electrical waves combining… However, these techniques reach their limit when very high accuracy is required (lower than 10 cm), especially in indoor environment.

In such a stringent requirement, low power radio as Ultra Wide Band (UWB) or Bluetooth based techniques have emerged as accurate solutions. UWB are radios having relative bandwidths larger than 20% or absolute bandwidths of more than 500MHz. Such radios combine low to medium rate communications with positioning capabilities using ranging techniques [1]. If UWB offers outstanding accuracy, the performance degrades significantly in severe environments, i.e. in presence of multipath due to crowded rooms or when machines generate impulsive noise in factories. On the other hand, since Bluetooth 5.x version, localization using Angle of Arrival (AoA) method is available. This technology improves already used methods based on Received Signal Strength Indication (RSSI) with Beacon Bluetooth anchors and tags that suffered from a lack of accuracy in crowded environment.

Moreover, to propose energy efficient solutions, a complete cross layer approach has to be envisaged, including hybrid algorithms combining Two Way Ranging (TWR), AoA methods [2] or Phase Difference of Arrival (PDoA), as well as dedicated MAC protocols. Combining multiple radio protocols may be considered depending of the application (energy efficiency, accuracy, range, …).

General objectives and expected results

The first phase of this work will be dedicated to the study on existing UWB and Bluetooth communications and localization techniques [3]. It exists many localization techniques among which we can cite the trilateration but some others have already been developed in the GRANIT team [4], either opportunistically selecting best anchors for trilateration [5] or using embedded Inertial Motion Unit that gives direction information [6]. 

A recent version of Decawave/Qorvo UWB transceiver embeds two antennas, offering the possibility to perform AoA or PDoA estimation. On the one hand, this technique simplifies the MAC protocol and increase the scalability because exhaustive TWR are no more required, but on the other hand this method does not support low angles. The candidate should then be able to improve these techniques considering application-specific aspects (number of persons/sensors, accuracy required, orientation, noise, speed of the persons…) and even propose a hybrid, adaptive localization scheme that takes profit of the complementarity of TWR, PDoA, AoA or direction-based approaches.

The advent of the UWB Active-Passive Two-Way Ranging (AP-TWR) protocol researched in [7] opens up a new way of providing robust positioning in the presence of NLoS conditions, and can provide dramatic energy gains. This new approach has to be considered in our hybrid localization scheme.

Then, these techniques will be implemented in a new version of the Zyggie [8] platform developed in the team, which embeds Decawave/Qorvo UWB transceivers.

Local context and regional expectations 

UWB is recognized as the most efficient solution for indoor positioning and was recently adopted by most of smart phones providers, offering numerous opportunities for localization-based applications and services. Several SMEs are therefore currently investigating the efficiency of this promising technology. GRANIT team was involved with Ticatag and Klaxoon, in Plug&Pos project whose aim was to bring additional content to a visitor in a museum once he is localized and we know his orientation. UWB solutions are also promising for Factory of the Future, to help industry manage the different flows of goods or persons. Orange Labs, Prolann, Axalon and GRANIT were associated to explore UWB positioning performance in such a context in Mo.Di.Flu project. Industry 4.0 is indeed considered by Orange Labs as a key target for offering new localization-based services for factories, but energy efficiency definitely remains the major challenge for a wider adoption.