Bluetooth® Channel Sounding

Achieving Exceptional Indoor Performance using the Metirionic Advanced Ranging Stack

The Bluetooth SIG just released the latest version of the core specification, version 6.0, which debuts a fundamental new capability offering secure fine ranging which is called Channel Sounding. Metirionic has been developing wireless distance measurement solutions since 2013 and is proud to showcase capabilities and measurable advantages of MARS for Bluetooth Channel Sounding.

 

MARS is now in its second generation (known as MARS 2.0), which is a complete rewrite from scratch. MARS 1.0, developed in 2019, was tied to individual devices, with each device having its own implementation. While v1.0 provided a good solution, MARS 2.0 introduces a clean separation of layers to offer a highly configurable version that is easier to port to different hardware platforms. This improvement enhances performance and flexibility across a variety of devices, making it a more efficient and scalable solution.

 

This new functionality shows the growing need for improved ranging accuracy and location services in Bluetooth LE (BLE) devices and applications. Further, it unleashes the creativity of engineers and product designers looking to deliver new capabilities including secure access, digital key, proximity detection, and location services for consumer, industrial, automotive and retail applications.

 

This article builds upon the previous article entitled “Metirionic’s Advanced Ranging Stack (MARS) Delivers 10cm Accuracy” offering detailed performance results captured in real-world indoor environmental conditions using the Metirionic cable car tester.

Streamlining Distance Calculation with MARS for Accurate Bluetooth LE Ranging

MARS: Simplifying Bluetooth LE Channel Sounding and Distance Calculation

The new Channel Sounding feature, as specified by the Bluetooth 6.0 specification, introduces significant potential for secure fine-ranging applications. However, it does not inherently provide distance results. Instead, the specification states: “It is the application’s responsibility to use the channel sounding data provided by the controller to calculate distance measurements.” This places the burden on the application developer or Bluetooth integrator to devise their own algorithm to calculate distances based on the interim results of the Channel Sounding procedure.

 

Recognizing this challenge, Metirionic offers a comprehensive solution with the Metirionic Advanced Ranging Stack (MARS). MARS is a highly configurable middleware software designed specifically to support Channel Sounding (CS) using Bluetooth Low Energy (BLE). It enables devices to characterize the propagation path between themselves and a connected peer, facilitating accurate distance estimation and angle of arrival for digital key applications, precise positioning, proximity services, and navigation solutions.

 

MARS is designed to take the results of a Bluetooth Channel Sounding procedure as input, specifically the Phase Correction Term (PCT) data containing in-phase and quadrature-phase (I/Q) values. These PCT data are passed over the Host Controller Interface (HCI) to the application layer, which typically resides on the host controller. The host controller could be a separate microcontroller, while the actual Channel Sounding procedure occurs at the Bluetooth device itself. This separation of responsibilities allows MARS to process the data in a flexible, modular architecture, offering seamless integration with various hardware configurations.

 

By leveraging the PCT data, MARS simplifies the process of distance calculation, ensuring accurate interpretation of Channel Sounding results. Developers no longer need to devise complex algorithms to extract distance measurements, as MARS efficiently processes the data to provide reliable results. This allows engineers to focus on integrating advanced Bluetooth LE functionalities into their products, while MARS takes care of the precision and performance behind the scenes.

 

With MARS, Metirionic has paved the way for developers to seamlessly integrate advanced ranging capabilities into their applications, eliminating the complexities of distance calculations and optimizing performance for Bluetooth LE-based devices.

Challenges of Multipath Propagation in a Real-World Office Setting

Indoor Test Environment

Indoor environments typically represent very challenging conditions with the effects of multipath propagation, absorption/attenuation, signal combining and cancellation, drastically and adversely affecting wireless technologies. The environment shared here is no different, given that it is a German office building consisting of prefabricated steel concrete slabs that notoriously reflect radio waves while simultaneously attenuating them as they pass through. Needless to say, this is a very practical environment to showcase the benefits of MARS enabling indoor BLE Channel Sounding measurements. The test corridor measures 14 meters (~46 feet) in length, 2.8 meters (~9 feet) in height, and 3 meters (~10 feet) in width, posing additional complexities with signal reflections and attenuation, factors critical for evaluating the real-world performance of MARS.

Precision Testing with the Metirionic Cable Car System for Accurate Distance Measurement

Metirionic Cable Car Tester

To capture accurate distance measurements, Metirionic custom-built the Metirionic Cable Car Tester as a precision system driven by a stepper motor, enabling the cable car to travel along a 12-meter track with sub-millimeter accuracy. The stepper motor, controlled by customized GRBL firmware — a common language for CNC machines — provides precise positioning of the car, ensuring the accuracy required for Channel Sounding experiments.

 

Here is a reference to a video:

YouTube

By loading the video, you agree to YouTube’s privacy policy.
Learn more

Load video

 

This video showcases the indoor testing of the Metirionic cable car system for Bluetooth Channel Sounding. Powered by a stepper motor connected to a toothed belt via a gear wheel, the cable car is capable of precise movements along a 12-meter track with sub-millimeter accuracy. Designed for repeatable and consistent measurements, these tests ensures optimal performance for Bluetooth Channel Sounding Devices and the Metirionic Advanced Ranging Stack (MARS) firmware.

Evaluating MARS Performance in Static and Dynamic Indoor Tests

Performance Results

Two types of tests were conducted to evaluate the performance of MARS in this challenging indoor environment. The first type involved static measurements, where distance measurements were taken at specific spots along the track while the cable car remained stationary. The second type of test was dynamic, with the cable car moving continuously along the track while distance measurements were captured in real time.

 

Leveraging the test devices running MARS in the highly reflective and absorbing indoor environment Metirionic utilized the cable car tester to measure distances between 0 and 11.5 meters, capturing results and comparing against reference distances to determine ranging accuracy.

Static Test

Reference vs Measured Distances (Figure 1)

In the static test, the cable car was positioned at specific points along the track.

Figure-1  illustrates the reference distances in blue, the measured distances in orange, and the corresponding error in green. As observed, the error margin fluctuates between 0 meters and 1.4 meters (~4.5 feet) across the measured distances corresponding with the challenging indoor environment. Considering the conditions, this is quite impressive because MARS is designed to mitigate the negative effects of multipath propagation, absorption, and attenuation that are commonly found in complex indoor environments. The reflective surfaces of the steel concrete slabs and narrow corridors create highly difficult testing conditions that typically cause severe inaccuracies in wireless ranging. However, MARS’s advanced algorithm compensates for these factors, enabling consistently high accuracy even in this demanding scenario.

Reference vs Measured Distances (Figure 1)
Static Test

CDF Analysis of Static Test Results (Figure2)

Figure-2 illustrates a Cumulative Distribution Function (CDF) of the distance errors showing just how accurate the solution is. Here, the x-axis represents the error in meters, while the y-axis indicates the probability. One sigma (68% of all errors) corresponds to an error smaller than 0.65 meters (~2 feet), shown by the orange line. Meanwhile, two sigma (95% of all errors) highlights that errors were smaller than 1.25 meters (~4 feet), shown by the green line.

 

CDF Analysis of Static Test Results  (Figure2)
Dynamic Test

Reference vs Measured Distances (Figure 3)

In contrast, the dynamic test involved moving the cable car continuously along the track while capturing real-time distance measurements. The results of the dynamic test are presented in Figures 3 and 4. While the static test results (Figures 1 and 2) show the reference distances in a “stair” pattern due to the fixed positioning of the cable car, the dynamic test results depict the reference distances as a smooth ramp, reflecting the continuous motion of the cable car.

Figure-3 illustrates the reference distances in blue for the dynamic test, showing a consistent increase as the cable car moves along the track, alongside the measured distances in orange and the corresponding error in green. The error margin fluctuates between 0 and 2 meters (~6.5 feet) across the measured distances, reflecting the challenging indoor environment. Despite the continuous motion of the cable car, the error remains within a similarly small range as observed in the static test. This is particularly impressive, as maintaining such accuracy while in motion demonstrates the robustness of MARS’s ability to handle dynamic conditions and mitigate the typical effects of multipath and signal interference.

Reference vs Measured Distances (Figure 3)
Dynamic Test

CDF Analysis of Dynamic Test Results (Figure 4)

Figure-4 illustrates a Cumulative Distribution Function (CDF) of the distance errors, showing just how accurate the solution is. Here, the x-axis represents the error in meters, while the y-axis indicates the probability. One sigma (68% of all errors) corresponds to an error smaller than 0.65 meters (~2 feet), shown by the orange line. Meanwhile, two sigma (95% of all errors) highlights that errors were smaller than 1.3 meters (~4 feet), shown by the green line.

The performance results from both static and dynamic tests demonstrate the consistancy and precision of MARS in complex indoor environments. MARS’s ability to maintain low error margins under challenging conditions, including multipath interference and signal attenuation, highlights its effectiveness in supporting Bluetooth Channel Sounding applications. Whether in stationary or moving scenarios, MARS delivers consistently high accuracy, making it an ideal solution for proximity services, navigation, and location-based applications. As Bluetooth technology continues to evolve, MARS is well-positioned to meet the increasing demand for precise ranging and positioning in a wide range of industries.

CDF Analysis of Dynamic Test Results (Figure 4)
Leading the Future of Bluetooth Channel Sounding with MARS

MARS Paves the Way for Channel Sounding

These results begin to demonstrate the consistancy  and accuracy that MARS brings to Bluetooth Channel Sounding solutions, especially in the more challenging indoor environments that suffering from substantial radio wave reflections that cause multipath. MARS’s ability to maintain high accuracy across varying distances, with most errors below 1.25 meters, positions it as a leading solution for Bluetooth Channel Sounding implementations.

 

As Bluetooth market adoption continues its astounding trajectory towards 7B annual shipments, the new capabilities afforded by Bluetooth 6.0 for secure fine ranging will only fuel that ambition and no doubt prove to be a game-changer in what we can expect from emerging devices and sensors embedding Bluetooth LE technology. Metirionic’s MARS stack is well-positioned to meet the demands of Channel Sounding and is offered for license in both binary and source code, sitting just above the Bluetooth protocol stack. MARS’s ability to deliver accurate ranging in complex indoor environments is critical for applications such as asset tracking, indoor navigation, proximity services, and Real Time Location Services.

 

Stay tuned for more performance results of MARS in additional environmental settings, including outdoors, extended range, and in Non-Line of Sight (NLoS) conditions. Metirionic’s commitment to refining and advancing BLE ranging solutions ensures that MARS will remain at the forefront of innovation in Bluetooth technology. Interested in understanding how your products may be able to leverage Metirionic’s software solutions, engineering services, technology licensing and demo kits, just reach out by emailing us contact@metirionic.com. In addition to discussing your interests and our capabilities, Metirionic can highlight how performance testing specific to your hardware and/or environment can be performed.