Bluetooth® CS using PBR

Why is the algorithm to calculate distance not part of the Bluetooth specification?

Bluetooth 6.0 introduces many exciting features, including Channel Sounding (CS), which enables applications to leverage phase-based ranging (PBR) to estimate distances. However, it’s crucial to clarify that the Bluetooth 6.0 specification itself does not include a direct algorithm for distance calculation. This responsibility falls on the application layer, which must use the raw data from the Channel Sounding procedure to compute distances.

So why did Bluetooth SIG leave out this algorithm? One reason lies in intellectual property (IP) protection. Many RADAR technologies and distance-calculation methods are already protected by patents, and the inclusion of a specific algorithm in the standard could potentially violate existing IP. By leaving the implementation of distance algorithms up to developers, the Bluetooth SIG avoids IP conflicts while allowing flexibility for innovation. Metirionic, for example, has invested heavily in its own distance calculation solutions that reside above the Bluetooth 6.0 protocol, and delivers precise distance measurements.

Advancing Distance Accuracy: From Bluetooth Channel Sounding insights to precision accuracy enabled with MARS

In our previous article on MARS Indoor Performance, we explored how the Metirionic Advanced Ranging Stack (MARS) achieves remarkable accuracy in indoor environments, even with the challenges of multi-path propagation. While that article focused on MARS’s potential in difficult environments, today we’ll discuss the distance calculation algorithm that is described in section 2.2.2.1, Phase-Based Ranging (PBR), from the Bluetooth® Channel Sounding – Technical Overview.

This time, using the previously explained cable car test setup, we captured the Channel Sounding PCT data (Mode 0 and Mode 2) of 10 Channel Sounding procedures at every centimeter along a 11.5-meter track. Using the phase difference between two received signals, as captured in the following formula from the Bluetooth Channel Sounding Technical Overview, we computed the distance:

Where c is the speed of light, (Pf2 – Pf1) is the phase difference and (f2 – f1) is the frequency separation.

Bluetooth CS using PBR: Results and Analysis

Figure 1 illustrates the measured distances compared to reference distances, showing that while cm-level accuracy was achieved in some positions, there were significant deviations in others. For the majority of the track, the distance error exceeded 1 meter when using PBR, with the worst point showing a 6-meter error.

Figure 2 breaks down the error per measurement index, and Figure 3 visualizes the cumulative distribution function (CDF) of these errors. The x-axis in the CDF represents the error (in meters), and the y-axis indicates the probability. The orange line at one standard deviation (one sigma) shows that 68% of errors were below 1.75 meters. The green line, representing two standard deviations (two sigma), illustrates that 95% of errors were within 4 meters.

While these results using PBR fall short of the accuracy we achieved when applying MARS to Bluetooth Channel Sounding, they still outperform traditional RSSI-based distance estimations. The key to this performance lies in the simplicity and efficiency of the PBR algorithm, which requires minimal computational effort and offers reasonably accurate results for a number of applications.

The Challenge of Multi-Path Propagation

Despite both devices maintaining line of sight throughout the test, multi-path propagation caused significant measurement errors. Radio signals radiate in all directions, reflecting off surfaces and objects, before merging with the direct signal traveling along the line of sight. The measured phase is thus an aggregate of all these paths, leading to inaccuracies—a phenomenon known as multi-path propagation.

This is where MARS shines. Our Pathfinder algorithm excels at separating the direct path from the multi-path signals, delivering superior accuracy even in challenging environments. Bluetooth Channel Sounding with PBR, while useful, struggles to achieve this same level of precision without additional algorithms like MARS.

Conclusion: Bluetooth Channel Sounding using MARS

At Metirionic, we’ve developed a patented portfolio that includes the ability to integrate Bluetooth Channel Sounding with MARS. If your application demands higher accuracy than what PBR can provide, ask your preferred silicon vendor whether MARS is available for your devices using the Bluetooth 6.0 protocol stack. MARS offers a unique, patented solution to overcome the limitations of PBR and multi-path propagation, delivering the performance you need.

For more information on our MARS technology and its integration with Bluetooth Channel Sounding, feel free to contact us!