“When we invest in infrastructure, we need it to be reliable!” – said every buyer of all infrastructure projects ever.
Of course, no one in their right mind would want to build business processes on a foundation of “unreliable” infrastructure. But what do we actually mean when we say “reliable”?This article will explore what the word means in the context of Bluetooth® data connectivity and positioning systems.
Accuracy (the quality of being precise), consistency (the quality of always performing in a similar way) and robustness (the quality of being strong and unlikely to break or fail) are all terms that you often hear used in relation to positioning systems, such as the Quuppa Intelligent Locating System™. The words are also at the core of what we mean when we talk about reliable systems.
Let’s first consider the broader definitions of these words through the example of archery, where an archer uses a bow to shoot arrows at a target. In this example, accuracy is the archers ability to hit the center of the target board and consistency is the archers ability to repeat this performance, hitting the same area on the target board time and again, shot after shot. Robustness on the other hand, is the archers ability to do both of the above despite changing environmental conditions (e.g. gusts of wind) or distractions (e.g. cheering from the audience).
Accuracy, consistency and robustness are key to the reliability discussion because they are the three pillars that form reliability. Using the archery example above, a reliable archer is able to keep hitting the target the same way despite changing conditions and distractions around them.
So how does this translate to the world of positioning systems? Well, let’s see:
- Accuracy is being able to provide the precise location (e.g. centimeter-level precision) for an object with minimal latency (e.g. within a fraction of a second). For more information, see our article Location Data – How accurate is Accurate Enough?
- Consistency means how well the system can continuously calculate a similar position (within a reasonable target area) for the same static tag. In other words, will the system continue to place the tag in the same location when the calculation is repeated.
- Robustness is the ability to continue providing accurate and consistent results even in suboptimal conditions (e.g. metallic environments or with high air interface traffic) and changing environments (e.g. moving heavy machinery, adaptable spaces or crowded areas).
Again, reliability is the combination of these three. From the positioning perspective, reliability is the ability of a positioning system to provide accurate positioning results consistently regardless of any environmental factors that may challenge the robustness of the system.
Reliability for Different Bluetooth Systems
Now that we have defined the word reliability to mean that the system can continue to deliver the promised result without fail even in challenging conditions, we can notice that the precise definition of a reliable system will depend on the industry, technology and type of system configuration in question. The key deliverables for a system effectively determine whether the system is reliable for its purpose or not. When we think about it like this, it is easy to see how the definition of reliability will also differ between Bluetooth systems designed for different uses, for example, Bluetooth communications systems as compared with Bluetooth positioning systems. This difference becomes even clearer when we take into consideration the latest advances introduced by Bluetooth Direction Finding (for more, see our article Bluetooth Direction Finding: Going Beyond Beacons). So let’s take a closer look at what reliability means for both Bluetooth communication and positioning systems.
Bluetooth was originally developed for communication systems to send data effectively and power efficiently between devices. For example, a song from your mobile phone to your headphones or a file from your computer to your friend’s mobile device. For such systems, accuracy is defined as the ability to successfully transfer the data from device A to device B. It doesn’t really matter how the data travels through the space between the devices, as long as the data is received successfully. From an application perspective, this effectively means that it is important for the Bluetooth communication layer (link layer) to be able to meet the quality of service requirements, such as the communication rate in bits per second (bps).
However, Bluetooth technology has since also been adopted for positioning systems, where the definition of accuracy is more stringent. In this case, receiving the data communication (radio packet) reliably from an object is not enough to ensure that its position can be calculated accurately. The question of how the signal gets from A to B becomes much more significant as it affects the positioning accuracy of the whole system. All positioning systems require this information, but it is good to remember that different technologies solve the challenges of radio propagation, such as multipaths, differently. For example, ultra-wideband technologies (UWB) exploit the time/delay domain while Bluetooth Direction Finding technologies use spatial information. As this information is so important for fulfilling the accuracy requirements of the system, it is crucial that the tag design is optimised for the purposes of positioning (for more on tag designs, please see our article Tags are not Beacons).
When we turn our attention to consistency, we notice that for communication systems, being able to provide consistency is to be able to continuously transfer data successfully from device A to device B over time. Doing so is again dependent on the quality of the Bluetooth communication layer.
As with accuracy, the challenge of consistency is more complex for positioning systems. For each new calculation, the positioning engine receives a range of new, often independent and uncorrelated, information that it needs to process and filter through to get to the relevant positioning data. This means that in order to provide consistent positioning results, the system must be intelligent enough to use advanced logic to pick out the relevant information from the baseline noise of data that it receives and combine the relevant data into a coherent best estimate.The quality of the positioning engine algorithm, its logic and processing capabilities are key to providing consistency for positioning systems.
All consistent positioning technologies implement their processing logic around physical measurements, but the types of measurements used vary from implementation to implementation. For example, UWB logic measures time (or in some implementations the delay or time difference) from wideband radio signals, effectively extracting range measurement estimates from the noisy data. For Bluetooth positioning systems, the introduction of Bluetooth Direction Finding opened new doors. By measuring the phase of the signal instead of it receiving signal strength indicators (RSSI), the positioning accuracy is no longer directly impacted by power level fluctuations. This means that the Bluetooth Direction Finding measures angles (or spatial directions) from narrowband radio signals to extract the relevant data from the phase of the received signal, rather than the power as is the case with beaconing technology. Overall, if well implemented, the performance of UWB and Bluetooth Direction Finding are very similar, optimally down to the centimeters level. Both can deliver results that are accurate and consistent. The important thing to remember is that by using advanced signal processing, it is possible to achieve consistent measurements from noisy data, regardless of whether data are measured in the time or space domain.
Now let’s take a look at the third variable of reliability: robustness. For communication systems, robustness is the ability of device A to successfully transfer data to device B regardless of environmental factors such as the number of users in the same area, the type of environment (e.g. indoor, outdoor, open or confined) or where the device is in the environment (e.g. in your pocket, in a bag or in your hand). In other words, this means that a robust communication system can transfer data successfully from device to device regardless of environmental factors.
For positioning systems, robustness is being able to provide accurate and consistent positioning results even in challenging environments. For example, in places that are constantly changing in layout with objects moving around dynamically, or in places with objects that impact the way radio signals propagate through the space, like machinery made of metal or elements made of absorbing materials. The challenges of radio propagation phenomena explored earlier also create additional requirements for the robustness of systems, putting additional pressure on the IT infrastructure, which needs to be carefully planned and designed to ensure that all areas of interest are well covered and so that any environmental challenges can be mitigated.
From the robustness perspective, the Bluetooth Direction Finding methodologies offer some great advantages over technologies based on trilateration principles. For example, in terms of easier deployments and cost effectiveness. Trilateration methods (e.g. those used by UWB) always require a minimum of three anchor points to calculate the location of the tracked object, while the angular based methods offered by Bluetooth Direction Finding allow each Locator to operate independently to compute the position of the target. This is because such systems (e.g. the Quuppa Intelligent Locating System™) can exploit two-dimensional angular measurements, i.e. measuring both the azimuth and elevation angles, to effectively create a spotlight over the tracked area. Like with actual spotlights on a stage, one light is enough to illuminate the performer, but it is common practice to use multiple lights together to create a stronger effect. One of the key advantages is the robustness that this builds into the system, as even if one of the multiple lights turns off the performer can still be seen, just like the tracked object can still be found by the system as long as one Locator can “see” the object. In the case of UWB, at least three devices always need to “see” the tracked object in order to compute its location. In practice, this means that positioning systems using angular information are inherently simpler, their deployment is easier and they are more intuitive to plan. This all adds up to a system that is more cost-effective and more robust to operate, even in harsh industrial environments.
Once again, reliability is a combination of these factors. From a communications perspective, the system needs to be able to always successfully communicate information repeatedly regardless of environmental factors to be reliable.
From a positioning perspective, the reliability requirements are more stringent, requiring the system to be able to provide accurate positioning information consistently in challenging and dynamic environments. All of the elements (i.e. tags, positioning engine algorithms and infrastructure) have to function harmoniously to be able to provide a reliable dot on the map. While positioning systems based on both UWB technologies and Bluetooth Direction Finding methodologies can provide similar levels of accuracy and consistency, the latter can also provide significant advantages for robustness, adding noticeably to their reliability levels.
As we can see, the use of the word reliability when it comes to systems is not quite as simple as it seems at first. If comes with a backdrop of other terminology that is essential for truly understanding what the concept means. And as if that were not enough, it even changes to mean different things in different industries, for different system configurations (e.g. communication vs positioning) and with different technologies. Of course, everyone wants a reliable solution for their business, but it is equally important to define what that reliability means for your business so that you can select the right positioning system for your use case.