8-June-2020 By Jeffrey Cooper

Sensored: A la Carte

Sensors are everywhere, and in Part 1 of 2 of this short series, we will examine the a la carte use of wearable sensors. What this means it typically a sensor is provided by a company with a bespoke app to use it. Exceptions to that are heart rate straps (chest or arm), which are ubiquitous, and Smart Watches, which are a complete platform.

For those that know me, it should come as no surprise that I am a fan of wearable sensors– smart watches and sensors galore. I’ve worn many to see how they work and what their potential is. Besides Samsung and Apple watches, I have used devices from LetsFit (low-cost, watch-like trackers), Moov (ankle sensor), Lumo Run (mid waist on the backside), smart ear buds from Samsung (heart rate and pedometer, but this was dropped in later versions), and heart rate straps from MyZone, Polar and Sports Tracker.

For all that I have used, there are still some I haven’t. I’ve not used a smart ring yet, a heart sensor that you wear on the forearm, smart glasses or smart shoes. That latter two are quite new and experimental. I don’t cycle, so I haven’t used cadence and power sensors. Smart Clothing is in its early days, too. And I’ve not used later devices from Polar or Garmin, and I’ve not used Suunto devices yet.

For this article, we’ll explore current devices and their a la carte use in common or bespoke applications.

A Matter of Perspective

For general workout analyses, it’s good to have the sensors appropriate to that workout. Using only devices widely available today, I would want to gird up with the following sensors depending on exercise:

• Running- Smart Watch (for GPS, steps), Chest or Forearm Strap for more accurate heart rate, and an ankle sensor from Moov or foot sensor from Boltt. You could also use a posture sensor from Lumo clipped on your waist below your lower back.
• Cycling- Smart Watch, Chest or Forearm Strap, ankle sensor from Moov or a cadence sensor on the bike, and a power sensor.
• Swimming- Smart Watch, underwater-capable heart rate sensor, Moov sensor
• Gym workouts- Smart Watch, Chest or Forearm straps

Running

Using sensors to measure aspects of a workout are valuable to improving how you do a particular type of workout. As your form and execution improve, you can measure how your health improves in response to that, looking at frequency and caloric burn, and comparing with heart rate recovery and resting heart rate. If you track sleep, you can also measure to see how sleep is impacted over the long term. Wearable close the feedback loop for you.

We’ll go through the cardio sports listed above. For running, though you might think a smart watch is good enough since it has heart rate and GPS, I would argue for two additions to it. First, a chest strap or arm strap. They are more accurate than a watch, but if budget’s an issue, the watch is good enough.

Secondly, if you can, use a foot sensor. When I started sprinting, I had already been using a Moov sensor, which for running, is worn around the ankle. When I examined my run data during a sprint, I was stunned to see I was pushing 18g’s for force on impact on the pavement. Armed with that data, I immediately went out and bought newer shoes for sprinting. Once I log a couple of hundred miles of sprints in them, those shoes are “retired” to the general running pool. The idea there is use them for the higher impact sprints while the padding is most effective.

The Power of Analysis

You can see how Moov analyzes runs in the picture above. I’ve combined a few screens and animated in the graphics that you normally would swipe through. If you’ve read my article on run tracking, you’ll see I use Strava and FITIV primarily. I used to use Moov as well until my second sensor from them died. I do, though, recommend you get one. The insight it provided was invaluable.

Regarding impact, which I said was invaluable in protecting my feet and joints, I’ve zoomed in on the Impact Graph and annotated it.

You can see that the average impact during the sprint portion of the run was between 16-17g. There was, however, a spike of 22g, on the 4th sprinterval (my word for these workouts).

Lumo is a device I used for a few months, which clips on your waistband in the small of your back. It primarily measures posture while you run, examining your tilt and form. It’s useful to correct your form. When I used it, the limiting factor was you had to use it on flat ground. Running up and down hills naturally tilted your body in response, and it mistakenly tried to correct posture as a result.

Swimming

Another great sport for Moov is swimming. The upside is that it’s quite accurate at capturing every nuance of your swim (almost- more in a minute). The downside is that you get no feedback during your swim, since the device is just a button. You can’t swim with your phone, so the button just gathers data for post-analysis. But that post analysis is great.

The other issue I have is that Moov completely misses Drill Mode, which is when you use a kickboard to isolate the legs for only kicking. For swimming, Moov is worn on the wrist. With no arm motion for Drill Mode, Moov misses that part of the workout. Also, Moov does not have, by itself, a heart rate sensor. They do sell a water-capable one, though I did not purchase one.

So, part of your workout will be missing. What is interesting is that Moov supports other sports, including cycling and cardio boxing. For boxing, Moov actually supports the use of two sensors, to capture the motion for both arms. If only they had considered the same for Swimming- one sensor for the wrist, and one sensor for, ideally, the opposite ankle.

I know there is a complication to that comparison. When cardio boxing, you are on dry land and have your phone present, so both sensors are coordinated in near-real time on the phone. When swimming, both sensors would operate independently, and the data would have to be fused and sorted after the fact. However, I am quite sure that, with the benefit post-analysis gives you, they could align that data as most of the kicking would line up with arm strokes, with a few gaps in there where there was just the kick board. Also, there will be gaps where there is no kicking, when you use hand paddles. But there should be plenty of data to align these two. During these gaps, there will still be some accelerometer data from launch, stop, and flip turns to properly sync up the data.

Moov isn’t the only way to measure swimming. Swim.com works on most smart watches- Apple Watch, Samsung watches, and Wear. Unlike Moov, because they are on a smart device, they display real time data. That’s largely a plus, with a small negative. Because they interpret the data in real-time, a phenomenally hard task, they don’t have the luxury Moov has of post-analysis, which can apply context to help make the reporting more accurate. But for the rare time Swim.com gets it wrong by missing a lap, they make up for it because of Drill Mode support. That allows you to capture kickboard only laps. They also have coached workouts, so have more utility beyond just measuring the workout. And they use heart rate from the watch, as well as it can be measured in the water (wear the watch snugly).

Cycling

There are quite a few sensors for cyclists, too. I’m not personally a cyclist, so I won’t go into depth. But if you want to cycle, besides tracking your bio parameters, you can get speed, power and cadence sensors for your bike to measure every aspect of it. These are not actually wearable sensors, but are close enough for inclusion in this article because they directly measure bodily effort.

Cadence sensors measure the RPM of your pedals, which of course changes when you change gear. Speed sensors measure RPM of the tire, and using the tire circumference, calculate distance and speed, independent of GPS. Power Sensors capture the energy you put into cycling through your feed. They are strain gauges and it is a pure measurement of power. Together, Cadence, Speed and Power are the main 3 aspects for cycle training. And of course your heart rate is important and will be relative to the power you put into it.

Smart Rings

Currently, I’m not aware of any smart rings that directly measure workouts. They are primarily used for what I call ambient data, like steps, heart rate sampling throughout the day, etc… Some models also track sleep, such as Oura’s latest model, and body temperature. They are useful if you don’t have a smart watch, or you have one that doesn’t monitor sleep.

Ear Buds

Ear buds are pretty passive devices overall. A few years back, Samsung tantalizingly introduced their first wireless buds that were loaded with functionality. They could store 4GB of music per ear (right and left channels), and had a pedometer and heart rate sensor built in. You could go running with only their buds- no phone, and no watch, and get a rough measurement of it. Distance wasn’t super accurate, since it was based off a pedometer, but it wasn’t too bad. And you got heart rate measurement, so you could calculate approximate calories burned. With all that storage, you could listen to a lot of music, too.

That did not last. The next model was stripped down, and Apple has chosen, so far, to not include such features.

The Gym

When it comes to gym workouts, there currently isn’t much in the realm of personal sensors that will help you out beyond a chest or arm strap for heart rate, and a smart watch. Newer cardio machines can pair with your phone and give you some of your data that you might outdoors at least. Currently there are a couple of options for selectorizer machines, but not much else. Hopefully that will change in the near future.

I have my own thoughts about Buds, Rings and other wearable sensors, which we’ll address next in the future-looking article Sensored: Fusion.

Footnote- Sleep

There is one other sensor I use- SleepTracker. It’s a sleep sensor that goes under the mattress below your pillow. They provide two sensors, and all me to track myself, while my wife track’s her sleep separately. Though both sensors are located under our respective pillows and just a few feet apart, they are able to track us separately. Through 18 inches of mattress, they track movement, breathing and pulse. I’m amazed how well it works and believe that, for the time being, it works better than wearables. SleepTracker was developed by FullPower, a leader in sleep measurement technologies and known for their early work with Jawbone.

The screen shots above were taken just now and show an analysis of my sleep last night. Ideally my sleep score should be higher as 75 is considered a healthy minimum. Getting to bed a little earlier would have helped as well as not drinking water just before bedtime. The little person is when you actually get up out of bed.

This is a good demonstration of my full-circle approach to fitness. If you want to improve your health, you should measure the energy that goes in (food, water), the energy you expend (exercise and ambient activity), and how well you sleep.