Nokia N97, Sports Tracker and Polar HRM: Export of workout data to Excel or Open Office

September 17, 2009

How can workout data from Nokia Sport Tracker be exported to Excel or OpenOffice, why would we like to export the data and what data is available. This is some of the questions that will be answered in this post.

Since the Polar heart rate monitor belt is now working on Nokia N97 and Sports Tracker can keep a diary of the workouts, sooner or later one would like to analyze the data in other ways that might be possible by looking at the different curves on the screen. It is very nice that these curves are available on the device, and especially if you are using the latest version 2.06 together with the route functionality in Sports Tracker and screen when doing your workout (I will post details on this later). As an example, the hart rate curve for a workout I did, is shown here on the Nokia N97. I will get back to this curve after analysis of some other data.

Heart Rate on N97

Heart Rate on N97

But even though there are important analytical tools available out of the box, it would be nice to cross-correlate the data with each other (for instance pace and altitude versus distance, or even with data not measured via Sports Tracker like humidity or temperature. The easiest way to look at several data sets at the same time is to upload the workout session to your account on the Sports Tracker web site. There is an example below. It shows the time on the x-axis and speed, altitude and heart rate on the y-axis. Very nice and certainly an improvement from looking at individual curves on the device only.

Workout Profile

Workout Profile

The first thing is to understand how the curves are created and how they compare to the curves on the device. From the following 3 pictures (altitude, speed and heart rate) which can be seen on the devise, it is easy to see that the overall envelope of the curves on the device and the curves on the workout profile on the web site is very similar. But is also seems that some kind of averaging has been done, as the curves are not completely the same. On the phone screen you have to reduce the data set because of the screen resolution, but on the uploaded data set shown on the Sports Tracker web site some averaging filters has been used as well – and not just because of screen resolution apparently.

SportsTracker Speed versus Distance SportsTracker Heart Rate versus Time SportsTracker Altitude versus Distance

On the Sports Tracker web site, just below the curves of the workout, a table list the data used for drawing the curves. Part of the table is seen below:
nokia-sportstracker-site-code-HR-Table
The workout was set with an automatic lap based on 500 meter interval. From the table it can be seen that there is a data point for each 500 meter (the lap distance). Then the speed and heart rate is listed as ”Average”, so it can be assumed that the speed at 500 meter is not the speed we had at the 500 meter mark, but some kind of average during that 500 meter distance. It could be a running average of N samples or it could be a complete average of all samples in the lap interval. To find out we have to export the workout data from the N97 to a comma separated file (csv file) which we can import into Excel or OpenOffice. I use OpenOffice exclusively, but it works the same way in Excel. In Sports Tracker the workout is exported by first selecting the session in the training diary, and then select the export function from the options menu.
Sports Tracker export menu on N97
The ”Export as CSV” is the one to select. By looking at the exported file in OpenOffice (or an text editor) it is easy to see the kind of data that are exported and the resolution of the log data (how often is data logged). Time data is exported, both real time (date and clock) and total time. The sampling (or log) interval is 1 second. For distance, relative distance (since last log point) and total distance is exported. GPS latitude, longitude and altitude data are exported. For speed, it is speed and pace that are exported. Beside the data numbers, units for the data are exported as well so the data can be interpreted correctly. But wait a second! Where is the heart rate data? Unfortunately it is not exported. It is easier to look at the xml file (export to xml) to see the exported data:

<eventlocation>
<totaltime format="HHMMSSSS">00:13:24.03</totaltime>
<realtime>2009-09-08 19:51:33.74</realtime>
<distance unit="km" >0.02</distance>
<totaldistance unit="km" >6.84</totaldistance>
<latitude>55.719252</latitude>
<longitude>11.951165</longitude>
<altitude unit="m" >57</altitude>
<speed unit="km/h" >37.6</speed>
<pace unit="min/km" >1:35</pace>
<fixquality>3</fixquality>
</eventlocation>

Each data point is exported as an ”eventlocation” and this corresponds to one row in the CSV file. No heart rate data is present . I have tried all the export formats and no one supports export of heart rate data. I will get the heart rate data from the Sports Tracker web site by using regular expression to extract the data from the html file on the site. And this I will use to analyse how the averaging on heart rate is done (if posible). But first I will find out how the 30.4 km is calculated (average speed at 0.5 km). From the CSV data I can see the speed at 0.5 km is 36 km/h. From the table above the first lap is after 59 seconds and since we have one log per second, there is 59 data points in the first lap. If all speed data is added together we get 1794.9 and divide that with the 59 data point and we get the 30.4 km/h. So the average is calculated as one number per lap, and it is based on all the data point in that interval (so it is not a running window).This “forced averaging” is also another reason for exporting the raw data to csv and analyse the data set with what ever algorithm one would like to use. If we make a plot in OpenOffice of the speed and altitude versus distance we can also see that at 0.5 km the speed is higher than 30.4 km/h. The curve is below. BTW, all dates on the OpenOffice curves are from 09-09-2009, but the correct date should have been 08-09-2009.
speed_altitude_unfiltered_from_csv-reduced
If we zoom in on the first 500 meter it is easier to see the speed which is well above the 30.4 km/h at 500 meter. There is also an average speed line, which is calculated automatically by OpenOffice, and this is similar to the 30.4 km/h as shown on the curve and table on Sports Tracker web site.
speed_altitude_unfiltered_first500_from_csv
The next is to see if the pulse rate data is available on the web site, so it could be extracted by a script, and if the curve is an average of the heart rate data available on the site, or if only the data points used for the curve is available on the site. If only the averaged data points are available we have to wait for a new version of Sports Tracker that supports exporting of heart rate data. Analysis based on pre-averaged data sets are not very useful. To get the data from the table or curve the browser was pointed to the web site, the workout was viewed and the html source for the page was opened and copied to Eclipse. In Eclipse a regular expresion was used on the html to get all the table data for heart rate. The result was:
HR-regexp-search

The file is called “NewFile.xml” because I just copied the html code in to a new (xml) file. The first number on each line in the list is the line number of the html source. Unfortunately only 20 data points where present in the html file. They are averaged heart rate, with one point per lap distance. The lap distance was 500 meter and the route was 10 km, so i total 20 data points are found. Unfortunately we have to wait for an update of Sports Tracker. But never the less, I added the data to OpenOffice to see if the averaging on heart rate was easy to sport. The result, inclusive the heart rate curve from the Nokia N97 overlayed, can be seen here:

heart-rate-versus-time-from-csv

The averaging is easy to spot. The heart rate average we have been looking at so far is based on the table. So I took the averaged heart rate and speed, and created a curve so they could be compared against the original speed data in the CSV file and also added the workout curves from the web site to the same picture, to make some final conclusions of the data set and correlation. The figure is here:

averaged_versus_workout_and_csv-reduced

First of all, the averaged heart rate is not the same as the curve seen on the low workout profile (the lower red curve). The upper red heart rate is more averaged. The upper averaged speed curve is also more averaged than the lower green speed curve from the workout profile. This tells us, that the table data is created from the complete data set, averaged over each lap distance, but the workout curves are exposed to a lighter form for averaging and not based on the table data, but on the real data set as exposed in the csv file, with the exception that we do not have the heart rate data in the csv file. I have tried to create a similar filter and use it on the csv speed data to see if we could get close to the speed curve on the workout profile, just as a ”prove” of the theory. It is seen below. The green curve is the original one from the workout profile on the web site, and the blue on eit the filtered speed data from the csv file. The shape is the same, although there are room for improvement.

speed-filter-approx-reduced

So the final conclusion is:

  • Heart rate date can not currently be exported from Sports Tracker

  • Heart rate data can be viewed via upload to Sports Tracker web site

  • The table data on Sports Tracker web site is averaged based on lap distance

  • Curves on Sports Tracker web site is based on lighter averaging of the complete log data set

29 Responses to “Nokia N97, Sports Tracker and Polar HRM: Export of workout data to Excel or Open Office”

  1. Thank you! I would now go on this blog every day!
    Saurooon

  2. I like this place very much.

    This is really a extraordinary web.
    And it is not like other money orientated website, the content here is genuinely useful.

    I am definitely bookmarking it as well as sharing it with my friends.

    :)

  3. I’m just wondering if someone has looked at this blog success review? people have said good things about the place, but it does not seem to be that known to many people?

    The person who did the review is also giving away a blog success bonus that seems pretty good. Do you guys think it’s worth the money?

    I’d love to make an income on the web from blogging, as it feels like the right thing to do considering my talents…

  4. Hello. It seems that the wearlink bluetooth transmitter is no longer available. At least that is what I am hearing. Would you know if there are other hrm belts that work with the N97?

  5. http://www.betextremesoft.com

  6. I don’t know if you are aware Lawrence Cleary, but the Polar Wearlink BT HRM belt is now available from the new Sport Tracking Technologies shop site.. I have just received mine (shipping began 1st July ’10).

    It is a shame that the HR data cannot be exported like the .gpx data..

  7. Yes, its big fail.

    What about the original data info on the phone?

    All session on sports tracker program creates a new .dat file. It contains all the session data, including all the heart rate data.

    Any idea to open, decode and export from this file?

  8. ценно , измените шрифты плохо видно

  9. It’s going to be ending of mine day, except before finish I am reading this great piece of writing to increase my know-how.

  10. 受此影响,国际现货黄金最高到达1247美元/盎司,最低则跌至1240美元/盎司,最终收盘于1245美元/盎司.

  11. Thanks very nice blog!|

  12. RE:Martin’s Blog » Nokia N97, Sports Tracker and Polar HRM: Export of workout data to Excel or Open Office НПП Валок дисковые ножи холодной резки металла колосники грохотов

  13. If you are an app developer and necessity expressive marketing then this article is championing you. We equip app downloads, app installs and app reviews so your app can escape a surmount the revelation it deserves. We eat helped thousands of customers to reach more users for their portable app. We grow your app veritable users, and then your app choice reach the google affect cooperate chart. This works after all motorized developers.

    mobile marketing vehicles

  14. http://forum307.pl/tag/warsztat-samochodowy-pruszkow
    Looking for a used or new vehicle can be a challenging method if you do not know what you really are doing. By educating yourself about automobile store shopping prior to visit the dealer, you could make issues much easier yourself. The following tips may help the next buying journey become more satisfying.

    Constantly bring a auto mechanic alongside when shopping for a fresh automobile. Car dealers are notorious for promoting lemons and you do not desire to be their next victim. Whenever you can not get a auto mechanic to look at automobiles with you, a minimum of make sure that you have him review your final selection prior to buying it.

    Know your limits. Prior to starting buying for your next auto or vehicle, decide what you can manage to shell out, and stay with it. Don’t forget to feature desire for your estimations. You can expect to pay around twenty percent as a payment in advance also, so be well prepared.

    Before attending a dealer, know what type of motor vehicle you need. Analysis all of you possibilities just before shopping so you can figure out what works for your financial budget and family members requires. Shop around to learn simply how much you ought to pay to get a prospective auto.

    Prior to signing any contract spend some time to read through each series, for example the fine print. If you find anything shown you do not fully grasp, will not indicator till you receive an answer that you fully grasp. Unsavory salesmen can make use of an agreement to put many fees which were not talked about.

    In the event you keep the previous assistance under consideration when which you go looking for a vehicle, you will end up prone to get a full offer. Getting a auto does not have to become headaches. Use the ideas with this post and you will receive the auto you need in a good value.

  15. Here and ended the craze week in fresh York in 2017. Immense week I’ll demand that you, but not without scandals and attractive spices! Signally outrageously roasting discussed models that gave us clothing. They assert many of them from elite escorts, and some outright whores. Which was picked up from the in someone’s bailiwick or from the site profound Dating. Who wants in you’ll view it can be viewed here http://sex911.top/?s=12184 There you intention see pictures and a quantity of other evidence but unfortunately solely after registration, if you are solo then you presumably will like it there).

  16. признать Белухина Д. Н. виновным в том что он простудился? нет, он взял себя в руки, и придумал нвоое лекарство, болезень не приговор!
    Зеленый чай, мед, лимон, все эти составляющие позволят победить многие инфекции не каких лекартсв и докторов !
    Врачи не всегда могут сражатся с приговором!

  17. Возможность работать дома, полная или частичная занятость. Собственный бизнес.

  18. https://georgiygrigorev.jimdo.com/

  19. Можно поподробней?
    П.С.
    купить полис осаго в Мытищи

  20. Я решила, своему мужу надо подарить коллекционную игру, только свободных денег не было и я обратилась в эту компанию. Может и вам повезет! Оформление заняло 5 минут и сразу на карту

  21. Покупайте качественные журнальные столы с БЕСПЛАТНОЙ доставкой и БЕЗ ПРЕДОПЛАТЫ.
    Подробнее на сайте: kupit-zhurnalnyj-stol.ru
    Внимание: акция -15% от стоимости производителя!

  22. ovtxhtpy search expired domains dnaytgsx

  23. Buying and selling is really a critical enterprise.
    Also then to determine whether it is very similar for not or poker we must compare trading with poker.
    What’s similar and what’s diverse?

  24. [...] between our citizens and our Government has by no means been so huge. The people are in search of trustworthy solutions, not simple solutions; clear management, not false claims and evasiveness and politics as [...]

  25. The URLs of visited websites would be the bookmarks and
    represent entire content. Follow users – this can be a usual activity in twitter,
    individuals follow each other. Brad states that all is required is usually to build your
    content and load it into Bookmarking Demon.

  26. Men who have recently undergone prostrate surgical treatment or penile surgery may also use the extender to accomplish stronger erection minimizing recovery time.
    Its delivery time period assists the sufferers to hold secrecy and privacy about
    the management of a mans impotence. Some flavors like regular tobacco,
    menthol, some camel flavor, Marlboro etc are widely used by many
    smokers.

    Here iis my site; view

  27. Fast forward a few years and possesses grow to be a hhighly effective
    means of driving traffic and creating wealth online.

    These social site’s links embody the foremost in fashion services like digg, delicious, and reddit, and abundance
    of niche and smaller amount in technique services.
    Bookmarking Demon is a software program bundle that means it is possible for that you
    address the technical element of issues.

  28. I believe everything posted made a bunch of sense. But, what about this?suppose you added a little content? I mean, I don’t wish to tell you how to run your blog, but supposeyou added something that grabbed folk’s attention? I mean Listen to the 2012 Holiday Show | The RadioTheatre Project | Live Radio Theatre at The Studio@620 is a little boring.You should glance at Yahoo’s front page and note how they createpost titles to grab people to click. You might try adding a video or a picture or two tograb readers excited about what you’ve written. In my opinion, it might bringyour blog a little bit more interesting.

  29. Доставка товара В Л Л ДЕ Волгоград осуществляется в предельно короткие временные промежутки.
    Наибольшее использование проверенных входе множества лет научно-технических приемов и методов решения проблем очистки Канализационных насосных станций с внедрением технологических и сегодняшних изобретений и достижений.
    Наш завод Е В Е Л Д Алексин непрерывно проводит работу над конечным качеством изгатавливаемой продукции и совершенствует свойства изготавливаемого оборудования,использовав при этом все более новые необычные технические тонкости и методики производства и сборки в Великий Устюг.
    Завод EELWD производит в Верхнеуральск системный научно-технический производственный цикл: от заготовочного до автоматизированной и механической обработки.
    Все изготовленные и сконструированные и готовые и выпущенные и созданные и сделанных и сделанные и произведённые и выработанные детали подвергаются контроль потребительского качества с применением современных точных мерных инструментов (изделий).

    механическая очистка сточных вод песколовки – механическая очистка сточных вод

Leave a Reply