Page 46 - ITU Journal Future and evolving technologies Volume 2 (2021), Issue 5 – Internet of Everything
P. 46

ITU Journal on Future and Evolving Technologies, Volume 2 (2021), Issue 5



                                                               5.   CONCLUSION
            • Room temperature and humidity, which indicate, for
              each room, the last detected information, while also
                                                               Early design and prototyping are fundamental to speed
              displaying a chart with the trend of the last hours.
                                                               up the development process of monitoring systems. To
              The chart toggles between temperature and humid-
                                                               achieve such a goal, proper tools and technologies must
              ity by pressing the appropriate current value
                                                               be adopted. In this respect, a domotics case study has
                                                               been pointed out in this paper, where different tech‑
            • Light status, which shows the current light state and
                                                               nologies, protocols and languages are grouped together
              the last date‑time when it toggled
                                                               without worrying about interoperability issues, thanks to
            • RaspberryPi fan control, which displays a gauge with   the capabilities of the adopted tools to interact between
                                                               themselves. In that sense, the presented approach repre‑
              the  current  real‑time  CPU  temperature.  This  is  di‑
              vided into three colour‑coded areas: (i) green as de‑   sents a viable solution for performing preliminary tests
                                                               on a domotics IoT‑based scenario. Note that other tech‑
              sired temperature; (ii) yellow above target; and (iii)
              red as over temperature.  The user can interact with   nologies could be adopted for the same purpose, such as
              the  dial  being  able  to  change  the  desired  tempera‑   MongoDB as data store, instead of In luxDB, and CoAP
                                                               (Constrained Application Protocol) as transmission pro‑
              ture. Also, it is possible to override the automatic fan
              controller with the appropriate switch and, then, use   tocol, instead of MQTT. Both such solutions are targeted
              the slider to manually set the fan speed.        to IoT and constrained scenarios; however MongoDB is
                                                               a document‑oriented (and not time based, as In luxDB)
                                                               database, while CoAP does not follow a publish and sub‑
                                                               scribe philosophy, as MQTT. For such reasons, in this
                                                               work InluxDBԀand MQTT have been preferred, since they
                                                               better  it the requirements of a domotics context, where
                                                               handling data following a topics’ hierarchy, instead of un‑
                                                               structured information, represents the most viable solu‑
                                                               tion.
                                                               Two important aspects still deserve attention, as an open
                                                               research activity: scalability and securit and privacy.
                                                               Such a kind of analysis could be carried out by de in‑
                                                               ing re‑usable modules and components to be integrated
                                                               (and replicated) in a more complex system. Instead, se‑
                                                               curity&privacy requirements can be achieved at various
                                                               levels: ranging from securing the MQTT communication
                        Fig. 6 – Domotics ‑ Vision HMI         exchange [17] [18], to protecting the data when they are
                                                               stored into the database [19] [20], or to providing secu‑
          As just anticipated, the historical database is hosted on
                                                               rity and privacy policy enforcement mechanisms at the
          an external VPS server, and it is paired with a data vi‑
                                                               IoT core platform’s level [21].
          sualization tool, named Grafana (see Section 3), that al‑
          lows the visualization of the gathered (and stored) data in
          various user’s de ined charts. The database used for this
          task is InluxDBԀ(see Section 3) mainly for two reasons:
          (i) it is designed speci ically to manage time‑based infor‑
          mation; and (ii) it is of icially supported by the Grafana
          suite. For security reasons, two separate accounts have
          been de ined on the historical database: (i) read/write,
          for the migration task; and (ii) read only, for the Grafana
          connection. As stated in Section 3, InluxDBԀde ines two
          types of data: tags and  ields. For this speci ic use case,
          the              and       ℎ       are added as tags, since they
          are mainly used in the            statement to  ilter data;
          while all the other information is stored into  ields. Two
          measurements are used: one for rooms and one for lights.
          Hence, as shown in Fig. 7, two types of panels are visible:
          (i) a chart to plot the temperature/humidity trend; and
          (ii) a panel to visualize how long the light states lasted.










        34                                   © International Telecommunication Union, 2021
   41   42   43   44   45   46   47   48   49   50   51