Separate the last part of the Requirements

The last part of the Requirements was moved to the separate section called
'System design' following the notes from the consultant.
master
Peter Babič 8 years ago
parent bbaf9693bb
commit 7194733be4
  1. 22
      analytical.tex
  2. BIN
      tukethesis.pdf

@ -48,40 +48,42 @@ They are also divided to a hardware part and software part. Software is easier t
\textbf{Optional:}
\begin{itemize}
\item Access to \gls{gui} outside of local network
\item Control wi-fi repeater mode to strengthen the signal for client nodes
\item Control Wi-Fi repeater mode to strengthen the signal for client nodes
\item Send measured data to the cloud storage
\item Separate administrator (view and change) and user (view only) privileges
\item Ability to set thresholds for measured data and notify user about crossing them via text based message
\end{itemize}
\subsection{System description}
\newpage
\section{System design}
\begin{figure}[ht!]
The first mandatory software requirement asks for a web server. It is entirely possible for every measurement device to contain its own web server. However, multiple points are requiring devices to work as a \textbf{system}. Two common system structures are \textit{centralised} and \textit{decentralised}.
Decentralised (peer-to-peer) systems are harder to build but are more fail-proof. Since fail-proofness is not mentioned in the requirements, centralised system might suffice.
\begin{figure}[ht]
\centering
\includegraphics[width=.75\textwidth,angle=0]{server_node_diag}
\caption{The block diagram of a \textit{server} node of a proposed system, including hardware requirements}\label{f:serv_node}
\end{figure}
The first mandatory software requirement asks for a web server. It is entirely possible for every measurement device to contain its own web server. However, multiple points are requiring devices to work as a \textbf{system}. Two common system structures are \textit{centralised} and \textit{decentralised}.
Decentralised (peer-to-peer) systems are harder to build but are more fail-proof. Since fail-proofness is not mentioned in the requirements, centralised system might suffice.
Using centralised system means, that the measurement devices will use one separate device, from now called the \textbf{server node}, to do most of the work on the software side. The work includes, but is not limited to, receiving the measured data, storing them, hosting the web server with the \gls{gui} containing all necessary options and information, handling the \gls{usb} or communication with a \gls{cloud} and so on. The block diagram for a server node, depicting required blocks can be seen in the figure \ref{f:serv_node})
Where there are at least two nodes in a system, they have to communicate together in a particular way, known to both of them. The web server naturally operates over \gls{tcpip}. Therefore, same networking stack (the way of comunication), that is used for communication between the server node and user can be used to communicate to client nodes as well. \Gls{tcpip} hardware is ready to be used and is supporting a full-blown networking \gls{stack}, powering communication over today's networks.
\begin{figure}[ht!]
\begin{figure}[ht]
\centering
\includegraphics[width=.75\textwidth,angle=0]{client_node_diag}
\caption{The block diagram of a \textit{client} node of a proposed system, including hardware requirements}\label{f:client_node}
\end{figure}
Where there are at least two nodes in a system, they have to communicate together in a particular way, known to both of them. The web server naturally operates over \gls{tcpip}. Therefore, same networking stack (the way of comunication), that is used for communication between the server node and user can be used to communicate to client nodes as well. \Gls{tcpip} hardware is ready to be used and is supporting a full-blown networking \gls{stack}, powering communication over today's networks.
The measurement devices, from now on called \textbf{client nodes} will consist of blocks of the remaining hardware requirements. The resulting block diagram can be seen in the figure \ref{f:client_node})
%The centralised system proposal appears to be more economical, than a decentralised system, which would require a separate copy of hardware for doing all the things for every client node, plus some clever way for communication between them.
%\item Split the device to the server (separate GL-inet router running a web-server, data processing and data storage) and to measurement nodes (electronics powered by an ESP8266 wi-fi module) to create a unique, replicable and efficient solution
%\item Split the device to the server (separate GL-inet router running a web-server, data processing and data storage) and to measurement nodes (electronics powered by an ESP8266 Wi-Fi module) to create a unique, replicable and efficient solution
%\item If reasonably accurate, use the inbuilt \gls{adc} of the ESP8266 for the voltage measurement in combination with a linear transformer (the transformer also powers the entire node)
%\item Use hall-effect sensor for a current measurement
%\item If possible, prevent the use of the external \gls{adc} as well as the use of the additional microcontroller to keep the cost down (if needed, use the microcontroller of the 8-bit AVR family produced by Atmel)

BIN
tukethesis.pdf (Stored with Git LFS)

Binary file not shown.
Loading…
Cancel
Save