Validating the INTERPRETOR Software Architecture for the Interpretation of Large and Noisy Data Sets

The Interpretor Software Architecture

Figure 1 shows the Context Diagram of the INTERPRETOR system. The INTERPRETOR system takes high frequency noisy data and other relevant data to assist in interpretation from various input sources and presents to various output sources an interpretation of the original data.

Figure 1.

Context diagram of the INTERPRETOR system

Figure 2 shows the data flow in the INTERPRETOR system of Figure 1. Data is initially filtered to get rid of noise; rather than reasoning on a point to point basis, the resulting data stream is then converted by a second process into abstractions – this is a form of data compression. A third process to provide an assessment of the original data interprets these abstractions.

Figure 2.

Data flow diagram of the INTERPRETOR system

We, therefore, derive the overall software architecture of the INTERPRETOR System in form of a Structure Chart as shown in Figure 3.

Figure 3.

Overall software architecture of the INTERPRETOR system

It can be seen that INTEREPRETOR is a data flow architecture and model of computation. The architecture is decomposed into three processes, which can be changed or replaced independently of the others - this makes INTERPRETOR a loosely coupled system. Indeed, each process of the INTERPRETOR performs one task or achieves a single objective - this makes the INTERPRETOR a highly cohesive system. INTERPRETOR can also be considered a pipe and filter architectural style because it provides a structure for systems that process a stream of data.