| Process control software presently being developed for microprocessor-based control systems is often coded in assembly language to achieve performance constraints. However, the existing microprocessors have primitive capabilities which do not facilitate development of process control software in assembly language. Typical tasks involved in a digital process control include arithmetic (e.g., proportional + integral + derivative control), and both synchronous and asynchronous data communication. In this dissertation, a procedure is developed for rating different microprocessors in terms of the relative difficulty of programming them for the same process control application in assembly language. The procedure is based on measuring the logical complexity of each microprocessor's instruction repertoire. A simple experiment is performed to test this procedure by programming a set of five kernel programs which are representative of typical tasks performed in a digital process control system. These kernel programs were coded for six different commercially available microprocessors. The kernel program development time and the complexity measure are very strongly correlated. The results should, however, be viewed with caution due to the limited size of the experiment.;*This research was supported in part by the National Science Foundation under Grants GK-38897X1 (ENG-7417742) and ENG-7681346. |
