Dsp System Software Development And Application-oriented

Until the fist digital signal processor (DSPs) was in use in 1980s,the theory of the digital signal processing was indeed applied to the real systems. From then on,DSP has been widely used for digital signal processing applications in the fields of communication,control,multimedia and so on.In the digital signal processing applications,software stored in the programmable DSP performs the algorithms. Thus,the software design is very important in the design of the whole DSP-based system. Assembly language is the most commonly used language in DSP programming. But as we know,the assembly language is so complicated that it is not easy to use even in the design of a less complicated system and is closely concerned with its own specific machine. So many DSP venders publish their DSP processors along with the software development environments (mainly C-Compiler) for the users programming in C language so that the applications can get more portability and maintainability. But most commercial C-Compilers are not as efficient as expected. So how to make a more efficient C-Compiler become a very important problem and a research hotspot.This paper does a research into the measures of DSP's C-Compiler design. After comparing the DSP with RISC processor in the respects of structure and application,and referring to the classic theory of compile for the general-purpose processors,some feasible schemes for DSP's C-Compiler design are given along with some optimization strategies suitable to the DSP's feature. A specific 16-bit DSP processor is taken as the example and the experiment result is also shown in this paper. After that,it gives the measures of designing DSP's assembler as a part of the DSP's software development environment together with the C-Compiler.Moreover,this paper explores the method of design the floating-point arithmetic unit. Referring to the IEEE754-1985 standard for binary floating-point arithmetic,the algorithm and the behavior description of floating-point adder and multiplier is given,and the simulation and verification is shown at the end of this paper.