DSP特点（Characteristics of DSP）.docVIP

DSP特点（Characteristics of DSP） DSP features For beginners who have not used DSP, the first confusion is DSP, what is the difference between other embedded processors, and what is the difference between SCM and ARM?. In fact, DSP is also an embedded processor, which can completely perform the functions of a single chip microcomputer. The only important difference is that DSP supports the multiply plus operation of a single clock cycle. This is a common feature of almost all manufacturers of DSP chips. Almost all of the DSP processor instruction set will have a MAC instruction, the instruction can take two operands out multiplication from RAM, and then added to an accumulator, all these operations are completed in one clock cycle. A processor with such an instruction has the DSP function. The reason that this instruction is called a digital signal processor is that the most common arithmetic operation in all digital signal processing algorithms is multiply - add. This is because a large number of operations are used in digital signal processing, or dot product. Whether it is FIR filtering, FFT, signal correlation, digital mixing, down conversion. All of these digital signal processing operation often is the input signal with a coefficient table or a local reference signal and integral (accumulation), which showed the two vector (or sequence) of the dot product, in the programming becomes the input sampling in a circular buffer. The coefficient table or local reference signal is also put in a buffer, and then use the two pointer to the two buffer. This allows you to use a MAC instruction within a loop to do dot product operations for the two. Such dot product operations are the fastest with the processor because it can be done only once with a single cycle. Understanding of the characteristics of DSP, when we design an embedded system, the first to consider whether the implementation of processor algorithm in dot product operation, whether it is frequently the two array