XAVIEN Technologies DSP Article

Digital Signal Processing (DSP)

in an Analog Application

By: Steven L. Kosmerchock CET

Digital signal processing (DSP) refers to different ways of improving the accuracy and reliability of digital communications. Basically, DSP works by clarifying, or standardizing the levels or states of a digital signal. A DSP circuit is able to differentiate between signals and noise. This improves the signal to noise (S/N) ratio while also improving the efficiency of the circuit. All communications circuits contain some noise. This is true whether the signals are analog or digital, and regardless of the type of information being generated. Figure 1.1 illustrates a typical DSP scheme in an analog application.

Figure 1.1

The first stage is an analog filter, designed to limit the frequency range and cleanup the signal before sampling. In most cases this circuit is also designed to filter out unwanted noise. The second stage is an analog-to-digital converter (ADC) that samples and converts the signal into a binary code. In figure 1.2 there is a typical analog signal. When the ADC goes to convert an analog signal to digital, the analog signal is sampled at a specified time rate. The ADC will then output digital signals (binary code) that represents the analog signal at a specified point as in figure 1.3. The binary code will represent the amplitude of the signal being sampled. For simplicity, I have assumed a 3-bit code (Figure 1.4), which can only represent 8 different values. Figure 1.3 is the analog signal of figure 1.2 with the points that are to be sampled. Each "diamond" on the graph represents a point that was sampled by the ADC.

---------------Figure 1.2-----------------------------------------Figure 1.3---------------

Figure 1.4

Here is a listing of the amplitude along with the binary code that represents it.

Amplitude:	Binary Code:
0 1 2 3 4 5 6 7	000 001 010 011 100 101 110 111

Figure 1.4 represents the binary code that the ADC outputs for a given amplitude. If the signal applied has an amplitude of "7" then the ADC will output 111.

After the conversion from analog to digital, the signals are then handled by the DSP circuit. The DSP circuit uses numerical techniques and digital processors to manipulate the signals. The DSP is programmed by an external party with the algorithm (program) that is required for the DSP to act accordingly to the specified situation. This allows the designers to change the software, not the hardware, if there needs to be a change in the circuit. S/N Ratio is also improved with the output being digital as well as a more accurate response. The next step is for the digital signals to be converted back into analog signals. This is accomplished with a digital-to-analog converter (DAC). Compare figure 1.5 to figure 1.6. Figure 1.6 is the analog signal as a DAC would represent it.

----------------Figure 1.5---------------------------------------------------Figure 1.6----------------

The final stage is often another analog filter, designed to remove the sharp transitions that will be produced by using a DAC.

The popularity as well as the practicality of digital signal processing has been growing strong over the last two decades. There are several reasons as to why converting analog to digital is becoming more accepted. One of the main reasons is because signals and data of many types are increasingly stored in digital devices, and transmitted in digital form from place to place. It makes sense in many cases to process them digitally as well. Another reason is that digital processing is more stable and reliable than its analog counterpart. Rapid advances in integrated circuit (IC) design and manufacture are producing ever more powerful DSP devices at a decreasing cost. Another important advantage of DSP is its ability to process a number of signals simultaneously, making it more efficient than analog signal processing. While DSP has not been perfected, a close look at the rapid advances in this technology will make you realize that in many ways this technology has no competitor in accuracy, price, and reliability.