What is the need of Digital Signal Processing?

Real world signal is analog. Digitized signal can approximate real world analog signal to a limited degree. Limited in terms of time resolution or accuracy. Digital signal processing is needed to improve the digital representation of the real world signal. Improvement can come in the form of accuracy, interpretation, transformation, or compression.

Many of these processing has an equivalent form in analog domain - ex. filtering. However, digital signal processing can typically be implemented cheaper, faster, and with more consistent result. 

Digital signal processing has a wide variety of applications, including:

Ø  Audio and video compression (the quality depends on the sampling rate chosen - higher sampling rate = higher quality. The file size can be compressed by applying source coding, such as Huffman coding.)

Ø  Audio signal processing (example: applying a low pass or bandpass filter to reduce external noise from an audio recording)

Ø  Image processing (example: using FFT, filtering and inverse FFT in order to remove noise from an image)

Ø  Medical applications (example: applying a histogram equalization to enhance an x-ray image)