Applications of DSP

The power of DSP is increasing exponentially with the scaling down of the transistors. We’ll now see how that impacts several applications in our daily life.

In essence, the main DSP operations only consist of addition, subtraction, multiplication, and division. It’s hard to believe that basic operations like these can be used to revolutionize how we live our lives. Let us discuss a few examples.


It’s impossible to transmit and receive analog signals without any degradation in quality. Since DSP represents signals with sequences of numbers, it becomes possible to pass these numbers through noisy communication channels in a perfect manner (not many fields or technologies in the world can claim this kind of perfection).

The physical layer (i.e., the network portion related to sending and receiving the actual signals) of all major communication standards is built through DSP operations. This is true for both wired and wireless communication media. Our WiFi modem, TV, 4G and 5G cellular telephones, and Bluetooth all employ DSP algorithms to efficiently implement highly complicated operations in the available silicon.

Audio, video, and image processing

Due to specific requirements of these applications (e.g., real-time computation, storage), information compression is a fundamental part of audio, image and video standards. The samples in digital signals represented by sequences of numbers are correlated with each other. This leads to the concept of source coding, a process in which the audio or video signal is analyzed for redundancy and then represented by a smaller number of bits that are actually sent over the channel for communication or stored in memory for later retrieval. These prediction algorithms enable us to rebuild the original signal from filter coefficients (numbers used to process the input samples through simple mathematical operations such as addition and multiplication) at the destination.

And it is not just about compression. Audio processing techniques include removing the background noise, increasing the audio quality, and enabling voice recognition technology. Moreover, there has been a steady rise in the integration of image processing in our lives. This isn’t surprising since vision is the most advanced of our senses. However, we are biologically limited to the visible portion of the electromagnetic spectrum. On the other hand, imaging techniques can process the entire spectrum from radio to gamma waves. These images appear the same to machines such as an ultrasound or an electon microscope—as non-visible bands in 1s and 0s.

Biomedical technologies

In the past few decades, technology has enabled a slow merger between human senses and artificial sensors powered by signal processing algorithms. It has enabled the use of millions of hearing aids throughout the world. DSP has also enhanced the image quality and resolution of medical imaging technologies like x-rays, CT scans, and MRIs. While artificial limbs and hearts driven by signal processing are more important, simple fitness tracking and the wearable market now generate billions of dollars in revenue annually. Biometric recognition including retina scans, facial expressions, and fingerprinting are now part of everyday life.

Finally, scientists believe that they are near the ultimate breakthrough—deciphering the inner working of a human brain. Once that happens, human thinking and signal processing can be merged in unprecedented ways that will result in both positive and negative unexpected outcomes.

Radar and sonar

Radar stands for radio detection and ranging. It’s a system that employs radio waves to determine the presence of faraway objects and estimate their range, angle, and radial velocity with respect to the site. These objects can be military or civil airplanes, spacecraft, ships in the ocean, missiles, or recently, driverless automobiles. Speed guns used by law enforcement agencies are based on this same principle. In the past decades, advanced DSP algorithms have been employed to build passive radars that pluck the waves out of the air for the analysis and detection of objects without having to send any signal at all!

Sonar is similar to radar in detecting objects but does so in underwater channels. It operates by transmitting sound waves (instead of radio waves that have poor propagation in water) in all directions and analyze the return signals to extract as much information about the reflecting object as possible. This helps in ocean exploration, submarine detection and ranging, and submarine to airplane communication.

Navigation systems

Global navigation systems utilize DSP techniques to determine the position of an object through navigation satellites and a specialized receiver. The most familiar of these is the Global Positioning System (GPS) built by the United States in 1995. Now other nations have their own satellite systems in place such as GLONASS (developed by Russia), Galileo (developed by Europe), and BeiDou (developed by China). India is also developing its own global positioning system.

When we ask our phone for directions to a particular place, its GPS receiver processes the signals from visible satellites and applies DSP methods to determine our own location. Then, it computes the best directions to the destination from the stored map. From there, the receiver tracks the signals from the satellites to update its position until our arrival at the destination.

Space exploration

DSP has helped shape our understanding of the universe. In just the past fifty years, it has revolutionized our knowledge of stars and galaxies that are millions of light-years away.

Our eyes can only see the visible portion of the electromagnetic spectrum and filter out much of the precious information arriving on the planet. Using modern telescopes, like the James Webb Space Telescope, astronomers capture the emissions at all spectrum frequencies from extremely low frequencies (high wavelengths) to extremely high frequencies (tiny wavelengths of gamma rays). And then, DSP allows us to process enormous amounts of information with great precision.

For instance, spectroscopes show us the absorption lines from which we can infer which elements in what proportions are present in a star. This is how we know precisely how much gold, silver, and other elements there are in our sun. And this is not much different from deep earth exploration in which accurate information about the underlying geology is required before expensive drilling operations take place.

In the words of David Christian, the author of Origin Story: A Big History of Everything:

Astronomers can tell a star’s surface temperature from the color (or frequency) of the light it emits, so we know that surface temperatures can be as low as 2,500 K and as high as 30,000 K. And, as we have already seen, they can calculate the total amount of light a star emits (its luminosity) by measuring its apparent brightness and then calculating how much brighter it would be close up. These two measurements—surface temperature and luminosity—provide the basic inputs for the Hertzsprung-Russell diagram. Finally, if we know a star’s luminosity, we can often estimate its mass. Similar techniques help us estimate the distance, size, motion, and energy of entire galaxies.

Autonomous vehicles

The mass adoption of driverless cars is just around the corner. These cars are even smarter than our smartphones. Working as an algorithm factory, they acquire all kinds of signals from the environment using tools such as ultrasound, radar, and cameras and process those signals to make intelligent decisions. While machine learning and artificial intelligence may be more popular buzzwords, these driverless cars rely on signal processing techniques for their clever integration into our society in a manner similar to biological organisms.