Analog to Digital Conversion: The Process of Digitising Analog Information

Every time you take a photo on your phone, record a voice note, or scan an old family document, something remarkable happens behind the scenes. A continuous wave from the real world is broken down into a stream of numbers that a computer can store and process. This is the heart of analog to digital conversion, the silent process that turns sound, light, temperature, and motion into the 1s and 0s that power modern information systems. Understanding how this works is essential for anyone studying how information moves through society, because almost every digital record you will ever handle began life as an analog signal.

Why convert analog to digital? 🔗

The world we experience is analog. Sound waves, light, temperature, and pressure all change smoothly and continuously, taking on an infinite number of possible values at every instant. The way people experience the world is mostly analog, think sound and light waves, but computers cannot work with these smooth, never-ending values. Digital electronics operate using discrete numbers, so any real-world signal must be translated into a form machines can understand.

The limitations of analog signals 🔗

Analog signals carry a serious weakness: they degrade over distance and time. When an analog signal travels through cables, repeaters, or storage media, it picks up noise and distortion that cannot easily be removed. Analog signals would not maintain enough integrity to be received when transmitted through many stations along a long stretch of distance. Each copy of an analog recording is also slightly worse than the original, a problem known as generation loss.

Digital signals solve this. Because they are simply sequences of 1s and 0s, they can be copied, transmitted, and stored with far greater reliability. Digital data can be shared and accessed more easily and, in theory, propagated indefinitely without generation loss, provided it is migrated to new, stable formats as needed. A digital file copied a thousand times remains identical to the first copy.

Why digital is preferred for processing and storage 🔗

Once information is digital, it becomes flexible in ways analog never could be. Text, audio, images, and video can all be reduced to the same underlying binary format, which means they can be intermingled, compressed, searched, and edited using the same tools. Digitisation allows information of all kinds in all formats to be carried with the same efficiency. For libraries, archives, and information centres, this is transformative. A digitised manuscript can be searched, shared across the world, and protected from the wear of repeated handling.

This is why memory institutions invest heavily in conversion. Libraries, archives, museums, and other memory institutions digitise items to preserve fragile materials and create more access points for patrons. Old audio and video tapes are especially urgent cases, because the equipment needed to play them is becoming obsolete. The availability of analog playback equipment is dwindling and the audiovisual industry has shifted to digital production, making timely conversion critical to avoid permanent loss.

It is worth noting one important distinction. Digitisation is the conversion process itself, while digital preservation is the ongoing effort to keep those files usable over decades. Digitisation is converting analog materials such as photographs, handwritten letters, film, and slides to a digital format, but a repository of outdated or corrupted files will not help anyone find information thirty years later. Converting is only the first step.

The role of ADCs and DACs 🔗

Since analog and digital signals are fundamentally different in nature, a device is needed to translate between them. These translators are the Analog to Digital Converter (ADC) and the Digital to Analog Converter (DAC). They sit at the boundary between the physical world and the digital one, and almost every device you use depends on them.

How an ADC works 🔗

An ADC takes a continuous analog signal and produces a stream of binary numbers. An analog to digital converter takes analog signals and, through a complex algorithm, converts the natural waveforms into digital binary bits. The ADC measures the incoming signal at regular intervals and assigns each measurement a numerical value. The classic everyday example is a phone call: when someone speaks into a microphone, an onboard ADC converts the analog voice signal into a digital bit stream that can be transmitted reliably across towers, satellites, and networks.

The ADC is what allows microprocessor-controlled circuits to communicate with the real world. Sensors that measure sound, light, temperature, or movement feed analog signals into circuits, and the ADC lets digital systems interact with that environment. Single-board computers like the Arduino and Raspberry Pi rely on ADCs to read sensor data from the physical world.

How a DAC works 🔗

The DAC performs the exact reverse function. It takes digital binary data and reconstructs a continuous analog signal from it. DACs are commonly used in music players to convert digital data streams into analog audio signals, and they appear in televisions and mobile phones to turn digital video into displayable images. Returning to the phone call example, at the receiving end a DAC converts the digital signal back into analog form so it can be amplified and heard as sound through a speaker.

Together, the ADC and DAC form a complete loop. Real-world signal enters through the ADC, gets processed and transmitted digitally, then exits through the DAC back into a form our senses can perceive. Internally, DACs typically use components such as operational amplifiers and resistive networks to convert an n-bit binary number into an equivalent analog output voltage or current signal.

Different ADC designs for different needs 🔗

There is no single best converter. Engineers choose an ADC architecture based on the speed, resolution, and cost the application demands. Pipelined converters are most popular for applications requiring high throughput rates with good resolution, while sigma-delta converters are usually the best choice when very high resolution is needed. A medical imaging system, an audio recorder, and an industrial sensor each have different priorities, which is why so many converter types exist.

Sampling and quantisation 🔗

The actual conversion inside an ADC happens through a defined sequence of steps. Analog to digital conversion is generally carried out in the order of sampling, quantisation, and coding. The first two of these steps, sampling and quantisation, are where the real magic and the real trade-offs happen. Sampling handles the time dimension, while quantisation handles the amplitude dimension.

Sampling: capturing moments in time 🔗

An analog signal is continuous in time, meaning it has a value at every possible instant. Sampling is the process of measuring that signal at fixed, regular intervals rather than continuously. In sampling, the analog signal is measured at a constant rate called the sample rate, which determines the number of samples taken per second. The more samples taken per second, the more closely the digital version mirrors the original.

But how many samples are enough? This is answered by one of the most important rules in signal processing, the Nyquist-Shannon sampling theorem. It states that the sample rate should be at least twice the highest frequency present in the analog signal to accurately represent it. This is why audio CDs use a sample rate of 44,100 samples per second: human hearing extends to roughly 20 kHz, so the sample rate is set above 40 kHz. If you sample too slowly, you create a distortion called aliasing, where high frequencies masquerade as false low frequencies, and the original signal can no longer be correctly reconstructed.

Quantisation: rounding to allowed values 🔗

Sampling alone is not enough. Each sample is still an analog measurement that could take any value. Quantisation is the step that maps these measurements onto a limited set of allowed levels. Quantisation involves mapping the infinite set of possible values of an analog signal to a finite set of levels in a digital representation. In simple terms, it rounds each sampled value to the nearest available step.

The number of available steps depends on how many bits are used to represent each sample. The number of voltage levels in a digital system equals 2 raised to the power of the number of bits. So an 8-bit converter offers 256 levels, while a 16-bit converter offers 65,536. More bits mean finer steps and a more faithful representation of the original signal. This is what people mean by “bit depth” when discussing audio or image quality.

Quantisation error and its trade-offs 🔗

Because quantisation forces every sample onto the nearest allowed level, there is almost always a small gap between the true value and the stored value. This process is lossy and not reversible, since the exact original value cannot be recovered once it has been rounded. This gap is called quantisation error, and it shows up in the final signal as a faint background noise known as quantisation noise.

The solution is to use more bits, but this comes at a cost. Higher bit depth reduces quantisation noise and enhances the fidelity of the digital representation, but it also requires more storage and processing power. Every digital conversion is therefore a balancing act between quality and resources. A high-resolution archival scan demands far more storage than a quick voice memo, and the engineer or archivist must decide what level of fidelity the purpose actually requires.

Sampling and quantisation also work together rather than in isolation. A higher sampling rate improves time resolution but offers little benefit if the quantisation levels are too few, so both must be optimised together to ensure the digital signal accurately represents the original input. After quantisation, the final coding step assigns each level a binary number, completing the conversion into a digital bit stream.

Bringing it together for information professionals 🔗

For students of information and society, the value of understanding this process goes beyond electronics. Every digitisation project in a library or archive is built on these same principles. The choice of sampling rate and bit depth determines whether a digitised audio recording captures the warmth of an old performance or reduces it to a flat, noisy approximation. These technical decisions shape what future generations will be able to hear, see, and study.

The stakes are real. Audio engineers and technical specialists have created standards for sampling rates, bit depths, and playback to guide the digital transfer of analog sound recordings. When you understand sampling, quantisation, and the role of ADCs and DACs, you understand why these standards exist and why getting the conversion right the first time matters so much. The analog original may not survive, but a well-made digital copy can carry its information forward for generations.

What do you think? If higher bit depth and sampling rates always produce better quality but cost more in storage, how should a library decide the right level of fidelity for digitising its collections? And as analog playback equipment becomes harder to find, which categories of analog records do you think deserve to be digitised first?

References

1. https://www.techtarget.com/whatis/definition/analog-to-digital-conversion-ADC
2. https://www.totalphase.com/blog/2022/06/analog-vs-digital-signals-adc-vs-dac/
3. https://en.wikipedia.org/wiki/Digitization
4. https://libguides.ala.org/libpreservation/digitization
5. https://siarchives.si.edu/what-we-do/digital-curation/digitizing-collections
6. https://preservica.com/resources/blogs-and-news/digitization-vs-digital-preservation
7. https://www.electronics-tutorials.ws/combination/analogue-to-digital-converter.html
8. https://en.wikipedia.org/wiki/Digital-to-analog_converter
9. https://www.analog.com/en/resources/analog-dialogue/articles/analog-to-digital-converter-architectures-and-choices.html
10. https://resources.pcb.cadence.com/blog/2023-the-sequence-of-analog-to-digital-conversion
11. https://www.vaia.com/en-us/textbooks/physics/inquiry-into-physics-7-edition/chapter-8/problem-17-what-is-analog-to-digital-conversion-and-how-is-i/
12. https://eureka.patsnap.com/article/how-do-sampling-and-quantization-affect-analog-to-digital-conversion
13. https://www.electronics-lab.com/article/analog-to-digital-conversion-part-1/
14. http://ohda.matrix.msu.edu/2012/06/preservation-of-analog-collections-through-digitization/