Analog vs Digital Signals

Two Worlds of Signals

In the previous lesson, we learned that signals are quantities that change over time and carry information. Now we face a fundamental question: how do we represent these signals? The answer splits into two distinct approaches — analog and digital — and understanding the difference between them is essential to everything that follows in signal processing.

The physical world is inherently analog. Sound waves, light, temperature, and electromagnetic radiation all vary continuously — smoothly and without gaps. But the computers, smartphones, and digital devices we use every day operate in a fundamentally different way: they work with discrete numbers. The story of modern technology is, in many ways, the story of how we learned to bridge these two worlds.

Analog Signals: The Continuous World

An analog signal is defined at every point in time, and its amplitude can take any value within a continuous range. Think of the sound of a violin: the vibrating string creates smooth, unbroken pressure waves in the air. At every instant — not just at specific moments, but truly every infinitesimal point in time — the air pressure has a precise value.

Analog signals are the native language of the physical world. Temperature doesn't jump from 20.0°C to 20.1°C — it passes smoothly through every value in between. Voltage in an electrical circuit varies continuously. Light intensity changes gradually as the sun moves across the sky. These are all analog quantities.

Historically, the first communication technologies were all analog. The telegraph (1830s), telephone (1876), AM radio (1906), and analog television (1920s) all transmitted and received continuous signals. Vinyl records store music as a continuous groove whose shape mirrors the original sound wave. Analog is beautiful in its fidelity to nature — but it has significant limitations.

Digital Signals: The Discrete World

A digital signal is a sequence of discrete values, defined only at specific points in time and restricted to a finite set of amplitude levels. Instead of the smooth, continuous curve of an analog signal, a digital signal is a series of numbers — a staircase rather than a ramp.

Digital signals are the language of computers. Every piece of digital information — text, images, audio, video — is ultimately represented as a sequence of binary digits (bits): ones and zeros. A digital audio signal on a CD, for example, consists of 44,100 numbers per second, each represented by 16 bits. These numbers approximate the original analog waveform closely enough that your ears can't tell the difference.

The digital revolution didn't happen because digital signals are inherently "better" than analog. It happened because digital signals can be processed, stored, copied, and transmitted with a reliability and flexibility that analog signals simply cannot match.

Key Differences

Noise immunity is perhaps the most important advantage of digital signals. When you copy an analog signal — say, dubbing a cassette tape — noise is added at every step, and quality degrades with each generation. Digital signals, by contrast, can be copied perfectly: a bit is either 0 or 1, and small amounts of noise don't change the interpretation. This is why a JPEG image looks identical whether it's the original or the millionth copy.

Flexibility is another decisive advantage. An analog filter is a physical circuit — changing its behavior means redesigning the hardware. A digital filter is software — you can change it by updating a few lines of code. This programmability is what makes modern devices so versatile: the same smartphone hardware can run thousands of different signal processing algorithms.

From Analog to Digital: Sampling & Quantization

Since the physical world is analog but computers are digital, we need a way to convert between the two. This analog-to-digital conversion (ADC) involves two key steps: sampling and quantization.

Sampling is the process of measuring the analog signal's amplitude at regular time intervals. Instead of knowing the signal at every instant, we capture "snapshots" at a fixed rate. The number of snapshots per second is called the sampling rate (or sampling frequency), measured in Hertz.

But how fast do we need to sample? This question has a precise answer, given by the Nyquist-Shannon Sampling Theorem:

If you sample too slowly (below the Nyquist rate), you get aliasing — the reconstructed signal will contain false frequencies that weren't in the original. This is why CD audio uses a 44,100 Hz sampling rate: human hearing tops out around 20,000 Hz, so 44,100 > 2 × 20,000, satisfying the theorem with a small margin.

Quantization is the second step: mapping each sample's continuous amplitude to the nearest value in a finite set of levels. With B bits per sample, we get 2^B possible levels:

Quantization inevitably introduces a small error — quantization noise — because we're rounding continuous values to discrete levels. The more bits we use, the smaller this error becomes. Each additional bit roughly doubles the number of levels and adds about 6 dB of signal-to-noise ratio.

Real-World Examples

The analog-to-digital transition reshaped entire industries. Here are some of the most striking examples:

Vinyl vs CD — A vinyl record stores sound as a continuous groove cut into plastic. A CD stores the same sound as 44,100 digital samples per second, each with 16-bit resolution. The vinyl is a perfect analog copy (in theory), but it degrades with every play. The CD is a quantized approximation, but it sounds identical on the first play and the ten-thousandth.

Analog vs Digital Radio — AM and FM radio transmit audio as continuous electromagnetic waves. Digital radio (DAB, HD Radio) transmits compressed digital streams. Analog radio degrades gracefully with distance (more static), while digital radio either works perfectly or cuts out entirely — there's no gradual degradation.

Film vs Digital Camera — Photographic film captures light on a continuous chemical emulsion with effectively infinite resolution. A digital camera sensor captures light in a grid of discrete pixels. Film has a beautiful analog quality, but digital cameras offer instant review, easy editing, and virtually zero cost per shot.

Analog vs Digital Telephony — Traditional landlines carried your voice as a continuously varying electrical signal. Modern phone systems (VoIP, cellular) convert your voice to digital samples, compress them, transmit packets over networks, and reconstruct the audio at the other end. The digital approach enables video calls, encryption, and global routing — none of which are practical with analog.

Why the World Moved to Digital

The digital revolution wasn't driven by a single advantage — it was the accumulation of many. Perfect copying means no generational loss. Error correction codes can detect and fix transmission errors automatically. Compression algorithms (MP3, JPEG, H.264) reduce storage and bandwidth requirements dramatically. Encryption protects privacy and security. Software flexibility means the same hardware can be reprogrammed for new tasks.

Perhaps most importantly, digital signals scale with Moore's Law. As transistors get smaller and cheaper, digital processing becomes more powerful. Analog circuits don't benefit from this scaling in the same way — a capacitor is still a capacitor, regardless of how small you make the transistors around it.

That said, analog hasn't disappeared. Every digital system still needs analog components at its boundaries — microphones, speakers, antennas, sensors, and displays all operate in the analog domain. The modern approach is hybrid: analog at the edges, digital in the middle.

Advantages and Trade-offs

Analog advantages: infinite resolution (no quantization error), no need for ADC/DAC conversion, naturally low latency, and graceful degradation in noisy environments (the signal gets noisier but doesn't disappear abruptly).

Digital advantages: perfect reproducibility, noise immunity, easy storage and transmission, programmable processing, error correction, encryption, and compression. Digital systems improve with computing advances and can be updated with software.

Digital trade-offs: Sampling introduces bandwidth limitations (you can't capture frequencies above f_s/2). Quantization introduces noise. ADC/DAC conversion adds latency and cost. And digital systems require significantly more bandwidth for raw transmission than their analog equivalents — though compression usually more than compensates.