Signals All Around Us

Everything is a Signal

Every sound you hear, every image you see, every message you send — it's all signals. The music streaming through your headphones, the electrical impulses in your heartbeat, even the fluctuations in stock prices — all of these are signals carrying information. This lesson explores what signals are, where they come from, and why learning to process them has become one of the most important skills in modern engineering.

Digital Signal Processing (DSP) is the art and science of manipulating these signals using computers. It sits at the heart of nearly every technology you use daily — from noise-cancelling headphones to medical imaging to 5G cellular networks.

What is a Signal?

At its simplest, a signal is any quantity that varies over time (or space) and carries information. Think of a sound wave traveling through the air: the air pressure changes over time, and those changes encode the information we perceive as speech or music.

Signals surround us in every aspect of daily life. A few examples:

Sound waves — Air pressure variations that your ears detect and your brain interprets as voice, music, or noise. ECG signals — Electrical voltages measured across the heart that doctors use to diagnose cardiac conditions. Stock prices — Financial values that fluctuate over time and can be analyzed for trends. Images — Brightness values that vary across two spatial dimensions, forming the pictures on your screen.

Mathematically, we represent a signal as a function. For a signal that varies over time, we write it as x(t) where t is the time variable.

Continuous vs Discrete

The physical world is inherently analog — sound, light, temperature, and pressure all vary smoothly and continuously. But computers are digital: they work with discrete numbers stored in finite memory. This fundamental mismatch is what makes signal processing both necessary and fascinating.

A continuous signal (analog) is defined at every instant in time. Sound in the real world is continuous — air pressure changes smoothly, without jumps or gaps. A discrete signal (digital) consists of individual samples taken at specific time intervals. When you record audio on your computer, you are sampling the continuous sound wave thousands of times per second.

The process of converting from continuous to discrete is called sampling, and it's governed by the Nyquist-Shannon sampling theorem (1928): to perfectly reconstruct a signal, you must sample at least twice its highest frequency. This theorem is the theoretical bridge between the analog and digital worlds.

Signal Properties

Every signal, no matter how complex, can be described by three fundamental properties:

Amplitude measures how strong a signal is — how "tall" the wave is. In audio, amplitude determines loudness. In radio, it determines signal strength. Frequency measures how fast the signal changes — how many cycles occur per second. In audio, frequency determines pitch. In radio, frequency determines the channel. Phase describes where in its cycle the signal currently is. Phase matters when combining or comparing signals.

The simplest and most fundamental signal is the sine wave. Any signal — no matter how complex — can be decomposed into a sum of sine waves at different frequencies, amplitudes, and phases. This remarkable fact, discovered by Fourier in 1822, is the foundation of all signal processing.

Why Process Signals?

Raw signals are rarely useful on their own. They're noisy, too large to store, or contain hidden information that isn't immediately apparent. Signal processing gives us the tools to:

Remove noise — Clean up a recording by filtering out background hiss, or improve a medical image by removing artifacts. Compress data — MP3, JPEG, and video codecs all use DSP techniques to reduce file sizes while preserving perceptual quality. Extract information — Detect a heartbeat anomaly in an ECG, identify a voice in a crowd, or find a trend in financial data. Transform for transmission — Modulate signals for wireless communication, enabling everything from WiFi to satellite TV.

Without signal processing, the modern digital world simply would not exist. Every phone call, every streamed video, every MRI scan relies on DSP algorithms running billions of operations per second.

DSP vs Analog Processing

Before digital computers, signal processing was done entirely with analog circuits — resistors, capacitors, inductors, and operational amplifiers. This approach has advantages: analog circuits are fast and introduce no quantization error.

But digital processing transformed the field. Digital advantages: perfectly repeatable results (the same input always gives the same output), easily programmable and reconfigurable, precise control over every parameter, and the ability to implement algorithms that are impossible in analog hardware. Analog advantages: no sampling artifacts, inherently fast processing, and no need for analog-to-digital conversion.

In practice, modern systems use both: analog circuits at the input and output (where signals interact with the physical world) and digital processing in between (where computation happens). This hybrid approach gives us the best of both worlds.

Signals in Daily Life

Signals and DSP are woven into virtually every technology we use. Audio — Music streaming services use DSP for compression (AAC, MP3), equalization, and noise cancellation. Your voice assistant uses DSP to understand speech. Video — Streaming platforms compress video using DCT-based codecs. Your camera uses DSP for image stabilization, autofocus, and HDR processing.

Medical — ECG machines filter heart signals. MRI scanners reconstruct images from frequency-domain data using the Fourier transform. Ultrasound uses beamforming, another DSP technique. Communications — WiFi, 4G, and 5G all rely on OFDM (Orthogonal Frequency Division Multiplexing), a DSP technique. GPS uses signal correlation to determine your position. Finance — Trading algorithms analyze price signals for patterns. Moving averages and filters smooth noisy market data.