Communication Systems Engineering: Modulation, Coding, and Wireless
Communication systems engineering enables the transmission of information over distances — from a few centimeters inside a medical implant to billions of kilometers across the solar system. Every phone call, every web page loading, every satellite navigation fix, and every Wi-Fi connection depends on the principles of modulation, coding, and propagation that communication engineers have developed over more than a century.
The fundamental challenge of communication is that all physical channels introduce noise, distortion, and interference. The signal that arrives at the receiver is never identical to the signal that left the transmitter. Communication systems engineering is the art and science of encoding information so that it can be recovered reliably despite these impairments. This guide covers the core concepts that make modern communication possible.
Information Theory and Channel Capacity
Claude Shannon’s information theory, published in 1948, provides the mathematical foundation for all communication. Shannon defined the bit as the fundamental unit of information and showed that any information source can be represented as a sequence of bits. He proved that the maximum rate at which information can be transmitted reliably over a channel is the channel capacity, given by C = B log(1 + S/N), where B is the bandwidth, S is the signal power, and N is the noise power.
The Shannon-Hartley theorem has profound implications. To increase the data rate, you can increase bandwidth, increase signal power, or reduce noise. But the logarithmic relationship means that doubling the data rate by increasing power requires squaring the power — an expensive proposition. Increasing bandwidth gives a linear increase in capacity but requires more spectrum, which is a finite and increasingly valuable resource.
Entropy measures the information content of a source. A source that produces equally likely symbols has maximum entropy — each symbol carries the most information. A source with predictable patterns has lower entropy, meaning its output can be compressed. This insight is the basis for all data compression, from ZIP files to JPEG images to video codecs.
Modulation Techniques
Modulation encodes information onto a carrier signal for transmission. The carrier is a sinusoidal waveform at the transmission frequency. By varying the carrier’s amplitude, frequency, or phase, or combinations of these, the transmitter impresses information onto the carrier.
Analog Modulation
Amplitude modulation varies the carrier amplitude in proportion to the modulating signal. AM is simple to implement and demodulate, which is why it was used for the first audio broadcasts. Its vulnerability to noise — noise adds to the amplitude, corrupting the signal — and its poor spectral efficiency have relegated it primarily to medium-wave and shortwave broadcasting.
Frequency modulation varies the carrier frequency in proportion to the modulating signal. FM is much more resistant to amplitude noise than AM, providing higher audio quality. The trade-off is increased bandwidth — wideband FM used for broadcast radio occupies about 200 kHz per channel, compared to 10 kHz for AM. Phase modulation varies the carrier phase and is closely related to frequency modulation.
Digital Modulation
Digital modulation maps digital bits onto the carrier. Binary phase-shift keying uses two carrier phases to represent 0 and 1, providing robust performance in noisy channels. Quadrature phase-shift keying uses four phases, transmitting two bits per symbol and doubling the data rate for the same bandwidth.
Quadrature amplitude modulation varies both amplitude and phase to represent multiple bits per symbol. 16-QAM transmits four bits per symbol, 64-QAM transmits six bits, and 256-QAM transmits eight bits. Higher-order modulation provides higher data rates but requires a higher signal-to-noise ratio to maintain the same bit error rate. Adaptive modulation, used in Wi-Fi and cellular systems, switches between modulation orders based on channel conditions.
Orthogonal frequency-division multiplexing divides the available bandwidth into many closely spaced subcarriers, each modulated with a low data rate. OFDM is robust against multipath interference and frequency-selective fading, making it the basis for Wi-Fi, 4G LTE, 5G, digital television, and DSL broadband.
Error Correction Coding
No communication channel is error-free. Forward error correction coding adds redundant bits to the transmitted data in a structured way that allows the receiver to detect and correct errors without retransmission. This is essential for applications where retransmission is impossible — deep-space communication, real-time voice and video — or where it would be inefficient.
Convolutional codes process the input through a shift register and produce output bits that depend on the current and previous inputs. The Viterbi algorithm provides efficient maximum-likelihood decoding. Turbo codes, invented in 1993, achieve performance close to the Shannon limit by using two convolutional encoders separated by an interleaver, with iterative decoding at the receiver.
Low-density parity-check codes were originally discovered in 1960 but were forgotten for decades until their rediscovery in the 1990s. LDPC codes achieve performance within fractions of a decibel of the Shannon limit and are used in Wi-Fi, digital video broadcasting, 5G, and satellite communication. Polar codes, the most recent breakthrough, are used for control channels in 5G and represent the first provably capacity-achieving code.
Wireless Propagation
Radio waves propagate through space according to the laws of electromagnetism. In free space, the signal power decreases with the square of the distance. In the real world, propagation is complicated by reflection, diffraction, scattering, and absorption. The path loss, or signal attenuation, depends on the frequency, distance, antenna heights, and environment.
Multipath propagation occurs when the signal arrives at the receiver via multiple paths, each with different delay and attenuation. The signals combine constructively or destructively, creating fading — rapid fluctuations in received signal strength. Fading can cause bit errors even when the average signal power is adequate.
Diversity techniques combat fading by providing multiple independently fading replicas of the signal. Spatial diversity uses multiple antennas separated by enough distance that their fading is uncorrelated. Frequency diversity transmits on multiple frequencies. Time diversity interleaves the data so that a fade affects only a few bits. MIMO systems use multiple antennas at both transmitter and receiver to exploit multipath for higher data rates rather than treating it as an impairment.
Multiple Access Techniques
Multiple access allows many users to share the same communication medium. Frequency division multiple access assigns each user a different frequency channel. Time division multiple access assigns each user a time slot. Code division multiple access assigns each user a unique spreading code that separates their signal in the code domain.
Orthogonal frequency division multiple access assigns different subcarriers to different users. It is the basis for 4G LTE and 5G cellular systems. OFDMA provides fine granularity for resource allocation and supports diverse traffic types from low-latency control signals to high-throughput data.
Cellular Systems
Cellular networks divide geographic areas into cells, each served by a base station. The same frequencies can be reused in non-adjacent cells, multiplying the system capacity. From the first-generation analog systems of the 1980s to the 5G networks of the 2020s, each generation has increased data rates, capacity, and efficiency by orders of magnitude.
Wired Communication
Wired communication systems transmit signals over guided media. Twisted-pair copper cables carry Ethernet at speeds up to 10 Gbps over short distances and telephone signals at lower speeds over longer distances. Coaxial cables provide higher bandwidth with better shielding and carry cable television and broadband internet.
Fiber optic communication uses pulses of light transmitted through glass fibers. The enormous bandwidth of optical fiber, exceeding 50 THz, supports data rates of hundreds of gigabits per second per wavelength, with multiple wavelengths multiplexed on a single fiber. Submarine fiber cables carry the vast majority of intercontinental internet traffic, with current systems supporting hundreds of terabits per second.
Frequently Asked Questions
What limits the maximum data rate of a communication channel?
The Shannon-Hartley theorem establishes the fundamental limit: C = B log(1 + S/N). The data rate is limited by bandwidth, signal power relative to noise power, and the efficiency of the modulation and coding scheme. Practical systems operate 3 to 6 dB below the Shannon limit because of implementation constraints.
How does 5G achieve higher speeds than 4G?
5G uses wider bandwidths, up to 100 MHz in sub-6 GHz bands and 400 MHz in millimeter-wave bands, higher-order modulation up to 256-QAM, massive MIMO with dozens of antenna elements, and more efficient channel coding with LDPC codes. Network architecture improvements reduce latency and increase spectral efficiency through denser deployments and beamforming.
What is the difference between analog and digital modulation?
Analog modulation varies a carrier parameter proportionally to a continuous analog signal, and the receiver recovers the original analog waveform. Digital modulation maps discrete digital symbols onto carrier parameters, and the receiver decides which symbol was transmitted. Digital modulation offers better noise immunity, enables error correction, and allows compression and encryption.
Why do we need error correction coding if we have good signal strength?
Error correction coding is needed because even good signal strength does not guarantee error-free reception. Fading, interference, and noise cause occasional errors that coding can correct. Coding also allows systems to operate with lower transmit power, reducing interference and extending battery life. Many systems would be impossible without coding — deep space probes transmit signals so weak that uncorrected error rates would be nearly 50 percent.