4.2 Signal Transmission

Signal transmission refers to the physical and technical process by which a signal—a physical variable carrying encoded information—is conveyed from a source location to a destination location through a medium or channel. In cybernetic communication theory and information theory, signal transmission is the technical substrate of communication: it is what must work for information to move from sender to receiver, and the characteristics of the transmission process—its capacity, its noise properties, its delay characteristics—determine what communication is possible and what constraints the system designer must work within.

The Transmission Model

Signal transmission is best understood within the context of Shannon's mathematical model of a communication system, which identifies five principal components:

1. Information source: Generates the message to be communicated.
2. Transmitter (encoder): Converts the message into a signal suitable for the channel.
3. Channel: The physical medium through which the signal travels.
4. Noise source: Adds unwanted disturbance to the signal during transmission.
5. Receiver (decoder): Reconstructs the message from the received (possibly corrupted) signal.

Signal transmission is the process governed by the channel and noise source—the physical journey of the signal from encoder to decoder.

Physical Mechanisms of Signal Transmission

Different types of channels use different physical mechanisms for signal transmission:

Electrical Transmission

Electrical signals are transmitted as variations in electrical current or voltage through conducting materials (metal wires, circuit boards). Electrical transmission is characterized by:

• Speed: electrical signals travel at speeds close to the speed of light in the conductor medium.
• Bandwidth: the range of frequencies that can be transmitted is determined by the electrical properties of the conductor and any associated filtering components.
• Noise: thermal noise (Johnson-Nyquist noise) arises from the random thermal motion of electrons; it is proportional to absolute temperature and bandwidth.
• Attenuation: electrical signals lose strength with distance due to resistance losses; amplifiers must be periodically inserted to restore signal strength in long-distance transmission.
• Interference: electromagnetic interference from other electrical sources can corrupt electrical signals; shielding and differential signaling reduce this problem.

Twisted pair cable, coaxial cable, and circuit board traces are common electrical transmission media.

Optical Transmission

Optical signals are transmitted as variations in light intensity or phase through optical fibers (glass or plastic) or through free space (laser communication). Optical fiber transmission has become the primary medium for long-distance high-capacity communication:

• Bandwidth: optical fibers can carry extremely wide bandwidths, supporting data rates in the terabits per second range through wavelength-division multiplexing.
• Attenuation: modern optical fibers have extremely low attenuation, allowing transmission over thousands of kilometers with periodic amplification.
• Noise: optical amplifiers introduce amplified spontaneous emission noise; photodetectors introduce shot noise.
• Immunity to interference: optical signals are unaffected by electromagnetic interference, making optical fiber superior to electrical cable in electromagnetically noisy environments.

Wireless/Radio Transmission

Radio frequency signals are transmitted as electromagnetic waves through free space:

• Propagation: radio waves travel at the speed of light; their propagation characteristics depend on frequency (lower frequencies diffract around obstacles, higher frequencies have more line-of-sight propagation).
• Bandwidth: the available bandwidth in any frequency band is limited; frequency allocations are managed by regulatory bodies.
• Path loss: signal strength decreases with distance squared in free space; reflections, diffraction, and absorption produce additional losses in real environments.
• Noise and interference: thermal noise at the receiver is unavoidable; interference from other transmitters using the same frequency is a major practical challenge.
• Multipath propagation: in terrestrial wireless transmission, signals arrive at the receiver via multiple paths (direct, reflected, diffracted), producing constructive and destructive interference that creates fading.

Acoustic Transmission

Acoustic signals are transmitted as pressure variations in a medium (air, water, solid material). Human face-to-face communication relies on acoustic transmission of voice signals through air:

• Speed: sound travels at approximately 343 meters per second in air at room temperature—much slower than electromagnetic transmission.
• Bandwidth: the human voice occupies roughly 300 Hz to 3400 Hz (telephone quality) to 20 Hz to 20,000 Hz (high fidelity audio).
• Noise: background noise (traffic, machinery, conversations) interferes with acoustic signal reception; reverberation in enclosed spaces causes additional distortion.
• Attenuation: acoustic signals attenuate rapidly with distance and are blocked by barriers.

Encoding and Modulation

For efficient and reliable transmission, messages are encoded into physical signal representations through modulation—the process of imposing the information-bearing variations onto a carrier:

Amplitude modulation (AM): the amplitude (strength) of the carrier signal is varied in proportion to the information signal. AM is simple and widely used in broadcast radio.

Frequency modulation (FM): the frequency of the carrier is varied in proportion to the information signal. FM is more resistant to amplitude noise than AM and produces higher fidelity audio.

Phase modulation (PM): the phase of the carrier is varied in proportion to the information signal. Phase modulation is used in digital communication systems.

Pulse code modulation (PCM): analog signals are sampled at regular intervals (at a rate at least twice the signal bandwidth, per the Nyquist-Shannon sampling theorem), the sample values are quantized to a finite set of discrete levels, and each level is encoded as a binary number. PCM is the basis of digital audio, digital telephony, and digital video.

The Nyquist-Shannon sampling theorem establishes that a band-limited signal can be perfectly reconstructed from samples taken at or above twice the signal's highest frequency:

where f_s is the sampling frequency and f_max is the highest frequency component in the signal.

Noise in Signal Transmission

Noise is any unwanted perturbation added to the signal during transmission that degrades the receiver's ability to reconstruct the original message. Different types of noise have different statistical properties:

Additive white Gaussian noise (AWGN): the most commonly analyzed noise model, in which noise adds to the signal with a Gaussian (normal) probability distribution and a flat power spectrum across all frequencies. AWGN models thermal noise in electronic circuits and many other physical noise sources.

Impulse noise: occasional large amplitude disturbances that are much stronger than the baseline signal level. Impulse noise is characteristic of electrical interference, atmospheric lightning, and certain types of equipment malfunction.

Multiplicative noise: noise that multiplies the signal rather than adding to it, as in fading in wireless channels where the signal strength varies with time. Multiplicative noise produces different distortion characteristics than additive noise and requires different analysis and mitigation techniques.

Crosstalk: interference from a nearby transmission channel. In electrical cables, capacitive coupling between adjacent conductors causes crosstalk; in wireless communication, transmissions on nearby frequencies cause interference.

Transmission Reliability and Error Correction

Shannon's noisy channel coding theorem establishes that reliable transmission is possible at any rate below channel capacity, through appropriate encoding. The practical implementation of reliable transmission uses error-correcting codes:

Block codes: information bits are encoded into blocks by adding parity bits that allow detection and correction of a specified number of errors per block. Reed-Solomon codes and BCH codes are widely used block codes.

Convolutional codes: information bits are encoded through a shift register whose output depends on the current and past input bits, creating a coded sequence with structured redundancy. Convolutional codes are widely used in wireless communication.

Low-density parity-check (LDPC) codes: codes with sparse parity-check matrices that can be decoded efficiently through iterative message-passing algorithms and approach channel capacity closely. LDPC codes are used in modern high-speed communication systems.

Turbo codes: codes constructed from parallel concatenated convolutional codes with interleaving, which achieve near-capacity performance through iterative decoding.

From Physical to Human Signal Transmission

The principles of physical signal transmission extend analogically to human communication, providing useful frameworks for thinking about communicative reliability and capacity:

Human speech as signal transmission: when a person speaks, they produce acoustic signals that are transmitted through the air (the channel) to the listener's ears, where they are decoded into perceived speech. The acoustic channel introduces noise (background noise, reverberation), attenuation (distance, barriers), and multipath effects (echoes). The speaker's articulation (encoding) and the listener's auditory processing (decoding) determine how well the original message is reconstructed.

Organizational communication as signal transmission: when messages pass through organizational hierarchies or networks, each transmission step introduces potential for distortion, delay, and information loss—the organizational equivalents of channel noise and attenuation. The design of organizational communication systems can be informed by transmission system design principles: reducing unnecessary relay stages, building in redundancy for important messages, providing feedback acknowledgment protocols, and designing channels appropriate to the information loads they carry.

Attention as bandwidth: human attentional capacity can be analyzed as a bandwidth-limited channel: we can process only a limited rate of information per unit time, and when the rate of incoming signals exceeds this capacity, we lose information through inattention, overload, and miscoding. Communication design that respects attentional bandwidth constraints—simplifying messages, pacing information delivery, providing appropriate redundancy—applies transmission system principles to human communication.