Section 3.2: Digital Modes and Signals #

In an analog voice signal, changes in the RF wave follow the sound you want to send. A digital mode instead represents information with discrete choices: one frequency or another, for example, or one of several phase positions. The receiver must recognize those choices and turn them back into data.

That opens up several kinds of communication. Radioteletype (RTTY) can carry a typed conversation, FT8 exchanges short structured messages at low signal levels, and packet systems can pass addressed messages through a network. They do not all provide the same speed, flexibility, or protection against errors. A noisy RTTY contact can produce garbled characters; other protocols add checks, correction, or retransmission to improve reliability.

The useful foundation is how information becomes a signal, how bits represent characters, and what happens when some of those bits arrive incorrectly. Station setup and on-air procedures build on these ideas in Sections 7.5 and 8.5.

Basic Digital Modulation Techniques #

Digital modes build on the same modulation principles we learned about in the previous section, but apply them in precise, computer-controlled ways to represent digital data.

A bit is one binary digit, 0 or 1. A symbol is one signal state sent during a defined time interval. One symbol can represent one bit or several bits. Baud counts symbols per second.

Frequency Shift Keying (FSK) #

One of the simplest digital modulation methods switches between two frequencies to represent the binary digits 0 and 1—it’s like having a very precise, computer-controlled FM system that instantly jumps between exact frequencies instead of gradually varying them.

Key Information:
Direct binary FSK modulation is generated by changing an oscillator’s frequency directly with a digital control signal.
The two separate frequencies of an FSK signal are identified as mark and space.

A receiver identifies each bit by detecting which of the two frequencies is present. The names “mark” and “space” come from telegraphy, where “mark” meant a signal was present and “space” meant no signal. In FSK, “mark” represents a binary “1” and “space” represents a binary “0”.

Binary FSK uses two frequencies. Here, time moves from left to right through eight equal symbol intervals. In this example, the upper frequency is mark, representing one; the lower frequency is space, representing zero. The bits are one, zero, one, one, zero, zero, one, zero. The frequency stays the same across the two adjacent ones and across the two adjacent zeros. It changes only when the bit value changes. Vertical position represents frequency, not signal strength: the space state still transmits a tone. — Binary FSK sends one of two frequencies during each symbol interval. Vertical position shows frequency, not amplitude.

Advanced FSK: Multiple Tones #

Why stop at two tones? Modern digital modes build on basic FSK by using more frequencies.

Key Information: FT8 uses 8-tone frequency shift keying modulation.

With 8 different tones, each data symbol can represent 3 coded bits (since 2³ = 8 possible combinations). It’s like having an 8-note musical scale where each note carries information! FT8 sends one tone at a time; its sequence also includes synchronization and error-correction information.

Phase Shift Keying (PSK) #

Instead of changing frequency like FSK, PSK takes a different approach—it changes the phase of the carrier to represent data. Remember our spinning wheel analogy from Section 1.2? PSK works by controlling exactly when each “wheel rotation” begins.

Binary PSK (BPSK) uses two phase positions—think of it as starting each cycle either “on time” or “half a cycle late.” The “B” in BPSK stands for “Binary,” meaning it uses two states, just like binary FSK uses two frequencies.

For even more efficiency, we can use more phase positions:

Key Information: QPSK modulation transmits digital data using 0-, 90-, 180- and 270-degree phase shifts to represent pairs of bits.

The “Q” stands for “Quadrature,” referring to a 90-degree relationship; QPSK uses four phase positions. Each phase position represents a unique 2-bit combination: 00, 01, 10, or 11.

Four points around a circle show QPSK’s four possible phase positions, 90 degrees apart. In this example, zero degrees represents zero zero; 90 degrees represents zero one; 180 degrees represents one one; and 270 degrees represents one zero. The points are equally far from the center: their angles differ, not their distances. Four distinct phase states allow each symbol to identify one of four bit pairs, carrying two bits of information. — Four phase states can represent two bits per symbol. This is one illustrative mapping.

Engineers call this a “constellation diagram” because the phase positions look like stars in specific locations around a circle.

Key Information: QPSK31 is sideband sensitive, its encoding provides error correction, and its bandwidth is approximately the same as BPSK31.

The mode can often detect and fix transmission errors automatically, even within its narrow bandwidth. “Sideband sensitive” means both stations must select the same sideband. The similar symbol rate and signal shaping keep QPSK31’s bandwidth close to BPSK31’s; its extra coding capacity helps correct errors instead of simply doubling typing speed. Ordinary BPSK31 has no error correction, so noise can garble its text.

Character Encoding: From Letters to Bits #

Digital modes need a way to convert text characters into binary data. Different modes use different encoding systems optimized for their specific requirements.

Baudot Code: The Classic System #

Key Information: Baudot code is a 5-bit code with additional start and stop bits.

Baudot is still used in RTTY, even though it predates modern computers. Each character has five data bits, plus the start and stop bits. With only 32 possible combinations (2⁵), the code has room for letters and control functions, but requires special “shift” characters to access numbers and punctuation.

Varicode: The Smart System #

Some modes use variable-length encoding:

Key Information:
PSK31 uses Varicode for sending characters.
PSK31 uppercase letters use longer Varicode bit sequences and thus slow down transmission.

Like Morse code, Varicode assigns shorter bit patterns to common letters (like ’e’ and ’t’) and longer patterns to rare characters (like ‘q’ and ‘z’). This optimization makes a real difference in transmission speed. Most experienced PSK31 operators type in lowercase to keep things moving quickly, saving uppercase for when they really want to emphasize something.

Weak Signal Modes: Pulling Signals from the Noise #

Some of the most impressive digital modes specialize in extremely weak signal communication.

WSPR: The Ultimate Weak Signal Explorer #

Key Information: WSPR is a digital mode used as a low-power beacon for assessing HF propagation.

WSPR (pronounced “whisper”) uses long transmissions and a very narrow signal to recover limited information. Stations transmit just their callsign, grid square, and power level using incredibly low power, sometimes just milliwatts. Yet receivers worldwide can decode these whisper-quiet signals and automatically report what they heard via the internet.

When conditions cooperate, it’s fascinating to run WSPR at 200 milliwatts and see reports from stations thousands of miles away that heard your tiny signal. The automatic reporting creates a real-time global map of propagation conditions.

FT8: Weak-Signal Contacts #

Key Information:
FT8 can receive signals with very low signal-to-noise ratios.
An FT8 signal report of +3 means the signal-to-noise ratio is equivalent to +3dB in a 2.5 kHz bandwidth.

Many operators are amazed the first time they successfully complete an FT8 contact with a signal they can’t even hear. The computer looks for known signal patterns, uses error-correction coding, and combines evidence over time to recover a short structured message.

FT8 signal reports use a standardized measurement system. From Section 1.5, +3dB means about twice the power. Here, signal power is about twice the reference noise power in 2500 Hz (2.5 kHz); it does not mean S3 or three watts. The 2500 Hz reference is not the width of a single FT8 signal. These reports estimate conditions at the receiving station, using a common reference so the numbers can be compared.

Error Detection and Correction #

A computer may detect a weak signal, but that does not guarantee it recovers every bit correctly. Noise or interference can change a received symbol. Digital protocols address that problem by adding information that helps detect or correct errors, or by arranging to send damaged information again.

Forward Error Correction (FEC) #

Key Information: Forward error correction (FEC) allows the receiver to correct data errors by transmitting redundant information with the data.

It’s like sending the same message in multiple slightly different ways—if one version gets garbled by noise or fading, the others provide enough information to reconstruct what was intended. No retransmission is required for errors the code can correct! For a small example, send 111 to represent 1. If noise changes it to 101, a majority vote still points to 1. Real codes use more efficient patterns, and too many errors can defeat them.

ARQ: When Accuracy Matters #

When a receiving station can request another transmission, Automatic Repeat reQuest (ARQ) provides a different approach:

Key Information:
In an ARQ mode, a NAK response means request retransmission of the packet.
Excessive transmission attempts in ARQ mode result in the connection being dropped.

When the receiving station detects corrupted data, it automatically sends a “Negative Acknowledgment” (NAK) requesting a do-over. It’s like having an automatic “say again?” response. Dropping the connection prevents systems from getting stuck trying forever when propagation just isn’t cooperating.

Packet Radio and Network Protocols #

Digital modes enable sophisticated networking capabilities impossible with analog voice.

Packet Structure #

Digital communication organizes information into neat, addressable packages:

Key Information: The header of a packet radio frame contains routing and handling information.

Like the addressing information on an envelope, the header tells the network where the packet should go and how to handle it along the way. This structure lets software route and handle the data without interpreting the message itself.

A simplified packet is divided into three fields. First comes the header, with routing and handling information. Next comes the data field, containing the message. The final field is an error check that helps the receiving system detect damaged data. The message is therefore only one part of what the packet carries. — The header tells the system how to handle the data; an error check helps detect corruption.

Mesh Networking: The Self-Healing Network #

One of the most impressive capabilities of digital networks is their resilience:

Key Information: In mesh networks, packets can reach their destination via alternate nodes if one node fails.

A microwave mesh network is like a city with multiple roads—if one route gets blocked, traffic automatically finds another way around. That alternate route must actually exist and be usable. When it does, automatic routing can keep messages moving despite a failed node.

Digital Voice: The Best of Both Worlds #

Digital techniques can encode voice as well as data, combining the naturalness of voice communication with digital advantages.

Key Information: DMR, D-STAR, and System Fusion provide digital voice modes.

These systems convert your voice to digital data using specialized codecs (coder-decoders), transmit it using digital techniques, and reconstruct natural-sounding voice at the receiving end. You get the error correction and efficiency benefits of digital modes while still enjoying the immediacy and personality of voice communication.

The systems have different compatibility requirements; a radio using one does not automatically decode the others.

Digital modes can extend range and reliability for particular kinds of messages and operating conditions. The gains often involve tradeoffs in message length, speed or flexibility.

Practical Advantages for General Class Operators #

As you move into General operating, digital modes can offer several practical advantages:

Extended Range: Digital modes often work with signals too weak for analog voice
Automated Exchanges: Software can handle parts of an exchange, while station-control duties still apply; see Section 9.4
Error Control: Depending on the protocol, error detection, correction, or retransmission can improve message reliability
Signal Reports: Digital reports provide numerical estimates using a defined reference
Network Capabilities: Digital modes can connect through networks and gateways
Emergency Communication: Digital modes work well in challenging conditions

Whether a signal carries voice or digital data, the receiver still has to select it from everything arriving at the antenna. In the next section, we’ll see how receivers can use frequency mixing to move the wanted signal to a frequency where filtering is practical—and why the same process can sometimes create interference.