Demodulation
Demodulation is the reverse operation of modulation. Modulation converts a low-frequency information (message or baseband) signal into a form suitable for transmission over a channel by superimposing it on a high-frequency carrier. Demodulation recovers the original information signal from the received carrier-containing waveform.
Basic concepts
- Message (baseband) signal: the original information to be transmitted (voice, data, video, sensor reading).
- Carrier signal: a high-frequency sinusoid used to carry the message over the physical medium.
- Received signal: the transmitted carrier plus message after propagation; it may be distorted and noisy.
- Demodulator: the receiver block that extracts the baseband message from the received waveform. Typical demodulator functions include synchronisation (carrier recovery), detection, filtering and sometimes error detection/correction.
Transmission process - brief
The transmitter, channel and receiver are the three core parts of a communication system. The carrier frequency is chosen much higher than the message bandwidth to allow efficient radiation, multiplexing and filtering. Common physical media include optical fibre, twisted pair, coaxial cable and wireless channels; each imposes its own distortions and noise that the demodulator must tolerate.
Types of analog modulation
Common analog modulation families are:
- Amplitude modulation (AM) and its variants
- Pulse modulation (pulse-amplitude, pulse-width, pulse-position)
- Angle modulation (frequency modulation and phase modulation)
Amplitude Modulation (AM)
In amplitude modulation the instantaneous amplitude of the carrier is varied in proportion to the instantaneous amplitude of the message signal. A general AM signal may be written as s(t) = [A_c + m(t)] cos(ω_c t), or using a modulation index μ, s(t) = A_c [1 + μ m_n(t)] cos(ω_c t), where m_n(t) is the normalized message (|m_n(t)| ≤ 1).
Demodulation methods for AM
- Envelope detector: a simple, non-coherent detector used when the carrier is present (conventional AM, i.e., carrier not suppressed). It typically uses a diode followed by an RC network. The diode rectifies the incoming waveform and the RC network tracks the envelope. The RC time constant must be chosen to follow the envelope but filter out carrier ripples.
- Coherent (synchronous) detector: used for suppressed-carrier forms such as DSB-SC and SSB. The received signal is multiplied by a locally generated carrier that is phase-and-frequency-synchronised with the transmitter carrier, then low-pass filtered to recover the message. Accurate carrier recovery is essential; phase or frequency error causes amplitude and phase distortion of the recovered message.
- Product detector: a practical implementation of the coherent detector. The received signal is multiplied by a recovered carrier (from a local oscillator). After multiplication, low-pass filtering extracts the baseband component.
- Carrier recovery: for coherent detection a local carrier must be generated in phase with the transmitter carrier. Methods include using pilot carriers, phase-locked loops (PLLs) and Costas loops (commonly used for DSB-SC and SSB to recover carrier phase and frequency).
Practical notes - envelope detector and coherent detection
- Envelope detector design: the diode charges a capacitor to the peak; when the input falls the capacitor discharges through the load resistor. The RC time constant should satisfy 1/ω_c ≪ RC ≪ 1/B_m, where ω_c is carrier angular frequency and B_m is message bandwidth, so the detector follows the envelope but smooths the carrier ripple.
- Coherent detector sensitivity: coherent detection provides better signal-to-noise performance than envelope detection when carrier recovery is possible; however it requires synchronisation hardware and is more complex.
Variants of AM
- Double Sideband (DSB): both upper and lower sidebands are present and symmetric about the carrier.
- Double Sideband with Carrier (conventional AM): carrier plus both sidebands; straightforward envelope detection is possible.
- Double Sideband Suppressed Carrier (DSB-SC): carrier suppressed; efficient in power but requires coherent detection at receiver.
- Single Sideband (SSB): only one sideband (upper or lower) is transmitted; spectral efficiency is improved and bandwidth is halved compared with DSB, but receiver requires frequency/phase accuracy or a carrier reinsertion method.
- Vestigial Sideband (VSB): a partial sideband is transmitted (a vestige of the opposite sideband) - used where perfect SSB is difficult while saving bandwidth (example: television broadcasting).
- Quadrature Amplitude Modulation (QAM): two independent message signals are transmitted on carriers 90° out of phase (in-phase and quadrature). QAM carries two signals in the same bandwidth and is widely used in digital communications (but conceptually an analogue combination of amplitude and phase weighting).
Pulse Modulation
Pulse modulation schemes encode the message into a train of pulses. The pulses occur at sampling instants; information can be carried by pulse amplitude, width or position.
Common pulse modulation types
- Pulse-Amplitude Modulation (PAM): the amplitude of each pulse is proportional to the sampled value of the message.
- Pulse-Width Modulation (PWM): the width (duration) of each pulse varies in proportion to the message amplitude.
- Pulse-Position Modulation (PPM): the position (timing) of each pulse within a time slot is varied according to the message amplitude.
Demodulation of pulse modulated signals
- PAM demodulation: sampling recovery followed by low-pass filtering reconstructs the continuous waveform from pulse amplitudes. A hold circuit or reconstruction filter is used.
- PWM demodulation: a ramp or integrator can convert pulse width to an amplitude; the output is then low-pass filtered to recover the message.
- PPM demodulation: timing extraction circuits detect the pulse time offsets and convert them to amplitude values, followed by filtering.
- Practical components: RC networks, transistor ramp generators, comparators and sample-and-hold circuits are commonly used in analytic demodulators for PWM/PPM/PAM.
Applications
- Pulse modulation is the basis for many digital transmission schemes and is used in telephony, Ethernet physical layers, remote control and optical transmission where pulses are natural carriers.
Angle Modulation
Angle modulation varies carrier phase or frequency with the message. The two principal forms are frequency modulation (FM) and phase modulation (PM). Both are examples of angle modulation because instantaneous frequency is the time derivative of phase.
Definitions
- Frequency modulation (FM): instantaneous frequency of the carrier varies with the message signal.
- Phase modulation (PM): instantaneous phase of the carrier varies with the message signal.
Demodulation methods for angle modulation
- Frequency discriminator: converts instantaneous frequency variations into amplitude variations which are then low-pass filtered to obtain the baseband message. Typical discriminator circuits are slope detectors, Foster-Seeley discriminators and ratio detectors (in FM broadcast receivers).
- Phase-locked loop (PLL): a widely used coherent method. The PLL locks a local oscillator to the incoming carrier; the control voltage used to tune the local oscillator is proportional to the instantaneous frequency (or phase) deviation and hence yields the demodulated output.
- Selective carrier (filter) method: uses a frequency-selective network followed by envelope detection to convert FM to AM and then detect; this method is less common and less linear than dedicated FM discriminators or PLLs.
- IC solutions: many integrated circuits combine limiter, discriminator and filtering stages to provide practicable FM demodulation with good sensitivity and noise performance.
Notes on PLL demodulation
- A PLL tracks phase and frequency of the received carrier. The loop error voltage (after loop filter) represents the instantaneous frequency deviation (for FM) and can be used as the recovered message after appropriate filtering and scaling.
- PLLs provide good capture and tracking in the presence of moderate noise and are suitable for carrier recovery in coherent demodulators.
Practical considerations for demodulator design
- Signal-to-noise ratio (SNR): demodulator choice affects SNR performance. Coherent detection generally gives better SNR than non-coherent (e.g., envelope) detection when carrier recovery is available.
- Carrier synchronisation: for suppressed-carrier schemes (DSB-SC, SSB) accurate carrier phase/frequency recovery is essential; Costas loops are often used for suppressed carriers.
- Filtering: after detection, baseband filters remove unwanted high-frequency components and noise; anti-aliasing and reconstruction filters are important in sampled/pulse systems.
- Distortion: detector nonlinearity, phase error and time-variant channel effects produce distortion; equalisation and adaptive filtering may be required for high fidelity.
- Implementation: demodulators can be realised in analogue hardware (diodes, discriminators, RC networks, PLL ICs) or in software/firmware after digitising the received waveform (software-defined radio approach).
Examples and typical applications
- AM broadcast: envelope detectors are used for simple receivers; SSB is used in long-range HF communication to save bandwidth and power.
- FM broadcast and two-way radio: discriminators and PLLs are employed; FM offers greater noise immunity for high-quality audio transmission.
- Telemetry and remote control: PWM and PPM are common where pulse timing or width carries control information.
- Digital communications: QAM and phase modulation techniques (QPSK, PSK) are demodulated using coherent receivers with carrier recovery and baseband digital signal processing.
Summary
Demodulation recovers the baseband signal from a modulated carrier. Selection of demodulation method depends on modulation type (AM, pulse, angle), presence or absence of carrier, required fidelity and robustness to noise, and implementation constraints. Key receiver functions include carrier recovery, detection (envelope, coherent/product, discriminator), and filtering/reconstruction. Practical demodulator design balances complexity, synchronisation requirements and signal-to-noise performance.