Synchronization and Clock Recovery#
The main process of QPSK demodulation is carrier synchronization and symbol synchronization.
There are two main methods for recovering carrier synchronization and symbol synchronization, one is based on feedforward compensation, and the other is based on feedback phase-locked loop, which correspond to open-loop and closed-loop methods in an all-digital receiver, respectively.
Closed-loop synchronization scheme is implemented using feedback control scheme. The Costas loop was first proposed in 1956. It has a certain frequency offset suppression capability and is the best device for tracking low signal-to-noise ratio and suppressing carrier signals. It can achieve signal tracking, acquisition, synchronization, and demodulation. The closed-loop synchronization takes a certain amount of time to enter a stable state, but the error after locking is very small, which is suitable for continuous demodulation schemes.
The open-loop synchronization scheme has fast synchronization speed and large capture bandwidth, but the locking accuracy is relatively poor, and it requires accurate estimation of the timing error during demodulation, which is suitable for fast capture in burst communication.
Symbol Synchronization#
Also known as symbol synchronization, baud synchronization, timing synchronization.
In a digital communication system, the information from the transmitter to the receiver is actually a series of symbol sequences. The transmitter first performs digital signal shaping and then modulation. The receiver recovers the digital signal through sampling and demodulation.
However, due to Doppler frequency shift and clock frequency offset between the transmitter and the receiver, there will be an error between the sampling decision timing at the receiver and the optimal sampling point of the data.
Therefore, in order to recover the data correctly, it is necessary to know the start and end time of each symbol. Therefore, the receiver needs a timing synchronization pulse sequence as a reference for sampling decision, and the process of generating the timing synchronization sequence is called symbol synchronization.
Symbol Synchronization Methods#
Data-Aided Method#
- Mueller-Muller Algorithm
- Zero-Crossing Detection Method
Non-Data-Aided Method#
- Phase Comparison Method
At the receiver, a sampling pulse signal to be synchronized is generated, and the phase error between this sampling pulse signal and the optimal sampling timing is continuously detected. The frequency and phase of the timing sampling timing are adjusted based on the error, so that the sampling timing can be stably maintained at the optimal sampling point.- Early-Late Gate
- Interpolation Method
- Filtering Method
[[Carrier Synchronization]]#
Carrier recovery refers to the generation of a local oscillation signal in the receiving equipment that is in the same frequency and phase as the received signal's carrier, for coherent demodulation by the demodulator. There are two commonly used methods for carrier synchronization:
- Pilot Insertion Method: The main principle is to add a pilot carrier signal at the modulation end, and then extract the pilot signal at the demodulation end, using correlation operation to make the local NCO generate a carrier with zero or small frequency difference from the pilot carrier signal.
- Direct Recovery Method: Recover the carrier directly from the modulated signal and allocate the signal power completely to the carrier signal carrying the baseband information.