Increasing Throughput with Polarization
In fiber-optic systems, capacity can be increased by improving . At comparable average symbol energy, moving from QPSK (2 bits per symbol) to 16-QAM (4 bits per symbol) increases the number of bits per symbol, but the smaller distances between constellation points increase the required signal-to-noise ratio and can limit transmission distance.
Dual-polarization QPSK (DP-QPSK) uses an additional physical dimension of the signal: polarization. Two independent QPSK components carry a combined 4 bits per symbol, matching the bit count of 16-QAM under the stated symbol-counting convention.
Polarization of Light: The Operating Principle
Light is an electromagnetic wave whose electric field has a defined polarization.
describes the orientation of the electric field's oscillations as the wave propagates. DP-QPSK uses two orthogonal polarization states (for example, vertical and horizontal) as separate, parallel communication channels in the same fiber and within the same band. Real fibers can introduce polarization coupling and polarization-mode dispersion, so the receiver separates the components and can compensate for channel effects.
How DP-QPSK Works: The Main Steps
The process of creating and transmitting a DP-QPSK signal involves sending two separate QPSK signals in parallel, one on each polarization.
- Split the Data Stream: The original high-speed stream of data is divided into two separate, lower-speed sub-streams.
- Independent QPSK Modulation: Each sub-stream is used to modulate its own, independent QPSK signal. Recall that a standard QPSK signal encodes 2 bits of information per symbol by shifting between four phase states.
- Assign to Polarizations:The first QPSK signal is transmitted on one polarization (e.g., horizontal or 'X'), while the second QPSK signal is transmitted on the orthogonal polarization (e.g., vertical or 'Y').
- Transmit Simultaneously: Both polarized signals are combined and transmitted down the same fiber optic cable, occupying the same frequency band at the same time.
- Separate and Demodulate: At the receiver, a special optical component called a polarization beam splitter separates the horizontal and vertical waves. Each wave is then sent to its own QPSK demodulator, and the two data sub-streams are recovered and recombined into the original high-speed stream.
Bit Count and Energy Normalization
Transmitting two independent QPSK streams in parallel within the same band doubles the information bit rate at a given symbol rate.
- A standard QPSK signal carries 2 bits per symbol.
- A DP-QPSK signal carries information from two parallel QPSK signals.
- Therefore, each combined symbol carries of information.
Power comparisons should state whether the energy refers to one polarization or to the combined signal. If the mean symbol energy in each branch is 1, the combined symbol has energy before any additional normalization. If the combined symbol is normalized to and the branches are equal, each branch has .
For ideal coherent DP-QPSK in AWGN with Gray mapping, polarization separation leaves two coherent QPSK decisions. With defined per information bit in one branch and two-sided noise power spectral density , the BER of each branch is . The same per-branch or combined-power convention must be used when reporting .
At the same symbol rate, 4 bits per combined symbol matches the bit count per symbol of 16-QAM or 16-PSK with a single polarization. This does not imply identical noise tolerance: DP-QPSK distributes the information across two QPSK constellations, and the requirements depend on power, filtering, receiver design, and channel. A QPSK constellation has four points, so the receiver distinguishes fewer states per polarization than with 16-QAM. The high throughput comes from parallel transmission on two orthogonal polarizations.