Energy-Efficient Implementation of Carrier Phase Recovery for Higher-Order Modulation Formats

Abstract

We introduce circuit implementations of one-and two-stage carrier phase recovery (CPR) for 256QAM coherent optical receivers. We describe in detail the optimizations of algorithms, such as modified Viterbi-Viterbi (mVV), blind phase search (BPS), and principal component-based phase estimation (PCPE), that are required to develop energy-efficient CPR circuits and show how design parameter settings and limited fixedpoint resolution affect the SNR penalty. 30-GBaud CPR circuit netlists synthesized in a 22-nm CMOS process technology allow us to study trade-offs between energy per bit and SNR penalty. We show that it is possible to reach an energy dissipation of around 1 pJ/bit at an SNR penalty of 0.6 dB for two-stage PCPE+BPS and mVV+BPS implementations, and that PCPE+BPS is the preferred choice thanks to its smaller area.

Transoceanic Phase and Polarization Fiber Sensing using Real-Time Coherent Transceiver

“…The equalizer used 9-bit resolution for both coefficients and error calculation. The carrier phase estimation (CPE) consisted of a principal component based phase tracker [5] , using 7 bits angle resolution and 64 symbols block averaging. To enable combined parallel DSP and environmental sensing, information must flow in real time from the hardware level (clocked logic) to the CPU and for post-processing/data management.…”

Multi-Format Carrier Phase Recovery Using a Programmable Circuit

“…Based on the work done in [3] and [4], we have identified principal component-based phase estimation (PCPE) [5] to be a good choice for a multi-format CPR implementation, as it can handle multiple square QAM formats without any modification. Other popular algorithms, such as blind phase search [6] or versions of the Viterbi-Viterbi estimator [2], need larger circuit extensions to support multiple formats, resulting in a larger total area.…”

“…This method has the added benefit of reducing the SNR penalty for both 16QAM and 64QAM. The mask and other control signals are generated and synchronized to the DSP pipeline in the control module, which also controls the window size, A, for the covariance calculation, making it possible to optimize the processing for different laser linewidths [4]. To reduce circuit complexity, the A setting is limited to multiples of P.…”

Benchmarking of Carrier Phase Recovery Circuits for M-QAM Coherent Systems

“…Using a combination of FPGA emulation and ASIC analysis, we present SNR performance penalty, power dissipation, latency, area usage, and cycle slip probability for five different CPR implementations. Using a consistent benchmarking methodology, for 16, 64, and 256QAM, we can extend our previous implementation work [1,2] and identify which CPR circuits are optimal for a particular M-QAM format.…”