Blind modulation format identification using nonlinear power transformation

Abstract

This paper proposes and experimentally demonstrates a blind modulation format identification (MFI) method delivering high accuracy (> 99%) even in a low OSNR regime (< 10 dB). By using nonlinear power transformation and peak detection, the proposed MFI can recognize whether the signal modulation format is BPSK, QPSK, 8-PSK or 16-QAM. Experimental results demonstrate that the proposed MFI can achieve a successful identification rate as high as 99% when the incoming signal OSNR is 7 dB. Key parameters, such as FFT length and laser phase noise tolerance of the proposed method, have been characterized.

Experimental Demonstration of Cognitive Provisioning and Alien Wavelength Monitoring in Multi-domain EON

“…Fig. 3a shows the schematic of MFI and SC-performance-monitoring blocks (please refer to [6], [7] for more details). At the ingress node, an FPGA with four WDM 10Gb/s TRXs acts as intra-domain traffic source.…”

“…The broker decides to route the request through link 1-3 (shortest path -lightly loaded) of the middle domain using QPSK. The MFI block recognizes that the modulation format is QPSK based on the presence of a spectral tone after the 4th power operation and FFT function in its DSP block (see the difference with scenario 2 where for 8PSK there is not tone after the power operation and FFT block [7]). Then, by applying a SC and measuring its BER at the ingress and egress nodes, the DM can monitor the alien wavelengths' QoT by using preexisting correlation BER data stored in a database in the DM.…”

“…The proposed cognitive multi-domain EON architecture makes use of a DNN machine-learning-aided RMSA and alien-wavelength monitoring. Monitoring is performed with a combination of a supervisory channel (SC) technique [6] and a novel modulation format identification (MFI) scheme [7] to discover the modulation format of alien signals. © 2018 Optical Society of America.…”

A Density Clustering Algorithm for Simultaneous Modulation Format Identification and OSNR Estimation

“…In recent years, researchers have utilized various technologies to achieve MFI such as signal cumulant and signal power distribution-based method [7,8], peak-to-average-power ratio of received data samples-based methods [9,10], Stokes space-based methods [11][12][13][14][15], and machine learning-based methods [16][17][18][19]. Various OSNR estimation techniques for coherent detection systems have been proposed recently including statistical moments [20], error vector magnitude [21], delay-line interferometer [22,23], Stokes parameters [24], Golay sequences [25], offset filtering, and optical power measurement [26], radio frequency (RF) spectrum [27], and amplitude noise correlation [28] based methods.…”

Modulation format identification assisted by sparse-fast-Fourier-transform for hitless flexible coherent transceivers

“…modulation (QAM) formats as the identification criteria [7][8][9][10][11][12][13][14][15]. The features include distributions or histograms of amplitude and phase etc.…”