Generalized Signal-to-noise Ratio Research Articles

On the Generalization of Cognitive Optical Networking Applications using Composable Machine Learning

On the Generalization of Cognitive Optical Networking Applications using Composable Machine Learning

Local and global optimization methods for optical line control based on quality of transmission

The ever-increasing demand for data traffic in recent decades has pushed network operators to give importance to the aspect of infrastructure control to facilitate its scalability and maximize its capacity. A generic lightpath (LP) is deployed starting from a traffic request between a given pair of nodes in a network. LPs are operated in the network based on an estimate of the quality of transmission (QoT), which is derived from the physical layer characteristics of a selected route. Regardless of the model used to estimate QoT, it is necessary to calibrate the model to maximize its accuracy and define minimum design margins. The model calibration process depends significantly on the type of data that can be collected in the field (i.e., type of metric, resolution) and therefore on the available monitoring devices. In this work, a systematic evaluation of the QoT estimation is carried out on a multi-span erbium-doped-fiber-amplified optical line system (OLS) using in the first case only total power monitors and in the second experimentally emulating optical channel monitors (OCMs). Given the type of monitoring devices available, three different physical models are calibrated, and six optimization methods are used to define the optimal configuration of the target gain and tilt parameters of the optical amplifiers, jointly optimizing the working point of all amplifiers (global approach) or proceeding span by span (local approach). Subsequently, the OLS was set in each configuration obtained, and the generalized signal-to-noise ratio (GSNR) profile was measured at the end.

Enabling seamless migration of optical metro-urban networks to the multi-band: unveiling a cutting-edge 6D planning tool for the 6G era

The relentless demand for high-bandwidth services and applications in the beyond-5G/6G era necessitates optical networks to provide ample spectral resources to accommodate the increasing traffic loads. This requirement is especially pronounced in metro and regional segments, where a projected traffic compound annual growth rate (CAGR) of over 40%, driven by content delivery network (CDN) traffic concentrated near the end users, threatens to push the fiber capacity to its limits on a number of heavily utilized links in the short to medium term. This limitation cannot be circumvented despite the utilization of spectrally efficient modulation formats designed for short-distance transmission. Thus, it is imperative to explore migration strategies to solutions that allow the expansion of spectral resources beyond the current usage of the extended C-band (4.8 THz). In this article, we present a comprehensive evaluation of multi-homed edge-to-core routing, modulation-level selection, and spectrum assignment simulations conducted over reference metropolitan area networks (MANs) by leveraging on a novel network planner called 6D-MAN, which we have developed to adapt to heterogeneous multi-layer networks based on multi-band and multi-fiber transmission. Our simulations consider coherent transmission with flexible modulation format transponders operating at capacities ranging from 100 to 400 Gbps on the 50 GHz grid and explore link-by-link migration strategies restricted to the C+L-band, ensuring optimal performance in terms of the generalized signal-to-noise ratio (GSNR). The obtained results demonstrate that deploying L-band equipment on a selected number of links effectively extends the lifespan of existing networks, enabling congestion-free operation with minimal intervention. In their entirety, our findings underscore the importance of contemplating alternative spectral resources beyond the conventional C-band. They unequivocally accentuate the potential of L-band equipment strategic deployment to tackle capacity constraints in MANs, resulting in a 60% reduction in fiber length and a remarkable 45% decrease in new fiber-pair link deployment needs. The profound implications of efficient fiber capacity utilization manifest themselves in substantial OPEX and optical total cost of ownership savings, amounting to approximately 71% and 17%–27%, respectively, over a 10-year period. This research provides valuable insights for MAN operators seeking sustainable strategies to support the growing demands of future beyond-5G/6G services.

QoT-aware tree selection, routing, modulation, and spectrum assignment for filterless EONs over the C + L-band

Filterless optical networks (FONs) as an economical solution use passive couplers/splitters rather than expensive active filter devices. In this way, after intermediate and destination nodes, lightpaths are dropped and continued, which generates leakage signals in other links and wastes spectrum. Thus, designing efficient resource allocation in FONs by considering the leakage signals and their interfering effects is of utmost importance. On the other hand, a gradual transition from wavelength division multiplexing optical networks to elastic optical networks (EONs) is occurring due to their efficient utilization of spectrum. Furthermore, utilizing the L-band beside the conventional C-band for spectrum assignment offers a wide range of frequency resources. Therefore, in this paper, we propose an integer linear program (ILP) to solve quality of transmission (QoT)-aware tree selection, routing, modulation, and spectrum assignment problems in filterless EONs over the C+L-band. Furthermore, we provide heuristic algorithms to deal with complex large-scale networks. The performance gap of the proposed ILP and heuristic algorithms is evaluated over a small-scale (5-node) network. The results show that the ILP and heuristic algorithms have almost the same performance in terms of spectrum usage and assigned modulation format, and ILP has a slightly higher generalized signal-to-noise-ratio (GSNR) (0.23 dB or 0.8% at optimum launch power). Furthermore, the heuristic algorithms are also examined over a large-scale network (TID region A topology). The results reveal that the GSNR estimation method severely affects the performance in terms of spectrum usage, blocking, and outage. Furthermore, by using the proposed MX5 method, as long as there is a fill margin of approximately 2 dB, there is no outage or blocking over the C+L-band, up to a network throughput of 110 Tbps and conventional C-band transmission with lower throughput (i.e., 40 Tbps). Finally, our extensive numerical results provide a rule of thumb for balancing blocking, outage, spectrum usage, and the number of expensive L-band transponders by selecting the appropriate modulation assignment method.

Enhanced DOA Estimation With Co-Prime Array in the Scenario of Impulsive Noise: A Pseudo Snapshot Augmentation Perspective

Co-prime array configuration is popular in the recent development of array signal processing by exploiting the sparse arrangement of arrays and the co-prime feature of the number of subarray elements. However, the assumption of Gaussian noise in most co-prime array processing research might lead to model mismatch in practical scenarios of impulsive noise, and therefore have an adverse impact on the estimation of direction of arrival (DOA). Moreover, co-prime array builds an enlarged virtual array by vectorizing the covariance matrix of the received signals, where the equivalent received signals of the virtual array have only a single snapshot. In this paper, we propose an enhanced fractional low-order method (EFLOM) for co-prime array configuration in scenarios of impulsive noise, from the perspective of pseudo snapshot increment. Since impulsive noise does not have finite second-order statistics or high-order cumulant, we construct a series of equivalent covariance matrices by using phased fractional low-order moments with different orders of the received signals. Then, the vectorization of multiple equivalent covariance matrices can be formed as the equivalent received signals of the virtual array with multiple snapshots. Since the reformed observations of the equivalent received signals are from the same sources, additional spatial smoothing preprocessing operations are still needed for decorrelation. While in this paper, we propose an improved spatial smoothing (ISS) technique by applying the information of both autocorrelation and cross subarray correlation of the covariance matrix. Afterwards, the classical multiple signal classification (MUSIC) is applied for the estimation of DOAs. The performance of the proposed method is theoretically verified and simulations are also provided to show its effectiveness in terms of the generalized signal to noise ratio (GSNR), parameter of impulsive noise, and angle separation.

Deep learning and deep transfer learning-based OPM for FMF systems

Optical performance monitoring (OPM) is indispensable to guarantee stable and reliable operation in few-mode fiber (FMF)-based transmission. OPM consists of measuring optical phenomena such as generalized signal-to-noise ratio (GSNR) based on analytical models. GSNR comprises nonlinear interference (NLI) noise which can be calculated either by exact analytical models e.g., enhanced Gaussian noise (EGN) model which is accurate but computationally complex, or asymptotic analytical models e.g., closed-form EGN model which are approximate but computationally fast. In this paper, we employ deep learning (DL) as an accurate and fast alternative for OPM in FMF-based transmission. However, DL-based OPM requires a large dataset to achieve proper performance whilst it is very difficult and time-consuming to obtain a large field or synthetic dataset. Regarding this issue, we develop deep transfer learning (DTL) for OPM in FMF to realize a fast response requiring a small training dataset and a few training epochs despite various changes in system/link parameters such as launched power, fiber type, and the number of modes. Results show root mean squared error of GSNR estimation is less than 0.02 dB for DL and DTL-based OPM methods. Compared to DL-based OPM, DTL-based OPM records 3 and 5 times reduction in required training dataset size and the number of epochs, respectively which is beneficial for real-time applications

DOA Estimation With Co-Prime Array by a Combined BNC-PFLOM Method in Presence of Impulsive Noise

Direction of arrival (DOA) estimation with co-prime array is well studied in the field of array signal processing, but most research assumes noise to be Gaussian white noise. In practical scenarios, there might be impulsive noise, which has neither second-order statistics nor high-order cumulant, therefore cannot be processed directly in co-prime array configurations. In addition, the signals received by the co-prime virtual array are of single snapshot, which can be considered as coherent virtual signals coming from the same source and therefore performance degradation in DOA estimation. To solve the above mentioned problems, a combined bounded non-linear covariance (BNC) and phase fractional low-order moments (PFLOM), called BNCPFLOM, is proposed in this paper. The proposed BNC-PFLOM combines the merits of both BNC and PFLOM by constructing equivalent data covariance matrices of the received signal to lower the impact of impulsive noise. Moreover, the proposed BNC-PFLOM can build multiple pseudo snapshots of virtual signals from the co-prime array by using different orders of the proposed method. With the increase of pseudo snapshots, the proposed method is more robust in DOA estimation. Besides, the pseudo snapshots are developed from different methods, which implies different virtual signals coming from different sources. Therefore, the correlation between the virtual co-array signals can be reduced; and the proposed method direct estimates the DOAs of signals without spatial smoothing when the number of incoming signals is small. The proposed method generates multiple pseudo snapshots, which improves the resolution and accuracy in DOA estimation with co-prime array, and can be used for radar, sonar, communication systems, etc. Simulation results demonstrate the effectiveness of the proposed method, which outperforms the conventional BNC, FLOM, and PFLOM with smaller root means square error (RMSE) in terms of the generalized signal to noise ratio (GSNR) and factor of impulsive noise.

Performance improvements by dynamic amplifier reconfigurations for C + L-band optical networks in the presence of stimulated Raman scattering

C+L-band optical networks suffer from transmission impairments including amplified spontaneous emission noise and Kerr-induced and stimulated-Raman-scattering-induced nonlinear interference. The generalized signal-to-noise ratio (GSNR) profiles across different links of the network play a crucial role in determining the overall network traffic load and operating margins, but the GSNR is highly dependent on the signal power profiles along the link and time-varying channel-loading conditions, particularly for future dynamic mesh networks. As the network is dynamically loaded, we monitor the GSNR across the active channels of each link and investigate the benefits of regularly reconfiguring the erbium-doped fiber amplifier (EDFA) gain and tilts to optimize either (1) the mean GSNR minus the GSNR standard deviation (i.e., to achieve a high and flat GSNR across channels) or (2) the minimum GSNR to improve the traffic load and/or GSNR margins. Different lightpath provisioning strategies are studied, and simulation results for COST 239, a German network, and a sample Alibaba production network show that, compared with static EDFA configurations assuming fully loaded conditions, the GSNR margin and hence network robustness with dynamic EDFA reconfigurations can be improved by at least 0.45, 0.57, and 0.72 dB, respectively, with similar or better network blocking probability. Alternatively, the improved GSNR profiles can enable an average network capacity improvement of at least 9%, 8%, and 16%, respectively, in principle when operating at the same GSNR margin at 30% spectral occupancy.

Employing channel probing to derive end-of-life service margins for optical spectrum services

Optical spectrum as a service (OSaaS) spanning over multiple transparent optical network domains can significantly reduce the investment and operational costs of the end-to-end service. Based on the black-link approach, these services are empowered by reconfigurable transceivers and the emerging disaggregation trend in optical transport networks. This work investigates the accuracy aspects of the channel probing method used in generalized signal-to-noise ratio (GSNR)-based OSaaS characterization in terrestrial brownfield systems. OSaaS service margins to accommodate impacts from enabling neighboring channels and end-of-life channel loads are experimentally derived in a systematic lab study carried out in the Open Ireland testbed. The applicability of the lab-derived margins is then verified in the HEAnet production network using a 400 GHz wide OSaaS. Finally, the probing accuracy is tested by depleting the GSNR margin through power adjustments utilizing the same 400 GHz OSaaS in the HEAnet live network. A minimum of 0.92 dB and 1.46 dB of service margin allocation is recommended to accommodate the impacts of enabling neighboring channels and end-of-life channel loads. A further 0.6 dB of GSNR margin should be allocated to compensate for probing inaccuracies.

Deep Neural Network-Based QoT Estimation for SMF and FMF Links

Quality of transmission (QoT) estimation tools for fiber links are the enabler for the deployment of reconfigurable optical networks. To dynamically set up lightpaths based on traffic request, a centralized controller must base decisions on reliable performance predictions. QoT estimation methods can be categorised in three classes: exact analytical models which provide accurate results with heavy computations, approximate formulas that require less computations but deliver a reduced accuracy, and machine learning (ML)-based methods which potentially have high accuracy with low complexity. To operate an optical network in real-time, beside accurate QoT estimation, the speed in delivering results is a strict requirement. Based on this, only the last two categories are candidates for this application. In this paper, we present a deep neural network (DNN) structure for QoT estimation considering both regular single-mode fiber (SMF) and future few-mode fiber (FMF) proposed to increase the overall network capacity. We comprehensively explore ML-based regression methods for estimating generalized signal-to-noise ratio (GSNR) in partial-load SMF and FMF links. Synthetic datasets have been generated using the enhanced Gaussian noise (EGN) model. Results indicate that the proposed DNN-based regressor can provide better accuracy along with less computation complexity, compared with other state-of-the-art ML methods as well as closed-form-EGN and closed-form-GN models.

Efficient Three-Step Amplifier Configuration Algorithm for Dynamic C+L-Band Links in Presence of Stimulated Raman Scattering

We propose an efficient three-step amplifier configuration algorithm to improve and maintain the generalized signal-to-noise ratio (GSNR) performance of dynamic C+L-band links in presence of amplifier spontaneous emission (ASE) noise, Kerr nonlinearity and stimulated Raman scattering (SRS) using erbium-doped fiber amplifiers (EDFA). In step 1, we derive sub-optimal signal power profiles at the beginning of each span using the local optimization global optimization (LOGO) strategy [1] that takes into account ASE noise and Kerr nonlinearity. The effect of SRS, which is the dominant impairment in wideband transmissions, is compensated through amplifier gain pre-tilt of each span in step 2. In step 3, we use the closed-form inter-channel stimulated Raman scattering Gaussian noise (ISRSGN) model [2] to compute the GSNR of a multi-span link as a function of the inline amplifier gains and tilts and propose an iterative gradient ascent optimization algorithm to maximize the mean GSNR and minimize its standard deviation across the C- or L-band. Simulation studies for homogeneous-/heterogeneous-span links, static full-channel loading and dynamic loading due to gradual channel additions or sudden fiber cuts show that compared with other GSNR profile optimization methods, the proposed algorithm can achieve similar GSNR performance but only require execution time in the order of seconds. The effectiveness of the proposed algorithm in restoring/improving GSNR performance for an 8-span C+L-band link is experimentally verified for a scenario imitating a sudden fiber cut induced-channel disruptions. We further show that different optimization functions such as maximizing the minimum GSNR across the C+L band can be used in practical settings, and thus the proposed algorithmic framework is adaptable to different network control and management objectives.

Raman Pump Optimization for Maximizing Capacity of C+L Optical Transmission Systems

Within a context of C+L band transmission, this work proposes a design approach for Raman pumps in hybrid fiber amplifiers (HFAs) with the goal of maximizing the total system capacity. First, the optimization problem is constructed. The capacity of a system can be equivalently assessed by the mean generalized signal-to-noise ratio (GSNR) of all channels, which is chosen as the optimization objective. The powers of Raman pumps are chosen as the decision variable, and constraints on the Raman pump powers are added. Then, an optimization framework for Raman pump powers is proposed. An artificial neural network (ANN) is used to establish a differentiable model for GSNR and signal power after Raman amplification (RA). The gradient descent algorithm is adopted to perform the optimization. Simulations are conducted on a single-span link for modeling and on an 8-span link for optimization. Results show that the ANN can reach a high modeling accuracy with a prediction error of 0.16 dB for GSNR and 0.06 dB for signal power with specific link and signal parameters. Results also demonstrate the superior performance of the proposed optimization framework, and a gain of 0.54 dB on the mean GSNR can be achieved by the designed HFA compared with that of the design aimed at a flat gain profile.

Super-resolution time delay estimation using exponential kernel correlation in impulsive noise and multipath environments

The time delay estimation (TDE) is an important and fundamental part in wireless communication signal processing, which has been widely used in localization, intelligent navigation, and radio monitoring. However, it remains a challenging task to support such TDE mechanisms with super-resolution in impulsive noise and multipath environments. Aiming at improving the TDE performance in both impulsive noise and multipath environments, in this paper, a novel super-resolution TDE method is proposed by using the defined exponential kernel correlation (EKC) function. In the proposed TDE method, EKC function is defined to suppress the non-Gaussian impulsive noise. Specifically, the theoretical analysis is also given to prove its boundedness and its effectiveness in non-Gaussian impulsive noise suppression. Further, a novel super-resolution TDE method is proposed by combining EKC function and multiple signal classification (MUSIC) method, which can significantly improve the estimation performance in terms of resolution and accuracy with both non-Gaussian impulsive noise and multipath environments. Finally, simulation experiments demonstrate that the proposed super-resolution TDE method maintains good performance even with a low characteristic exponent and low generalized signal-to-noise ratio (GSNR).

A Study of Power Efficiency Maximization in Submarine Cable Systems

Power efficiency is a key concept in submarine cable transmission systems because of fixed and limited electrical power supplies. Electrical power is provided by a DC voltage applied across the cable from power feed equipment (PFE) located at the terminal ends and is used to power all optical amplifiers throughout the entire link which may be many thousands of kilometers. We examine here via system modeling aspects of power efficiency in the context of maximizing this quantity, primarily with respect to the optimal generalized signal-to-noise ratio (GSNR). We explore the dependence on capacity metric, span loss, link length, and fiber attenuation. We also compare the optimal GSNR predicted using pump power as a measure of power consumption in a single fiber with total cable capacity predicted by application of a pump sharing model and the resulting electrical-to-optical conversion efficiency predicted as a function of amplifier output power, span loss, and repeater power available. We find that the cable capacity measure predicts somewhat higher optimal GSNR than the single fiber measure based on amplifier pump power. The optimal GSNR decreases with longer link length and optimal span loss is somewhat higher than previous results based on different analyses.

Heuristic-based optimization framework for customizable design of long-haul data center interconnect networks

With the widespread deployment of data centers, Internet service providers are expecting more efficient design strategies to build long-haul data center interconnect (DCI) networks. In this paper, we propose a heuristic-based optimization framework to design these networks. Through this framework, network designers can obtain a site-type design scheme that arranges customized site types such as in-line amplifiers, dynamic gain equalizers, optical terminal multiplexers, and electrical regenerators, and three strategies are provided for reference. Taking the quality of transmission as the main metric, and the overall cost of the network as the ancillary measurement, we compare the schemes obtained by the proposed framework against the baseline scheme obtained by a traditional periodic design strategy. Simulations are conducted on a topology of the Tencent DCI network. Under the condition that all schemes ensure that the minimum general signal-to-noise ratio (GSNR) remains above the given GSNR threshold, the schemes designed by our framework can achieve overall cost savings up to 25.73%.

Simple amplifier configuration algorithm for dynamic C + L-band systems in the presence of stimulated Raman scattering

We propose a simple two-step amplifier configuration algorithm based on signal power across different channels to improve the generalized signal-to-noise ratio (GSNR) performance of dynamic C + L-band links in the presence of amplifier spontaneous emission (ASE) noise, Kerr nonlinearity, and stimulated Raman scattering (SRS) using erbium-doped fiber amplifiers (EDFA). In step 1, ASE noise and Kerr nonlinearity are taken into account to derive sub-optimal signal power profiles at the beginning of each span using the local optimization global optimization (LOGO) strategy. The effect of SRS is compensated through amplifier gain pre-tilt in step 2. Simulations for links with homogeneous/heterogeneous spans, static full-channel loading, and dynamic loading due to gradual channel additions for C + L-band upgrades show that the proposed algorithm can achieve similar GSNR performance, but requires much less execution time, compared to other iterative methods that target for improving the GSNR across the C + L band, thus making it a fast and efficient GSNR management strategy for future dynamic C + L-band networks.

Research on geometric algebra-based robust adaptive filtering algorithms in wireless communication systems

Noise and interference are the two most common and basic problems in wireless communication systems. The noise in wireless communication channels has the characteristics of randomness and impulsivity, so the performance of adaptive filtering algorithms based on geometric algebra (GA) and second-order statistics is greatly reduced in the wireless communication systems. In order to improve the performance of adaptive filtering algorithms in wireless communication systems, this paper proposes two novel GA-based adaptive filtering algorithms, which are deduced from the robust algorithms based on the minimum error entropy (MEE) criterion and the joint criterion (MSEMEE) of the MEE and the mean square error (MSE) with the help of GA theory. The noise interference in wireless communication is modeled by alpha-stable distribution which is in good agreement with the actual data in this paper. Simulation results show that for the mean square deviation (MSD) learning curve, the GA-based MEE (GA-MEE) algorithm has faster convergence rate and better steady-state accuracy compared to the GA-based maximum correntropy criterion algorithm (GA-MCC) under the same generalized signal-to-noise ratio (GSNR). The GA-MEE algorithm reduces the convergence rate, but improves the steady-state accuracy by 10–15 dB compared to the adaptive filtering algorithms based on GA and second-order statistics. For GA-based MSEMEE (GA-MSEMEE) algorithm, when GA-MSEMEE and the adaptive filtering algorithms based on GA and second-order statistics keep the same convergence rate, its steady-state accuracy is improved by 10–15 dB, and when GA-MSEMEE and GA-MEE maintain approximately steady-state accuracy, its convergence rate is improved by nearly 100 iterations. In addition, when the algorithms are applied to noise cancellation, the average recovery error of the two proposed algorithms is 7 points lower than that of other GA-based adaptive filtering algorithms. The results validate the effectiveness and superiority of the GA-MEE and GA-MSEMEE algorithms in the alpha-stable noise environment, providing new methods to deal with multi-channel interference in wireless networks.

Experimental validation of GNPy in a multi-vendor flex-grid flex-rate WDM optical transport scenario

We experimentally test the accuracy of a quality of transmission estimator (QoT-E) within a laboratory flex-grid flex-rate framework, considering eight multi-vendor transceivers (TRXs) with symbol rates ranging from 33 to 69 Gbaud, and variable constellations [quadrature phase shift keying, 8-quadrature amplitude modulation (QAM), and 16-QAM probabilistic constellation shaping], for data rates of 100 Gbits/s up to 300 Gbits/s, and a flex-grid wavelength division multiplexed (WDM) spectrum, with channel spacings of 50 and 75 GHz. As a QoT-E, we utilize an enhanced implementation of the open-source GNPy project. We demonstrate that this QoT-E provides a high level of accuracy in generalized signal-to-noise ratio (GSNR) computation, with an average error value not exceeding 0.5 dB, for the scenario under investigation. These values are computed with respect to the measured bit-error ratio converted to the GSNR using the TRX model obtained via back-to-back characterization. These results demonstrate that the optimal management of flex-grid flex-rate WDM optical transport arises by managing power spectral densities instead of power per channel, as in traditional fixed-grid systems.

Cross-feature trained machine learning models for QoT-estimation in optical networks

The ever-increasing demand for global internet traffic, together with evolving concepts of software-defined networks and elastic-optical-networks, demand not only the total capacity utilization of underlying infrastructure but also a dynamic, flexible, and transparent optical network. In general, worst-case assumptions are utilized to calculate the quality of transmission (QoT) with provisioning of high-margin requirements. Thus, precise estimation of the QoT for the lightpath (LP) establishment is crucial for reducing the provisioning margins. We propose and compare several data-driven machine learning (ML) models to make an accurate calculation of the QoT before the actual establishment of the LP in an unseen network. The proposed models are trained on the data acquired from an already established LP of a completely different network. The metric considered to evaluate the QoT of the LP is the generalized signal-to-noise ratio (GSNR), which accumulates the impact of both nonlinear interference and amplified spontaneous emission noise. The dataset is generated synthetically using a well-tested GNPy simulation tool. Promising results are achieved, showing that the proposed neural network considerably minimizes the GSNR uncertainty and, consequently, the provisioning margin. Furthermore, we also analyze the impact of cross-features and relevant features training on the proposed ML models’ performance.

DOA Estimation for Coprime Array With Mixed Noise Scenario via Phased Fractional Low-Order Moment

Direction of arrival (DOA) estimation with coprime array is an important topic in array signal processing. In this letter, the coprime array technique is extended to mixed noise (white Gaussian noise and impulsive noise). Since the second order covariance matrix does not exist in presence of mixed noise, the phased fractional low-order moment (PFLOM) is used to replace the data covariance matrix of the received signals from coprime array, and then a DOA estimation approach based on toeplitz method with coprime array is applied. The feasibility of PFLOM with mixed noise scenario of coprime array is theoretically derived, and the simulation results show that the proposed method has better estimation performance than that of the existing methods, especially in the low generalized signal-to-noise ratio (GSNR) environment.