Next Article in Journal
Boundary Integral Equations Approach for a Scattering Problem of a TE-Wave on a Graphene-Coated Slab
Previous Article in Journal
Enhancing Optoelectronic Performance Through Rare-Earth-Doped ZnO: Insights and Applications
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Time Parameter Optimization for the Semiconductor Laser-Based Time-Delay Reservoir Computing System

1
School of Mathematics and Information Science & Technology, Hebei Normal University of Science and Technology, Qinhuangdao 066004, China
2
Hebei Innovation Center for Smart Perception and Applied Technology of Agricultural Data, Qinhuangdao 066004, China
3
College of Computer Science, Nankai University, Tianjin 300071, China
4
Artificial Intelligence and Big Data College, Chongqing Polytechnic University of Electronic Technology, Chongqing 401331, China
*
Author to whom correspondence should be addressed.
Photonics 2025, 12(5), 455; https://doi.org/10.3390/photonics12050455
Submission received: 2 April 2025 / Revised: 4 May 2025 / Accepted: 7 May 2025 / Published: 8 May 2025

Abstract

Time-delay reservoir computing (RC) systems, particularly those based on semiconductor lasers (SLs), have gained attention due to their low energy consumption, high response rates, and rich nonlinear dynamics. This work investigates the influence of key time parameters—virtual node interval (θ), delay feedback (τ), and data injection period (T) on the performance of SL-based time-delay RC systems. Using the Santa Fe time series prediction task and memory capacity evaluation task, we analyze how these parameters affect prediction accuracy and memory capability. The results reveal that θ = 0.2Tro (where Tro is the relaxation oscillation period of the SLs) optimizes prediction performance, while θ = 0.5Tro maximizes memory capacity. Additionally, feedback delay τ significantly impacts system performance. Shorter τ values (e.g., τ = 0.54T) enhance prediction accuracy, whereas longer τ values (e.g., τ = 1.74T) improve memory capacity. These findings provide valuable insights for optimizing time-delay RC systems, enabling better task-specific performance and stability.

1. Introduction

Reservoir computing (RC) is a neural-inspired computational framework designed to address the challenges associated with training recurrent neural networks (RNNs). In 2001, Jaeger et al. and Maass et al. proposed the Echo State Network (ESN) [1] and the Liquid State Machine (LSM) [2], respectively, both of which share the common feature of employing a training-free network to simplify the training process. In 2007, Verstraeten et al. unified these approaches under the term RC [3]. Both RNNs and RC employ a recursive structure and nonlinear activation functions, which not only grant them memory capabilities but also enhance their ability to learn and model complex nonlinear systems. However, RNNs typically update all network weights using backpropagation algorithms, which often lead to issues such as gradient vanishing or gradient explosion. In contrast, RC processes information through a sparsely connected network with fixed weights, known as the reservoir [4,5]. Notably, in RC, only the output weights require training, while the input weights and internal reservoir weights remain unchanged. As a novel paradigm of artificial neural networks, RC has demonstrated exceptional performance in various tasks, including time series prediction, nonlinear channel equalization, handwritten digit recognition, and speech recognition [6]. Moreover, RC has garnered significant attention due to its inherent advantages, such as low energy consumption, high computational speed, and ease of hardware implementation [7,8].
In 2011, Appeltant et al. pioneered a time-delay RC system using a Mackey–Glass circuit [9]. This approach utilized a single physical node combined with a feedback loop, significantly simplifying the hardware implementation of RC. Since then, extensive research has been conducted on time-delay RC, leading to the development of RC systems based on electronic circuits, optoelectronic systems, and all-optical systems [10,11,12,13,14]. For instance, in 2012, Larger et al. proposed an optoelectronic time-delay RC system [15], which demonstrated excellent performance in tasks such as digital speech recognition and chaotic time series prediction. Similarly, Duport et al. developed an all-optical RC system by employing a semiconductor optical amplifier (SOA) as the nonlinear node of the reservoir [16]. Compared to electronic and optoelectronic devices, semiconductor lasers (SLs) have emerged as a superior choice for constructing time-delay RC systems due to their low energy consumption, high response rates, and rich nonlinear dynamics. In 2013, Brunner et al. implemented the first all-optical RC system using an SL as the nonlinear element and evaluated its performance through tasks such as time series prediction and speech recognition. The results demonstrated that the system could achieve high-speed operation with excellent performance [17]. In 2015, Romain et al. constructed a time-delay RC system based on a semiconductor ring laser (SRL) capable of handling two tasks simultaneously, significantly improving the computational efficiency of RC [18]. In 2018, Hou et al. proposed a time-delay optical RC system using an SL with dual optical feedback loops. This system exhibited enhanced prediction performance due to the increased complexity of nonlinear states compared to single-loop configurations [19]. In 2020, Guo et al. introduced an electrically modulated RC system based on a semiconductor nanolaser (SNL), which further improved the information processing rate of optical RC systems [20]. More recently, extensive research has focused on time-delay RC systems based on SLs, covering laser type selection, system architecture optimization, key parameter tuning, and task-specific applications [21,22,23,24,25]. For instance, in 2025, Li et al. successfully constructed a time-delayed RC system to process image recognition task based on mutually coupled multimode SLs [26]. By leveraging the multi-mode characteristics of SLs, they effectively reduced the feedback loop length and therefore enhanced the information processing rate.
Previous studies have shown that the performance of SL-based time-delay RC systems is influenced by several factors. Firstly, it depends on the system architecture and the type of SL used [27,28,29]. Secondly, it is significantly affected by parameters such as feedback strength, optical injection strength, and several time-related parameters [30,31,32,33,34,35,36]. These time parameters include the interval between virtual nodes (θ), the delay feedback (τ), and the data injection period (T), all of which play a crucial role in determining the performance of the RC system. Currently, the optimal values for these parameters are primarily determined through empirical testing, which makes the tuning process of SL-based time-delay RC systems highly complex and time-consuming.
In this work, we investigate the relationship between these time parameters in an SL-based time-delay RC system under different types of tasks. By employing the Santa Fe time series prediction task and the memory capacity evaluation task, we provide a detailed analysis of how these time parameters influence the prediction performance and memory capabilities of the system. Our findings offer valuable insights into optimizing the configuration of time parameters in time-delay RC systems, thereby enhancing their computational performance and stability.

2. System Model

Figure 1 illustrates the structure diagram of the SL-based time-delay RC system. In the system, the drive-SL (D-SL) is used as a light source, and the input signal converts the amplitude variation in the electrical signal into the phase variation in the optical signal through the phase modulator (PM). The optical signal, carrying the input information, is injected into the Response-SL (R-SL), where a variable attenuator (VA1) adjusts the injection intensity. R-SL acts as the physical node of the reservoir layer, receiving signals from both the Drive-SL and its own feedback loop to generate a nonlinear response. The optical circulator (OC), along with a second variable attenuator (VA2) and a fiber coupler (FC1), forms a feedback loop, where VA2 regulates the feedback intensity. Part of the output of FC1 is fed back to R-SL, and the other part is collected in time as the state of the node in the reservoir.
According to the information processing flow, this RC system can be divided into three main parts: the input layer, the reservoir layer, and the output layer, as illustrated in Figure 2. In the input layer, the information to be processed first undergoes mask preprocessing. Since the input data are typically in discrete form, denoted as u(n), they needs to be converted into a continuous signal u(t), where each data point remains constant over time T. The time parameter T usually ranges from a few nanoseconds to tens of nanoseconds, determining the data input period. Within a period T, the response state of the SL is sampled n times to obtain n reservoir states. It is noteworthy that the relaxation oscillation period of the SL is generally less than 1 nanosecond. Therefore, if u(t) is directly injected into the SL within each period T, the SL output will quickly reach a steady state after a brief transient response. This leads to a homogenization of subsequently collected n reservoir states, which is unfavorable for effective feature extraction from the data. To achieve richer dynamic responses, the input layer performs mask processing on u(t). The data input period T is divided into n intervals, each of length θ (θ = T/n). Within each interval, a random weight is assigned to construct the input weight signal, known as the mask. This process is known as mask operation, where u(t) is multiplied by the mask signal to produce the final input signal j(t). The parameter θ significantly impacts the diversity of reservoir states. If θ is too large, it leads to homogenization of reservoir states, On the other hand, if θ is too small, it may result in an insufficient laser response and the loss of input information features.
Furthermore, to enhance the system’s nonlinear characteristics and endow the reservoir with memory capabilities, the SL is subjected to delay feedback with a delay feedback of τ. Although previous studies have conventionally set the time parameter τ to a value comparable to the data injection period T, theoretically it can be set to any arbitrary value if the impact of τ on the system’s memory capacity and nonlinear characteristics is not taken into account.
In the output layer, the discrete reservoir states x(n) are obtained by collecting the output light intensity of the SL, which are then used to train the readout weights Wout. The Wout can be solved using the ridge regression method. After training, the output of the reservoir y(n) is obtained by multiplying x(n) with Wout.
In this RC system, the nonlinear dynamics of the R-SL under optical injection and time-delay feedback can be expressed as [37]:
d E r ( t ) d t = 1 + i α 2 G N ( N r ( t ) N 0 ) 1 + ε | E r ( t ) | 2 1 τ p E r ( t ) + κ i n j E d ( t ) exp ( i ω t ) + κ E r ( t τ ) exp ( i ω r τ )
d N r ( t ) d t = J r N r ( t ) τ s G N ( N r ( t ) N 0 ) 1 + ε | E r ( t ) | 2 | E r ( t ) | 2
where Nr is the carrier density of the response SL, α is the linewidth enhancement factor, GN is the gain coefficient, N0 is the transparency carrier density, ε is the saturation coefficient, and τp and τs are the photon and carrier lifetimes, respectively. κinj is the optical injection strength, κ is the feedback strength, ωd, and ωr, are the optical angular frequencies of the D-SL and R-SL, respectively, Δω is the angular frequency detuning (Δω = 2πΔf = 2π(fdfr)), and τ is the delay time of the feedback loop. Jr is the injection current of the response laser, defined as Jr = j × Jth, where Jth is the threshold current and j is the normalized pump current. Ed and Er are the slowly varying electric fields of the D-SL and R-SL, respectively. Ed can be expressed as:
E d ( t ) = I d exp ( i π j ( t ) )
where Id is the light intensity of the D-SL, and j(t) is the information injected into the reservoir.
The delay differential equations are solved using the fourth-order Runge–Kutta method with an integration step of 2ps. The parameter values used are as follows: α = 3.0, N0 = 1.4 × 1024, GN = 8.4 × 10−13, ε = 2.0 × 10−23, τp = 1.927 × 10−12, τs = 2.04 × 10−9, ωd = 1.23 × 1015, Δf = −4 × 109.

3. Results and Discussion

3.1. Santa Fe Time Series Prediction

We first investigate the optimization method for the time parameters of the RC system using the Santa Fe time series prediction task. The Santa Fe dataset consists of a chaotic time series with 9000 sample points. The objective of this task is to predict the next data point based on historical data. The prediction performance is evaluated using the normalized mean square error (NMSE) which can be expressed as:
N M S E = 1 L n = 1 L [ y ( n ) y d ( n ) ] 2 / var ( y d )
In Equation (4), L is the length of the test data, y(n) is the system output, yd is the desired output, and var denotes the variance. It is generally considered that the RC system exhibits good prediction performance when NMSE ≤ 0.1. In this work, 2000 data points are used as the training set and another 1000 data points are used as the testing set.
Among the three time parameters T, θ, and τ, T is denoted as T = n × θ, where n represents the number of virtual nodes. In this work, we set n = 50. Therefore, our focus is on discussing the impact of θ and τ on the system’s performance.
In previous studies, θ has often been measured by the relaxation oscillation period Tro of the SL under free-running. It is suggested that θ should take values within the range of [0.1Tro, Tro], where Tro can be expressed as:
T r o = 2 π G N J t h ( j 1 )
In Equation (5), Jth = Nth/τs, where Nth is the threshold carrier density (Nth = N0 + 1/(GN τp))). Throughout the testing, j is set to 1.05. Therefore, we know that the relaxation oscillation period Tro is 0.97 ns.
Next, we discuss the impact of θ on the system’s prediction performance. As θ changes, both the injection strength κinj and feedback strength κ affect the system’s prediction performance, we analyze the evolution of NMSE in the κinj-κ parameter space for different θ values.
Figure 3 illustrates the evolution maps of the prediction error NMSE in κinj-κ parameter space as θ varies from 0.1Tro to 0.6Tro under τ = T. Different colors in the figure indicate different NMSE values. To enhance color recognition, only regions with good prediction performance (NMSE < 0.1) are displayed. The white dashed lines in the figure outline areas where NMSE ≤ 0.02. In the six figures, the regions with good prediction performance are predominantly located in the lower-right portion of the scanning interval. This indicates that a certain injection strength κinj is required to achieve good prediction performance, ensuring that the SL responds sufficiently to input signals. Additionally, under a given injection strength, the feedback strength k should not be excessively high, as this could induce system oscillations or even drive the system into a chaotic state. From Figure 3a–f, it is evident that as θ increases, the regions with good prediction performance gradually shrink. The minimum NMSE values (NMSEmin) in the six figures are 0.011, 0.009, 0.009, 0.011, 0.017, and 0.016, respectively. Notably, when θ is 0.2Tro and 0.3Tro, the system achieves the lowest prediction error of 0.009 under τ = T.
Figure 4 illustrates the minimum prediction error NMSEmin obtained by scanning the κinj-κ parameter space as θ increases from 0.1Tro to Tro. The subplot in the figure shows a magnified view as θ increases from 0.1Tro to 0.8Tro. The NMSEmin values in Figure 4 is defined as the minimum NMSE obtained in evaluation map of kinj-k parameter space. For instance, at θ = 0.1Tro we scanned a parameter space of 0 ns−1k ≤ 5 ns−1 and 0 ns−1kinj ≤ 10 ns−1 (shown in Figure 3a). Within this space, the minimum NMSE is 0.011, which is defined as the NMSEmin for θ = 0.1Tro. It can be observed that as θ increases, NMSEmin first decreases and then increases. The minimum value of 0.009 is achieved at 0.2Tro and 0.3Tro. When θ exceeds 0.8Tro, the system fails to achieve good prediction performance (NMSE > 0.1), indicating that larger θ values lead to reduced prediction accuracy. It is evident that determining θ requires a comprehensive consideration of the results from Figure 3 and Figure 4. On one hand, a larger interval in the evolution map that yields good prediction performance indicates relatively simpler parameter adjustment in hardware implementation. On the other hand, a smaller NMSEmin signifies a stronger capability of the system to make accurate predictions. Based on the regions of good prediction performance and the NMSEmin values, we conclude that θ = 0.2Tro is the optimal choice for maximizing prediction accuracy in SL-based time-delay RC systems.
Next, we investigate the impact of delay feedback τ on the prediction performance, here we set θ = 0.2Tro, κinj = 4.2 ns−1, κ = 0.8 ns−1. In previous studies, τ was conventionally set either synchronized with T (τ = T) or not synchronized with T (τT, with a tendency for τ = Tθ), but there was no in-depth analysis of more specific scenarios. Figure 5 depicts the variation in NMSE as τ increases from 0.1T to 2T. Throughout the change in τ, the NMSE values remain below 0.04, fluctuating mainly around 0.02. When τ is 0.4T, 0.54T, 0.8T, 0.9T, 0.98T, and T, NMSE is below 0.01, and the minimum NMSE of 0.0055 is achieved at 0.54T. When τ > T, it is almost impossible to achieve a prediction error of NMSE < 0.01. It is evident that when adopting the non-synchronous scheme, apart from the traditional setting of τ = T−θ (τ = 0.98T), there are more options available. Particularly, setting τ equal to 0.54T is more conducive to achieving relatively fewer prediction errors.

3.2. Memory Capacity Evaluation

Subsequently, we investigate the influence of time parameters on the system’s memory capacity based on the memory capacity evaluation task. This task assesses the system’s ability to store and recall the previously input information. In this task, the data to be processed is a set of random numbers uniformly distributed within the interval of [−0.5, 0.5], and the system is required to output the data from j steps earlier. The memory performance of the system is evaluated using MC, which can be expressed as follows [38,39]:
M C = j = 0 + m ( j )
In Equation (6), m(j) represents the correlation coefficient between the system output sequence and the expected values when the delay step is j. m(j) can be expressed as:
m ( j ) = < [ v ( i j ) < v ( i j ) > ] [ y j ( i ) < y j ( i ) > ] ) < | v ( i j ) < v ( i j ) > | 2 > < | y j ( i ) < y j ( i ) > | 2 >
where i represents the current input step, y is the estimated value, and v is the target value. The symbol <.> denotes the time average. In this test, we use 2000 data points for training and 1000 data points for testing.
Figure 6 illustrates the evolution of MC within the parameter space of injection strength κinj and feedback strength κ under different θ. Different colors in the figure represent varying MC values, with the black dashed lines outlining the regions where MC ≥ 10. The parameter intervals displayed here are 15 ns−1κinj ≤ 40 ns−1 and 4 ns−1κ ≤ 18 ns−1. Compared to the Santa Fe time series prediction task, achieving a high MC requires greater injection strength and feedback strength. We hypothesize that higher injection strengths induce more significant response states in the SL, which persist across multiple feedback cycles but gradually diminish over time. The regions outlined with high MC values (MC ≥ 10) are located at the edge of chaos, indicating that, under certain injection strengths, larger feedback can enhance the system’s memory capacity, but excessive feedback that causes system oscillations can reduce memory capacity. From Figure 6a–f we can see that as the θ increases from 0.1Tro to 0.6Tro, the maximum MC values (MCmax) obtained in each evolution map are 13, 11, 12, 12, 16, and 13, respectively. It is evident that the system’s maximum MC of 16 occurs in Figure 6e at θ = 0.5Tro under τ = T. Additionally, we can observe that when θ = 0.5Tro, the region with MC ≥ 10 is more extensive, which is beneficial for parameter adjustment in hardware implementation.
Figure 7 depicts the variation in MCmax as θ increases from 0.1Tro to Tro, where each value represents the maximum MC obtained by scanning the κinj-κ parameter space. When 0.1Troθ ≤ 0.4Tro, MCmax fluctuates around 12. When θ = 0.5Tro, MCmax reaches its peak value of 16. When θ > 0.5Tro, MCmax gradually decreases to around 11. Combining the insights from Figure 6 and Figure 7, we find that θ = 0.5Tro is optimal for achieving high memory capacity, as it provides a balance between parameter tuning simplicity and system performance. We find that 0.5Tro represents an intermediate phase of the system’s relaxation oscillations, during which the system state is characterized by a combination of nonlinearity, rapid dynamics, energy release, and transient instability. Compared to the early relaxation oscillation phase at 0.2Tro, the system exhibits a more comprehensive response and tighter coupling between input signals. We hypothesize that this is the reason for the emergence of the MC peak at 0.5Tro.
Next, we investigate the influence of τ on memory capacity. Here, we set the parameters as θ = 0.5Tro, κinj = 19 ns−1, κ = 16 ns−1. Figure 8 illustrates the variation in MC as τ increases from 0.1T to 4T. When τ < 1.74T, MC fluctuates with an upward trend, and MC is 16.5 at τ = T. When τ = 1.74T, MC reaches its maximum value of 21.7. When 1.74T < τ≤ 3.72T, MC oscillates between 14 and 21, and the value of 21 are observed at τ = 2.48T, 2.92T, and 3.6T. When τ > 3.72T, MC oscillates and decreases to around 13. Comparing Figure 5 and Figure 8, it can be observed that in the Santa Fe time series prediction task, NMSE fluctuates around 0.02 as τ increases from 0.1T to 2T, with NMSE below 0.01 occurring at several specific points where τT. In contrast, in the memory capacity test task, a relatively large τ tends to favor higher MC values, but MC decreases when τ > 3.72T. We can infer that a relatively large delay feedback is conducive to achieve high memory capacity, as the current data injection is superimposed with the reservoir state from several cycles earlier and injected into the reservoir together. As τ becomes excessively large, the system’s memory capacity gradually declines, likely due to the reservoir’s struggle to forge a linkage between the current input and the reservoir state from a too distantly delayed cycle. The difference in the trend of the curves in Figure 5 and Figure 8 can enlighten us that when dealing with different tasks, it is necessary to adjust the delay feedback according to the task’s requirements for memory capacity and nonlinear transformation capability.

3.3. Discussion

In the above tests, the optimal time parameters θ and τ for the Santa Fe time series prediction task and the memory capacity evaluation task were identified. We recognize that during hardware implementation, certain parameters of the SL such as the linewidth enhancement factor (α) and the gain coefficient (GN), may deviate from their nominal values. Therefore, the robustness of the identified optimal parameters needs further discussion.
Figure 9a shows the NMSE variation with respect to α and GN for Santa Fe prediction task under κinj = 4.2 ns−1, κ = 0.8 ns−1, θ = 0.2Tro, τ = 0.54T. It shows that the NMSE remains below 0.01 when α varies within −2% to +3% of its nominal value, and when GN fluctuates within ±6% of its nominal value. Figure 9b shows the MC variation versus α and GN for memory evaluation task under κinj =19 ns−1, κ =16 ns−1, θ = 0.5Tro, τ = 1.74T. Compared to Figure 9a, the MC exhibits significantly stronger robustness against parameter drift in both α and GN. Specifically, the MC maintains values above 16 when α varies from −10% to +6% or GN varies from −5% to +8% of their nominal values. These findings indicate that the RC system’s performance can tolerate moderate parameter variations during hardware implementation, which is advantageous for maintaining stable long-term operation.
It should be noted that this work does not directly address the influence of injection strength (κinj) and feedback strength (κ) on the time parameters θ and τ. However, previous studies have demonstrated that a higher injection strength can enhance the reservoir’s response bandwidth [40], which facilitates a reduction in the virtual node interval and consequently improves the information processing rate while maintaining comparable reservoir performance. In this study, our investigation consistently focused on optimizing reservoir performance. Therefore, in Figure 3 and Figure 6, when evaluating system performance across the κinj-κ parameter space for different θ values, we selected θ based on configurations yielding optimal performance. Under this research framework, we observe that although the input data from both Santa Fe prediction task and MC evaluation task were normalized before being injected into the system, the parameter ranges (kinj, k) for achieving good performance are distinct. For prediction tasks, achieving good performance requires relatively small values of κ and κinj (0 ns−1κ ≤ 5 ns−1, 2.5 ns−1κinj ≤ 10 ns−1). In contrast, memory evaluation tasks demand significantly larger κ and κinj to attain a high MC (4 ns−1k ≤ 18 ns−1, 15 ns−1kinj ≤ 40 ns−1). According to previous studies on the nonlinear dynamical states of SLs, increasing feedback strength enriches the nonlinear states of SLs and can even drive them toward chaotic regimes. Numerous studies on SL-based reservoir computing have demonstrated that optimal reservoir performance is typically achieved near the edge of chaos [17]. Furthermore, we recognize that optical feedback is crucial for enabling memory capacity in RC. Comprehensive analysis of these factors reveals that achieving high MC generally requires strong feedback, while maintaining the system at the edge of chaos under high feedback conditions necessitates correspondingly high injection strength.
This work achieves a minimum NMSE of 0.0055 for the Santa Fe time series prediction task under optimized parameters θ = 0.2Tro and τ = 0.54T, while the memory capacity evaluation task reaches its peak performance of 21.7 at θ = 0.5Tro and τ = 1.74T. The obtained results outperform previous approaches using τ = T or τ = T ± θ configurations. Specifically, for the Santa Fe prediction task, our NMSE of 0.0055 represents an improvement over previously reported values of 0.0293 [20], 0.008 [41], and 0.0075 [33]. Similarly, in the memory evaluation task, the achieved MC of 21.7 exceeds the prior results of 10 [42] and 17.6 [33]. The research results demonstrate that adaptively adjusting the virtual node spacing θ according to specific task requirements can enhance RC performance. Furthermore, the feedback delay τ need not be synchronized with the data injection period T. Compared to conventional approaches that limited τ to a few predetermined values, our findings provide new insights for implementing high performance RC systems.

4. Conclusions

This work systematically explores the impact of time parameters of θ, τ, and T on the performance of SL-based time-delay RC systems. For the Santa Fe time series prediction task, we found that θ = 0.2Tro and τ = 0.54T yield the lowest prediction error of NMSE = 0.0055, highlighting the importance of balancing virtual node interval and delay feedback for accurate forecasting. In contrast, for memory capacity evaluation task, θ = 0.5Tro and τ = 1.74T maximize memory performance (MC = 21.7), as longer delay feedback allow the system to retain information over multiple cycles. However, excessively large τ (τ > 3.72T) values degrade memory capacity, likely due to the reservoir’s inability to link current inputs with distant past states.
With the progress in research on SL-based RC, the processing of diverse tasks based on RC has attracted extensive attention. Unlike previous studies that primarily focused on using a single RC system to handle diverse tasks to demonstrate its universality, our findings reveal that different types of tasks actually require different nonlinear transformation characteristics and memory capacities from the system. The results indicate that employing task-specific time parameters (θ and τ) can enhance system performance. Alternatively, one could design systems with tunable feedback delays, adaptively adjusting the feedback delay according to task requirements while employing different θ values at the input layer for data preprocessing to achieve optimal RC performance. In the future, we will conduct in-depth research on time parameter optimization of RC, including a theoretical analysis of the system and its hardware implementation.

Author Contributions

Conceptualization, D.Y. and Y.H.; methodology, D.Y., Q.L. and X.G.; software, K.L.; validation, C.H. and K.L.; formal analysis, C.H.; investigation, D.Y., Q.L., Y.H. and X.G; resources, Q.L. and Y.H.; data curation, C.H. and K.L.; writing—original draft preparation, D.Y. and Q.L.; writing—review and editing, Q.L., Y.H. and X.G.; visualization, Y.H. and X.G.; supervision, D.Y. and C.H.; project administration, Q.L. and D.Y.; funding acquisition, D.Y., K.L. and C.H. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by Natural Science Foundation of Hebei Province, grant number F2022407007; Hebei Innovation Center for Smart Perception and Applied Technology of Agricultural Data, grant number ADIC2023Y007, ADIC2024Y005. Youth Project of Science and Technology Research Program of Chongqing Education Commission of China, grant number KJQN202403114.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Data confirming the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Jaeger, H.; Haas, H. Harnessing nonlinearity: Predicting chaotic systems and saving energy in wireless communication. Science 2004, 304, 78–80. [Google Scholar] [CrossRef] [PubMed]
  2. Maass, W.; Natschläger, T.; Markram, H. Real-time computing without stable states: A new framework for neural computation based on perturbations. Neural Comput. 2002, 14, 2531–2560. [Google Scholar] [CrossRef] [PubMed]
  3. Verstraeten, D.; Schrauwen, B.; D’Haene, M.; Stroobandt, D. An experimental unification of reservoir computing methods. Neural Netw. 2007, 20, 391–403. [Google Scholar] [CrossRef]
  4. Soriano, M.C.; Brunner, D.; Escalona-Morán, M.; Mirasso, C.R.; Fischer, I. Minimal approach to neuro-inspired information processing. Front. Comput. Neurosci. 2015, 6, 68. [Google Scholar] [CrossRef] [PubMed]
  5. Kacper, P.; Ewelina, W.; Dawid, P.; Andrzej, B.; Agnieszka, P.; Vasileios, A.; Zoran, K.; Konrad, S. Molecules, semiconductors, light and information: Towards future sensing and computing paradigms. Coord. Chem. Rev. 2018, 365, 23–40. [Google Scholar]
  6. Gouhei, T.; Toshiyuki, Y.; Jean, B.; Ryosho, N.; Naoki, K.; Seiji, T.; Hidetoshi, N.; Daiju, N.; Akira, H. Recent advances in physical reservoir computing: A review. Neural Netw. 2019, 15, 100–123. [Google Scholar]
  7. Francesco, M.; Cristina, R.; Avishek, N.; Irene, M.; Darko, Z.; Marco, R.; Massimo, T. An overview on application of machine learning techniques in optical networks. IEEE Commun. Surv. Tutor. 2019, 21, 1383–1408. [Google Scholar]
  8. Kitayama, K.; Notomi, M.; Naruse, M.; Inoue, K.; Kawakami, S.; Uchida, A. Novel frontier of photonics for data processing—Photonic accelerator. APL Photonics 2019, 4, 090901. [Google Scholar] [CrossRef]
  9. Appeltant, L.; Soriano, M.C.; Van der Sande, G.; Danckaert, J.; Massar, S.; Dambre, J.; Schrauwen, B.; Mirasso, C.R.; Fischer, I. Information processing using a single dynamical node as complex system. Nat. Commun. 2011, 2, 468. [Google Scholar] [CrossRef]
  10. Cheng, T.Y.; Chou, D.Y.; Liu, C.C.; Chang, Y.J.; Chen, C.C. Optical neural networks based on optical fiber-communication system. Neurocomputing 2019, 364, 239–244. [Google Scholar] [CrossRef]
  11. Argyris, A.; Cantero, J.; Galletero, M.; Pereda, E.; Mirasso, C.R.; Fischer, I.; Soriano, M. Comparison of photonic reservoir computing system for fiber transmission equalization. IEEE J. Sel. Top. Quantum Electron. 2020, 26, 5100309. [Google Scholar] [CrossRef]
  12. Stelzer, F.; Röhm, A.; Lüdge, K.; Yanchuk, S. Performance boost of time-delay reservoir computing by non-resonant clock cycle. Neural Netw. 2020, 124, 158–169. [Google Scholar] [CrossRef] [PubMed]
  13. Ying, D.H.; Yanne, K.C. Classification of IO-modulated signals based on reservoir computing with narrowband optoelectronic oscillators. IEEE J. Quantum Electron. 2021, 57, 5000408. [Google Scholar]
  14. Jieru, S.; Jialin, M.; Chen, L.; Tianyu, W.; Changjin, W.; Hao, Z.; Qingqing, S.; David, W.; Lin, C. Self-powered optoelectronic synaptic device for both static and dynamic reservoir computing. Nano Energy 2025, 134, 110574. [Google Scholar]
  15. Larger, L.; Soriano, M.C.; Brunner, D.; Appeltant, L.; Gutierrez, J.M.; Pesquera, L.; Mirasso, C.R.; Fischer, I. Photonic information processing beyond Turing: An optoelectronic implementation of reservoir computing. Opt. Express 2012, 20, 3241–3249. [Google Scholar] [CrossRef]
  16. Duport, F.; Schneider, B.; Smerieri, A.; Haelterman, M.; Massar, S. All-optical reservoir computing. Opt. Express 2012, 20, 22783–22795. [Google Scholar] [CrossRef]
  17. Brunner, D.; Soriano Miguel, C.; Mirasso Claudio, R.; Fischer, I. Parallel photonic information processing at gigabyte per second data rates using transient states. Nat. Commun. 2013, 4, 1364. [Google Scholar] [CrossRef]
  18. Nguimdo, R.; Verschaffelt, G.; Danckaert, J.; Van der Sande, G. Simultaneous computation of two independent tasks using reservoir computing based on a single photonic nonlinear node with optical feedback. IEEE Trans. Neural Netw. Learn. Syst. 2015, 26, 3301–3307. [Google Scholar] [CrossRef]
  19. Hou, Y.S.; Xia, G.Q.; Yang, W.Y.; Wang, D.; Jayaprasath, E.; Jiang, Z.F.; Hu, C.X.; Wu, Z.M. Prediction performance of reservoir computing system based on a semiconductor laser subject to double optical feedback and optical injection. Opt. Express 2018, 26, 10211–10219. [Google Scholar] [CrossRef]
  20. Guo, X.; Xiang, S.; Zhang, Y.; Lin, L.; Wen, A.; Hao, Y. High-speed neuromorphic reservoir computing based on a semiconductor nanolaser with optical feedback under electrical modulation. IEEE J. Sel. Top. Quantum Electron. 2020, 26, 1500707. [Google Scholar] [CrossRef]
  21. Bogris, A.; Mesaritakis, C.; Deligiannidis, S.; Li, P. Fabry-Perot lasers as enablers for parallel reservoir computing. IEEE J. Sel. Top. Quantum Electron. 2021, 27, 7500307. [Google Scholar] [CrossRef]
  22. Zhong, D.Z.; Yang, H.; Xi, J.T.; Zeng, N.; Xu, Z.; Deng, F.Q. Predictive learning of multi-channel isochronal chaotic synchronization by utilizing parallel optical reservoir computers based on three laterally coupled semiconductor lasers with delay-time feedback. Opt. Express 2021, 29, 5279–5294. [Google Scholar] [CrossRef] [PubMed]
  23. Guo, X.; Zhou, H.; Xiang, S.; Yu, Q.; Zhang, Y.; Han, Y.; Wang, T.; Hao, Y. Short-term prediction for chaotic time series based on photonic reservoir computing using VCSEL with a feedback loop. Photon Res. 2024, 12, 1222–1230. [Google Scholar] [CrossRef]
  24. Hasegawa, H.; Kanno, K.; Uchida, A. Parallel and deep reservoir computing using semiconductor lasers with optical feedback. Nanophotonics 2023, 12, 869–881. [Google Scholar] [CrossRef] [PubMed]
  25. Qu, Y.; Lian, H.; Shao, R.; Liu, J.; Liu, L.; Yang, J. Time series analysis for financial indices using optical reservoir computing. Opt. Eng. 2024, 63, 054108. [Google Scholar] [CrossRef]
  26. Li, Z.; Xie, Y.; Xu, F.; Ye, Y.; Jiang, X.; Su, Y.; Li, L.; Chen, Z.; Tang, Y. Time-delayed reservoir computing using mutually coupled multimode semiconductor laser for high-speed image recognition. Opt. Laser Technol. 2025, 187, 112774. [Google Scholar] [CrossRef]
  27. Nguimdo, R.M.; Erneux, T. Enhanced performances of a photonic reservoir computer based on a single delayed quantum cascade laser. Opt. Lett. 2019, 44, 49–52. [Google Scholar] [CrossRef]
  28. Yang, Y.G.; Huang, Y.; Zhou, P.; Li, N.Q. Photonic reservoir computing for parallel task processing based on a feedback-free spin-polarized VCSEL. Opt. Commun. 2025, 574, 131225. [Google Scholar] [CrossRef]
  29. Harkhoe, K.; Verschaffelt, G.; Katumba, A.; Bienstman, P.; Van der Sande, G. Demonstrating delay-based reservoir computing using a compact photonic integrated chip. Opt. Express. 2020, 28, 3086–3096. [Google Scholar] [CrossRef]
  30. Cai, D.; Huang, Y.; Yang, Y.; Zhou, P.; Li, N. Enhanced performances of photonic reservoir computing using a semiconductor laser with random distributed optical feedback. Opt. Lett. 2023, 48, 6392–6395. [Google Scholar] [CrossRef]
  31. Zhu, P.J.; Wang, H.X.; Ji, Y.F. Performance enhancement reservoir computing system based on combination of VCESL optical feedback and mutual injection structure. IEEE J. Sel. Top. Quantum Electron. 2025, 31, 7400112. [Google Scholar] [CrossRef]
  32. Yamazaki, Y.; Kinoshita, K. Photonic physical reservoir computing with tunable relaxation time constant. Adv. Sci. 2024, 11, 2304804. [Google Scholar] [CrossRef] [PubMed]
  33. Yue, D.; Wu, Z.; Hou, Y.; Cui, B.; Jin, Y.; Dai, M.; Xia, G. Performance optimization research of reservoir computing system based on an optical feedback semiconductor laser under electrical information injection. Opt. Express. 2019, 27, 19931–19939. [Google Scholar] [CrossRef]
  34. Ortin, S.; Pesquera, L. Tackling the trade-off between information processing capacity and rate in delay-based reservoir computers. Front. Phys. 2019, 7, 210. [Google Scholar] [CrossRef]
  35. Ortin, S.; Pesquera, L. Reservoir computing with an ensemble of time-delay reservoirs. Cogn. Comput. 2017, 9, 327–336. [Google Scholar] [CrossRef]
  36. Rohm, A.; Ludge, K. Multiplexed networks: Reservoir computing with virtual and real nodes. J. Phys. Commun. 2018, 2, 085007. [Google Scholar] [CrossRef]
  37. Lang, R.; Kobayashi, K. External optical feedback effects on semiconductor injection laser properties. IEEE J. Quantum Electron. 1980, 16, 347–355. [Google Scholar] [CrossRef]
  38. Dejonckheere, A.; Duport, F.; Smerieri, A.; Fang, L.; Oudar, J.; Haelterman, M.; Massar, S. All-optical reservoir computer based on saturation of absorption. Opt. Express 2014, 22, 10868–10881. [Google Scholar] [CrossRef]
  39. Goldmann, M.; Köster, F.; Lüdge, K.; Yanchuk, S. Deep time-delay reservoir computing: Dynamics and memory capacity. Chaos. 2020, 30, 093124. [Google Scholar] [CrossRef]
  40. Estébanez, I.; Schwind, J.; Fischer, I.; Argyris, A. Accelerating photonic computing by bandwidth enhancement of a time-delay reservoir. Nanophotonics 2020, 9, 4163–4171. [Google Scholar] [CrossRef]
  41. Nakayama, J.; Kanno, K.; Uchida, A. Laser dynamical reservoir computing with consistency: An approach of a chaos mask signal. Opt. Express 2016, 24, 8679–8692. [Google Scholar] [CrossRef] [PubMed]
  42. Sugano, C.; Kanno, K.; Uchida, A. Reservoir computing using multiple lasers with feedback on a photonic integrated circuit. IEEE J. Sel. Top. Quantum Electron. 2020, 26, 1500409. [Google Scholar] [CrossRef]
Figure 1. Structure diagram of the time-delay RC system based on SL. D-SL: drive-SL, R-SL: response-SL, PM: phase modulator, VA: variable attenuator, OC: optical circulator, FC: fiber coupler.
Figure 1. Structure diagram of the time-delay RC system based on SL. D-SL: drive-SL, R-SL: response-SL, PM: phase modulator, VA: variable attenuator, OC: optical circulator, FC: fiber coupler.
Photonics 12 00455 g001
Figure 2. Information processing flowchart of RC.
Figure 2. Information processing flowchart of RC.
Photonics 12 00455 g002
Figure 3. Evolution map of the NMSE in κinj-κ parameter space under different θ. (a) θ = 0.1Tro; (b) θ = 0.2Tro; (c) θ = 0.3Tro; (d) θ = 0.4Tro; (e) θ = 0.5Tro; (f) θ = 0.6Tro.
Figure 3. Evolution map of the NMSE in κinj-κ parameter space under different θ. (a) θ = 0.1Tro; (b) θ = 0.2Tro; (c) θ = 0.3Tro; (d) θ = 0.4Tro; (e) θ = 0.5Tro; (f) θ = 0.6Tro.
Photonics 12 00455 g003
Figure 4. NMSEmin varies with θ under τ = T.
Figure 4. NMSEmin varies with θ under τ = T.
Photonics 12 00455 g004
Figure 5. NMSE varies with τ, where the red dots mark the points with NMSE less than 0.01.
Figure 5. NMSE varies with τ, where the red dots mark the points with NMSE less than 0.01.
Photonics 12 00455 g005
Figure 6. Evolution maps of MC in the κinj-κ parameter space under different θ. (a) θ = 0.1Tro; (b) θ = 0.2Tro; (c) θ = 0.3Tro; (d) θ = 0.4Tro; (e) θ = 0.5Tro; (f) θ = 0.6Tro.
Figure 6. Evolution maps of MC in the κinj-κ parameter space under different θ. (a) θ = 0.1Tro; (b) θ = 0.2Tro; (c) θ = 0.3Tro; (d) θ = 0.4Tro; (e) θ = 0.5Tro; (f) θ = 0.6Tro.
Photonics 12 00455 g006
Figure 7. MCmax varies with θ under τ = T.
Figure 7. MCmax varies with θ under τ = T.
Photonics 12 00455 g007
Figure 8. MC varies with τ, where the red dots mark the points with MC higher than 21.
Figure 8. MC varies with τ, where the red dots mark the points with MC higher than 21.
Photonics 12 00455 g008
Figure 9. RC performance varies with deviation percentage of α and GN. (a) Santa Fe prediction task, where the pink dotted line indicates the NMSE = 0.01 level; (b) Memory evaluation task.
Figure 9. RC performance varies with deviation percentage of α and GN. (a) Santa Fe prediction task, where the pink dotted line indicates the NMSE = 0.01 level; (b) Memory evaluation task.
Photonics 12 00455 g009
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Li, Q.; Hou, Y.; Li, K.; Guo, X.; Hu, C.; Yue, D. Time Parameter Optimization for the Semiconductor Laser-Based Time-Delay Reservoir Computing System. Photonics 2025, 12, 455. https://doi.org/10.3390/photonics12050455

AMA Style

Li Q, Hou Y, Li K, Guo X, Hu C, Yue D. Time Parameter Optimization for the Semiconductor Laser-Based Time-Delay Reservoir Computing System. Photonics. 2025; 12(5):455. https://doi.org/10.3390/photonics12050455

Chicago/Turabian Style

Li, Qiudi, Yushuang Hou, Keqiang Li, Xiaoyu Guo, Chunxia Hu, and Dianzuo Yue. 2025. "Time Parameter Optimization for the Semiconductor Laser-Based Time-Delay Reservoir Computing System" Photonics 12, no. 5: 455. https://doi.org/10.3390/photonics12050455

APA Style

Li, Q., Hou, Y., Li, K., Guo, X., Hu, C., & Yue, D. (2025). Time Parameter Optimization for the Semiconductor Laser-Based Time-Delay Reservoir Computing System. Photonics, 12(5), 455. https://doi.org/10.3390/photonics12050455

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop