1 Introduction

In recent decades, mobile telecommunications witnessed a large increase in the number of subscribers and mobile broadband services. The new demands for broadband and multimedia applications are a necessity of the times, especially in tablets and smart phones. 5G systems can meet an exponential increase in traffic volumes and transactions for user demands (Jonathan 2015; Kholidy et al. 2023). 5G has been classified by The International Telecommunication Union (ITU) into three categories (1) massive Machine Type Communications (mMTC), (2) enhanced Mobile broadband (eMBB) and (3) ultra-Reliable and Low-latency communications (uRLLC) (He and Hu. 2020). According to ITU classification, 5G provides many advantages such as high transmission rate, large capacity, and low latency (D. Acatauassu et al. 2020; Yuan et al. 2019). In 5G networks, RAN is evolutional to Centralized Radio Access Network (CRAN) (Abdalla et al. 2020). CRAN has the advantages of reduced cost and less interference for users. This can be achieved by separating it into two parts: BaseBand unit (BBU) and the Remote radio unit (RRU). The connection between the two parts is called fronthaul segment. The separation between BBU and RRU gives the operator flexibility to manage RRUs and BBU pool that can contain centralizing the baseband processing functions (Sekander et al. 2018; Liu and Effenberger 2016).

A popular interface as Common Public Radio Interface (CPRI), is used in the fronthaul segment (Wey and Zhang 2018; Barzaq et al. 2021). Fronthaul bit rates recommended for 5G interface are 25 Gb/s, 50 Gb/s, n × 25 Gb/s (He and Hu 2020). Recently, an innovation has been adopted in 5G is an Open technologies Radio Access Network (O-RAN). Its concept is splitting the RAN to many parts, thus executing many functions (Wu et al. 2015). The O-RAN architecture enables many 5G operators use of non-proprietary subcomponents from a variety of vendors. The BBU is divided into Centralized Unit (CU) and a Distributed unit (DU). The new disaggregated functions can also be virtualized or containerized between the new parts. In Open RAN (oRAN) architecture, the vDU (virtualized Distributed unit) and RRU (Remote radio unit) are key components of the radio access network. The vDU is responsible for processing and managing the baseband signals, while the RRU handles the radio frequency (RF) transmission and reception (Li et al. 2014).

The optical fiber connection between the vDU and RRU allows for efficient transmission of baseband signals and control data between these two components. The vDU provides the baseband processing capabilities, including functions like channel coding, modulation/demodulation, and beamforming, while the RRU handles the RF signal transmission and reception at the cell site (Parvez et al. 2018). A fronthaul segment that connects BBU and RRU is a part of an optical transport network which plays a very important role in 5G networks. It has many useful technologies in 5G such as Passive optical network (PON), point-to-point fiber access, and Optical transport network (OTN). The use of optical fiber in the vDU-RRU link in oRAN deployments is crucial for delivering the high-performance, low-latency, and scalable connectivity required by 5G networks (Liu and Effenberger 2018; Shbair and Nahal 2019). In Semi-active WDM, a passive WDM module can be installed in Active Antenna processing unit (AAU) site, no need for power supply and active WDM module at BBU site. One fiber can be used to access between the AAU and BBU with 25G eCPRI services (https://forum.huawei.com/enterprise/en/5g-fronthaul-solution/thread/624437-875 2022). By enhancing the optical fiber performance for the vDU–RRU link, oRAN architectures can achieve greater flexibility and scalability. The fiber connection enables the deployment of RRUs at remote locations, such as cell towers or lamp posts, while the vDU can be centralized in a data center or cloud environment. This centralized and virtualized approach allows for more efficient resource allocation, network optimization, and easier maintenance (Hu et al. 2019).

Fiber bragg gratings (FBGs) are optical devices that can be used to enhance the performance of optical fiber systems, including those used in 5G networks. FBGs are periodic variations in the refractive index of an optical fiber, which act as wavelength-selective mirrors or filters (Sayed et al. 2017; Sharma and Sharma 2014; Sharma and Pancharya 2015; Sharma 2016; Sharma et al. 2018, 2021; Sharma and Subhashish 2020; Sharma and Asha 2022; Sharma 2023). FBGs can be utilized in various ways to improve the performance of optical fiber systems where laser linewidth (Sharma and Sharma 2014, Sharma and Banerjee 2018) can be optimized by FBG apodization functions (Sayed et al. 2020).

Many researches are performed to optimize the laser linewidth where dispersion can be reduced. In (Sayed et al. 2021a, b), a cascaded FBG model is constructed to reduce the linewidth for the transmitted optical signal at the transmitter. In (Bhaskar et al. 2022a, b), a cascaded apodized FBG is used as a passive optical tool to enhance performance and to narrow the band filter in C-band. In (Bhaskar et al. 2022a, b), two cascaded chirped FBGs are used, one with an increasing chirp and the other with a decreasing chirp. System reflectivity and Full width at half maximum (FWHM) are noticed to get optical narrow band pass filter in C-band.

In (Mustafa et al. 2023) four stages of FBGs are used to enhance Maximum Time Division Multiplexing (MTDM) system. The model is used in different schemes pre, post and symmetrical and common apodization functions. The performance is enhanced using a symmetrical scheme. In (Sayed et al. 2021a, b), FBGs are investigated as symmetrical and post dispersion compensation schemes to optimize the optical signal linewidth, Δλ, for Wave Division Multiplexing system. The performance is enhanced using a symmetrical scheme. In (Mustafa et al. 2022), FBG with different common apodization functions are investigated in WDM system not only for enhancing performance but also to reduce the system power budget. The dispersion is compensated by a chirped FBG as a post compensator where the performance is enhanced, and the power budget is reduced.

This work aims to enhance performance of the fronthaul segment as an optical fiber Giga Ethernet link between vDU and RRU in a simulated O-RAN 5G network. 4mm small apodized FBG is proposed with specific parameters and simulated with Optisystem. The proposed apodized FBG is connected with the Giga-Ethernet transceiver module with a data rate of 25 Gb/s in three position pre, post, and symmetrical to optimize its linewidth of the transmitted optical signal which reduces dispersion. Gaussian, Tanh, and uniform are common apodization functions widely to enhance its performance which optimizes laser linewidth for a distance 10 km to the RRU transceiver module. Quality factor and BER are measured at receiver by the BER as performance parameters. The paper contribution in enhancing fronthaul performance for 5G applications involve optimizing various aspects: (a) Low-Latency Transport: reducing signal processing delays and optimizing the network architecture to minimize the number of network elements and transmission hops. (b) Time and Phase Synchronization: Ensure accurate synchronization between the vDU and RRU by implementing precise timing protocols, such as Precision Time Protocol (PTP) or Synchronous Ethernet (SyncE). (c) Network Slicing: Implement network slicing to create dedicated virtual networks for specific 5G services or applications.

This paper is organized as follows. After the introduction in Sect. 1, Sect. 2 presents the proposed model. The simulation results are displayed and discussed in Sect. 3. Finally, Sect. 4 is devoted to the main conclusions.

2 Proposed model

Small apodized FBG is designed and simulated with Optisystem to optimize the optical signal linewidth for the fronthaul Ethernet segment between vDU and RRU in the proposed open RAN 5G systems, which enhances its performance. Three common apdization functions Gaussian, tanh, and uniform are used in FBG to enhance its performance to optimize the optical signal linwidth. The proposed FBG is connected with optical transceiver module at transmitter in vDU and the receiver at RRu for a distance 10 km. The proposed apodized FBG is used in the optical link with three position pre, post, and symmetrical. In each case, performance is measured by Quality factor and BER as performance parameters.

Figure 1 shows the proposed open RAN 5G layers. First, the internet layer is connected to the network core cloud which is connected to the virtual centralized unit (vCU) which represents the backhaul connections. Then, the optical transport layer gives the midhaul connection between (vCU) and the virtual distribution unit (vDU), which is connected to open RAN RU multi-vendors layers as the fronthaul connections. Finally, the eNodeBs and macro RUs are connected to open RAN RU multi-vendors layers to broadcast to different multi-vendor unit equipment (UEs).

Fig. 1
figure 1

proposed O-RAN 5G model layers

Full size image

Figure 2 shows the proposed O-RAN model components represented by the layers. In the central office, the 5G cloud supports the IP/MPLS protocols connected to the internet.

Fig. 2
figure 2

Proposed 5G components

Full size image

The vCU is connected as a backhaul connection to 5G cloud and midhaul connection to external vDU edge cloud with the optical WDM network. Many eNodeBs and macro RUs cells are connected to the vDU edge cloud as fronthaul Ethernet segments. The fronthaul segments have a bit rate of 25 Gb/s Ethernet Optical transceiver modules at distances 10 km each. Both eNodeBs and macro RUs broadcast to multi-vendor UEs.

2.1 Proposed AFBG design

The AFBG designed parameters with Optisystem are illustrated in Table 1. The AFBG length is 4 mm. the modulation refractive index is 0.0003 and effective refractive index 1.47. The operation wavelength is 1550 nm. The Parameters for Gaussian and tanh apodization functions are 0.5.

Table 1 Designed AFBG parameters (Sayed et al. 2020; Sayed et al. 2021)
Full size table

Figure 3a, b and c shows the reflectivity for each apodization functions Gaussian, tanh, and uniform respectively. The Gaussian one is shown in Fig. 3a where the reflectivity 70% and its linewidth 0.408 nm. Also, for Fig. 3b tanh and (c) uniform the reflectivity are 93% and 94% respectively. The linewidth for tanh is 0.418 nm and 0.42 is for the uniform. All apodization function reflectivity and linewidth are summarized in Table 2.

Fig. 3
figure 3

Relation between reflectivity and wavelength for each apodization function

Full size image
Table 2 Comparison between Reflectivity and Linewidth for each apodization function
Full size table

2.2 Proposed optical fiber link without AFBG

The proposed optical fiber link between vDU and RU is illustrated in Fig. 4. The proposed link consists of three parts (1) vDU, (2) the optical fiber, and (3) RU.

  1. 1.

    At the vDU side, the RLC layer receives data packets from higher-layer protocols. It segments these packets into smaller ones suitable for transmission over the radio interface. The segmented are then passed to the Media Access Control (MAC) layer. The MAC layer provides control information to the RLC layer, enabling coordination and synchronization between the two layers. This includes information about the allocated radio resources, scheduling parameters, and Hybrid Automatic Repeat Request (HARQ) related information (Lipovac et al. 2021). The RLC layer utilizes this information to adjust its operation and optimize the transmission process. The MAC layer multiplexes multiple RLC from different logical channels or users into a single MAC for transmission. The physical layer high in the vDU performs modulation and coding operations on the data to be transmitted. It converts the digital data into a suitable format to transmit it over the optical transmitter pluggable module with 25 Gb/s data rate which is connected to the optical fiber (Ahmed Abdul Mateen et al. 2021; https://www.5gtechnologyworld.com/open-ran-functional-splits-explained/ 2024).

  2. 2.

    At the optical fiber, the optical transmitter (Tx optical module) is connected to the G.652 Single Mode Fiber (SMF) for a wavelength 1550 nm and distance 10 km to RU.

  3. 3.

    At RU side, the optical receiver (Rx optical module pin) receives the optical signal to electrical to the physical-low layer. The bit error rate analyzer (BER) measures the optical signal quality factor and bit error rate as performance parameters. The Physical-low layer handles the digital processing of data, while the RF layer is responsible for the conversion of digital signals into analog radio waves for transmission over 5G antennas (https://www.5gtechnologyworld.com/open-ran-functional-splits-explained/ 2024).

Fig. 4
figure 4

Proposed optical Ethernet link between vDU edge cloud and RUs site

Full size image

Table 3 summarizes all simulated optical link parameters. The SMF G.625D optical fiber attenuation coefficient and Dispersion coefficient at 1550 nm are 0.24 dB/km and 18 ps/nm km respectively. The transmitter module has − 5 dBm transmitted power with a bit rate 25 GB/s. The PIN receiver responsivity is 1 A/W with a dark current 10 nA and cut off frequency 75 × Bit rate Hz.

Table 3 Simulated optical fiber link parameters (Sayed et al. 2020, 2021a, 2021b)
Full size table

2.3 Proposed optical link with a pre-AFBG connection

The pre-AFBG connection is illustrated in Fig. 5. The AFBG is connected between the optical transmitter and the optical fiber to optimize the optical signal linewidth at transmitter side.

Fig. 5
figure 5

Proposed optical Ethernet link with a pre-AFBG connection

Full size image

2.4 Proposed optical link with a post-AFBG connection

The linewidth of the optical signal is optimized at the receiver side. The AFBG is placed between the optical fiber and optical receiver as shown in Fig. 6.

Fig. 6
figure 6

Proposed optical Ethernet link with a post-AFBG connection

Full size image

2.5 Proposed optical fiber link with a symmetrical-AFBG connection

Finally, the optical signal linewidth is optimized in both transmitter and receiver sides. It can be considered a combination between pre and post as in Fig. 7.

Fig. 7
figure 7

Proposed optical Ethernet link with a symmetrical-AFBG connection

Full size image

3 Simulation results and discussion

Now, we present the simulation results of the AFBG utilized at pre, post, and symmetrical connections. Maximum Q-factor and minimum BER are measured by BER analyzer and compared for each apodization function. Finally, the transmitted power influence on the system performance under AFBG apodization functions is discussed and the proposed model linewidth is evaluated and compared with the other in the related work.

3.1 Quality factor

3.1.1 Without AFBG

The Q-factor results for the optical link without AFBG are illustrated in Fig. 8. It shows the relation between the Q-factor and time (ns).

Fig. 8
figure 8

Q-factor for the optical link without AFBG

Full size image

The Q-factor is sharply increased as the time increases. It has a maximum peak at 7.8 between time 5 and 6 ns, where it is sharply decreased as the time increases.

3.1.2 Pre and post AFBG connection

Figure 9a and b shows the Q-factor results for the pre and post AFBG connections respectively.

Fig. 9
figure 9

Q-factor (a) for the pre-AFBG and (b) the post

Full size image

The two connections pre and post are close to each other for all apodization functions except Gaussian. Each function has its maximum peak. The Gaussian gives 8.16 for the pre and 8.17 for post connections. The Tanh achieves 9.27 and 9.24 for the uniform at time between 0.4 and 0.6 ns. It gives the best Q-factor of 9.27 for the pre and post AFBG.

3.1.3 Symmetrical AFBG connection

The obtained results for the symmetrical AFBG are shown in Fig. 10. Each function has its maximum peak between time 0.4 and 0.6 ns. The Gaussian one gives a Q-factor of 7.75 while the tanh one achieves 9.36. The uniform gives the best results with a Q-factor of 9.4 for the symmetrical connection.

Fig. 10
figure 10

Q-factor for the symmetrical AFBG

Full size image

3.2 BER

3.2.1 BER without AFBG

Figure 11 shows the relation between BER and time (ns). BER decreases as the time increases. The BER has a peak down at 2.43e− 15 between time 5 and 6 ns. After that, the BER increases as the time increases.

Fig. 11
figure 11

BER without AFBG

Full size image

3.2.2 Results for pre and post AFBG connection

Figure 12a and b show the BER results for the pre and post AFBG connections for all functions respectively.

Fig. 12
figure 12

BER (a) for the pre-AFBG and (b) the post

Full size image

For all apodization functions in the pre and post connections, the BER is decreased as the time is increased. Different BER peaks down for each function is found. The Gaussian gives BER 1.67e− 16 for a pre and 1.54e− 16 for a post. Tanh and uniform give the same peak down BER 9.3e− 21. After that, the BER increases as the time increases.

3.2.3 Symmetrical AFBG connection

The BER for the symmetrical results are shown in Fig. 13. All functions have a peak down BER at time between 4 and 6 ns.

Fig. 13
figure 13

BER for the symmetrical AFBG

Full size image

Finally, the BER increases as the time increases. For the gaussian, the minimum BER is 4.59e− 15. For, the tanh and the uniform are close to each other. The tanh gives minimum BER 3.98e− 21 and the uniform gives the best results 2.72e− 21.

3.3 Comparison for the best Q-factor and BER

Figure 14 and Table 4 show the best Q-factor results for the different AFBG connections.

Fig. 14
figure 14

Comparison for the best Q-factor for AFBG connections

Full size image
Table 4 Comparison for the best Q-factor for AFBG connections
Full size table

The comparison shows that the best Q-factor result is 9.4 for the symmetrical connection using uniform function. The second is 9.27 for the pre and post using tanh. The lowest is 7.83 without using AFBG.

For the BER, Fig. 15 and Table 5 show the best BER comparisons for AFBG connections.

Fig. 15
figure 15

Comparison the best BER for AFBG connections

Full size image
Table 5 Comparison for the best BER for AFBG connections
Full size table

The comparison shows that the best BER is 2.72e− 21 for the symmetrical connection using uniform function. After that, 9.3e− 21 is for the pre and post using tanh. The maximum one is 2.43e− 15 without using AFBG.

3.4 Effect of transmitted power on system performance

5G infrastructure components depend on optical network connections such as transceivers, amplifiers, and switches. So, optical networks play a crucial role in connecting various elements of the 5G infrastructure (such as base stations and data centers). The power-saving measures implemented in these networks primarily target energy efficiency, cost reduction, and environmental sustainability, rather than directly impacting the functionality or performance of the 5G technology itself. So, it is very important to note that energy efficiency and power-saving measures are essential considerations for all components of the telecommunications ecosystem, including both the optical and wireless segments, to ensure the sustainable growth and operation of 5G networks and to minimize their environmental impact. In the following section, the achieved power saving in our proposed model is clarified.

Figure 16 shows the relation between the transmitted power and log(BER) as an evaluating parameter. A wide range of transmitted power from − 18 to 17 dBm is applied to the symmetrical connection, which is the best case, to show many aspects of transmitted power influences on the system performance.

Fig.16
figure 16

log (BER) versus transmitted power

Full size image

The figure also shows this influence at each apodization function. For all apodization profiles, the transmitted power range is − 18 to − 10 dBm. The BER decreases slowly to 1 × 10−20. After that, the BER is slightly decreased under 1 × 10−20 as the power increases, where uniform and tanh profiles are close to each other. After 0 dBm, the BER starts to decrease slowly to its minimum values. For uniform and tanh profiles, the minimum BER at 15 dBm transmitted power are 1 × 10−30 and 1 × 10 −31, respectively. Finally, the BER starts to increase as transmitted power increases; however, the Gaussian profile still moves to minimum a BER of 1 × 10−32, which consumes more transmitted power to get the minimum BER than both uniform and tanh profiles. At a BER of 1 × 10−20, the system can save more power which agrees with 5G goals. The transmitted power can be saved to − 10.51 and − 9.7 dBm using uniform and tanh functions, respectively. However, the Gaussian profile uses 12 dBm transmitted power to keep the same performance of 1 × 10−20.

3.5 Proposed model evaluation

The linewidth at FWHM achieved by the apodized FBG is evaluated through a comparison with the related work (Bhaskar et al. 2022a, b) in Fig. 17. The minimum linewidth 0.408 nm is achieved with the Gaussian apodization profile in the proposed model, while 1.56 nm is obtained by the related work (Bhaskar et al. 2022a, b). This confirms the superiority of our proposed function.

Fig.17
figure 17

Proposed model evaluation

Full size image

4 Conclusion

The 5G network is investigated and simulated with Optisystem to enhance performance of the fronthaul optical fiber link between BBU and RRU. A designed AFBG is connected to the optical Ethernet module with 25 Gb/s data rate, which is used in fronthaul segment to connect BBU pool to RRUs. The AFBG is connected as pre, post, and symmetrical connections. Different commonly used apodization profiles are investigated as uniform, tanh, and Gaussian. The obtained results reveal the minimum linewidth 0.408 nm is obtained by Gaussian function. The Q-factor and BER without AFBG are 7.83 and 2.43 × 10−15, respectively. The pre and post connections give the maximum Q-factor and BER 9.24 and 9.3 × 10−21, respectively obtained from tanh function. The best results for the maximum Q-factor and minimum BER are 9.2 and 2.72 × 10−21, respectively obtained from uniform function in the symmetrical connection. The transmitted power can be saved to − 10.51 and − 9.7 dBm using uniform and tanh functions, respectively. However, the Gaussian profile uses 12 dBm transmitted power to keep the same performance of 1 × 10−20. Finally, the proposed model linewidth is 0.408 nm and the evaluated by related work is 1.56 nm.