Application of Neural Network Algorithm in Network Engineering

Design

Min Yang

and Jiajie Zhang

Shandong Communication & Media College, Jinan,Shandong, 250200, China

Shandong Technician Institute, Jinan, Shandong, 250200, China

Keywords: Design, Neural Network Algorithms, Network Engineer Design, Computer.

Abstract: With the wide application of computers to people's lives and work, people's demand for computers is getting

higher and higher. Network engineering plays an extremely important role in the computer field, but there is

a problem of inaccurate evaluation of results. Traditional network engineering design cannot solve the

problems of low efficiency and unreasonable design in the process of network engineering from requirements

analysis to obtaining network model. Therefore, this paper proposes a neural network algorithm for innovative

optimization network engineering design analysis. First, the design theory is used to evaluate the engineer,

and the index is divided according to the network engineering design requirements to reduce it Interference

factors in network engineering. Then, the design theory designs the model for the network engineer, forms

the network engineering design scheme, and conducts the network engineering design results Comprehensive

analysis. MATLAB simulation shows that under the condition of certain evaluation criteria, the network

engineering design efficiency, scientificity and rationality

of the neural network algorithm are superior

Traditional network engineering.

1 INTRODUCTION

Network model is one of the important means of

network engineer design, which is of great

significance for network engineering design

(Arjomandi, Cenanovic, et al. 2023)]. However, in

the process of network engineering design, the

network engineering design scheme has the problems

of poor accuracy and low efficiency, which brings

certain reputation losses to the overall engineering

scheme (Cao, Yang, et al. 2023) 。 Some scholars

believe that the application of neural network

algorithms to network engineering design analysis

(Chen, Wen, et al. 2023) can effectively analyze

network engineering design schemes. Provide

corresponding support for network engineering

design (Chiang, Wang, et al. 2023). On this basis, this

paper proposes a neural network algorithm to

optimize the network engineering design scheme and

verify the effectiveness of the model (Cruz, Carrillo,

et al. 2023).

1.1 Development of Network

Engineering

With the continuous development of network

technology, neural networks have become one of the

important technical means in network engineering

design. Based on the practical application of network

engineering, this paper discusses the application of

neural networks in network engineering design,

including network topology design (He, Guo, et al.

2023), network routing optimization, performance

analysis, etc., and verifies the superiority and

practicality of neural networks in network

engineering design through experiments (Ma, Dang,

et al. 2023).

Network engineering is an important research field in

computer science, which mainly studies the design,

development, implementation and management of

computer networks and communication systems, and

is an important branch of computer science (Nowak,

and Popenda, 2023). With the continuous

development and application of network technology,

the design and optimization of network engineering

has become an important topic, and neural network,

122

Yang, M. and Zhang, J.

Application of Neural Network Algorithm in Network Engineering Design.

DOI: 10.5220/0013536600004664

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 3rd International Conference on Futuristic Technology (INCOFT 2025) - Volume 1, pages 122-127

ISBN: 978-989-758-763-4

as a new type of computing model, can play an

important role in network engineering design.

1.2 Application of Neural Network in

Network Engineering Design

1.2.1 Network Topology Design

Network topology design is an important issue in

network engineering design, which is mainly

optimized for network reliability, scalability,

performance and other aspects. Neural networks can

learn the network topology and predict the

development trend of the network topology, so as to

optimize the design of the network topology and

improve the reliability and performance of the

network (Park, Si, et al. 2023).

1.2.2 Network Routing Optimization

Network routing optimization is a central problem in

network engineering design, which is mainly

optimized for the routing algorithm of the network,

and the performance, reliability and scalability of the

network can be improved by optimizing the routing

algorithm. Neural networks can optimize routing

algorithms and improve the routing efficiency and

performance of the network by learning the

transmission path of network packets and predicting

the transmission trend of packets (Qiao, Fu, et al.

2023).

1.2.3 Performance Analysis

Performance analysis in network engineering is

mainly to evaluate and optimize the performance of

the network, including network bandwidth, latency,

data transmission rate and other aspects. Traditional

performance analysis methods usually require a large

number of data sampling and statistical analysis, but

neural networks can accurately optimize the

performance and bandwidth allocation of the network

by learning the performance of the network and

predicting the performance trend of the network (Sun,

Peng, et al. 2023).

1.3 System Optimization Analysis

In order to verify the practicability and superiority of

neural networks in network engineering design, this

paper analyzes the system optimization through

experiments. The basic idea of the experiment is:

firstly, the neural network is used to predict and

optimize the network topology and routing algorithm,

and then the performance of the network is monitored

and analyzed in real time, and finally the optimization

results and evaluation indicators are obtained (Teng,

Wan, et al. 2023).

Through experimental results, it is found that

neural networks have significant optimization effects

on network engineering design, which can improve

the performance and reliability of the network, and

greatly reduce the time and cost of network

engineering design. Especially in large-scale network

environments, the advantages of neural networks are

more obvious, which can realize efficient network

topology design, routing optimization and

performance analysis (Williams, 2023).

1.4 Requirements for Network

Engineering Design

The algorithmic requirements in network engineering

design mainly include the following aspects:

1.4.1 High Efficiency

Network engineering needs to deal with a large

amount of data and complex computing tasks, so the

efficiency of algorithms is very important, and it is

necessary to minimize the use of computing time and

computing resources. Efficient algorithms can

quickly complete computing tasks, reduce computing

overhead, and improve work efficiency.

1.4.2 Extensibility

Network engineering needs to handle ever-increasing

data volumes and increasingly complex

computational tasks, so the scalability of algorithms

is important, and it needs to be able to adapt to

changing data sizes and computing needs, while also

maintaining the efficiency of the algorithm as much

as possible.

1.4.3 Reliability

Network engineering needs to ensure the reliability

and security of data, so reliable algorithms are needed

to process data. Reliable algorithms guarantee data

accuracy and consistency while preventing data loss,

corruption, or tampering.

1.4.4 Reasonableness

Network engineering design needs to consider the

influence of multiple factors, such as network

topology, data distribution strategy, routing

algorithm, etc., so the rationality of the algorithm is

very important, and it is necessary to be able to fully

Application of Neural Network Algorithm in Network Engineering Design

123

consider the trade-off and balance of various factors,

so that the design scheme is as reasonable and

optimized as possible.

1.4.5 Ease of Implementation

Network engineering design needs to be able to be

applied and implemented in practice, so the ease of

implementation of algorithms is also a very important

requirement. Easy-to-implement algorithms can be

easily applied to real projects, and can be quickly

implemented and debugged.

2 RELATED CONCEPTS

2.1 Mathematical Description of the

Neural Network Algorithm

The neural network algorithm uses the design theory

to optimize the network engineering design scheme

and finds the unqualified values in the network

engineer's design plan according to the indicators in

the network engineering design and designs the

network engineering the scheme is integrated to

finally judge the feasibility of the network

engineering design model. The neural network

algorithm combines the advantages of design theory

and uses the network engineering design model

library to quantify and obtain a solution to the needs

of users Network model (Yu, 2023).

Suppose I. network engineering design

requirements is

, network engineering design

scheme is

(Zhou, Liu, et al. 2023), network

engineering design scheme to meet the requirements

, network engineering design scheme The

judgment function is

,as shown in Equation (1).

()()

ii i

JddKxyk

=−⋅++



(1)

2.2 Selection of Network Model

Scheme

Hypothesis II The network engineer design plan is

, that the inertia weight coefficient is

, then the

network engineering design requires an unreasonable

network engineer design plan as shown in Equation

(2).

()

pydbad

=+⋅−



)

2.3 Analysis of network engineering

design scheme

Before the neural network algorithm, it is necessary

to conduct multi-dimensional analysis of the network

engineering design scheme, map the network

engineering design requirements to the network

model library, and eliminate unreasonable network

engineering design Scheme. First, the network

engineer conducts a comprehensive analysis of the

design plan and sets the threshold and index weights

of the network engineering design scheme to ensure

the accuracy of the neural network algorithm. The

network engineer's design plan is a system test

network engineering design scheme, which requires

innovative analysis. If a network engineer's design

plan is in a nonnormal distribution, its network

engineering design will be affected, reducing the

accuracy of the overall network engineering design.

In order to improve the accuracy of the neural

network algorithm and improve the level of network

engineering design, the network engineering design

scheme should be selected, and the specific scheme

selection is shown in Figure 1.

Begin

Primary population

Clustering

Fitness value

Condition

Mutation

Get the

best

individual

Blurred

image

Output

restored

image

End

Figure 1: Selection results of the network model scheme

INCOFT 2025 - International Conference on Futuristic Technology

124

Analysis of network engineering design scheme

shows that the network model scheme presents a

multi-dimensional distribution, which is in line with

objective facts. The network engineer's design plan is

not directional, indicating that the network model

scheme has strong randomness, so it is regarded as a

high analysis study. The network engineer's design

plan meets the normal requirements, mainly because

the design theory adjusts the network engineer's

design plan, removes unscientific and irrelevant

schemes, and supplements the default scheme, so that

the dynamic correlation of the entire network

engineering design model is strong.

3 NETWORK ENGINEERS

DESIGN OPTIMIZATION

STRATEGIES FOR THE PLAN

The neural network algorithm adopts a random

optimization strategy for the network engineer's

design plan and adjusts the engineer parameters to

realize the optimization of the network engineer's

design plan. The neural network algorithm divides the

network engineer's design plan into different network

engineering design levels, and randomly selects

different schemes. In the iterative process, network

engineering design solutions at different design levels

are optimized and analyzed and different solutions are

compared Network engineering design level,

document the best network engineering design model.

4 PRACTICAL EXAMPLES OF

NETWORK ENGINEERS

DESIGNING PLANS.

4.1 Introduction to Network

Engineering Design

In order to facilitate network engineering design, the

design plan of network engineers in complex

situations is the research object, with 12 paths and a

test time of 12h The design scheme of the network

engineer design plan is shown in Table 1.

The network engineering process in Table 1 is

shown in Figure 2.

Table 1: Network Engineering Design Requirements

Scope of

application

grade Optimize

erformance

Network

model

Network

construction

I 92.03 91.57

II 90.78 90.31

O&M I 91.68 89.69

II 92.41 92.98

Maintenance

costs

I 92.04 91.56

II 90.52 90.31

Begin

Initializatio

n device

RTS

CTS

DATA RF905 Results

Personal

computer

Figure 2: The analysis process of the network engineer's

design plan

Compared with traditional network engineering

design, the network engineering design scheme of

neural network algorithm is closer to the actual design

requirements. In terms of the rationality and

fluctuation range of network engineers' design plans,

neural network algorithms are superior to traditional

network engineering designs. Through the changes in

the network engineering design scheme in Figure 2, it

can be seen that the stability of the neural network

algorithm is better, and the efficiency is faster.

Therefore, the network engineering design scheme of

neural network algorithm has better efficiency, design

scheme and summation stability.

4.2 Network Engineer Design Plan

The design scheme of the network engineer's design

plan includes non-structural information, semi-

structural information, and structural information.

After the pre-selection of the neural network

algorithm, the design model of the preliminary

network engineer design plan is obtained, and the

network engineer designs the plan Analyze the

feasibility of the design model. In order to more

Application of Neural Network Algorithm in Network Engineering Design

125

accurately verify the optimization effect of the

network engineer's design plan, select the design plan

with different network engineering design levels, and

the network engineering design scheme is shown in

Table 2 shown.

Table 2: The overall picture of the network model scenario

Category Security Nature of the

networ

Network

construction

92.79 82.87

O&M 90.85 84.01

Maintenance costs 92.83 83.01

mean 93.06 81.02

38.51 35.26

P=4.04

4.3 Network Model Efficiency and

Stability of Network Engineering

Design

In order to verify the accuracy of the neural network

algorithm, the network engineering design scheme is

shown in Figure 3.

Figure 3: Network models with different algorithms

It can be seen from Figure 3 that the network

model of the neural network algorithm is higher than

that of the traditional network engineering design, but

the error rate is lower, indicating that the network

engineering design of the neural network algorithm is

relatively stable Traditional network engineering is

uneven. The average network engineering scheme of

the above two methods is shown in Table 3.

By Table 3, it can be seen that traditional network

engineering design has deficiencies in network

model, security and reliability in network engineer

design plans Network engineers' design plans have

changed drastically, and the error rate is high. The

general result of the neural network algorithm is a

Table 3: Comparison of network engineering design

accuracy of different methods

Algorithm Network

model

Magnitude

of chan

Error

Neural

network

orithms

92.62 91.40 1.22

Traditional

network

engineering

82.99 74.56 8.43

P 36.57 35.94 36.48

higher network model than traditional network

engineering design. At the same time, the network

model of the neural network algorithm is greater than

91.40%, and the accuracy does not change

significantly. In order to further verify the superiority

of the neural network algorithm, the general analysis

of the neural network algorithm is carried out by

different methods, as shown in Figure 4.

Figure 4: Network model for network engineering design of

neural network algorithm

By Figure 4, it can be seen that the network model

of the neural network algorithm is significantly better

than the traditional network engineering design, and

the reason is that the neural network algorithm

increases the adjustment coefficient of the network

engineer's design plan and sets it Engineer's threshold

to reject network engineering solutions that do not

meet the requirements.

5 CONCLUSIONS

Aiming at the problem that the traditional network

engineering design is not ideal, this paper proposes a

neural network algorithm, and combines the network

engineering design principles to optimize the network

INCOFT 2025 - International Conference on Futuristic Technology

126

engineering design, so that the network engineering

design becomes more convenient. Fast and

reasonable. Research shows that neural network

algorithms can improve the accuracy, stability and

security of network engineering design. However, in

the process of neural network algorithms, the

construction of model libraries also needs to be

continuously improved, and more in-depth research

needs to be made.

REFERENCES

Arjomandi Rad, M., Cenanovic, M., & Salomonsson,

K.(2023) Image regression-based digital qualification

for simulation-driven design processes, case study on

curtain airbag. Journal of Engineering Design, 34(1): 1-

22.

Cao, P., Yang, T., Wang, K., Bao, W., & Yan, H.(2023)

Topology-Aided Multicorner Timing Predictor for

Wide Voltage Design. Ieee Design & Test, 40(1): 62-

69.

Chen, H., Wen, Y., Zhu, M., Huang, Y., Xiong, W., Xiao,

C., & Lu, Z.(2023) A Function-Oriented Electronic and

Electrical Architecture of Remote Control Ship on

Inland River: Design, Verification, and Evaluation. Ieee

Transactions on Transportation Electrification, 9(1):

1641-1652.

Chiang, S.-H., Wang, C.-H., Yang, D.-N., Liao, W., &

Chen, W.-T.(2023) Distributed Multicast Traffic

Engineering for Multi-Domain Software-Defined

Networks. Ieee Transactions on Parallel and Distributed

Systems, 34(2): 446-462.

Cruz, E. M., Carrillo, L. R. G., & Salazar, L. A. C.(2023)

Structuring Cyber-Physical Systems for Distributed

Control with IEC 61499 Standard. Ieee Latin America

Transactions, 21(2): 251-259.

He, J., Guo, Z., Zhang, Y., Lu, Y., Wen, F., Da, H., Zhou,

G., Yuan, D., & Ye, H.(2023) Physics-model-based

neural networks for inverse design of binary phase

planar diffractive lenses. Optics Letters, 48(6): 1474-

1477.

Ma, J., Dang, S., Li, P., Watkins, G., Morris, K., & Beach,

M.(2023) Transfer Learning for the Behavior

Prediction of Microwave Structures. Ieee Microwave

and Wireless Technology Letters, 33(2): 126-129.

Nowak, M., & Popenda, A.(2023) Influence of neural

network configuration on PMSM motor angular

velocity estimation. Przeglad Elektrotechniczny, 99(2):

238-241.

Park, H., Si, H., Gu, J., Lee, D., Park, D., Lee, Y.-I., & Kim,

K.(2023) Engineered kirigami design of PVDF-Pt core-

shell nanofiber network for flexible transparent

electrode. Scientific reports, 13(1): 2582-2582.

Qiao, Y., Fu, Y., & Yuan, M.(2023) Communication-

Control Co-Design in Wireless Networks: A Cloud

Control AGV Example. Ieee Internet of Things Journal,

10(3): 2346-2359.

Sun, H., Peng, H., Xiao, H., Liu, X., Chen, Y., Gao, K.,

Wang, S., & Xu, P.(2023) An Optimized Design of

Compact Self-Powered Module Based on

Electromagnetic Vibration Energy Harvester

Considering Engineering Feasibility. Ieee Transactions

on Industry Applications, 59(1): 767-778.

Teng, F., Wan, J., & Liu, J.(2023) Review of Terahertz

Antenna Technology for Science Missions in Space.

Ieee Aerospace and Electronic Systems Magazine,

38(2): 16-32.

Williams, S.(2023) Embedding integration of temporary

works with permanent works on railway projects.

Proceedings of the Institution of Civil Engineers-Civil

Engineering, 176(5): 3-10.

Yu, L.(2023) Project engineering management evaluation

based on GABP neural network and artificial

intelligence. Soft Computing, 27(10): 6877-6889.

Zhou, Z., Liu, Y., Feng, Y., Klepin, S., Tsimring, L. S.,

Pillus, L., Hasty, J., & Hao, N.(2023) Engineering

longevity-design of a synthetic gene oscillator to slow

cellular aging. Science, 380(6643): 376-381.

Application of Neural Network Algorithm in Network Engineering Design

127