Priority-Aware Genetic Algorithm for Multi-Constraint Dynamic

Spectrum Allocation in Logistic Centers

A V S S Sampath

, Nischal S Malagati

, V Kanapathi

, Shinu M Rajagopal

and Prashanth B. N

Department of Computer Science and Engineering,

Amrita School of Computing, Bengaluru,

Amrita Vishwa Vidyapeetham, India

Department of Mechanical Engineering,

Amrita School of Engineering, Bengaluru,

Amrita Vishwa Vidyapeetham, India

Keywords:

Genetic Algorithm, Graph Coloring, Interference of IoT Devices, Dynamic Spectrum Allocation.

Abstract:

With the growing technology in the ﬁeld of IoT (Internet of Things), the study aims to improve the technology

related to spectrum and channel allocation. Dynamic spectrum allocation is pivotal and useful in logistics

centers, where the IoT device provides real-time tracking, monitoring, and data exchange. This study presents

various models and algorithms that are implemented in MATLAB for dynamic spectrum allocation tailored

to logistic environments, addressing challenges to spectrum allocation. The proposed solution to the problem

improves the bandwidth and channel allocation and reduces conﬂicts between different IoT devices. The

results demonstrate the efﬁciency of the proposed models, which paves the way for improvement in advanced

spectrum management in IoT-based logistics.

1 INTRODUCTION

In the growing world of technology, the Internet of

Things (IoT) has revolutionized industries worldwide,

including logistics, where data exchange in real time

is crucial for efﬁcient operations. IoT devices de-

ployed in logistics centers are used for communica-

tion between the inventory management system, de-

vice tracking, and other operational equipment. But

with increasing number of these IoT devices, the con-

ﬂicts between them also increase, which demands

more spectrum allocation. This poses a challenge to

ensuring communication between all IoT devices and

optimal resource allocation in logistic networks.

This study aims to abolish this complication by

modeling various models for the optimal bandwidth

allocation. Usage of genetic algorithms can be effec-

tively applied to various such problems, including the

0/1 knapsack problem. This problem, which is solved

in exponential time, can also be done in linear time

using genetic algorithms. The analysis later on of

various crossover functions and mutation operations

throws more light on the enhanced way of allocating

our IoT devices (Ramana, Shilpa, et al. 2023). An-

other study also highlights the optimization in appli-

cations of the genetic algorithms. The study solves

the 0/1 knapsack problem using a multi-objective al-

gorithm. This highlights the effectiveness of genetic

algorithms (Ramana, Shilpa, et al. 2023). Another

study also illustrates the evolution of help desk sys-

tems and the integration of advanced techniques like

genetic algorithms. The literature also points out the

limitations of genetic algorithms and how to deal with

them (Lekshmy, Anusree, et al. 2018).

Resource allocation is also a critical aspect of

network management, particularly in environments

such as a logistic center. It involves the efﬁcient

distribution of resources, i.e., bandwidth, here to re-

spected IoT devices to ensure reliable and efﬁcient

communication. A similar study emphasizes the us-

age of bipartite graph matching techniques to enhance

the system capacity in device-to-device communica-

tion. Two primary algorithms have been discussed:

the Hopcroft-Karp (HK) algorithm and the Kuhn-

Munkres (KM) algorithm. The study also identiﬁes

challenges such as deep fading and outage conditions

affecting the 10-15% of overall resource (Vaishnav

and Panda, 2017). These IoT devices can also be

Sampath, A. V. S. S., Malagati, N. S., Kanapathi, V., Rajagopal, S. M. and B N, P.

Priority-Aware Genetic Algorithm for Multi-Constraint Dynamic Spectrum Allocation in Logistic Centers.

DOI: 10.5220/0013584200004664

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 715-721

ISBN: 978-989-758-763-4

715

RFID (Radio Frequency Identiﬁcation); similarly, in a

study, supply chain management using RFID tags was

discussed. The study provides an overview of SCM,

discussing the main factors that govern the product

cycle (Ravi, 2010). This article aims to study the

chromatic polynomial of some families of graphs and

to describe the properties of the chromatic polynomial

of some graph operations (Sony and Manjusha, 2023).

Our study aims to work on the limitations of all the

above literature and to build an optimized model for

bandwidth allocation.

2 EXISTING SYSTEMS

Dynamic allocation of spectrum being an emerging

topic in the networking phase, there is minute imple-

mentation of these models in real time. A study pro-

posed a working real-life model that is based on the

idea of WSN’s (wireless sensor networks). WSN’s are

distributed networks with ﬂexible wireless communi-

cation. The limitation of this research is that the lim-

ited spectrum of resources poses challenges for WSN

development. With an increase in the resources, the

more the chances of collision of bandwidth between

two devices (Xiaomo, Cai, et al. 2022).

3 RELATED WORKS

The authors propose a novel channel allocation al-

gorithm designed to maximize both frequency and

time continuity in spectrum sharing between mobile

network operators (MNOs) and incumbent radio sys-

tems. Unlike traditional methods that only focus on

meeting MNO demand, this approach considers con-

tinuity across time and frequency to boost spectrum

efﬁciency by up to 6 percent and channel continuity

by 23 percent (Ikami, Hayashi et al. 2020). The au-

thors present an advanced dynamic spectrum alloca-

tion algorithm that uses multiple fairness indicators

to allocate spectrum equitably among mobile network

operators (MNOs) sharing resources with incum-

bents. This approach improves fairness by 16 percent

and satisfaction by 44 percent, addressing the lim-

itations of previous single-indicator models (Ikami,

Hayashi et al. 2020). The authors of this paper ad-

dress spectrum sharing issues among multiple net-

work operators. It proposes a spectrum allocation al-

gorithm incorporating priority-based sharing and ne-

gotiation to dynamically manage spectrum resources.

Key innovations include using a Spectrum Sharing

Metric (SSM) to account for class service priori-

ties, urgent bandwidth requests, and long-term spec-

trum occupation ratios among operators (Kim, Lee et

al. 2005). This research develops a game-theoretic

model for dynamic spectrum allocation, allowing sec-

ondary users (SUs) to adjust spectrum requests based

on primary user (PU) pricing and SU competition.

The model achieves faster convergence to Nash equi-

librium and offers improved allocation stability, simu-

lating real-world competitive conditions among SU’s

(Zhao, Liu et al. 2018). The authors identify the key

factors affecting inventory management in Thailand’s

construction industry using exploratory factor analy-

sis (EFA). Four main factors were extracted: perfor-

mance, cost, strategy, and inventory policy, with a to-

tal of 15 associated items. The ﬁndings aim to pro-

vide insights for Thai construction companies on im-

proving inventory-related ﬁnancial performance and

project cost management (Jakkraphobyothin, Srifa, et

al. 2018). This research describes how 5G base sta-

tions provide multiple services in various scenarios,

which provides the authors an opportunity to enhance

spectrum efﬁciency. Base stations can ﬂexibly uti-

lize the idle frequency band for spatiotemporal low-

demand services and guarantee services with high pri-

ority. We use the Lyapunov optimization method to

solve the problem (Zhang, He, et al. 2020). This

research study presents a review on signal process-

ing techniques used for performing spectrum detec-

tion in CR networks Cognitive Radio helps to allocate

the available radio frequency spectrum of an essential

client and an auxiliary client. In the cognitive radio

technology also known as Dynamic Spectrum Access

(DSA), secondary clients may take advantage of the

numerous spectrum gaps in allowed spectrum groups

(Poonguzhali, Rekha, et al. 2023). The authors ex-

plore game theory models and graph-coloring mod-

els, which can improve spectrum efﬁciency and solve

the deﬁciency of spectrum resources (Yan, Song, et

al. 2009). Game theory-based cognitive radio dy-

namic spectrum allocation is one of the hot researches

of the ﬁeld of cognitive radio. Taking into account

the differences in the spectrum, using a Cournot game

model, and adding the spectral similarity matrix to the

original pricing function, a new utility function that

makes the spectrum allocation more close to the ac-

tual network (Zhang, He, et al. 2020).

4 METHODOLOGY

Fig. 1 explains how the workﬂow of the code goes.

First, initialize the speciﬁc IoT devices, channels, and

time slots. Next, initialize the population of random

channel allocation. Calculate the penalties for inter-

ference and priority mismatch. Select top solutions

INCOFT 2025 - International Conference on Futuristic Technology

716

Figure 1: Workﬂow Diagram

with the best ﬁtness, combine the solutions to gener-

ate new ones, and introduce diversity by modifying

solutions. Optimize channel usage to avoid interfer-

ence. We then evaluate ﬁtness after graph coloring.

We repeat the above steps until the output is achieved.

The priority-based approach in the dynamic spectrum

allocation helps in ensuring optimal channel usage for

efﬁcient bandwidth utilization and reliable data trans-

mission to the cloud. The initial attempt at under-

standing how this spectrum allocation is done can be

simulated using the help of MATLAB Simulink. The

authorized devices in the context of the IoT devices

dealing with the RFID scanners would be those de-

vices that deal with the

• Dock Door RFID scanners: Track items during

loading and unloading.

• Conveyor belt RFID scanners: Scan items on

the conveyor for automated sorting.

• Handheld RFID scanners (critical areas): used

to check high-value or time-sensitive shipments.

• Checkpoint RFID scanners: Monitor movement

of priority goods like fragile or perishable items.

These devices hold a great signiﬁcance in the deliv-

ery and smooth functioning of the process conducted

at the logistic centers. Hence, these devices have

the channel allocated to them at the ﬁrst. A signal

strength threshold of greater than 0.7 ensures that only

these critical devices get the spectrum and have unin-

terrupted communication.

However, there are also unauthorized devices and

low-priority devices that handle less important data.

In relation to the designated area, these devices could

• Backup Dock Door Scanners: Used for redun-

dancy or in low-trafﬁc areas.

• Inventory Shelf RFID Scanners: Periodically

scans shelves to ensure stock is correctly placed.

• Secondary Checkpoint RFID Scanners: Used

to validate movement of non-priority goods

within the facility.

Although the devices that take care of these also help

in the logistic center’s orderly operation, they are not

crucial during situations. Although this data may con-

tribute to the center’s greater functionality, it is only

somewhat signiﬁcant. Only when the primary users

or authorized devices are served are these devices as-

signed to a channel for communication. It is also pos-

sible to use the idea of spectrum holes to facilitate data

sharing for secondary users. In order to accommo-

date non-critical devices and maintain optimal chan-

nel use, a lower threshold of 0.3 or less is used.

The output of the following methodology is a binary

vector, indicating whether each device is allocated a

channel. But nonetheless, the output shows an equal

share of the channel to all the devices while maintain-

ing priority for critical devices.

Fig. 2 explains how the channel is allocated to the

Figure 2: The graph of allocation

devices based on the priority levels.

The graph here displays the dynamic spectrum allo-

cation for IoT devices with different priority levels.

Time or device IDs are displayed continuously on the

Priority-Aware Genetic Algorithm for Multi-Constraint Dynamic Spectrum Allocation in Logistic Centers

717

X-axis, while the spectrum allocation status is shown

on the Y-axis. Devices with different priorities are

represented by different colored lines on the graph;

high-priority devices are shown as green, low-priority

devices are shown as orange or red, and secondary

or waiting-for-allocation devices are shown as blue.

The splitting of spectrum resources among devices

according to their priority levels is shown in the al-

location pattern. While low-priority devices only re-

ceive spectrum when resources are available or on an

as-needed basis, high-priority devices receive spec-

trum more regularly or continuously.

While secondary devices are given spectrum accord-

ing to availability, this priority-based method guaran-

tees those devices essential to fast operations and cor-

rect shipments receive bandwidth ﬁrst.

The previous methods must have introduced the con-

cepts of how the primary users are preferred and their

data to be sent is vital. But these assign the chan-

nels based on some round-robin fashion where de-

vices are given the channels one by one in a repeating

order based on their priority. This method is simple

and straightforward and ensures that all devices get a

chance to use the channels, but this doesn’t handle the

conditions of interference between the channels. For

example, two devices assigned to the same channel at

the same time can cause interference, which lowers

performance.

This problem is more critical in logistic centers,

where devices need to communicate over a limited

number of channels. The devices with high priority

need reliable and hassle-free channels for data trans-

mission, while the secondary devices should still be

assigned channels that don’t disrupt the high-priority

ones.

To solve these issues, we use a more advanced

method, a genetic algorithm (GA) combined with a

graph coloring model. The GA helps us optimize to

the best channel assignments by evolving solutions

over time. The graph coloring model ensures that de-

vices that might interfere with each other are not as-

signed the same channel. This combined approach

also provides better performance as it considers both

device priority and the need to minimize interference.

The genetic algorithm is a search heuristic inspired by

the method of natural selection, where the best traits

are passed on to future generations. It is a method

used to ﬁnd approximate solutions for optimization

and search problems. It works by imitating how nat-

ural evolution happens. The algorithm runs over a

population of possible solutions, applying selection,

crossover, and mutation to evolve solutions toward

better ones. Fig. 3 displays the ideal genetic algo-

rithm ﬂowchart and illustrates the iterative process of

Figure 3: Flowchart of a Genetic Algorithm

evolving solutions through selection, crossover, and

mutation to optimize complex problems.

• Population Initialization: Random channel as-

signments are created for all IoT devices, where

each channel assignment represents a possible so-

lution

• Fitness Function: This is used to evaluate how

good each solution is by calculating the interfer-

ence between devices. Devices assigned to the

same channel at the same time step cause inter-

ference, which is penalized in the ﬁtness function.

The function also ensures that high-priority de-

vices are given better channel assignments.

• Selection: The best solutions, based on the least

interference and highest priority, are chosen for

the next generation

• Crossover and Mutation: New solutions are

created by combining parts of selected solutions

(crossover) and introducing random changes (mu-

tation). This keeps the population diverse and pre-

vents the algorithm from getting stuck in subopti-

mal solutions.

INCOFT 2025 - International Conference on Futuristic Technology

718

To further reduce interference and optimize channel

assignment, a graph coloring model is integrated into

the genetic algorithm. This model ensures that de-

vices likely to cause interference are not assigned to

the same channel. Using graph coloring as a pre-

optimization step reduces the likelihood of starting

with high-penalty solutions.

• Graph Representation: Each IoT device is rep-

resented as a node. An edge is drawn between the

two nodes if they could interfere with each other.

• Graph Coloring: A unique color is assigned to

each device so that no two devices that interfere

share the same channel. This helps reduce inter-

ference, especially for high-priority devices.

• Integration with GA: The graph coloring model

is used for each generation before evaluating the

ﬁtness function. It ensures that devices with pos-

sible conﬂicts are allocated distinct channels.

It can be highlighted that the genetic algorithms

might be able to offer a novel approach for dynamic

spectrum allocation within the logistic centers. This

variant of genetic algorithms uses priority-based

optimization, ensuring critical devices are assigned

optimal channels with minimum interference. A

ﬁtness function that focuses on interference reduction

is emphasized to support efﬁcient spectrum usage

in resource-constrained IoT settings. Genetic algo-

rithms are potential candidates to be applied to large

logistic centers with thousands of simultaneously

operating devices.

In fact, the integration of graph coloring into the

framework of the genetic algorithm is a ground-

breaking hybrid strategy. It derives all the beneﬁts

from the adaptive exploration by genetic algorithms

combined with organized conﬂict resolution through

graph coloring. It actually enhances the overall

efﬁciency of spectrum allocation with this minimum

interference and observation of priority of various

devices.

To summarize what has been discussed so far,

the ﬁrst approach to allocating channels to devices

was based on the devices’ priorities and signiﬁcance

in the logistics environment. Then it was discovered

that the method of assigning channels to higher

priority devices ﬁrst and then lower priority devices

was very similar to round-robin. This method could

not handle the situation of interference, which occurs

when two devices compete for a single channel to

send data. As a result, a newer algorithm, known

as the genetic algorithm, was developed with some

modiﬁcations in order to handle the interference situ-

ation smoothly while providing the least-interference

combination and a low penalty score situation. The

above technique is combined with a graph-coloring

model, in which no two devices are assigned to the

same channel. The optimized channel assignment is

the result of multiple checks and runs of the genetic

algorithm, and the graphs displayed in the results

section represent the best channel assignment for that

speciﬁc scenario.

5 RESULTS AND OUTPUT

The genetic algorithm optimizes channel assignments

for devices, reducing interference by prioritizing

high-priority devices for the best channels and as-

signing remaining channels to lower-priority devices

to minimize conﬂicts. Interference is measured by

penalties added when devices share the same channel

at the same time, with high-priority devices receiving

less penalty. An interference graph identiﬁes potential

conﬂicts between devices, and graph coloring ensures

connected devices use different channels, improving

efﬁciency and reducing interference.

We take in a hypothetical scenario of having 25 IoT

devices with a total of 10 channels present, and they

compete for the spectrums based on their priorities

over a time step of 20 units of duration.

Fig. 4 depicts the scenario of interference among de-

Figure 4: Interference Graph

vices and allocates channels based on priority levels.

The graph displayed gives an idea of how channels

are allocated to devices over time. The x-axis rep-

resents the time steps, whereas the y-axis indicates

the channel numbers assigned to them. We move for-

ward with the assumption that a channel is allocated

to one device for performance reasons. Each line cor-

Priority-Aware Genetic Algorithm for Multi-Constraint Dynamic Spectrum Allocation in Logistic Centers

719

responds to a speciﬁc device, and the points along the

lines show the channel assigned to that device at a

given step. The different patterns of the lines show

dynamic channel allocation, ensuring devices use dif-

ferent channels over time to minimize interference

and optimize spectrum usage. Fig. 5 depicts the inter-

Figure 5: Interference graph with data

ference scenario among devices and shows their pri-

orities and the data they carry.

The graph above depicts the scenario of having 10

channels and 15 devices over a 10-unit time period.

Each device’s priority and RFID tag data are dis-

played, and if there is interference, the devices are

assigned different channels based on their priorities

and ﬁtness score evaluation.

The graph depicts the best-case scenario of low inter-

ference and a lower penalty score among all combina-

tions of assigning 15 devices to 10 channels in a time

frame of 10 units. This could be achieved through

optimal channel assignments and the evolution of so-

lutions over time.

Realistic application of the proposed Genetic

Algorithm (GA) with Graph Coloring can be im-

plemented by developing an IoT central controller.

Several IoT devices, for instance, sensors and

communication nodes, are connected to a central

server to develop an interference graph in a conﬂict

discovery process. The GA assigns channel priority

to critical devices and reduces channel interference,

while graph coloring makes sure that two interference

devices will not be assigned the same channel. These

optimal channel allocations are therefore transmitted

to the devices through wireless communications. A

performance monitoring system is implemented, and

the algorithm can dynamically adjust the channel if

devices are added or change the interference pattern

to guarantee effective and stable communications.

6 CONCLUSION AND FUTURE

SCOPE

This study presents us with a MATLAB-based frame-

work for dynamic spectrum allocation to meet the de-

mands of logistic centers. Our study addressed many

challenges, such as spectrum scarcity, resource allo-

cation, bandwidth allocation, and reliable communi-

cation between IoT devices. The integration of algo-

rithms like Round Robin, priority-based algorithms,

Genetic Algorithms with graph coloring ensures all

the challenges are tackled, even in high-density IoT

environments. This work highlights the importance

of dynamic and adaptive spectrum management tech-

niques in overcoming the limitations of existing sys-

tems.

Future scope of the study would be

• Increasing the scale of the IoT devices, extending

our framework to accommodate a large number of

IoT devices.

• Incorporating machine learning techniques and al-

gorithms to predict the spectrum usage patterns.

• Utilizing energy-harvesting techniques to support

sustainable IoT devices.

• Integrating the spectrum with security systems to

safeguard against unauthorized access and inter-

ference in logistic networks.

REFERENCES

I. Y. CR, A. R, P. T. Ramana, P. Shilpa and J.

G, ”Comparative Performance Analysis of Ge-

netic Algorithm Variants on Solving 0/1 Knap-

sack Problem,” 2023 IEEE World Conference

on Applied Intelligence and Computing (AIC),

Sonbhadra, India, 2023, pp. 125-130, doi:

10.1109/AIC57670.2023.10263822.

A. R, I. Y. CR, P. T. Ramana, P. Shilpa and J. G,

”Optimal Ship Route Search Based on Multi-

Objective Genetic Algorithm,” 2023 14th Inter-

national Conference on Computing Communica-

tion and Networking Technologies (ICCCNT),

Delhi, India, 2023, pp. 1-5, doi: 10.1109/ICC-

CNT56998.2023.10307569.

U. S. Vaishnav and M. Panda, ”Resource alloca-

tion for device-to-device (d2d) communication

INCOFT 2025 - International Conference on Futuristic Technology

720

in underlaying cellular network,” 2017 Interna-

tional Conference on Advances in Computing,

Communications and Informatics (ICACCI),

Udupi, India, 2017, pp. 1424-1438, doi:

10.1109/ICACCI.2017.8126040.

R. Ravi, ”Supply chain management in retail us-

ing Radio frequency identiﬁcation(RFID),” 2010

IEEE 17Th International Conference on In-

dustrial Engineering and Engineering Manage-

ment, Xiamen, China, 2010, pp. 1907-1911, doi:

10.1109/ICIEEM.2010.5645898.

V. G. Lekshmy, P. K. Anusree and V. S. Varunika, ”An

Implementation of Genetic Algorithm for Clus-

tering Help Desk Data for Service Automation,”

2018 International Conference on Advances in

Computing, Communications and Informatics

(ICACCI), Bangalore, India, 2018, pp. 952-956,

doi: 10.1109/ICACCI.2018.8554532.

M. Sony and R. Manjusha, ”Chromatic Partition-

ing of a Network and an Optimum Schedul-

ing Problem,” 2023 Third International Con-

ference on Advances in Electrical, Computing,

Communication and Sustainable Technologies

(ICAECT), Bhilai, India, 2023, pp. 1-6, doi:

10.1109/ICAECT57570.2023.10117620.

A. Ikami, T. Hayashi and Y. Amano, ”Dynamic Chan-

nel Allocation Algorithm for Spectrum Shar-

ing between Different Radio Systems,” 2020

IEEE 31st Annual International Symposium on

Personal, Indoor and Mobile Radio Commu-

nications, London, UK, 2020, pp. 1-6, doi:

10.1109/PIMRC48278.2020.9217152.

A. Ikami, T. Hayashi and Y. Amano, ”Fair Spectrum

Allocation for Dynamic Spectrum Sharing

between Different Radio Systems,” 2020

23rd International Symposium on Wire-

less Personal Multimedia Communications

(WPMC), Okayama, Japan, 2020, pp. 1-6, doi:

10.1109/WPMC50192.2020.9309506.

Hoon Kim, Yeonwoo Lee and Sangboh Yun, ”A dy-

namic spectrum allocation between network op-

erators with priority-based sharing and negotia-

tion,” 2005 IEEE 16th International Symposium

on Personal, Indoor and Mobile Radio Commu-

nications, Berlin, 2005, pp. 1004-1008 Vol. 2,

doi: 10.1109/PIMRC.2005.1651592.

P. Li, Z. Zhao, D. Liu and D. Hou, ”The Research of

Dynamic Spectrum Allocation Based on Game

Theory,” 2018 IEEE 3rd Advanced Information

Technology, Electronic and Automation Control

Conference (IAEAC), Chongqing, China, 2018,

pp. 14-17, doi: 10.1109/IAEAC.2018.8577243.

K. Jakkraphobyothin, S. Srifa and T. Chinda,

”Factor Analysis of Inventory Management in

Thai Construction Industry,” 2018 3rd Tech-

nology Innovation Management and Engineer-

ing Science International Conference (TIMES-

iCON), Bangkok, Thailand, 2018, pp. 1-5, doi:

10.1109/TIMES-iCON.2018.8621817.

Y. Zhang, D. He, W. He, Y. Xu, Y. Guan and W.

Zhang, ”Dynamic Spectrum Allocation by 5G

Base Station,” 2020 International Wireless Com-

munications and Mobile Computing (IWCMC),

Limassol, Cyprus, 2020, pp. 1463-1467, doi:

10.1109/IWCMC48107.2020.9148584.

M. Poonguzhali, M. Rekha, A. Sumathi, N. Duraichi,

K. Jeyalakshmi and S. Sattanathan, ”Study of

Cognitive Radio Techniques for Dynamic Spec-

trum Allocation,” 2023 4th International Confer-

ence on Smart Electronics and Communication

(ICOSEC), Trichy, India, 2023, pp. 488-494,

doi: 10.1109/ICOSEC58147.2023.10276258.

X. Yan, Q. Song, H. Zhang and L. Shao, ”Dy-

namic Spectrum Allocation based on Cog-

nitive Radio,” 2009 5th Asia-Paciﬁc Con-

ference on Environmental Electromagnet-

ics, Xi’an, China, 2009, pp. 254-257, doi:

10.1109/CEEM.2009.5303943.

Xin-chun Zhang, Shi-biao He and Jiang Sun, ”A

game algorithm of dynamic spectrum allocation

based on spectrum difference,” The 19th Annual

Wireless and Optical Communications Confer-

ence (WOCC 2010), Shanghai, 2010, pp. 1-4,

doi: 10.1109/WOCC.2010.5510648.

Yu, Xiaomo, Cai, Yonghua, Li, Wenjing, Zhou,

Xiaomeng, Tang, Ling, Research on Dynamic

Spectrum Allocation Algorithm Based on Cyclic

Neural Network, Wireless Communications and

Mobile Computing, 2022, 7928300, 14 pages,

2022.

Priority-Aware Genetic Algorithm for Multi-Constraint Dynamic Spectrum Allocation in Logistic Centers

721