Under various deployment circumstances, fifth-generation (5G) telecommunications delivers improved network compound management with fast communication channels. Due to the introduction of the Internet of Things (IoT) in data management, the majority of the ultra-dense network models in 5G networks frequently have decreased spectral efficiency, weak handover management, and vulnerabilities. The majority of traditional handover authentication models are seriously threatened, making them vulnerable to a variety of security attacks. The authentication of networked devices is the most important issue. Therefore, a model that incorporates the handover mechanism and authentication model must be created. This article uses a fuzzy logic model to create a handover and key management system that focuses on cloud handover management and authentication performance. In order to decrease delays in 5G networks, the fuzzy logic is built with multiple criteria that aim to reduce the number of executed handovers and target cell selection. The simulation is run to evaluate the model’s performance in terms of latency, spatial complexity, and other metrics related to authentication attack validation.

The process of making medical decisions is naturally hazy. When determining the diagnosis and prognosis, the doctor uses linguistic concepts. Fuzzy logic design is becoming increasingly important in the context of the steady development of artificial intelligence in healthcare for enhancing the quality of therapeutic and diagnostic effectiveness. Fuzzy logic (FL) offers a useful method for addressing uncertainties in the process of making health-related decisions; as a result, FL-based design becomes a very potent instrument for data and knowledge management, allowing users to think like expert clinicians. This proposed work suggests employing a fuzzy logic design to construct systems that tend to lessen the effects of handover and effectively authenticate the data.

The 3GPP standards allow advanced long-term evolution (LTE-A) in 5G networks to communicate with other wireless networks, resulting in improved coverage, cost, and efficiency [

The fundamentals of key agreement and authentication protocols are used in 5G networks, so authentication protocols must be strengthened to meet the demands of this technology [

The existing protocols used in 5G networks have been reinvented in this article to present an authentication mechanism [

In this paper, we develop a handover and key management system using a fuzzy logic model that tends to address the performance of handover management and authentication in the cloud. The fuzzy logic is designed in a multi-criterion manner that reduces the handover executed and targets the cell selection in 5G networks to reduce delays. The simulation is modelled to examine how well the suggested handover method performs in terms of latency, spatial complexity, and other criteria for authenticating an assault.

The following is the paper’s primary contribution:

In order to address the performance of handover management and authentication in the cloud, the authors develop a fuzzy logic model for a novel handover and key management system.

Ozhelvaci et al. [

Singh et al. [

Alezabi et al. [

Torroglosa-Garcia et al. [

Ozhelvaci et al. [

Nyangaresi et al. [

Huang et al. [

In this paper, a model was developed using a fuzzy logic design that tends to reduce the effects associated with handover and authenticating the data in an effective manner.

The model considered for the analysis is treated as a multi-objective function that holds power density, received carried power, traffic intensity, power density, the velocity of user mobility, and probability of call blocking. Initially, the received power is defined as below:

λ-Wavelength of signal (m),

d-Distance of UE of gNB (m),

_{t}

_{t}

_{r}

The power density _{D}

The study uses modified Stanford University Interim (MSUI) to estimate the path loss, whose expression is given below

α-Slope correction factor

_{0}-Reference distance

_{L}_{0})-Path loss of gNB,

_{f}

_{h}

The study adopts the traffic intensity model _{i}

The Erlang C formula is the model for the probability of blocking _{b}

Here,

The target gNB (TgNB) selection is made available via fuzzy logic control that optimizes the selection via predefined rules. The fuzzy logic model is designed as a five-layered model as in

The initial layer for the fuzzy systems is the fuzzy layer and it uses input variables (Crisp one). These variables are translated into a linguistic variable that involves three fuzzy sets that include low, medium, and high. After translation, the membership function is applied to each input translated variable.

The fuzzy set

Henceforth the truth level derivation for the rules is carried out with the following expression:

The antecedent of rule (say

where,

The triggered rule output is computed with reference to the membership function and rule base. Upon computation of the triggered rule output, each rule is aggregated into a unique set as expressed below:

The unique output is thus transformed into a crisp output value as defined below:

where,

The crisp output is computed at each node and it is then sent to the product layer with the following criteria:

_{i}_{i}

The rule firing strengths are estimated using the following equation:

The above equation is normalized and it is given as follows:

The nodes are made adaptive by computing the dynamic function in the de-fuzzy layer:

Finally, the outputs of the de-fuzzy layer are summed and it is given below:

While handover operations are carried out, the study further considers key management as an essential part of the operation to authenticate the data from IoT devices in a secured manner. To do this, we use an intra-3GPP [

Source gNB (SgNB)

TgNB

Authentication Server Function (ASF)

Mobility Management Function (MMF)

Authentication processing function (APF)

The architecture with 3GPP specifications is given in

Consider a next-hop parameter for estimating the key (KMMF with K being the pre-stored secret key in USIM and APF) at UE and MMF. The MMF sustains both key management and authentication when combined with UE in relation with ASF and APF. To secure the chaining counter at the next hop, the study derives the integrity key (_{I}_{C}_{MMF}_{S}_{ASF}_{SF}_{AMF}_{AMF}

The communication link between the non-access network and UE is secured using _{NF}_{NASe}_{RRCe}_{gNB}_{gNB}

In this section, we present the discussion on various performance metrics via simulation of the proposed fuzzy logic system. The simulation is conducted in a MATLAB environment to simulate the behavior of the fuzzy model. The proposed method is compared with state-of-art handover and authentication models in terms of space complexity, communication overheads, executed handovers, and latency in handover. The list of parameters considered for designing the system model is given in

Parameters | Value |
---|---|

MSUI-Reference distance | _{0} = 1 |

SUI-Reference distance | _{0} = 1 |

Slope correction factor | α = 0.9 |

Shadowing correction | |

Maximum distance of eNB-UE | |

Frequency of transmission | |

Transmit power of gNB | _{t} |

Height of transmitter antenna | _{t} |

Height of user | _{0} = 1.5 m |

Gain of transmitter antenna | _{t} |

Coverage height of receiving antenna | _{h} |

Frequency correction | _{f} |

Path loss exponent | |

Path loss on free space |

Requires MATLAB

Simulink required to use Fuzzy Logic Toolbox block library

Global Optimization Toolbox recommended for fuzzy inference system tuning

The other parameters used in the simulation are mentioned in

The method in a and c records the highest communication overhead, whereas methods with fuzzy logic offer reduced overhead, where the least overhead is recorded by the proposed model. The higher overhead is effectively managed by the proposed model, where the data is exchanged between three network entities that include the SgNB, UE, and TgNB. However, in the existing models, the data is exchanged only between the SgNB and TgNB, and without including the UEs, the rate of communication overhead appeared high. The results show that the proposed fuzzy logic design on handover modeling exhibits an improvement of 25% than the other methods in terms of reduced communication overhead.

The space complexity between the proposed and conventional models is estimated in terms of overall message size i.e., it is estimated between 108 bits to 2048 bits. The result of space complexity is given in

The result of space complexity shows that the proposed fuzzy logic design obtains a least message size than the existing methods. The size of overall messages with 3GPP R16 specification has a reduced complexity of 42.9% than the other methods. The reduction in the space complexity is due to efficient design and low complex mechanisms in handling the handover and authentication protocols. Such effective design reduces the power and resource constraints at the access points of 5G networks, which makes the radio transmission to be efficient than other models. Such achievement in space complexity shows the efficient design with reduced memory requirement for handling the handover execution.

The result of the handover latency is given in

The reduction in latency while handling the handovers is due to proper measurement of network parameters and optimal utilization of TgNB by the fuzzy logic controller. However, the existing models tend to handle the TgNB, UE and SgNB in handling the handover latency.

With the 3GPP R16 specification, the proposed method is again tested with existing models in terms of the total number of handovers executed. It is seen that the proposed model exhibit few handovers than other existing models in

With multiple parameters, while considering the simulation of the proposed fuzzy logic design, the triggering decision on handover is made effective by the proposed model, where the total number of handovers is utilized in a minimum manner. Such minimization in handover helps to reduce the risk of handling the entire handover process.

In fuzzy logic design, the pseudo-identity of UE has improved its authentication and anonymity after handover, where it reduces the possibility of attacks in the network. In conventional models, it is seen that the pseudo-identity of UE is constant over the longevity of time and hence the possibility of attack is high in conventional models and in 5G systems. The proposed fuzzy logic model is carried out in such a way that it avoids the possibility of de-synchronization attacks. This is due to vertical key derivation during the key management, where the existing model uses horizontal techniques. The encapsulation of NCC in EEs enables the system to avoid eavesdropping and de-synchronization attacks, where the validity of encapsulation is tested in terms of three entities that include SgNB, UE, and TgNB using timestamp agreement and random parameters.

In this article, we develop a handover and authentication model using the fuzzy logic design that tends to reduce the effects associated with handover and authenticating the data in an effective manner. With suitable authentication using the fuzzy logic, the vulnerabilities are curbed making the system resilient to DoS attacks in making the HO’s fail and jamming attacks in modifying the NCC, and other attacks that include traceability, confirmation, de-synchronization, and replay attacks. Comparing the study to previous methods, the handover delay is reduced by between 3 and 4 percent. The work successfully lowers the overhead and latency related to calculation for handover. Additionally, it can be shown that the space complexity is substantially smaller than that of the alternative approaches, demonstrating the effectiveness of the handover mechanism over cutting-edge models. The use of deep learning models in future work may achieve a stronger focus on controlling mobility and handover on UE and gNB.

The authors received no specific funding for this study.

The authors declare that they have no conflicts of interest to report regarding the present study.