Mobile Networks and Applications最新文献

In response to the local path planning issue encountered by Autonomous Underwater Vehicle (AUV) during autonomous navigation when facing sudden threats or obstacles, a rapid path planning repair solution based on the IRRT*-VSRP method is proposed in this paper. This method combines an enhanced RRT* algorithm with a threat-based variable step-size receding horizon predictive strategy, effectively reducing the search space in three-dimensional environments. Its notable features include rapid local path repair and generation, thereby improving the success rate and efficiency of planning. Simulation results demonstrate that the IRRT*-VSRP algorithm significantly reduces the time required for planning repair and enhances the directionality of tree expansion, rendering it suitable for complex underwater three-dimensional environments and enhancing the efficiency of AUV planning repair.

This article presents a performance analysis of wireless energy harvesting (WEH)-enabled sensor networks that extract energy from ambient radio frequency (RF) signals prior to uplink transmission. A time-switching (TS)-based protocol is utilized to alternate sensor nodes between energy harvesting (EH) and data transmission modes. Implementing the non-orthogonal multiple access (NOMA) technique aims to boost the sensor network’s performance regarding uplink sum rate and outage probability. To optimize resource allocation, we propose an unequal operating time frame (OTF) scheme that determines data transmission and energy harvesting intervals based on channel gain quality. Simulation results affirm the superiority of NOMA over orthogonal multiple access (OMA), with NOMA enabling higher sum rates by accommodating more signals within the same frequency band, though at the expense of slightly degraded outage performance.

With the rapid development of cognitive computing and the Internet of Things (IoT), sensing systems have produced a wide range of real-time communication applications. They use 5G/6G-enabled technologies to connect to the outside world to collect data and process different end-user requests. Wireless systems and artificial intelligence (AI) have led to significant development in the optimization process of network communication. Due to various constraints of wireless systems, many solutions have been presented to cope with routing and connectivity concerns. However, topology awareness and attaining management of quality of services are still demanding research challenges for sustainable development. This study proposes an AI-assisted routing model for mobile wireless sensor networks (MWSN) to optimize energy and detect communication link failures. Moreover, the proposed intelligent security approach increases the trustworthiness of the constraint devices on unpredictable routes. Firstly, it explores a genetic algorithm, a metaheuristic optimization technique to determine the feasible solutions, and based on independent metrics it generates an optimal set of routes. In the proposed model, the genetic algorithm provides a fault-tolerant solution for dynamic environments, specifically under unpredictable conditions. Second, new routes are established using dynamic decisions that satisfy the energy considerations. In the end, the proposed model performs regular auditing to detect malicious devices based on unexpected behavior. The proposed model is tested and it outperforms IMD-EACBR and AGRIC in terms of realistic performance metrics.

The rising incidence of brain tumors in the medical field necessitates the development of precise and effective diagnostic tools to assist the medical experts especially neurosurgeons as well as radiologists in early diagnosis and treatment recommendations. This study introduces a unique resource-efficient inceptor model utilizing computer-vision techniques for diagnosing presence of abnormal tissues inside brain MRI scans. The proposed model utilizes strengths of the inception architecture and incorporate resource-efficient design principles for optimizing its performance for healthcare applications. The model has been trained on a distinct dataset with different sizes where it is further processed, trained and validated on InCeptor model. Features are extracted by transfer learning process namely InceptionV3 for leveraging prior knowledge learnt from imagenet which is further integrated with support vector machines for performing binary classification to have accurate and efficient outcomes for giving timely recommendation and treatment to patients suffering from such disorder. The architecture of the proposed model has been designed in such a way that model should be computationally efficient for making it suitable in healthcare especially for brain tumor diagnostic purpose with limited resources. Experimental results demonstrates accuracy of 98.31%, precision of 99.09%, recall of 98.91%, specificity of 95% and F1- Score of 99% over state of art techniques.

Owing to the variety of ways to view the Internet and the changes in user behavior on the Internet, network traffic has been explosively growing in recent years. Users can watch high-quality videos on the Internet; it is a critical issue to reduce network traffic and increase the user's quality of experience (QoE). Therefore, there have been in-network caching services that cache the content that had been fetched by the user in a proxy server. Meanwhile, the software-defined network (SDN) has been developed to implement the network function through the virtualization function. Programmers can implement customized network functions using the SDN architecture. In this paper, we proposed an Extended SDN Cache service architecture (ESC). The ESC decomposes the function of inspecting incoming traffic, making cache decisions, and caching content to three network entities. This design can reduce the load of a single network entity. To reduce the load of the SDN controller, we utilize an extended OpenFlow switch named the DPI (deep packet inspection) switch, which can inspect the incoming traffic. The ESC designed a mechanism that can cache the different parts of a video in distinct cache nodes. The distributed content storage mechanism can increase the cache capability and the system's flexibility. We use the M/M/1 queuing model to analyze the average queuing delay time and compare the ESC queuing delay time with the C-flow and the OpenCache. The numerical analysis results show that the ESC queuing delay is shorter than the other two.

With the development of cloud computing and edge computing technologies, these technologies have come to play a crucial role in the field of autonomous driving. The autonomous driving sector faces unresolved issues, with one key problem being the handling of latency-sensitive applications within vehicles. Cloud computing and edge computing provide a solution by segmenting unresolved computing tasks and offloading them to different computing nodes, effectively addressing the challenges of high concurrency through distributed computing. While the academic literature addresses computation offloading issues, it often focuses on static scenarios and does not fully leverage the advantages of cloud computing and edge computing. To address these challenges, a multivariate particle swarm optimization (MPSO) algorithm tailored for the cloud-edge aggregated computing environment in the autonomous driving domain is proposed. The algorithm, grounded in real-world scenarios, considers factors that may impact computation latency, abstracts them into quantifiable attributes, and determines the priority of each task. Tasks are then assigned to optimal computing nodes to achieve a balance between computation time and waiting time, resulting in the shortest total average weighted computation latency time for all tasks. To validate the effectiveness of the algorithm, experiments were conducted using the self-designed CETO-Sim simulation platform. The algorithm’s results were compared with those of simulated annealing, traditional particle swarm optimization, purely local computation, and purely cloud-based computation. Additionally, comparisons with traditional algorithms were considered in terms of iteration count and result stability. The results indicate that the MPSO algorithm not only achieves optimal computation offloading strategies within specified time constraints when addressing computation offloading issues in the autonomous driving domain but also exhibits high stability. Furthermore, the algorithm determines the processing location for each computing task, demonstrating significant practical value.

Next-generation wireless networks (NGWN) consist of the integration of various technologies, such as Mobile ad-hoc networks (MANET), Wi-Fi, WiMAX, and LTE which are connected to the internet. Switching off the nodes among networks with same or different technology is handled by mobile IP. The determination of hand-off is not solely reliant on received signal strength, as relying solely on this metric could result in unnecessary hand-offs. Various factors, such as power consumption in communication, delay, traffic load, and network bandwidth, also play crucial roles in ensuring successful transmission. This paper introduces a seamless hand-off technique based on Markov processes (S-MSH), which takes into account different network properties that impact the Quality of Experience (QoE) for mobile terminals (MT) during communication. The proposed approach focuses on creating a Markov Decision Process (MDP) model for the system, considering user traffic requirements. The Q-learning algorithm is applied to the model to predict whether a hand-off is beneficial. An integrated similarity index-based approach, termed S-MSH, has been introduced to expedite the convergence rate of MSH. Simulation and numerical results demonstrate that the proposed approach surpasses the performance of the Network Priority Multicriteria Vertical Handover Decision Algorithm (NPMH) and the Simple Additive Weighing Algorithm (SAW) in terms of total reward and the number of handoffs.

Abstract

Aiming at the connectivity failure caused by the failure of mobile network nodes in complex application scenarios, a minimum cascade repair method for the failure of mobile network nodes under time–frequency feature fusion is proposed. The short-time energy and zero-crossing rate of time-domain features of mobile network nodes were extracted.The frequency domain features of network nodes are extracted by fully connected long and short time memory network. In the form of weighting and series, the time domain and frequency domain features are fused to obtain the time–frequency fusion features. The failure node is detected by using the adaptive mechanism of node clustering feature combined with time–frequency fusion feature. Judging whether the failed node is a cut point according to the rule; If the failed node is a cut point, the network needs to be repaired. Considering the impact of faulty nodes on network communication and connectivity in the real world, an innovative minimum cascade repair algorithm is introduced to establish communication links to repair the network topology without moving nodes, achieving optimization of repair rules. Calculate the energy consumption of the repair moving distance of the failed node, based on the optimized repair rules. If the energy consumption is less than or equal to the energy consumption of the communication radius of the neighboring node, the mobile network can be repaired statically by directly establishing a communication link between the neighboring nodes. Otherwise, the best candidate node is selected and its location to be moved is calculated to complete the mobile network mobile repair. Experimental results show that this method can effectively repair the network after node failure. This method reduces the energy loss of the repair node and improves the status indicator value of the mobile network.

After the low-earth orbit (LEO) satellite Internet has gone through the two stages of competing with the terrestrial network and supplementing the terrestrial network, it has begun to enter the third stage of constructing the satellite-ground integrated network with the terrestrial network to provide seamless global coverage. 5G New Radio (NR) is one of the core enabling technologies of the third stage of satellite Internet. This paper focuses on how to make full use of the power and bandwidth resources on the LEO satellite by using adaptive transmission scheme to maximize the throughput of the user downlink based on 5G NR. To solve the problem that the ultra-long propagation delay, outdated channel state information (CSI) and dynamic multi-scenario of LEO satellite will lead to the high implementation cost and greatly reduced performance when applied the 5G adaptive transmission scheme to LEO satellites, we optimized the adaptive transmission scheme of 5G NR based LEO satellite from multiple dimensions such as adaptive transmission process, signal to noise ratio (SNR) prediction and modulation and coding scheme (MCS) adaptive switching strategy. The simulation results show that compared with the fixed threshold switching strategy based adaptive transmission scheme, the proposed scheme can improve the average throughput of the system by 26.6% under the dynamic multi-scenario environment served by the LEO satellite.

A large amount of randomly generated noise in mobile networks leads to a lack of targeting and gaming processes in the speech enhancement process, and the enhancement process from the perspective of acoustic features alone suffers from major drawbacks. Propose a single-channel speech quality enhancement method based on generative adversarial networks in mobile networks. Explain the principle of generative adversarial network to realize single-channel speech quality enhancement in mobile networks and clarify its shortcomings. Design an improved Mel frequency cepstral coefficient extraction method to extract 12 base features as the enhancement basis. Use the relative average least squares loss instead of the traditional loss function to enhance the training efficiency, use the hybrid penalty term to enhance the generator's ability to generate single-channel speech, and optimize the discriminator through the multi-layer convolution and the addition of fully connected layers to enhance the speech quality enhancement ability of adversarial generative networks in various aspects, forming a relative average generative adversarial network (RaGAN) based on hybrid penalty term to realize single-channel speech quality enhancement processing. Through the experiment, when the discriminator is applied with the size of a 3*3 convolutional kernel, the best effect of speech quality enhancement is achieved in the mobile network. This method can complete the enhancement of single-channel speech quality in the mobile network, and the effect is significant, which can effectively reduce the noise in the original single-channel speech.