查看更多>>摘要：In this paper two new Cyclic Prefix (CP) based blind Carrier Frequency Offset (CFO) estimations for Orthogonal Frequency Division Multiplexing (OFDM) transmission over multipath channels are proposed. In doing so, we first estimate the maximum delay of the fading channel. We borrow the concept of remodulation introduced in earlier works and use the repetitive structure of CP to calculate a maximum-likelihood-based measure. In the first proposed method we use Particle Swarm Optimisation (PSO) aided search on all possible samples to find the optimal set. This technique provides performance improvement at the expense of more complexity. Then, in a second proposed method we average over the optimal set of samples to estimate CFO. The second technique provides a major improvement over previous works while offering less complexity. Simulation results corroborate that both our proposed methods significantly decrease the Mean Square Error (MSE).
查看更多>>摘要：Vehicular Named Data Networks (VNDN) is a content centric approach for vehicle networks. The fundamental principle of addressing the content rather than the host, suits vehicular environment. There are numerous challenges such as interest/data flooding, routing, naming schemes, channel constraints and security, among others. The proposed work aims to analyse the behaviour of the naive VNDN under diverse scenarios, which is crucial for devising effective forwarding, caching and security strategies to enhance VNDN's performance. In this paper, the naive VNDN architecture is implemented and its performance is evaluated by considering the metrics such as Throughput, Packet Delivery Ratio (PDR), Interest Satisfaction Delay (ISD), Interest Satisfaction Ratio (ISR), Copies of Data Packets Processed (CDPP) and Hop Count Difference (HCD). These metrics are tested against varying vehicle speed, network size and interests generated by each vehicle in the network. The simulation results provide insights that should be considered while developing the forwarding strategies and caching mechanisms that will improve the performance of VNDN.
查看更多>>摘要：Wireless communication has grown tremendously in recent years, impacting nearly every feature of our lives. The increased exigency for wireless broadband services leads to a huge demand for dynamic spectrum access, such as TV White Spaces (TVWS). The possibility of interference increases with the high density of users, reducing the Spectrum Efficiency (SE). SE improvement techniques have been utilised to resolve network congestion issues. This paper presents the Orthogonal Frequency Division Multiplexing (OFDM) wireless communication model with the Quadrature Amplitude Modulation (16-QAM) technique for TVWS SE gain. The performance analysis of the proposed model is based on parameters such as Packet Loss, Bit Error Rate (BER) and Signal-to-Noise Ratio (SNR). The constellation diagrams and signal trajectories are used to evaluate the efficacy of the modulation technique. Zero packet loss is achieved by implementing the 16-QAM OFDM, demonstrating efficient usage of TVWS with extended coverage and almost homogeneous data distributions.
查看更多>>摘要：The traditional fault localisation methods are always based on rough modelling of causal relationship between fault types and performance, ignoring the complex network dynamic characteristics of complex systems, especially the influence of community structure on fault propagation, which leads to the poor fault localisation result. In this paper, a new fault propagation model based on community structure is proposed, firstly the structure and characteristics of complex equipment system are transformed into a complex network, and then the community structure is detected. Finally, we realise time-continuous fault diagnosis, incorporate the dynamics of the immune networks to the fault propagation model. In the stage of detecting community structure, the improved SimRank method is adopted and the fault rate is set according to the calculated object, which effectively improves the defect that the traditional algorithm is easy to be disrupted by independent objects. The experimental results show that the algorithm based on community structure can achieve accurate fault localisation.
查看更多>>摘要：Rice plays a crucial role in agriculture, but a major issue is that various disasters and diseases of rice will greatly reduce rice production, especially affecting rice required for human consumption, and the rice seeds sown in the next year will also encounter problems. The learning nature of convolutional neural networks is used to identify rice ear diseases, and rice ear disease recognition rate is improved by modifying the network structure and integrating other network structures that can enhance deep learning for picture recognition. In this study, MobileNetV2 is used as the main network and trained on ImageNet using migration learning. The underlying convolutional layer weights are frozen to conserve resources. Then, the pre-trained MobileNetV2 network is fused with BAM blocks to develop a new network. Experiments show that the efficiency and recognition rate of this method are improved, with an average recognition rate of 98.18%. The generalisation ability of the model is then tested on the PlantVillage data set, with an average recognition rate of 98.7%. The results of the experiments show that the model can effectively improve image recognition, and the generalisability of the model is also guaranteed.
查看更多>>摘要：The RPL specification encrypts control messages, but internal attackers and self-serving behaviour can still exploit RPL. To encounter the issues of the lack of robust security mechanisms in RPL, this paper concentrates on the modelling of energy and trust-aware RPL routing models in mobile IoT via an advanced hybrid optimisation algorithm. This proposes the algorithm Manta-Ray Updated White Shark Optimisation (MRUWSO). The four stages are: DIO messages are broadcast by the root node, and every node that receives a DIO message updates its routing tables with the root. The optimal route establishment is done by a proposed MRUWSO algorithm. Thirdly, optimal routing will be done under the improved trust evaluation function that ensures the authentication. Lastly, if the preferred parent changes or a route is updated, every time, a DAO message needs to be sent. The proposed MRUWSO attained a minimal delay of 1.21×e~(-07) under 500 nodes which is superior.
查看更多>>摘要：This paper studies the vehicle scheduling problem of feeder bus, constructs a mixed integer nonlinear programming model to optimise the departure schedule of electric bus, and uses the improved artificial bee colony algorithm to solve the model. Then, based on the 5G communication system, an urban rail transit vehicle transportation scheduling management system is constructed to achieve transportation bus data collection, task scheduling management, vehicle management, station management, etc. The experimental results obtained by solving the model through parameter calibration show that the total cost of the optimised timetable bi-directional operation route is reduced by an average of about 22.83%, which verifies the good effect of the optimisation model constructed in the study. The experimental results show that the model can handle non-linear objectives and constraints to reflect the non-linear relationships in reality, and has good application effects for the management of transportation bus scheduling.
查看更多>>摘要：To achieve accurate Direction-of-Arrival (DOA) estimation of far-field targets, distributed aperture arrays are employed in this work. Compared with traditional aperture array radars, distributed aperture arrays can expand the physical aperture of the array, leading to improved angle estimation performance. However, the spacing between sub-arrays in distributed aperture arrays is significantly greater than half a wavelength, violating the Nyquist sampling theorem. This results in multiple grating lobes in the echo directional diagram, causing ambiguity in angle estimation. A novel high-resolution angle estimation method based on subarray angle estimation for distributed aperture arrays is proposed. The method follows a two-step coarse-fine process and is applicable to arbitrarily configured distributed arrays and effectively addresses the issue of angle ambiguity. The estimation performance surpasses existing methods, achieving an estimation error of about 0.04° at a signal-to-noise ratio of 10 dB.
查看更多>>摘要：One key reason for the gap between deep reinforcement learning goals and applications is that trained intelligences overfit local training data features, and models trained with perception-decision in a single network are sensitive to small environmental changes. This restricts the agents from learning real rules from experience. In this paper, we use a target detection model to recognise objects in a visual scene, learn perceptual level knowledge and use the generalisation ability of the target detection model to reduce the observation overfitting of the agent and improve the robustness of the agent in vision. After the object instances are extracted by the target detection model, we use a model structure equipped with a relational inference mechanism to model the relationships between the objects, which further improves the model's ability to generalise to unseen scenes by introducing a relational inductive bias.
查看更多>>摘要：Estimation of precise and cost-effective mapping of soil textures is crucial to track the soil properties and geographical distribution of heavy metals in soil for sustainable soil utilisation. Conventional Digital Soil Mapping (DSM) algorithms face difficulty in modelling the soil spatial variation resulting in suboptimal performance. Hence, a novel Corvis-Echo Optimised Multitasks Learning-based Deep Convolution Neural Network (CEO-based MTL Deep CNN) is proposed for generating DSMs effectively. In the proposed research, the MTL-based Deep CNN model is exploited to capture the local patterns and spatial relationships from the input data and facilitates simultaneous modelling of various soil properties. Specifically, the Corvis-Echo Optimisation (CEO) is applied for fine-tuning the hyperparameters of the classifier to attain better performance. The experimental results demonstrate that the proposed model attains the accuracy, precision, recall, F1-measure, and delay of 98.81%, 99.55%, 98.75%, 99.15%, and 317.61 ms respectively by varying the epochs 100.
NETL
NSTL