DeepGCNs-Att: Point cloud semantic segmentation with contextual point representations

Graph Convolutional Networks are able to characterize non-Euclidean spaces effectively compared with traditional Convolutional Neural Networks, which can extract the local features of the point cloud using deep neural networks, but it cannot make full use of the global features of the point cloud for semantic segmentation. To solve this problem, this paper proposes a novel network structure called DeepGCNs-Att that enables deep Graph Convolutional Network to aggregate global context features efficiently. Moreover, to speed up the computation, we add an Attention layer after the Graph Convolutional Network Backbone Block to mutually enhance the connection between the distant points of the non-Euclidean space. Our model is tested on the standard benchmark S3DIS. By comparing with other deep Graph Convolutional Networks, our DeepGCNs-Att’s mIoU has at least two percent higher than that of all other models and even shows excellent results in space complexity and computational complexity under the same number of Graph Convolutional Network layers.

Improved Unequal-Clustering and Routing Protocol

We propose improved unequal-clustering and routing protocol (IUCR) protocol to solve both of these problems jointly. IUCR provide fixed area clustering derived from transmission range of network nodes. This clustering also develops strong network backbone that provides fail-over-proof routing. Efficient routing path is achieved by finding minimal hop-count with availability of alternate routing path.

Improved Unequal-Clustering and Routing Protocol

We propose improved unequal-clustering and routing protocol (IUCR) protocol to solve both of these problems jointly. IUCR provide fixed area clustering derived from transmission range of network nodes. This clustering also develops strong network backbone that provides fail-over-proof routing. Efficient routing path is achieved by finding minimal hop-count with availability of alternate routing path.

COMPARISON OF HOP COUNT ON WIRELESS MESH NETWORK

Wireless Mesh Network (MWN) is a self-configured and self-organized network that can typically be implemented on 802.11 hardware. It consists of several nodes that make up the network backbone in a multi-story and sealed room, in contrast to building a hall or a place without bulkheads. This experiment uses an odd and even number scheme with a maximum number of routers of 8 pieces. In a sealed room, the performance of the method of installation of the number of strange Hops is better than the number of even Hops, with throughput calculation of 2665.19 KB, delay 0.25 s, data lost 0.60 %, and jitter 0.01 s and the best scheme that is with the number of Hops as much as five pieces, with the calculation of the number of throughput 7001.88 KB, delay 0.51s, data lost 0.47%, and jitter 0.002 s. In the free spaces, it can produce the better performance of the even hop count calculation scheme than the odd hop count by building throughput 16709.8 KB, delay 0.2 s, data lost 0.08 %, and jitter 0.03 s. and the best scheme that is with the number of throughput 68975,2 KB, wait for 0.0148 s, data lost 0 %, and jitter 0.0014 s. WMN performance in unshared space is more maximized than the version in a sealed area, with throughput values of 11786.82 kbps, delay of 2.08 ms, and data lost by 0.08 %, and jitter 0.03 s.it can produce the better performance of the even hop count calculation scheme than the odd hop count by producing throughput 16709.8 KB, delay 0.2 s, data lost 0.08 %, and jitter 0.03 s. and the best scheme that is with the number of throughput 68975,2 KB, wait for 0.0148 s, data lost 0 %, and jitter 0.0014 s. WMN performance in unshared space is more maximized than the version in sealed space, with throughput values of 11786.82 kbps, delay of 2.08 ms, and data lost by 0.08 %, and jitter 0.03 s. and data lost by 0.08%, and jitter 0.03s.

Decentralised Learning in Federated Deployment Environments

Decentralised learning is attracting more and more interest because it embodies the principles of data minimisation and focused data collection, while favouring the transparency of purpose specification (i.e., the objective for which a model is built). Cloud-centric-only processing and deep learning are no longer strict necessities to train high-fidelity models; edge devices can actively participate in the decentralised learning process by exchanging meta-level information in place of raw data, thus paving the way for better privacy guarantees. In addition, these new possibilities can relieve the network backbone from unnecessary data transfer and allow it to meet strict low-latency requirements by leveraging on-device model inference. This survey provides a detailed and up-to-date overview of the most recent contributions available in the state-of-the-art decentralised learning literature. In particular, it originally provides the reader audience with a clear presentation of the peculiarities of federated settings, with a novel taxonomy of decentralised learning approaches, and with a detailed description of the most relevant and specific system-level contributions of the surveyed solutions for privacy, communication efficiency, non-IIDness, device heterogeneity, and poisoning defense.

Binarization of music score with complex background by deep convolutional neural networks

Binarization is an important step for most of document analysis systems. Regarding music score images with a complex background, the existence of background clutters with a variety of shapes and colors creates many challenges for the binarization. This paper presents a model for binarization of the complex background music score images by fusion of deep convolutional neural networks. Our model is directly trained from image regions using pixel values as inputs and the binary ground truth as labels. By utilizing the generalization capability of the residual network backbone and useful feature learning ability of dense layer, the proposed network structures can differentiate foreground pixels from background clutters, minimize the possibility of overfitting phenomenon and thus can deal with complex background noises appearing in the music score images. Comparing to traditional algorithms, binary images generated by our method have a cleaner background and better-preserved strokes. The experiments with captured and synthetic music score images show promising results compared to existing methods.

Smart Grid Using Internet of Things

The IoT (internet of things) is a network of people and stuff at any moment, anytime, for anyone, with any network or service. IoT is therefore a major complex worldwide network backbone for online service providers. The smart grid (SG) is one of IoT's main applications. SG is an interconnected data exchange network that gathers and analyzes data obtained from transmission lines, generation stations, and customers through the power grid. The internet of things has risen as the basis of creativity for energy grids. The chapter is based on the idea that, if one grid station transmitting electricity to customers is cut off due to some defects of IoT-based systems, all grid station loads can be connected to another system so that power is not disrupted. The authors discuss the IoT and SG and their relationship in this chapter. The best advantages for SG and specifications can be addressed in the SG works, creative innovations using IoT in SG, IoT software, and facilities in SG.

Design of the LAN Network of Hospital Comandante Manuel Piti Fajardo

The design of the network of a health institution is a complicated task due to all the aspects that it encompasses, to satisfy the consumption needs of digitized services, using minimal time and at the lowest possible cost. The Manuel Piti Fajardo Hospital, according to the scope of health services it provides, has a system for Hospital Management, the Galen Clinica. This system requires a well-structured network design, with the appropriate equipment that responds quickly and efficiently to the traffic generated in the network. In line with this objective, a cost-benefit study was carried out, after applying the LAN design methodology and calculating approximately the traffic generated on the network in the main departments of the hospital, whether it be the Matrix or the Unit. Surgical and Imaging. With this, the result was to locate the network backbone, and determine the network components that should be replaced according to the financial budget of the hospital, a better response to the requests made by users and according to the evolution of technologies. of information and communications.