Search

An accurate and efficient path loss modeling method for millimeter wave communications plays a significant role in the large-scale deployment of a fifth-generation (5G) mobile communication system. Conventional path loss modeling methods such as deterministic methods, empirical methods, and machine learning-based methods, in practice, cannot achieve the desired level of performance in terms of accuracy. This paper proposes a novel path loss model for 5G communications in suburban scenarios using deep learning with dilated convolution and attention. Dilated convolution is used to alleviate the locality of feature extraction and capture the global information on input images. Attention rendered by global context blocks helps the attention-enhanced convolutional neural network (AE-CNN) model utilize the global information on inputs to extract essential features of propagation environments. A distance-embedded local area multi-scanning algorithm which generates input images that can improve learning the latent features with dependency on distance is proposed. The experimental results indicate that the AE-CNN model can outperform state-of-the-art deterministic and empirical methods in terms of root mean square error in test scenarios.

Millimeter waves (mmWaves) of the 28 GHz frequency bands have been selected for the 5G communications with special usage scenarios such as smart factories. Indoor path loss prediction plays an important role in configuring a base station to be able to utilize the full capacity of the new technology. Although machine learning has attracted much attention recently in path loss modeling thanks to its ability to make accurate predictions, its performance can be limited by the size of available measurement data set used for training. In this paper, we propose a new training strategy to train path loss models based on convolutional neural network (CNN). The proposed strategy is based on meta-learning which performs well in few-shot learning scenarios with multiple tasks comprising a meta-task. It is shown that the indoor path loss model based on a CNN configured as a metatask of multiple beams can outperform the CNN models by a conventional training algorithm as well as empirical models.

The most of sentiment analysis on the user reviews only aim to classify the review as overall positive, negative, and neutral sentiment in terms of the overall polarity score. Such overall sentiment score cannot provide detailed information. This study proposes a method of automatic extraction of features and their modifying sentiment words in user reviews, thus a recommendation system provide more detailed sentiment and information. To parse such sentiment and information, we propose a general method for extracting modification relations between nouns and adjectives in Korean sentences, which are corresponding to features and their modifying (sentiment) words, respectively.

Convolutional Neural Networks (CNNs) have shown surprisingly good performance in both classification and regression problems. Given a floor plan of a building and indoor measurement data of Wi-Fi received signal strength (RSS) at discrete locations, a CNN can be trained to approximate the underlying functions of path loss. In this article, we propose a novel CNN-based indoor path loss modeling approach. Based on a floor plan and measurement data, input images are generated for training a CNN, which can make predictions of the RSS of 5 GHz Wi-Fi in an indoor usage scenario. Experiment results show that CNNs can be used for indoor path loss modeling with the encouraging performance with the Root Mean Square Error (RMSE) of 3.9404 dBm and good generalization property.

Traditional low-cost GPS installed on vehicles and other equipment has a tolerance of tens of meters. With the help of auxiliary devices and/or methodologies such as Differential GPSs (DGPSs), Assisted GPSs (AG-PSs), Real-Time Kinematic (RTK), computer vision and etc., the positioning accuracy of GPS receivers increased a lot, but a certain amount of cost increased at the same time. In this paper, we propose a new cooperative vehicular localization scheme for improving the accuracy of GPS fixes based on Vehicle-to-vehicle (V2V) communication. The proposed scheme first estimates the distances between neighboring vehicles using WLS-DD scheme [1] and uses a machine learning technique called Constrained Self-Organizing Map (C-SOM) with a set of GPS fixes collected over time to generate the final estimates of GPS locations with much lower errors. We present simulation results that demonstrate the superior performance of the proposed Constrained-SOM.

With the advent of multiple radio interfaces on modern smart phones, simultaneous transmission and reception of data through via both mobile networks and WiFi networks has gained a considerable attention as a cost-effective means for mobile data offloading. In this paper, we present a practical SDN-based data offloading framework based on a Femto-WiFi Integrated network. The proposed framework take advantage of the flow handling and analysis capability of SDN controllers to make decisions on how to distribute data flows between mobile core networks and broadband IP networks. It follows the device-initiated, network-assisted approach, where a User Equipment (UE) make selection on the radio interfaces to be used based on the network traffic flow information provided by an Access Network Selection Information (ANSI) server.

In many nations of Europe, tram is still used as main public transportation system. Despite of its many advantages, overhead lines installed along the track line cause severe problems, such as safety issue, maintenance and difficulty in renewing road system. To solve these problems, wireless charging segment is developed to replace the overhead lines, catenary, and pantograph. With wireless charging segment installed underground, tram can operate freely without any support from overhead lines. In this paper, a model of wireless charging tram system is suggested considering all main parts, such as tractive effort, battery modeling, and driving cycle. In order to find out the minimum cost, optimization of battery capacity, power track allocation and pickup capacity of wireless charging tram system is conducted using heuristic algorithm. Moreover, the optimized result and location of power track installation based on Prague tram line are derived. Lastly, initial installation cost between conventional and proposed tram system is evaluated.

This paper presents a location-based interactive model for Internet of Things (IoT) and cloud integration (IoT-cloud). In the model, IoT-cloud provides sensing service on demand based on interest and location of mobile users. Cloud plays a role as a virtual sink which makes schedule for physical sensor networks on-demand based on locations of mobile users. Through extensive analysis results, we show that the model scheme achieve a significant improvement compared to traditional wireless sensor networks (WSNs).

Energy efficiency is a very important requirement in designing a MAC protocol for wireless sensor networks (WSNs) using battery-operated sensor nodes. We proposed a new energy-efficient MAC protocol, RIX-MAC, based on asynchronous duty cycling and receiver-initiated scheme. In this paper, we analyze the performance such as throughput, delay, and energy consumption of RIX-MAC with modeling and simulation. Our modeling is validated by comparing it with the simulation results in NS-2. As the result, we obtain average 95% confidence interval of the simulation results of this paper.

Due to excessive endeavor to reduce greenhouse gas caused by internal combustion engine-powered vehicles, battery-powered electric vehicles (EV) are arising as a solution. However, EV possess numerous limitations, such as short travel distance and long charging time. The On-Line Electric Vehicle (OLEV) is developed to overcome the limitations that the current generations of the EV have. The OLEV is battery-powered electric vehicle which can be charged wirelessly from the transmitter embedded under the road. The cost of the OLEV system is dependent on battery capacity, pickup capacity and cable length. To investigate the key factors of optimizing the cost of OLEV system, it is necessary to know the interrelations among above three main variables. This paper derives the cost function of the OLEV system, considering the three main components, such as battery, pickup and power cable. Also, it shows the change of the vehicle cost and infra cost with varying the capacity or length of OLEV system's components.

The specification of IEEE 802.16j defines the operation of multi-hop relay stations (RSs) with cooperative communication as an optional functionality to achieve space diversity gain. One of the key issues in cooperative communication using RSs is to determine which RSs should be used as cooperative nodes, i.e., cooperative relay selection. In this paper, we investigate the conditions that should be tested for RSs to operate as cooperative nodes based on the signal intensity data. We also propose two cooperative RS selection schemes for determining a group of cooperative RSs out of all candidate RSs: max-min-max cooperative RS selection and joint cooperative RS-path selection. The first scheme selects a group of RSs with the highest diversity gain, whereas the second scheme selects the cooperative RSs and paths simultaneously that could yield the highest system throughput. Simulation results are presented to evaluate and compare the performance of the two schemes.

This paper presents a seamless handover scheme for the IEEE WAVE (Wireless Access for Vehicular Environment) networks based on proactive caching of data packets. In the proposed scheme an On-broad Unit (OBU) uses the IEEE 802.11 disassociation message to inform the old Road-Side Unit (RSU) about its departure from the coverage area such that the old RSU can forward the buffered packets to candidate RSUs for caching. A multicast-based forwarding scheme is used to forward the buffered packets when there are multiple candidate new RSUs around crossroads or diverging roadways. One of the candidate RSUs becomes the new RSU which transmits the cached packets to the OBU upon its entry into the radio coverage. The IEEE 802.11f Move-notify message is used by the new RSU to request the rest of candidate RSUs to purge the cached packets in order to prevent the waste of radio resource. Simulation results show the proposed scheme outperforms a simple multicast-based scheme.

This paper presents general software optimization techniques to improve software performance and energy consumption in embedded systems. Software optimization can be categorized into three levels: algorithmic, source code-level. Then these techniques are applied to optimize our tactical message processing software, which is a message encoder-decoder for tactical communication equipped on weapon systems. The optimized software achieved performance increase of about 12%, memory access decrease of 72% and memory usage decrease of 35% over the original software.

The ad-hoc network is a collection of wireless mobile nodes without any support of a stationary infrastructure. Due to the node's mobility and the network topology changes to be kinetic, the network disconnection occur frequently. This paper proposes a variation of the AODV protocol, which selects CORE nodes using path's accumulation information, and then establishes multi paths based on CORE-node's information. When network is disconnected, our proposed protocol saves path reset time and reduces the reset control messages by using the information held by the CORE-node.

The cooperative communication makes wireless transmission reliable utilizing special diversity, where the relay selection is one of important issues. In this paper, we propose a relay selection scheme exploiting SNR and success probability of the link between nodes in network, especially focus in the link between the source and the relay. Proposed scheme is compared with previous schemes by the simulation. The simulation results show that our scheme perform better than other schemes in terms of BER and throughput.

As an emerging technology, wireless mesh networks are making significant progress in the area of wireless networks in recent years. However, the nature of sharing transmission medium makes it challenging to fully exploit these networks. In an attempt to improve the radio resource utilization, several routing metrics have been specifically designed for wireless mesh networks. However none of these routing metrics efficiently tackles interference issues. In this paper, we propose a routing metric for wireless mesh networks, which takes account of the amount of interference by considering the traffic situation of adjacent nodes and selects less interfered path for transmission. The aim of this metric is to route the traffic through interference areas and balance the load amongst the network nodes. Simulation results show that the proposed routing metric reduces the end-to-end delay when compared to the MIC, ETT and hop-count routing metrics.

With the fast proliferation of mobile Internet, the wireless community has been increasingly looking for a framework that can provide seamless mobility. In this paper, we propose a fast cross-layer handover scheme based on movement prediction in mobile WiMAX environment. Prediction is achieved by linear regression model with keeping track of the signal strength of mobile users. With the help of the prediction, layer-3 handover activities are able to occur prior to layer-2 handover, and therefore, total handover latency can be reduced. The experiments conducted with system parameters and propagation model defined by WiMAX Forum demonstrate that the proposed method predicts the future signal level accurately and reduces the total handover latency.

Although the Distributed Coordination Function is the fundamental access protocol of IEEE 802.11, it cannot meet the quality of service (QoS) requirements in general. So, the Point Coordinate Function is provided to support QoS related services. However, it has inherent problems. Access point (AP) has no knowledge of the queue status and instantaneous channel condition of stations in the system. In this paper we propose an efficient and versatile polling scheduler that shows excel-lent throughput and fairness performance. Comparison with well known polling schemes is provided through computer simulation under various channel situations including error prone environments.

Conventional topology-based routing protocols such as AODV, DSR and ZRP are not suitable for highly mobile ad hoc network, where frequent route re-establishment may cause high end-to-end delay and packet loss. This paper presents a new ad hoc routing protocol called the multi-hop MAC forwarding protocol (MMFP). The MMFP avoids explicit path setup in order to reduce the control overhead associated with it, but instead uses the reachability information towards the destination at each hop. Next-hop nodes are determined on-the-fly by contention based on a priority value. The basic operations of the MMFP are conceptually similar to that of MAC bridges and position-based ad-hoc routing protocols. The MMFP is designed to be integrated with the IEEE 802.11 MAC protocol in order to achieve higher efficiency and accuracy in its time-critical operations. Simulation results show that the MMFP outperforms the AODV in a practical ad hoc network scenario in terms of both the end-to-end delay and packet delivery ratio.

Traditional neural network training techniques do not work well on problems with many discontinuities, such as those that arise in multicomputer communication cost modeling. We develop a new algorithm to solve this problem. This algorithm incrementally adds modules to the network, successively expanding the 'window' in the data space where the current module works well. The need for a new module is automatically recognized by the system. This algorithm performs very well on problems with many discontinuities, and requires fewer computations than traditional backpropagation. >