Cognitive Radio

Computing Techniques, Network Security and Challenges

Budati Anil Kumar, Peter Ho Chiung Ching, Shuichi Torii, Budati Anil Kumar, Peter Ho Chiung Ching, Shuichi Torii

• Publisher: CRC Press
• Year: 2021
• eBook ISBN: 9781000488609
• Topic: Technology & Engineering
• Subtopic: Computer Networking
• Index: Technology & Engineering

The scarcity of radio spectrum is one of the most urgent issues at the forefront of future network research that is yet to be addressed. To address the problem of spectrum usage efficiency, the cognitive radio (CR) concept was proposed. The challenges of employing CRs include ensuring secure device operations and data transmission with advanced computing techniques. Successful development of CR systems will involve attainment of the following key objectives:

• Increasing the rate and capacity of CR-based networks
• How the power is utilized in CR hardware devices with CMOS circuits
• How the framework is needed in complex networks
• Vedic multipliers on CR networks
• Spatial analysis and clustering methods for traffic management
• To transmit a large volume of data like video compression
• Swarm optimization algorithms
• Resource sharing in peer-to-peer networking

This book gathers the latest research works focusing on the issues, challenges, and solutions in the field of Cognitive Radio Networks, with various techniques. The chapters in this book will give solutions to the problems that Industry 4.0 faces, and will be an essential resource for scholars in all areas of the field.

1 A Framework for Identification of Vehicular Traffic Accident Hotspots in Complex Networks

Mohd. Minhajuddin Aquil and Mir Iqbal Faheem

Career Point UniversityAbstract

DOI: 10.1201/9781003102625-1

Contents

1. 1.1 Introduction
2. 1.2 Overview on Vehicular Traffic Accident Hotspot Techniques
3. 1.3 Kernel Density Estimation
4. 1.4 K-Means Clustering
5. 1.5 Point Density Method
6. 1.6 Line Density Method
7. 1.7 Interpolation Method
8. 1.8 Kriging Method
9. 1.9 Spline Method
10. 1.10 Natural Neighborhood Method
11. 1.11 Mapping Cluster
12. 1.12 Moran’s I Method
13. 1.13 Getis-Ord GI* Method
14. 1.14 Earlier Studies
15. 1.15 Research Directions
16. 1.16 Research Framework
17. 1.17 Conclusions
18. References

1.1 Introduction

Accidents have been a major social problem in developed countries for over 50 years. Since 2001, there has been a growth of 202% of two-wheeler and 286% of four-wheeler vehicles with no road development. Approximately 1.35 million people die each year as a result of road traffic crashes as per a report submitted by the World Health Organization (WHO). The 2030 Schedule for Workable Progress has set an ambitious goal of halving the global number of fatalities and injuries from road traffic crashes by 2020. More than half of all road traffic fatalities are among susceptible road users like pedestrians, cyclists, and riders. It is only in the past decade that developing countries like India have begun to experience a significant increase in the number of road accidents taking place and have found it necessary to institute road safety programs. It is strongly felt that most of the road traffic accidents, being a multi-factor incident, are not only due to a driver’s fault on account of the driver’s negligence or ignorance of traffic rules and regulations, but also due to many other related parameters such as changes in road geometrics, flow characteristics, road user’s behavior, environmental conditions, visibility and absence of traffic guidance, and control and management devices. The Geographical Information System (GIS) emphasizes on providing services on a location scale, and it merely enables the operators to use spatial information and descriptive data to make plans, tables, and diagrams. This system accurately provides search tools; data analysis and results are displayed. The GIS is an organization and decision support system that contains graphic data and site maps that are productive for traffic accident information organization. In the management of road safety, a road traffic accident hotspot is a place where accidents occur frequently. It may have occurred due to various parameters like poor road geometrics, environmental factors, driver’s characteristics, and so on. Since few decades, treatment of accident black spot has been the spine of road safety management. In the current scenario, road safety is a major concern. Road safety measures can be adopted by implementing various steps.

1.2 Overview on Vehicular Traffic Accident Hotspot Techniques

To analyze vehicular traffic accident hotspots, various methods have been discussed in this chapter such as Kernel Density Estimation (KDE), Nearest Neighborhood Hierarchy (NNH), Inverse Distance Weighted (IDW), and Kriging Kim and Levine (1996) described the traffic safety GIS prototype, which performed a spatial analysis of traffic accidents that are developed for Honolulu, Hawaii. Many types of spatial analysis methods based on point, segment, and zone analyses have been developed. Affum and Taylor (1998) introduced a method for traffic management, which is based on a GIS package for studying accident patterns over time. The different methods/techniques are shown in Table 1.1.

Table 1.1 Methods/Techniques for Hotspot Detection

S. No	Technique	Formula	Purpose
1.	KDE	f(x) = $\frac{3}{n{h}^2\pi }{{\displaystyle \sum }}_{i=1}^n{\left\{1-\frac{1}{h^2}\left[{\left(x-x_i\right)}^2+{\left(y-y_i\right)}^2\right]\right\}}^2$	For smoothing effect within a particular radius and cell size
2.	Point Density	f(x) = $\frac{1}{n}{{\displaystyle \sum }}_{i=1}^n\frac{1}{h}w\left(\frac{x-x_i}{h}\right)$	Calculates magnitude per unit area using neighborhood operation for a given cell size
3.	Line Density	f(x) = $\frac{1}{n}{{\displaystyle \sum }}_{i=1}^n\frac{1}{h}w\left(\frac{x-x_i}{h}\right)$	Calculates magnitude per unit area for the radius of the cell size
4.	IDW	$$z=\frac{{{\displaystyle \sum }}_{i=1}^s{z}_i\frac{1}{d_i^k}}{{{\displaystyle \sum }}_{i=1}^s\frac{1}{d_i^k}}$$	For classifying within the max and min values
5.	Kriging	$$f\left(x\right)={{\displaystyle \sum }}_{i=1}^n{\surd }_i\left(x^{*}\right)f\text{(}{x}_i\text{)}$$	For assuming spatial variation of attributes
6.	Spline	Q (x, y) = $\sum A_i{d}_i^2\text{log}{d}_i+a+bx+cy$	For a smoothing effect
7.	Moran’s I	I = $\frac{{{\displaystyle \sum }}_{i=1}^n{{\displaystyle \sum }}_{i=1}^n{w}_{ij}\left(x_i-x\right)\left(x_j-x\right)}{s^2{{\displaystyle \sum }}_{i=1}^n{{\displaystyle \sum }}_{i=1}^n{w}_{ij}}$	For detecting the presence of the clustering of similar values
8.	Getis-Ord GI*	G(d) = $\frac{\sum \sum w_{ij}\left(d\right)x_i{x}_j}{\sum \sum x_i{x}_j},i\ne j$	For separating the clusters of high and low values

1.3 Kernel Density Estimation

KDE is one of the significant spatial analysis tools in the commercially available GIS software package. K divides the entire study area into a pre-determined number of cells. It uses a quadratic kernel function to fit a smoothly elongated surface to each accident location. The surface value reduces from the highest at the incident location point to zero when it reaches a radial distance from the incident location point. The value of the kernel function is assigned to every cell as individual cell values. The resultant density of every cell is computed by adding its i cell values individually. To account for the road accident severity, the weight of each accident is represented as its Identification Number (ID). This facilitates the counting of each accident according to its weight assigned. In case of no injury or severity, according to incident points, the population field is selected as “None”. Kernel function can be defined as stated in Eq. (1.1).

$$f\left(x\right)=\frac{3}{n{h}^2\text{π}}{{\displaystyle \sum }}_{i=1}^n{\left\{1-\frac{1}{h^2}\left[{\left(x-x_i\right)}^2+{\left(y-y_i\right)}^2\right]\right\}}^2$$(1.1)

where h is the bandwidth, f is the estimator of the probability density function, π is a constant, x_i and y_i are the deviations of x- and y-coordinates between a point and a known point that is within the bandwidth, and n is the number of known points. The kernel estimator depends upon the choice of bandwidth (h), and hence, suitable bandwidth should be determined according to the purpose of the study. Density values in the raster are predicted values rather than probabilities. It generally gives a smoother surface than the normal estimation method.

1.4 K-Means Clustering

These hotspots are classified using the clustering process and are organized into classes (or clusters) based on similar attributes. These clusters are then arranged into groups, based on the similarity of the clusters. This hierarchical process allows spatial classification based on the similarity of either characteristic of the accidents within the hotspots or the environmental factors. When determining the database used to build the classification, it is essential to assess the type of data which would be collected and would have the potential of having an impact on accident density.

1.5 Point Density Method

The point density tool estimates the density of point features around each output raster cell. Neighborhood can be shown around each raster cell center, and the number of points that come within it is totaled and divided by the area of the neighborhood. If other than NONE is used in the population field setting, the item’s value determines the number of times to count the point. For instance, an item with a value of three would cause the point to be counted as three points. The values can be floating-point or integer. If a unit is selected within the area, then the estimated density for the cell is multiplied by the suitable factor before it is written to the output raster. Although more points will fall inside the broader neighborhood, this number will be divided by a more extensive area when calculating density. It can be evaluated using the equation given below:

$$f\left(x\right)=\frac{1}{n}{{\displaystyle \sum }}_{i=1}^n\frac{1}{h}w\left(\frac{x-x_i}{h}\right)$$(1.2)

The main consequence of a larger radius is that density is calculated considering an additional number of points, which can be farther away from the raster cell. This results in a more generalized output raster as stated in Eq. (1.2), where h is the bandwidth, w is the weight, and n is the number of known points.

1.6 Line Density Method

In this method,...

Table of contents

• Cover
• Half Title
• Title Page
• Copyright Page
• Table of Contents
• Editors
• Contributors
• Chapter 1 A Framework for Identification of Vehicular Traff ic Acc ident Hotspots in Complex Networks
• Chapter 2 Traff ic Safe ty Management Using Spatial Analysis and Clustering Methods
• Chapter 3 Machine-Learning Algorithms on Cognitive Multisensor Image Application
• Chapter 4 Performance Evaluation of Video Compression Tec hniques: x.263, x.264 and x.265 to Improve Video Streaming Quality
• Chapter 5 Design of CMOS Circuits for Cognitive Radio Application with Power Analysis
• Chapter 6 A Nove l Design of 16 Bit MA C Unit Based on Vedic Mathematics Using FPGA Hardware for Cognitive Radio Application
• Chapter 7 Hybrid Optimization Tec hnique Using Particle Swarm Optimization and Firefly Algorithm using the Chord Protocol
• Chapter 8 MESH -DHT Approach for Eff icient Resource Sharing in P2P-Based Wireless Mesh Cognitive Networks
• Chapter 9 Design of the VLSI Tec hnology for Cognitive Radio
• Index