Web Based Visualization of Outdoor Path Loss Models

Web Based Visualization of Outdoor Path Loss Models

Saifuddin Abdullah & FuadAlnajjar

The City College of City University of New York (CUNY)

Keywords:path loss charting, path loss models, online data charting.

Abstract

This paper covers the formulation of a web based path loss charting application designed to approach the concept of online data charting. It basically approaches the given scenario by integrating a database with a chart-generating algorithm, which takes its numerical parameters from the users through a web interface.Thisis the first of its kind of application, which utilizes 10 different path loss model equations encoded in Python, and gives a concept of modeling path loss progression through a web based GUI interface over a specified distance range.

1.Introduction

Visualization and simulation of raw data into an interactive illustration is always an insightful experience. Data ranging from statistical values to the randomized figures yielding from the combination of some formulae, visualization and illustration always provides a quick look inside the data trend and harmony. For these days, there are multiple applications and software programs through which simulation and illustration of raw data values is carried out very effectively in both 2D and 3D geometrical models. Moreover, these programs not only offer a way to simulate the pre-recorded data, but also provide a way to visualize the real-time data. However, in all these cases, there is a very little room for UI (user interface) and UX (user experience) compatibility practices, since these applications are always targeted towards a specific set of individuals or ‘data-people’ working under domains like network administration, data analytics, process engineering or statistical evaluation. However, what if this whole idea of data illustration and visualization can be brought to the general public or to basic science students with the help of the most commonly shared entity – the web? In addition to this, what if these users only have to enter a few values in a user friendly web form to get their desired data simulated or illustrated? This is the basic motivation behind this paper and the project presented within it as a supporting scheme, which utilizes a Python based charting wrapper to illustrate the provided path loss data (as an example) for outdoor environments.

2.BACKGROUND

Path loss is a phenomenon which is widely attributed to the outdoor and indoor wireless communication networks (like cellular networks, home WiFis, long distance p2p links, etc.). This phenomenon actually indicates the ‘amplitude’ or level of loss incurred by an electromagnetic wave (or a signal) travelling from transmitter to receiver. During the phase of design and development of a communication network, this quantity is usually pre-established by estimating the probable loss that can be incurred by the objects and physical entities in the surrounding environments. However, since this estimation is very demanding (takes a lot of time and resources for each site), therefore, engineers utilize a few pre-existing equations that actually predict the path loss for any particular distance point, considering the properties of the surrounding environment as input variables. These equations are commonly known as path loss models, and most of them are empirically evaluated, based on different urban, sub-urban and rural terrains across the globe. For instance, if an electrical engineer is trying to establish a communication link between two points in a certain environment, he or she is ideally supposed to utilize any of the given path loss modeling equations to estimate the maximum and minimum path loss values in different environments. Since path loss is simply the amount of loss in any given signal’s power level, therefore, it is in direct proportion with the increasing distance and frequency, and in inverse proportion with the number of object in the surrounding environment, and is represented in the units of decibels (dB).

There are numerous applications in the market which empower students and engineers to study the path loss of any given environment by defining the physical characteristics and network equipment features, which are (actually) the most commonly utilized applications in outdoor network development areas. In contrast to these, the area of focus of this project is to integrate such a simulating mechanism in a web-based application, which a user can access through his/her device of choice (i.e. a smartphone or a laptop) and can calculate the required path loss by specifying the values of his/her choice.

3.PATH LOSs models

The following path loss models (or equations) were included in the current project:

1. Hata Urban Path Loss Model [1]

1. Hata Rural Path Loss Model [2]

1. Hata Sub-Urban Path Loss Model [2]

1. SUI (Stanford University Interim) Path Loss Model [3]

e.Erceg-Greenstein Path Loss Model [4]

f.3GPP (Third Generation Partnership Protocol) Path Loss Model [5]

g.Egli Path Loss Model [6]

h.COST Hata Path Loss Model [6]

i.Green-Obaidat Path Loss Model [7]

j.ECC – 33 (Electronic Communication Committee - 33) Path Loss Model [8]

3.1.PROGRAMMING MODELS

All the programming and development of these models were carried out using Python as the primary programming language. In more details, since each model is actually represented in the form of a mathematical equation (containing multiple variables); therefore, programming of these models was basically the programming of their associated equations. In classical or manual settings, each of these models can be abstractly represented as:

PL (db) = Variable A + Variable B + Variable C x d

In the equation specified above, variables A, B and C can be simply any variables comprising of elements like Tx and Rx antennae heights, Tx and Rx gains, Tx frequency and wavelength, and several other elements – all bounded together with the key element of the distance. In the conventional settings (or manual experiments), this distance (d) is usually a single distance point for which the path loss is calculated. In our application, this is kept as a 1D array, representing 100 different distance point under the 50 kilometers. This led to a rich visualization or simulation of the programmed path loss models, and gave a great insight into how these path loss model results cover their progressions respectively with increasing distance along with increasing frequency and antennae heights.

As mentioned earlier, the project represented by this paper is entirely based on Python. The process with which contained model equations are developed goes as follows:

a.First, each model equation was developed as a separate program, with each having two basic variables as X and Y (since the visualization was supposed to be the 2D path loss chart). The X-axis was as strictly kept as a variable 1D array containing 100 discrete distance point separated by an interval of ‘a’ from distance d1 to d2 (d1d2). This array was established using numpy.array() function of a Python library called NumPy (Numerical Python). On the other hand, the Y-axis was handled by a single variable (Y), which comprised of the original model equation of the path loss model.

b.Once the variables pertaining to X and Y are developed, they are integrated into the pre-defined charting functions of a Pythonic library called Matplotlib. This library is especially developed for the charting of scientific raw data, and offers a rich set of data plotting and analysis functions. As described earlier, the X-axis plot was kept as independent distance points, while Y-axis plot was integrated as a path loss equation, primarily dependent and varying according to the progressing distance range.

1. Once the plot code was completed for each model, a special matplotlib function called plt.save() was used to generate visualization images on the local drive.

In the above settings, the variables like distance range, transmission frequency, transmission and reception gains, and antennae heights were set as user defined inputs. In other words, whenever the code is loaded, it asks the user to define these parameters as input values.

3.2.WEB INTEGRATION

Once the code was ready, the main issue of web integration arose. To accomplish this, the following approach was adapted:

a.A basic web form was developed using PHP/JavaScript, which was supposed to format and post the user defined input to the database.

b.A SQL database was developed with a table having all the columns corresponding to the inputs expected from the user. This database was supposed to receive values from the PHP/JS web form described in (a).

c.The data values stored in the database table (under a unique row) are then transferred to the original Python code resting on the server. The Python code retrieves the code using MySQLDB library distribution for Python. At this stage (before processing), each value retrieved from the database is formatted accordingly to serve as respective numerical data in the code.

d.Once the formatting is completed, this data is fed to the Python code, which is triggered once formatted data is received. This code then processes the input data and returns an image file stored (under the designated image format) on the local server repository on which the code itself is resting.

e.The generated image is finally integrated in another PHP/JS page, which formats it and integrates it within the HTML of the designated page to show it to the user.

This entire process can be illustrated as follows (see Figure 1):

Figure 1: Process of integrating the charting algorithm into web

Some sample of a pathloss charts produced by our application can be viewed in Figure2 and Figure3:

Figure 2: Generated visualization showing path loss progression according to studied models on 1800 MHz with Tx antenna height as 60 meters from the ground.

Figure 3: Generated visualization showing path loss progression according to studied models on 900 MHz with Tx antenna height as 60 meters from the ground

3.2.1.Addition

Another fairly simplistic addition was made to the path loss charting app by introducing a third Z-axis to the charting algorithm, which was finalized to be the array of transmitter antenna height (in meters). This led to the creation of a 3D path loss points’ illustration, with respect to increasing and decreasing transmitter antenna height. This was done after following the established fact that increasing transmitter height has a significant impact on the variation of power loss or path loss in transmitted data signal [9][10]. Moreover, all of the programed models were also combined in a single program in order to create a collective visualization of all the path loss models depicting a comparison between signal’s power loss (Y - axis), increasing distance (X - axis) and increasing transmitter antenna height (Z - axis). The generated visualization through this additional method helped a lot in understanding the impact of increasing and decreasing antenna heights on all the included path loss predicting models.

4.CURRENT LIMITATIONS

This project currently misses two major aspects (which are worked-on): real time data updating and user-interaction with the chart through browser’s interface. A prototype algorithm to generate a pre-defined number of charts and to create a moving visualization through them was tested locally, and is currently in progress of its implementation on the browser’s interface. Once this is successfully achieved, users will be enabled to interact with the chart and to view the progression of their charted data (for instance, path loss models) in the form of an animation.

5.future Course

There is a valid need for user-interactive web charting systems, which is the next possible course for this project. In more simple words, the user should be made able to interact with the chart, or provide some new values on the chart-screen to have it updated in a real time session. Moreover, this idea can be transformed into a fulltime data charting app (not restricted to path loss illustration) through which users can chart their respective datasets in any type of plotting scheme (like histograms, 3D contour plots, scatter plots, etc.). This is the next step for this project, which is currently under the investigation of authors.

6.conclusion

The sole idea behind this project was the creation of a Python based online data charting app, which was successfully completed by integrating the created path loss predicting algorithm (as an example) into web’s interface. This algorithm utilizes Python as its primary language for data sorting and processing (with the help of libraries like NumPy, SciPy and Matplotlib), while uses PHP/JS to capture the user input from a web based form. The process of online data charting (in the case of this web app) is very simplistic: users enter their desired input values, which are captured using a web based PHP/JS form. These values are stored in a SQL database, and are retrieved from the charting algorithm, which sorts and filters them accordingly. Finally, these values are fed to the charting code which generates the respective image file (shown on the web browser).

References

[1] Hecker, A., Neuland, M. & Kuerner, T. (2006): Propagation models for high sites in urban areas, Advances in Radio Sciences, Vol. 4, pp. 345-349.

[2] Hata, M. (1980): Empirical Formula for Propagation Loss in Land Mobile Radio Services,IEEE Transactions on Vehicular Technology, Vol. 29, pp. 317 - 325.

[3]Abhayawardhana, V.S., Wassell, I.J., Crosby, D., Sellars, M.P. & Brown, M.G. (2005): Comparison of Empirical Propagation Path Loss Models for Fixed Wireless Access Systems, Vehicular Technology Conference Publication, Vol. 1, pp. 73 – 77.

[4]Erceg, V. & Greenstein, L. J., et al.(1999): An Empirically based path loss model for wireless channels in suburban environments, IEEE Journal on Selected Areas of Communications, Vol. 17, pp. 1205-1211.

[5]Saleh, A.B., Redana, S., Hamalinen, J. & Raaf, B. (2010): On the Coverage Extension and Capacity Enhancement of Inband Relay Deployments in LTE-Advanced Networks, Hindawi Publishing Corporation.

[6]Mardeni, R. & Priya, T.S. (2010): Optimised COST-231 Hata Models for WiMAX Path Loss Prediction in Suburban and Open Urban Environments, Modern Applied Sciences, Vol. 4 (9), pp. 75-89.

[7]Green, D.B. & Obaidat, A.S. (2002): An accurate line or sight propagation performance model for ad-hoc 802.11 wireless LAN (WLAN) devices, IEEE International Conference on Communications Publication, Vol. 5, pp. 3424 – 3428.

[8]Sharma, P.K. & Singh, R.K. (2010): Comparative Analysis of Propagation Path Loss Models with Field Measured Data, International Journal of Engineering Science and Technology, Vol. 2(6), pp. 2008-2013.

[9]Ain, K., Tarafder, M.M.R., Khan, S.A. & Ali, M.L. (2011): Path Loss Compensation Technique for WiMax Technology based Communication System, International Journal of Engineering Science and Technology, Vol. 3(9), pp. 7226- 7233.

[10] Abdullah, S. & Alnajjar, F. (2012): A Collective Statistical Analysis of Outdoor Path Loss Models, International Journal of Computers & Technology, Vol. 03 (01).