Analysis Of Electromagnetic Wave Propagation In Indoor .
Recent Researches in Communications, Signals and Information TechnologyAnalysis of Electromagnetic Wave Propagation in Indoor EnvironmentsPaulo G. Esperante, Raquel Cymrot, Paulo A. Garcia, Marcos S. Vieira, Marcelo PerotoniSchool of Electrical EngineeringMackenzie UniversityRua da Consolação 930, ZIP CODE 01302-907SÃO PAULO - SP - BRAZILpaulo.esperante@mackenzie.br, raquel.cymrot@mackenzie.br, paulo.garcia@mackenzie.com.br,marcos sv@mackenzie.com.br, marcelo.perotoni@cst.comAbstract: The wireless networks have been the object of many studies and analyzes of current technologyindustry, providing not only communication with mobility to end users, but also incorporating newapplications. One such technology is known as Wireless Fidelity (Wi-Fi) 802.11. This paper aims to present thebehavior of the propagation of electromagnetic waves radiated from an access point Wi-Fi (AP), with differentpositions of the antenna. These analyses were based on measurements taken in an environment considering lineof sight (LOS) at different distances from the AP, but in a confined environment. Consequently, although therewere no obstacles (LOS) between transmitter and receiver, there were conditions of confinement on thepropagated signal, given the characteristics of this environment (such as the height of the ceiling and the wallsthemselves). Thus, measurements were made by switching the antenna positioning of the access point,vertically and horizontally. From the analysis of these measurements, it was possible to verify the positioningof the antenna in the access point that generated improved signal coverage, even moderately. In addition, fromthe measured data (statistically based) the technique of linear regression was used in order to generatemathematical models for each specific situation measured. These proposed models were compared to the Friismodel plus correction factors, and were used both for validating the measurements, as a basis for installation ofnew access points in similar environments.Key-Words: antenna, electromagnetic waves, Friis, line of sight, propagation, statistic analysis, wireless, Wi-Fimalls, schools, universities, offices, hospitalsetc. that offer Internet access, often free ofcharge. For the user to connect to this networkit is necessary to have a notebook, desktop,smartphone, or any compatible Wi-Fi ic waves. The propagation ofelectromagnetic waves might be analyzed andmodeled by many models and equations. Oneway can be Friis Equation that is amathematical model that represents a signaltransmitted in a medium in which there is a lossof signal power as a result of free space loss.Friis Equation is relevant once other modelsthat explain these phenomena are based on it.This equation is based on transmission in LineOf Sight (LOS). Other models ofelectromagnetic wavespropagation arecompounded by variables that attenuate thesignal due to obstacles, number of buildingfloors, kind of flooring, kind and number of1 IntroductionTwo decades ago, telecommunicationsnetworks were predominantly wired, implyingsome limitations due to the fact there was aneed for a physical connection among theparties. With the growth of telecommunications,wireless networks have taken over many of theof communications services, since that kind ofcommunications offers services that facilitatepeople and companies life [1][2].The Wireless Fidelity (Wi-Fi) is a set ofspecifications for wireless local area networks(WLAN - Wireless Local Area Network) basedon IEEE 802.11g standards (Institute ofElectrical and Electronics Engineers, 2009) [3].Due to the flexibility of Wi-Fi, it has becomewidespread the implementation of networks thatuse this technology in different places, mainlyas a result of the previous mentionedadvantages and by the reduction of costs. It iscommon to find available Wi-Fi services inhotels, airports, bus stations, bars, restaurants,ISBN: 978-1-61804-081-7101
Recent Researches in Communications, Signals and Information Technologyused the Wireless-Mon software, which ensures themeasurement of only one Access Point (AP's) evenin the presence of others signal sources.The regression analysis aims to model andinvestigate the relationship between a responsevariable and one or more predictor variables(regressor). The linear models coefficients areestimated to minimize the quadratic errors sum(least squares methods) [9].A way of quantifying part of the explainvariability by the model is calculating thecoefficient of determination, R2, which is a numberthat varies from 0 to 1.walls, furniture lay-out, opening of doors anddistance from the antenna [4][5].It is extremely important the comprehensionof the factors that cause the multiple paths,however they are practically inevitableespecially in indoor environments. Themultipath effect occurs due to four phenomena:diffraction, scattering, reflection and refraction.The prediction of such phenomena might bebest addressed with the design of transmitters ina confined environment.The Line Of Sight (LOS) situation does notconsider multipath effects. It is really necessaryto specify the kind of multipath, since it is notunusual for the electromagnetic wave to beblocked by doors, walls, floors and otherobstacles [4][5][6].There are simple mathematical models, forexample, the Line Of Sight (LOS), in whichthere is no obstacle between the transmitter andthe receiver. That model is the Friis Equation.Performing a series of transformations andsubstitutions [6][7][8], we obtain the followingfree space attenuation model:Fig. 1 – Wireless Transmitter - starting point and finalpoint in the corridor to 30 metersThe closer this variable is to one the better themodel is. The addition of new variables alwaysincreases the value of R2, regardless of the relevanceof this variable. It has to be thought over if theincrement of this variable in the regression isworthwhile, because of the model’s degrees offreedom lost in the residual. After the model is built,it has to be done the residual analysis to verify if thetheory model suppositions’ were satisfied [9].It was realized an experiment in line of sight(L.O.S.) condition, the independent variable was thedistance and the dependent variable was the signallevel. After that, it was done an analysis in theobtained data as to find out the linear regressionmodel that best models or represents the measuredresults. The computed of electromagnetic wavepropagation model was then compared with preexistent models like Friis equation.In order to perform a regression analysis,ANOVA tables were built and tested thesignificance of each variable model, comparing themeasured variability for each factor with theresidual variability. The factors that are consideredsignificant are the ones with descriptive level2 Problem FormulationThe objective of this paper is to analyze the WiFisignal, and the behavior of electromagnetic wavepropagation in indoor environment. Measuring theattenuation levels during a certain time then basedon the collected data, later it was performed a linearregression analysis that was compared with the FriisEquation. For transmitting the Wi-Fi signal a WiFiAccess Point D-LINK model DL-624 AirPlusXtreme G with a monopole-type antenna of 10 cm,using the channel frequency 6 (2.437 GHz).To measure the received signal level, an ItautecInfoWay model N8610 series notebook was used.Any kind of notebook could have been used sincethey do not impose any changes in the performedmeasurements.The measurements in Line Of Sight occurred inthe hallway on the ground floor of the buildingEngineering School of Mackenzie PresbyterianUniversity. Measurements started after 3 meters in ahallway 30 meters wide. A total of 11 measurementpoints were defined. The access point antenna wasinstalled at 1.15 meters of height from the ground.To check the received power on the notebook it wasISBN: 978-1-61804-081-7102
Recent Researches in Communications, Signals and Information Technology(P-value) less than the significance level fixed foreach test, in general equal to 5% [9].The suppositions that have to be verified about theresiduals are that they have Normal distribution,with zero mean, constant variability and theirindependent relation.The Normal distribution of the residuals isverified using the Normal Probability Plot. Usingthe “thick pencil rule”, it can be asserted that theresiduals obeyed a Normal distribution when all thepoints can be hidden by a straight line drawn with a“thick pencil”. This is a descriptive test. It is desiredthat statistical tests for testing normality areperformed, with the hypothesis of Normaldistribution not rejected when the descriptive level(P-value) test is larger than the significance level.The histogram helps to test the data normality;however its use is only meaningful when thenumber of collect data is large enough, such as thedistribution form is very well represented in thisgraph.The supposition that the residual mean is zero isverified looking at the residuals versus fit graph. Itis required an approximate symmetry of the dots inrelation of the origin in the x axis.The supposition of constant variances is verifiedlooking at the same residuals versus fit graph. Whenit is not noted an extreme difference when the fittedvalues changes, it can be concluded that thedispersion is approximately constant.The supposition of independence is verified bythe residual versus order graph, i.e., when the orderof the collected data is unknown, this graph ismeaningless. The residuals have to be distributedrandomly as the time goes on, without anyestablished pattern.The uses of graphs are important to interpret aresult. The Boxplot is a statistical tool that e helpsthe results analysis and contain information in agraphical format. This graph has a box format, withthe superior level given by the 3rd quartil and theinferior level given by the 1st quartil. The median isrepresented by a line in the box interior and two linesegments are drawn in the extreme of the box untilmaximum and minimum. Values that are not insidethe box are called outliers and are shown asasterisks. The graphic representation given byboxplot provides, among other information, the datavariability and symmetry.By performing the regression in order to obtainthe regression models, it was found equation 2, withthe antenna in a horizontal position and equation 3with the antenna in a vertical position. Then Friisequation is presented (equation 4), which representsthe theoretical model. The coefficient ofdetermination for the regressions presented in 2 and3 equations were respectively equal to 84,7% e92,2%, results that shows how the total variability ofthe measures were explained by the models.The equation 2 and the equation 3 differ slightly,but these differ from the theoretical Friis equation.This happened because in indoor narrowenvironments with Line Of Sight, it is very likelythat multipath effects take place must occur.The data were analyzed using the softwareMinitab 16.Figure 1 presents the results of the residualanalysis with the antenna positioned horizontally. Itis noted that the residual s of the experimentpresented a Normal distribution with P-value 0,425. The other model assumptions were alsoverified, that is, the residuals had approximatelyzero mean, approximately constant variance, asshown in the graph of residual versus fitted values,and they seem apparently independent since thegraph of residues versus order did not present astandard.Residual Plots for attenuation with antenna horizontallyNormal Probability istogram705ResidualFrequency60Fitted ValueVersus Order1510505010200-5-10-9-6-30Residual361 10 2 0 3 0 4 0 50 60 7 0 8 0 9 0 00 10 2 0 3 0 4 0 501 1 1 1 1 1Observation OrderFig.1 – Residual plots for line of sight with antennapositioned horizontallyFigure 2 shows the results of residual analysiswith the antenna positioned vertically. Theexperiment did not show Normal distribution withP-value 0.007, when the Anderson-Darling testwas applied. However, when the Ryan-Joiner test3 Problem SolutionThe experiment can be divided in two analyses.At first the antenna was in the horizontal positionand in the second analysis, the antenna was invertical position.ISBN: 978-1-61804-081-7Versus FitsN150AD0,368P-Value 0,42599Residual99,9103
Recent Researches in Communications, Signals and Information Technologywas performed, the adherence to Normaldistribution was proved, with P-value 0.053. Theothers model assumptions were verified, so themodel can be used.Figure 3 presents the test of equal variances forall the measurements in all of the distances andconsidering the two antenna positions. The Barletttest was used, once it adhered to the Normaldistribution and such hypothesis was not rejected (P 0,055).the dispersion of the data, which is bigger as thespacing of the boxes grows. There is also thepresence of some outliers in the graph, which arerepresented by asterisks. There are lines of segmentsjoining the averages for each distance with theantennas in both positions. Note that the locationinfluences the measures, which are always higherwhen the antenna is vertically positioned.Interaction Plot for mediçõesData Means80Residual Plots for attenuation with antenna veticalNormal Probability Plot90501087040MeanPercentVersus FitsN150AD1,090P-Value stogram60Fitted Value708050Versus sidual464043069121518Distance21242730-4Fig.4 – Interaction plot for analysis in line of sight withantenna vertically and horizontally positioned-81 1 0 20 30 40 5 0 6 0 7 0 80 90 00 1 0 2 0 3 0 40 501 1 1 1 1 1Observation OrderFig.2 – Residual plots for line of sight with antennapositioned vertically, use of the Anderson-Darling testBoxplot of V27HV30HVValues of measurementsTest for Equal Variances for mesurementsDistanceBartlett's TestTest StatisticP-Value29,710,056Levene's TestTest e (m)H V3H V6H V9H V12H V15H V18H V21H V24H V27H V30Fig.5 – Boxplot of data for line of sight with antennavertically and horizontally positioned024681095% Bonferroni Confidence Intervals for StDevsFig.3 – Test for equal variance for measurements withdata of the antenna vertically and horizontally positioned4 ConclusionWith the calculated mean, figure 4 wasgenerated, which presents the interaction betweenthe variables distances and positions of the antenna,which shows the behavior of attenuation for bothantenna vertically and horizontally positioned.It can be seen that there is little differencebetween the signals received by 12 meters and 15meters. This phenomenon probably occurredbecause there is a lowering in the ceiling to 12meters where the multipath probably masked themeasurements to 12 meters. As can be seen inFigure 4.Figure 5 shows the Boxplot chart that allowsanalyzing the symmetry of the measurements andAnalyzing the behavior of electromagnetic wavein free space, which means has no obstacle betweentransmitter and receiver, the distance factor hadexpected behavior since the greater the distance thegreater the attenuation.Performing comparisons with the Friis model,one can verify that the created models have differentequations, which is justified since the measureswere obtained in a hallway, which leads to themultipath signal. However, in both cases (APantennas in the vertical and horizontal position), theattenuation showed similar behavior. The generatedmodels also showed constant variance attenuationwith distance, similar to the Friis model. This showsthat the models created can be used to scale theISBN: 978-1-61804-081-7104
Recent Researches in Communications, Signals and Information Technologyplacement of AP’s in confined ambient with a gooddegree of accuracy and, in its absence, the basicmodel of Friss serve at least as an initial basis forthis purpose.References:[1] LOPES, R.; FREIXO, P.; SERRADOR, A.,Indoor Propagation Models and Radio PlanningforWLANs,INTERNATIONALCONFERENCE ON E-BUSINESS ANDTELECOMMUNICATION NETWORKS, Vol.3,2004, pp. 87-92.[2] LASSABE F.; CANALDA P.; CHATONNAYP.;SPIES F., Indoor Wi-Fi nications, Vol. 64, No 9-10, 2009pp.651-664.[3] IEEE 802.11 Working Group, 2010 http://grouper.ieee.org/groups/802/11/index.html .[4] NAJNUDEL, M., Estudo de propagação emambientes fechados para o planejamento deWLANs. Disertação (Mestrado em EngenhariaElétrica)-Pontifícia Universidade Católica doRio de Janeiro, 2004. Disponivel em: http://www.maxwell.lambda.ele.pucrio.br/Busca etds.php?strSecao resultado&nrSeq 5069@1 .[5] SARKAR, T. K.; JI, Z.; KIM, K.; MEDOURI,A.; SALAZAR-PALMA, M., A Survey ofvarious propagation models for mobilecommunication,IEEEAntennasandPropagation Magazine, v. 45, n. 3, p. 51-82,June 2003.[6] SAUNDERS, S. R. Antennas and Propagationfor Wireless Communication Systems, JohnWilley&Sons, London, 1999.[7] RAPPAPORT, S.T. Wireless Communications:Principles and Practice, Pearson Prentice Hall,2008.[8] DAVIES, J.N.; GROUT, V.; PICKING, R.,“Prediction of Wireless Network SignalStrength within a Building”, Proceedings of theSeventh International Network Conference(INC 2008), University of Plymouth, 8-10 July2008, pp193-207.[9] MONTGOMERY, D. C.; RUNGER, G. C.,Applied Statistics and Probability forEngineers, John Wiley & Sons, 2006.ISBN: 978-1-61804-081-7105
Access Point D-LINK model DL-624 AirPlus Xtreme G with a monopole-type antenna of 10 cm, using the channel frequency 6 (2.437 GHz). To measure the received signal level, an Itautec InfoWay model N8610 series notebook was used. Any kind of notebook could have been used since they
wave propagation, including ground wave and ionospheric propagation, goes on to make this text a useful and self-contained reference on antennas and radio wave propagation. While a rigorous analysis of an antenna is highly mathematical, often a simplified analysis is sufficient for understanding the basic principles of operation of an antenna.
Ground Wave Ground wave propagation occurs at low frequencies. Typically 4 MHz and below. In ground wave propagation, the magnetic field ofIn ground wave propagation, the magnetic field of the RF signal couples with the earth. A vertically polarized antenna works well for this type of propagation.
Ground-Wave Propagation Ground-wave propagation involves the transmission of a radio signal along or near the surface of the earth. The ground-wave signal is divided into three parts: the direct wave, the reflected wave, and the surface wave. The direct wave travels through the atmosphere from one
Motive Wave. It is a five wave trend but unlike a five wave impulse trend, the Wave 4 overlaps with the Wave 1. Ending Diagonals are the last section ("ending") of a trend or counter trend. The most common is a Wave 5 Ending Diagonal. It is a higher time frame Wave 5 trend wave that reaches new extremes and the Wave 3:5 is beyond the .
on radio propagation. This handbook also provides basic information about the entire telecommunications environment on and around Mars for propagation researchers, system . 1.2 Radio Wave Propagation Parameters. 4 2. Martian Ionosphere and Its Effects on Propagation (Plasma and Magnetic Field). 7
The speed of propagation of an electromagnetic wave modulus and atmospheric pressure and temperature is in free space is a constant, c, which is equal to 3 x 108 m/s. In a material medium such as the atmosphere, the speed P of propagation varies. Even small variations in speed pro- Nx 77.6 P/T (19.3)
Basics and Application of Ground-Penetrating Radar as a Tool for Monito ring Irrigation Process 157 2.1 Electromagnetic principles of GPR 2.1.1 Electromagnetic wave propagation in soil The propagation velocity v of the electromagnetic wave in soil is characterised by the dielectric permittivity and magnetic permeability µ of the medium: 00 11
Ground Wave Propagation Follows contour of the earth Can Propagate considerable distances Frequencies up to 2 MHz Example oAM radio. Sky Wave Propagation. Sky Wave Propagation Signal reflected from ionized layer of atmosphere back down to earth Signal can travel a number of hops, back and
