Development of Propagation Path Loss Prediction Model for Mobile Communications Network Deployment in Osogbo, Nigeria

Transcription

1 Development of Propagation Path Loss Prediction Model for Mobile Communications Network Deployment in Osogbo, Nigeria Hammed Lasisi, Yinusa A. Adediran, and Anjolaoluwa A. Ayodele Abstract Path loss, a major parameter in the analysis and design of the link budget of a telecommunication system, could be explained as the reduction in power density of an electromagnetic wave as it travels through, over a distance. Path loss prediction s are therefore vital tools in cell planning, cell parameter estimation, frequency assignments and interference evaluation. This paper reports on the development of a path loss prediction that describes the signal attenuation between transmitting and receiving antennas as a function of the propagation distance and other parameters for Osogbo, Nigeria. The is extensively useful for conducting feasibility studies for signal prediction, coverage optimization and interference analysis during the initial phase of network planning in the study area and other areas with similar environmental and propagation characteristics. Index Terms Interference; Path Loss; Frequency Assignment; Cell Planning; Link Budget. I. INTRODUCTION In mobile communication, path loss s are necessary for proper planning, interference estimation, frequency assignments, and cell parameter estimation which are basic for network planning process as well as Location Based Service (LBS) techniques that are not based on Global Positioning System (GPS) system. Estimation of the path loss is usually termed prediction. Exact prediction is possible only for simpler cases, such as free propagation or the flat earth. For practical cases the path loss is calculated using a variety of approximations such as: i. Empirical : It is derived from in-depth field measurements, that is, measured and averaged losses along typical classes of radio links. It is efficient and simple to use. The most commonly used of this method are Okumura-, the COST and the. ii. Deterministic : This is derived from and depend on the detailed and accurate description of all objects in the propagation. This method is expected to produce more accurate and reliable prediction of the path loss than the empirical method; however, it is significantly more expensive in computational effort. Thus, it is used predominantly Published on November 6, 7. H. Lasisi is a lecturer at Osun State University, Osogbo. He completed his BS, MS, and PhD Degree at University of Ilorin. Y. A Adediran is a Professor at University of Ilorin in the Department of Electrical and Electronic Engineering. A.A Ayodele is a first class graduate of Osun State University in Electrical and Electronic Engineering. iii. for short propagation paths. Semi-deterministic : This is based on both the empirical s and deterministic aspects []. II. PATH LOSS MODELS Six path loss s are considered in the developmental process of the new. They are: free path loss, Okumura-, COST 3, Ericsson, Egli and the. A. Space Path Loss Model (FSPL) path loss is the simplest way to path loss as all hindrances that might affect signal propagation are neglected. Path loss in free defines how much strength of the signal is lost during propagation from transmitter to receiver across a free devoid of vegetation. FSPL is dependent on frequency and distance, and is given as []. P LFSLP = log d + log f () where PLFSPL is free path loss (in db), f is frequency (MHz), d is distance between transmitter and receiver (m). B. Okumura- Model Okumura is based upon an extensive series of survey and measurements made in and around Tokyo city in 968 between MHz andmhz. Depending on the area of interest, Okumura path loss approximations could be stated as [3]. Urban area: L = A + Blog R E () Suburban area: L = A + Blog R C (3) Open area: L = A + Blog R D (4) A = log f 3.8log H b () B = log H b (6) C = (log ( f 8 )) (7) D = 4.78(log f) log f (8) E = 3.(log (.7H m ) 4.97 for large cities f > 3MHz (9) E = 8.9(log (.4H m ). for large cities f < 3MHz () E = (.log f.7)h m (.6log f.8)for medium to small cities () DOI: 3

2 where L is Path loss, A, B, C, D and E are the antenna correction facto. R is the radial distance between the transmitter and the receiver. H b and H m are the base station antenna and mobile antenna heights respectively, and f is the frequency of transmission (MHz). The is valid for the carrier frequency range MHz f MHz, base station height in the range 3 m H b m, and mobile station height in the range m H m m [3]. C. COST 3- Model Committee 3 of the European Cooperation in the field of Scientific and Technical Research (COST) in 993 extended the frequencies of interest. The, which was renamed COST 3-, is applicable in the frequency range of MHz to MHz and can predict the median path loss for the distance of up to km between the transmitter and the receiver with transmitter antenna height of 3 m to m, and receiver antenna height of m to m []. COST 3- can be used to calculate path loss in three different environments; urban, suburban and rural (flat) areas. It can be expressed as [6]. P LCM = log f 3.84log H b ah m + ( log H b )log d C m () where P Lcm is the path loss, d is the distance between transmitting and receiving antennas (km), f is frequency (MHz), H b is the transmitter antenna height (m), ah m is the antenna height correction factor. The parameter C m has different values for different environments. It is given as db for rural as well as suburban and 3dB for urban areas. The parameter ah m (mobile antenna height correction factor) is defined in urban areas as ah m = 3.log (.7H r ) 4.8 (3) The value for ah m in suburban and rural (flat) areas is given as ah m = (.log f.7)h r (.log f.8) (4) H r is the receiver antenna height in metres. D. Ericsson Model The Ericsson software from Ericsson Company can be used to predict propagation path loss. Network planning engineers use it. This also builds on the modified Okumura- to allow room for changes in parameters according to the propagation environment. Path loss according to this is given as (Jalel etal., 3) P L = a + a log o d + a log H b + a 3 log (H b + d) 3.(log (.78H r )) + g(f) () g(f) = 44.49log f 4.78(log f) (6) where P L is the path loss, and f is frequency (MHz), H b is transmission antenna height (m), H r is receiver antenna height (m). The default values of these parameters a, a, a and a 3 for different terrains are given in Table I. TABLE I: VALUE OF a,a, a AND a 3 PARAMETERS FOR ERICSSON MODEL [] Environment a a a a 3 Urban Suburban Rural E. The Model Reference [4] proposed this in 98. In a very short time it became widely popular among researchers and system engineers mainly because the parameters of the can be easily adjusted to the local environment by additional field calibration. In the beginning, the was developed for use at 9 MHz and has two modes: areato-area and point-to-point. Built as two different s, this includes an adjustment factor that can be used to make the more flexible to different regions of propagation. is given as []. Scenario : Urban area, PL = log d + n log ( f 9 ) α (7) Scenario : Surburban, PL = log d + n log ( f 9 ) α (8) Scenario 3: Rural area, PL = log d + n log ( f 9 ) α (9) Where α is the correction factor which accounts for the base station and mobile station antenna heights, transmit powers and antenna gain and it is given as α = α + α + α 3 + α 4 + α () α = ( h b 3.48 ) () α = ( h m 3 ) k () α 3 = ( P t ) (3) α 4 = G b 4 (4) α = G m () P L is the path loss, n is the path loss exponent chosen to be 3, f is frequency, d is distance, h b is base station antenna height, h m is mobile station antenna height, P t is transmitter power, G b transmitter antenna gain, G m is mobile antenna gain. DOI: 4

3 F. Egli Model The Egli is a terrain for radio frequency propagation. It predicts the total path loss for point-to-point link (line of sight transmission). Typically, it is suitable for cellular communication scenarios where one antenna is fixed and another is mobile in the frequency range of 3 MHz to 3 GHz and where the transmission has to go over an irregular terrain. Egli is expressed as [3]. PL = { log f + P , h r log f + P , h r > (6) P = 4 log d log H b log H r (7) where H b is the height of the base station antenna, H r is the height of the mobile station antenna in meters, d is the distance from base station antenna to mobile station antenna in meters, and f is the frequency of transmission in megahertz (MHz). III. METHODOLOGY The method used in collecting data is the Drive Test (DT) technique and the resources required are: i. Vehicle to convey other mounted DT equipment around the test area at a recommended speed of 8 - km/hr. ii. Laptop to serve as interface between the TEMs software and TEMs mobile station and record the measured data such as base station ID, base station antenna height, frequency of transmission, received signal level (RLx), signal quality index (SQI), signalto-interference ratio (C/I), number of calls attempt, number of calls blocked, number of dropped calls as well as handover information during the drive test. iii. Drive Test Software (TEMS, Version..4 used) interfaces with TEMs mobile station for data measurement. iv. TEMS Mobile Station (MS) interfaces with base station through RF for data measurement. v. Global Positioning System (GPS) for positioning measurement during the DT. A. Calculation of Path Loss from Measured Data Let E min = Equivalent minimum field strength, A e = Effective antenna aperture, G = Antenna gain, c = signal speed, f= signal frequency, Ф min = Minimum power flux density at receiving place, P L = path Loss, P T = BTS power transmitted, P R = Mobile Station Power received. Then, given [3]. E min = Φ min + + log π = Φ min (8) Φ min = P smin A e (9) Putting (9) into (8), E min = P R (dbm) A e (3) But, P L = P T P R (3) Putting (3) into (3) gives E min = P T P L A e (3) But, A e = G + log (.64(C f ) ) (33) 4π Substituting (33) into (3) gives P L = P T (dbm) G + log (.64(c f ) ) E min (34) B. Models Performance Testing Criteria The performance of considered s were weighed in comparison with the experimental path loss using different statistical evaluation criteria such as Sum Square Error (SSE), Average Absolute Relative Error (AARE) and Nash- Sutcliffe efficiency coefficient (E). C. Sum Square Error (SSE) This describes the total deviation of the calculated path loss values (X C) from the existing path loss s output values (X E). Sum Square Error is given as []. SSE = N i= (X C X E ) (3) where N is the total number of data points predicted, X E is the existing values of path loss and X C is the calculated value of path loss. The with the lowest SSE is considered to be the most relevant to the study area. D. Average Absolute Relative Error (AARE) This is the average of the relative errors in the prediction of a particular variable and it is expressed as a percentage. Lower values of average absolute relative error (AARE) indicate better performance. It can be computed as follows []. AARE = N N X C X E i= X (36) X E E. Nash-Sutcliffe Efficiency Coefficient (E) The Nash-Sutcliffe efficiency coefficient is used to describe the accuracy of outputs in relation to calculated data. A value of E equal to depicts a perfect match between calculated path loss and existing path loss outputs; therefore, the closer the efficiency is to unity (+) the more accurate the []. E is computed as follows: E = (X C X E ) (37) (X E X ) E where X E is the average value of the path loss given by a particular existing. 4π DOI:

4 Sum Square Error (SSE) Carrier to Interference ratio (Unitless) Average Absolute Relative Error AARE) Nash-Sutcliffe Efficiency Coefficient (E) Path loss IV. RESULTS AND DISCUSSION The field strength data collected during the drive test were converted into path loss and were subjected to statistical performance evaluation criteria. A path loss prediction was therefore developed. Results are as presented Fig. to ,,,3,4,,6,7,8, Distance (km) Calculated path loss Okumura hata Egli Ericsson Model Fig.. Carrier to Interference ratio COST-3,,4,6,8 Distance (km) C/I Best (A) C/I Worst (A) C/I Best (C) C/I Worst ( C) Fig.. Plots of the values for experimental and considered path loss s over distance From the graph, the equation to fit the graphs is y =.4x.987 (38) where, y is the required modified path loss and x is COST 3 for path loss. However, the regression coefficient, R of the graphs is calculated to be R² =.999. This indicates a good relationship between the graphs because the closer the regression ratio to unity the better. Substituting (), that is, COST 3 path loss for x in the curve fitting equation ((38)) and simplify gives a modified path loss prediction for mobile communication for the studied area as P LCMM = log f 4.log H b.ah m + ( log H b )log d.c m (39) where P Lcmm is the modified path loss prediction for Osogbo, Nigeria. d the is distance between transmitting and receiving antennas (km), f is frequency of transmission (MHz), H b is transmitter antenna height (m), ahm is antenna height correction factor The Parameter C m has different values for different environments. It is given as db for rural as well as suburban and 3dB for urban areas. The parameter ah m (mobile antenna height correction factor) is defined in urban areas as ah m = 3.log (.7H r ) 4.8 (4) H r is the receiver antenna height in metres. Its coverage is for frequencies between MHz and MHz, mobile station antenna height between m and m and base station antenna height between 3 m and m., -, - -, - -, -3-3, -4-4, - -, -6 Fig. 3. SSE for the considered s and experimental path loss 3 Okumura Egli Cost-3 Ericsson Models Fig. 4. AARE for the considered s and experimental path loss Okumura Egli Cost-3 Ericsson Models Fig.. E for the considered s and experimental path loss DOI: 6

5 Path Loss (Calculatd) 6 9 Path Loss (Cost-3 Model) Calculated Path Loss Versus Model Linear (Calculated Path Loss Versus Model) Fig. 6. Curve-fitting of experimental path loss and COST 3 path loss V. CONCLUSION AND RECOMMENDATION Propagation path loss prediction has been developed for the study area using drive test technique in conjunction with six existing s. The indeed works very fine because there exists a high degree of correlation between its predicted path loss and experimental values when the was verified. The method is therefore recommended because it is less tedious and less expensive compared to deterministic method which requires detailed and accurate description of all objects in the propagation, and more expensive in computational effort. The produces relatively accurate results. The is extensively useful and highly recommended for conducting feasibility studies for signal prediction, coverage optimization and interference analysis during the initial phase of network planning in the study area and other areas with similar environmental and propagation characteristics. Dist ance (km) APPENDIX PATH LOSS VALUES FOR EXPERIMENTAL AND SELECTED MODELS Experi mental path loss Okumur a hata Calculated Path Loss Versus COST 3- Model Egli COST y =,4x -,987 R² =,999 Eric sson mode l Mode l REFERENCES [] Abidoye, L. K and Das, D. B, (4). Artificial Neural Network Modelling of Two Phase Flow in Porous Media. Journal of Hydro Informatics, doi:.66/hydro [] Abhayawardhana V.S., Wassel I.J., Crosby D., Sellers M.P., and Brown M.G., (). Comparison of Empirical Propagation Path loss s for fixed wireless access systems. 6th IEEE Technology Conference, Stockholm, pp ,. [3] Chhaya Dalela., (). Propagation Path loss Modeling for Deployed WiMAX Network. International Journal of Emerging Technology and Advanced Engineering. Vol., Issue 8, pp [4] Danladi T.A., Lawa A.U., and Aderinola M., (3).Studies on Effects of Building Internal Pattern on Downlink Mobile Phone Signal Strengths and Power Loss. The International Journal of Engineering and Science (IJES). Vol., Issue, pp [] Egli J.J., (97). Radio Propagation above 4 MC over Irregular Terrain. Proceedings of the IRE. Vol.4, Issue, pp [6] Goldsmith A., (). Wireless Communication. Cambridge University Press, New York. [7] Isabona Joseph., Konyeha. C. C., Chinule. C. Bright., and Isaiah Gregory Peter., (3). Radio Field Strength Propagation Data and Pathloss calculation Methods in UMTS Network. Advances in Physics Theories and Applications, Vol., 3. [8] Jalel Chebi., Ali K. Lawas., and Rafiigul Islam M.D., (3). Comparison between Measured and Predicted Path Loss for Mobile Communication in Malaysia. World Applied Sciences Journal (Mathematical Applications in Engineering). Vol., pp.3-8. [9] Josip Milanovic, Rimac-Drlje S., and Bejuk K., (7). Comparison of propagation accuracy for WiMAX on 3.GHz. 4th IEEE International conference on electronic circuits and systems, Morocco, pp. -4. [] Mohammad Shahajahan and Abdulla Hes-Shafi A.Q.M., (9). Analysis of Propagation Models for WiMAX at 3.GHz. MSc, Blekinge Institute of Technology, Karlskrona, Sweden. [] Segun I.P., and Olasunkanmi F.O., (4). Empirical Path Loss Models for GSM Network Deployment in Makurdi, Nigeria. International Refereed Journal of Engineering and Science (IRJES). Vol.3, Issue6, pp Lasisi Hammed is a lecturer and researcher in the department of Electrical and Electronics Engineering, Osun state University, Osogbo, Nigeria. He holds doctorate (Ph.D) and Master s degrees in Electrical and Electronics Engineering from University of Ilorin, Nigeria. His area of specialization is Telecommunication Engineering. He is a registered member of the Council for the Regulation of Engineering in Nigeria (COREN). He can be contacted on lashammed@yahoo.com.biography. Personal hobbies will be deleted from the biography. DOI: 7