Path-Loss Estimation for Wireless Cellular Networks Using Okumura/Hata Model

Transcription

1 Science Journal of Circuits, Systes and Signal Processing 018; 7(1): doi: /j.cssp ISSN: (Print); ISSN: (Online) Path-Loss Estiation for Wireless Cellular Networks Using Okuura/Hata Model Ibrahi Mohaed Electrical and Electronic Departent, Faculty of Engineering, Oar Al-Mukhtar University, Al-Beida, Libya Eail address: To cite this article: Ibrahi Mohaed. Path-Loss Estiation for Wireless Cellular Networks Using Okuura/Hata Model. Science Journal of Circuits, Systes and Signal Processing. Vol. 7, No. 1, 018, pp doi: /j.cssp Received: Noveber 30, 017; Accepted: Deceber 9, 017; Published: January 11, 018 Abstract: In cellular networks, accurate path-loss estiation is highly desirable not only to iprove the perforance but also to realize precise estiation of financial feasibility. In other words, while accurate path-loss estiation helps to achieve an acceptable perforance and reasonable cost, inaccurate path-loss estiation would either lead to degrade the perforance, or increase the cost. Different odels were introduced in the literature for achieving accurate path-loss estiation. One of theses odels is the Okuura/Hata odel which is recoended for being used to estiate the path-loss in the cellular systes that eploying icro-cells. It is characterized by its suitability for being used in a variety of environents. The ain objective in this paper is to provide a guide line for path-loss estiation analysis using Okuura/Hata odel. Matlab software was used to perfor this analysis. Copared with free space odel in which frequency and separation distance are the only contributors for path-loss, ore accurate estiation can be achieved when Okuura/Hata odel is used as it includes further correction factors, such as obile station antenna height and base station antenna height. Keywords: Cellular Networks (CNs), Path-Loss Estiation Models, Free Space Propagation Model, Okuura/Hata Model 1. Introduction One of the ain objectives of cellular systes is to achieve high capacity (increase the nuber of user). This objective can be realized by creating a distinctive way by which the liited frequency spectru assigned to the cellular systes is exploited efficiently. The concept of frequency reuse in which a segent of specific frequency spectru can be used several ties is the key to achieve an efficient exploit of the assigned spectru. In frequency re-use theory, each cell within a single cluster represents a replica of one of six adjacent cells (co-channel cells) surrounding it. Each cochannel cell belongs to a neighbor cluster. Although a lowpower transitter is coonly eployed in each cell, a sufficient distance should be deterined for co-channel cells separation to guarantee iniu interference. Although interference can also coe fro the second and higher tires co-channel cells, such interference can be ignored as it often contributes by less than one percent (1%) of the total interference [1]. As a type of counication syste, the perforance of a cellular syste is indicated by its signal-tonoise ratio SNR. While the ter signal refers to the power of an intended carrier, noise represents the power produced at the receiver due to theral effect plus the su of the unwanted but received powers produced by the co-channel cells. For acceptable perforance, a cellular syste should eet a specific value of SNR. SNR ay differ fro syste to another. For exaple, approxiately, a iniu SNR value of 18 db or 1 db is specified for acceptable perforance when advanced obile phone syste AMPS or global syste for obile counication GSM is considered, respectively []. In cellular systes, a wireless channel is ainly characterized by its effects of dispersion and attenuation. While dispersion is analyzed to verify the kind of signal distortion encountered during propagation, and thus take the proper solution to iniize it, attenuation is analyzed to estiate the path-loss. Dispersion can be categorized as either tie-based dispersion or frequency-based dispersion. In a wireless channel, a signal ight be attenuated due to different phenoena, such as reflection, diffraction, or scattering. Reflection occurs when a signal that is propagating in a ediu passes to another ediu with different properties. Attenuation in this case can be attributed to that a part of the incident signal energy is either absorbed

2 Science Journal of Circuits, Systes and Signal Processing 018; 7(1): or propagated into the reflecting ediu. Diffraction is defined as the deviation of a signal fro its path. It occurs when a signal that is propagating in a ediu passes into a shadow region created by an obstruction. Scattering is defined as the spreading of energy out of its intended path. It occurs when a signal hits an object whose size is uch saller than or on the order of the signal wavelength. Accurate path-loss estiation is highly recoended in the initial stages of cellular network design to precisely deterine the nuber of cell sites required for providing coverage in a given area, which leads to achieve a siple estiation of financial feasibility. Additionally, accurate path-loss estiation can help to realize an acceptable syste perforance. Figure 1(a), (b), and (c) clarifies the aforeentioned situations in which the estiated path-loss is alost equal to the actual path-loss in (a), lower than the actual path-loss in (b), or higher than the actual path-loss in (c). Moreover, accurate path-loss estiation can help to achieve an optiu cell site location []. Figure 1. Effect of accurate and inaccurate path-loss estiation. Different odels were introduced in the literature for achieving accurate path-loss estiation [4] [5] [6] [7] [8]. Path-loss odels can be categorized according to the separation distance as either long-distance prediction odels intended for acro-cells or short-distance prediction odels intended for icro-cells. Pico-cells (cells that cover part of a building and ainly span fro 30 to 100 eters) can be included in the short-distance prediction odels. Indoor prediction odels were introduced to estiate the path-loss in this case. Figure provides a diagraatic categorization of the path-loss odels. They started with the siplest lineof-sight path-loss odel; also referred to as free space propagation odel. In free space propagation odel, no obstructions due to earth surface or other obstacles are encountered during propagation. Aong these odels is the Okuura/Hata odel which is characterized by its suitability for being used in a variety of non-line-of-sight environents (i.e. typical urban, typical suburban, and rural environents). Figure. Widely used path-loss odels. Unlike free space odel in which frequency and separation distance (distance between transitter and receiver) are the only contributors used for estiating the path-loss, further factors such as obile station antenna height and base station antenna height are included in the Okuura/Hata odel to achieve ore accurate path-loss estiation. In this paper, the author ais to provide a guide line for cellular network designers and operators to estiate the path-loss using Okuura/Hata odel. This paper is organized as follows: Section provides theoretical background of Free-space path-loss estiation odel and Okuura/Hatapath-loss estiation odel. Results are discussed in section 3. Section 4 concludes the paper.

3 Ibrahi Mohaed: Path-Loss Estiation for Wireless Cellular Networks Using Okuura/Hata Model. Theoretical Background In cellular networks, accurate path-loss estiation helps to precisely deterine the nuber of cell sites required for providing coverage in a given area. While accurate path-loss estiation leads to realize an acceptable syste perforance and achieve reasonable cost estiation, inaccurate path-loss estiation would either lead to degrade the syste perforance or increase the cost. Various path-loss estiation odels were introduced in the literature; each of which aiing to achieve accurate path-loss estiation. In this section, theoretical background of free-space path-loss estiation odel and Okuura/Hata path-loss estiation odel is provided. frequency carrier used to achieve accurate path-loss estiation. The path-loss in db according to Okuura/Hata odel for typical urban environent is given as [9] [10].1. Free-Space Path-Loss Estiation Model Free space odel is one of the well-known long-distance prediction odels. In free space odel, a line-of-sight propagation path is assued. i.e., no obstructions exist between transitter and receiver albeit by the surface of earth. It is ostly applicable for wireless channels in which a transission syste such as icrowave or satellite-tosatellite is eployed. The path-loss in this case is referred to as free space path-loss ( L Pfree ). Free space path-loss is given as [9] 4πd LPfree = λ where d and λ represent the separation distance (i.e., distance between the transitter and receiver) and wavelength, respectively. Given λ = c/ f yields 4π fd LPfree = c where c and f represent the speed of light (3x10 8 /sec) and frequency in hertz Hz, respectively. Substituting d in kiloeter and f in egahertz, free space path-loss can be expressed in db as LPfree[ db] = Logf + 0Logd (1) It can be observed fro (1) that duplication of frequency or distance would lead an increase in the path-loss by 6 db... Okuura/Hata Model Okuura/Hata odel is one of the well-known non-loneof-sight odels. It was priarily developed for path-loss estiation in typical flat urban environent. Then it was iproved to include path-loss estiation in the typical suburban and rural environents. A correction factor that is related to the obile station antenna height was included in different fors according to the environent considered and Figure 3. Sequence of the progra operation where typical urban environent Okuura/Hata odel for large cities is considered. L ( urban) = Logfc + ( Logh ) Logd 13.8 Logh a( h ) () 50 [ db] b b Where fc is the carrier frequency in MHz, d is the separation distance between the obile station and base station in K, h b and h are the base station and obile station antennas height in eter, and a( h ) is correction factor that is related to the obile station antenna height. a( h ) is given in different fors according to the size of the intended area. For large cities and at fc 00 MHz, a( h ) is given as

4 Science Journal of Circuits, Systes and Signal Processing 018; 7(1): a( h ) = 8.9[ Log(1.54 h )] 11 (3) For large cities and at fc 400 MHz, a( h ) is given as a( h ) = 3.[ Log(11.75 h )] 4.97 (4) For sall and ediu-sized cities a( h ) = [1.1 Log( fc) 0.7] h [1.56 Log( fc) 0.8] (5) The path-loss in db according to Okuura/Hata odel for the typical suburban environent is given as fc L50 ( suburban) [ db] = L50( urban) Log The path-loss in db according to Okuura/Hata odel for the rural environent is given as (6) 3. Results and Discussion This section is divided into two subsections. Results that estiate the path-loss for Free-space path-loss prediction odels were provided in the first subsection whereas; results that estiate the path-loss estiations for Okuura/Hata prediction odel were provided in the second subsection. In both subsections, graphical representations of path-loss versus separation distance at different effective factors were provided. In all cases, path-loss at any intended distance can be readily estiated by drawing a vertical line at that distance such that it intersects the path-loss versus distance curve and then draw a horizontal line fro the intersection point; the value of path-loss in this case is the intersection point with the path-loss axis. Carrier frequencies of 900 MHz, 1800 MHz, and 100 MHz were selected in the analysis as they represent the carrier frequencies used in the GSM standard in its fors of nd, and 3 rd generations cellular networks, respectively. The Matlab software (The Mathworks, Inc., Natick, MA, USA) was used for realizing the graphical representation Path-Loss Estiations Considering Free-Space Model Figure 4. Path-loss vs separating distance at different carrier frequencies: Free space odel was considered. 50 ( )[ db] = 50( ) 4.78( ) (7) L rural L urban Logfc Logfc In this part of results, the path-loss is estiated by considering equations (1). Figure 4 shows the graphical representation of (1). It represents the path-loss vs separating distance at different carrier frequencies, where free space odel is considered. It can be obviously seen fro the graph that the path-loss starts to appear at a certain value of separation distance and continues to increase rapidly with a slight increase in the separation distance. However, it starts to increase steadily at higher values of separating distance. It can be obviously observed fro the graph that the sallest value of path-loss is encountered when the lowest carrier frequency is assigned. 3.. Path-Loss Estiation Considering Okuura/Hata Model Table 1 provides the range of paraeters for which Okuura/Hata odel is valid [9]. Table 1. The range of paraeters for which Okuura/Hata odel is valid. Miniu value Maxiu value Carrier frequency fc (MHz) Base station height h b () Mobile station height h () 1 10 Separation distance d (K) 1 0 Going through details of sections., one can note that it ight be needed to perfor a long and coplex calculation approach to estiate the path-loss when Okuura/Hata odel is used. To avoid coplexity in calculations and save the tie, the approach we used to estiate the path-loss was by coposing MATLAB coputing progras and involving the loop echanis in these progras. Figure 3 is a sapled flowchart describing the sequence of progra operation. Figure 5. Path-loss vs separating distance at different carrier frequencies: typical urban environent Okuura/Hatta odel for large cities is considered..

5 4 Ibrahi Mohaed: Path-Loss Estiation for Wireless Cellular Networks Using Okuura/Hata Model In this part of results, the path-loss is estiated by considering equations () to (7). Graphical representation of the path-loss versus separation distance (1K to 0 K) was provided. Further graphical representations that include significant paraeters which ight affect the path-loss estiation, such as carrier frequency (900, 1800, and 100 MHz), and base station height (0, 80, 140, and 00 ) were provided. Figure 5 shows the graphical representation of equations (), (3) and (4). It represents the path-loss vs separating distance at different carrier frequencies, where the typical urban environent Okuura/Hata odel for large cities is considered. It can be obviously seen fro the graph that the path-loss starts to appear at a certain value of separating distance in which the odel is valid (1 K) and continues to increase with the increase of the separating distance. Figure 6 (a, b and c) shows the path-loss versus separating distance at different values of base station heights where typical urban environent Okuura/Hata odel for large cities is considered. It can be obviously seen fro the graph that the path-loss starts to appear at a certain value of separating distance in which the odel is valid (1 K) and continues to increase with the increase of the separating distance. However, an increase in the base station height could lead to reduce the path-loss. Figure 7 shows the graphical representation of equations () and (5). It represents the path-loss vs separating distance at different carrier frequencies, where the typical urban environent Okuura/Hata odel for sall and id-sized cities is considered. It represents the path-loss vs separating distance at different carrier frequencies, where the typical urban environent Okuura/Hata odel for sall and id-sized cities is considered. The graph follows the sae behavior as in Figure 5 with a slight difference can be observed. For exaple, a path-loss value of db was easured at 0 K separation and 100 MHz, when typical urban environent Okuura/Hata odel for large cities was considered whereas a path-loss of db was easured at an identical separation distance and frequency carrier when typical urban environent Okuura/Hata odel for sall and id-sized cities was considered. Figure 6. Path-loss versus separating distance at different base station heights: Okuura/Hatta odel for typical urban environent was considered. Figure 7. Path-loss vs separating distance at different carrier frequencies: typical urban environent Okuura/Hatta odel for sall and id-sized cities is considered.

6 Science Journal of Circuits, Systes and Signal Processing 018; 7(1): and id-sized cities is considered. The graph follows the sae behavior as in Figure 6 with a slight reduction in the path-loss can be observed. For exaple, a path-loss value of db was easured at 00 base station height, 0 K separation, and 100 MHz when typical urban environent Okuura/Hata odel for sall and id-sized cities was considered whereas a path-loss of db was easured at an identical base station height, separation distance, and frequency carrier when typical urban environent Okuura/Hata odel for large cities was considered. Figure 9 shows the graphical representation of equation (6). It represents the path-loss vs separating distance at different carrier frequencies, where the typical suburban environent Okuura/Hata odel is considered. The graph follows the sae behavior as in Figures 5 and 7 with a slight difference can be observed. For exaple, a path-loss value of db was easured at 0 K separation and 100 MHz, when typical suburban environent Okuura/Hata odel was considered whereas a path-loss of db, or db were easured at an identical separation distance and frequency carrier when typical urban environent Okuura/Hata odel for sall and id-sized cities, or typical urban environent Okuura/Hata odel for large cities were considered, respectively. Figure 9. Path-loss vs separating distance at different carrier frequencies: typical suburban environent Okuura/Hatta is considered. Figure 8. Path-loss versus separating distance at different base station heights: Okuura/Hatta odel for typical urban environent was considered. Figure 8 (a, b and c) shows the path-loss versus separating distance at different values of base station heights where typical urban environent Okuura/Hata odel for sall Figure 10 (a, b and c) shows the path-loss versus separating distance at different values of base station heights where typical suburban environent Okuura/Hata odel is considered. The graph follows the sae behavior as in Figures 6 and 8 with a decrease in the path-loss can be observed. For exaple, a path-loss value of db was easured at 00 base station height, 0 K separation, and 100 MHz when typical suburban environent Okuura/Hata odel was considered whereas a path-loss of db or db were easured at an identical base station height, separation distance, and frequency carrier when typical urban environent Okuura/Hata odel for sall and id-sized cities or typical urban environent Okuura/Hata odel for large cities were considered.

7 6 Ibrahi Mohaed: Path-Loss Estiation for Wireless Cellular Networks Using Okuura/Hata Model Figure 11. Path-loss vs separating distance at different carrier frequencies: rural environent Okuura/Hatta is considered. Figure 10. Path-loss versus separating distance at different base station heights: Okuura/Hatta odel for typical suburban environent was considered.

8 Science Journal of Circuits, Systes and Signal Processing 018; 7(1): contributors for path-loss, ore accurate estiation can be achieved when Okuura/Hata odel is used as it includes further correction factors, such as obile station antenna height and base station antenna height. Going through results obtained, there were soe significant observations that can be exploited advantageously. For exaple, results on Figures 6, 8, 10, and 1 confir that an increase in the base station height leads to reduce the path-loss. This can be exploited advantageously and thus give the opportunity to achieve a lower path-loss. However, uch higher base station would affect the cost effectiveness of the cellular syste. On other words, the base station height should be chosen such that it is coparable to the surrounding buildings. References [1] Jochen H. Schiller, Mobile Counications, PEARSON EDUCATION LIMITED, 003 Figure 1. Path-loss versus separating distance at different base station heights: Okuura/Hatta odel for rural environent was considered. Figure 11 shows the graphical representation of equation (7). It represents the path-loss vs separating distance at different carrier frequencies, where the rural environent Okuura/Hata odel for is considered. The graph follows the sae behavior as in Figures 5, 7 and 9. However a notable decrease in the path-loss can be observed. For exaple, a path-loss value of db was easured at 0 K separation and 100 MHz, when rural environent Okuura/Hata odel was considered whereas a path-loss of db, db, or db were easured at an identical separation distance and frequency carrier when typical urban environent for sall and id-sized cities, typical urban environent for large cities, or typical suburban environent Okuura/Hata odels were considered, respectively. Figure 1 (a, b and c) shows the path-loss versus separating distance at different values of base station heights where rural environent Okuura/Hata odel is considered. The graph follows the sae behavior as in Figures 6, 8, and 10. However a notable decrease in the path-loss can be observed. For exaple, a path-loss value of db was easured at 00 base station height, 0 K separation, and 100 MHz when rural environent Okuura/Hata odel was considered whereas a path-loss of db, db, or db were easured at an identical base station height, separation distance, and frequency carrier when typical suburban environent, typical urban environent for sall and id-sized cities, or typical urban environent for large cities Okuura/Hata odels were considered, respectively. 4. Conclusions A guide line for path-loss estiation using Okuura/Hata odel was provided. Copared with free space odel in which frequency and separation distance are the only [] Andrea Goldsith, WIRELESS COMMUNICATIONS, Cabridge University Press, 005 [3] Yoo-Seung Song and Hyun-Kyun Choi, Analysis of VV Broadcast Perforance Liit for WAVE Counication Systes Using Two-Ray Path Loss Model, ETRI Journal, Volue 39, Nuber, pp 13-1, April 017 [4] Divya Kurup, Maria Scarpello, Günter Vereeren, Wout Joseph, Kristof Dhaenens, Fabrice Axisa, Luc Martens, Dries Vande Ginste, Hendrik Rogier, Jan Vanfleteren, In-body path loss odels for iplants in heterogeneous huan tissues using iplantable slot dipole conforal flexible antennas, EURASIP Journal on Wireless Counications and Networking 011 [5] Haipeng Ding, Zhengyuan Xu, Brian M. Sadler, A Path Loss Model for Non-Line-of-Sight UltravioletMultiple Scattering Channels, EURASIP Journal onwireless Counications and Networking Volue 010, Article ID 59857, 1 pages [6] Mario Versaci, Salvatore Calcagno, Fabio La Foresta, Biagio Caaroto, PATH LOSS PREDICTION USING FUZZY INFERENCE SYSTEM AND ELLIPSOIDAL RULES, Aerican Journal of Applied Sciences, Science Publication, 01, 9 (1), [7] A. Bhuvaneshwari, R. Healatha, T. Satyasavithri, Sei Deterinistic Hybrid odel for Path Loss prediction iproveent, nd International Conference on Intelligent Coputing, Counication & Convergence (ICCC-016), pp [8] Yun-Jie Xu, Wen-Bin Li, Propagation path loss prediction odel of ulti-sensor network in forest, Advanced in Control Engineering and Inforation Science, pp-06 10, published by Elsevier, 011 [9] Vijay K. Garg, Wireless Counications and Networking, Elsevier, 007 [10] Kazunori Uchida; Naoto Hadano; Masafui Takeatsu; Junichi Honda, Propagation Estiation by Using Building Coverage and Floor Area Ratios Based on 1-Ray Model Cobined with Okuura-Hata Mode, 17th International Conference on Network-Based Inforation Systes, 014, pp: