Statistical Prediction Model of Rain and Dust Storm Worst Month in Microwave - Millimeter Wave Band

Transcription

1 Statistical Prediction Model of Rain and Dust Storm Worst Month in Microwave - Millimeter Wave Band Mohd. Ghazali Hamza Razak School of Engineering and Advanced Technology ehkmohamed2@live.utm.my Yusnaidi Md Yusof Malaysia-Japan International Institute-Technology Liza.A.Latiff Razak School of Engineering and Advanced Technology Yusaini H. Mohamed, M.F. Camara Faculty of Mechanical Engineering Abstract: - Recent development in the field of advanced communication services and applications has triggered interest in super high frequencies and significantly increased research activity in millimeter wave. In this paper, a new method of calculation of worst month in the statistical model has been developed for use in the rain and dust storm to predict the probability of exceedance are presented. A thorough review of various prediction models previously developed for rain or sand and dust storm are comprehensively carried out. Thus, annual rain rate and dust storm monthly variation of rate is predicted for different measurement of time of the year from the recorded data. Methodology for studying the annual rain and dust process are presented and applied to data collected from Sudan in five main climatic zones at specific areas between E and N. The rain and dust attenuation was predicted for some station of Sudan, namely Dongala, Port Sudan, Khartoum, ElFasher, and wad Madina. Moreover, formula for prediction of the worst- month long-term statistics for different weather induced propagation impairments was developed. The main contribution of this study is to assist in the design of both terrestrial and space radio links using frequencies above 10 GHz in the range of microwave and millimeter wave bands. Keywords: Worst- Month Propagation, Rain Attenuation, Sand and Dust storm; Microwave & Millimeter-wave. recent years have become parts and different regions of the world are suffering from climate change, which in turn became the affect of service operations, which must be studied and know the deal, including phenomena such as rain and dust storms, and snow [5-6]. Rainfall and dust storm statistically observed are most imperatives, important and demanding operational tasks and challenge made by meteorological services around the world Increasing in spectrum occupancy and demand for high bandwidth for the evolution of complex radio access network, the need to explore the advantages of millimeter wave band has become imperative. The advantages of the millimeter wave band are, 1) large spectrum availability, 2) high frequency, 3) small antenna, 4) equipment size. It is necessary to see the mathematical predictions or calculations of different parameters before going to design such type of systems. The quick growth in Links in wireless communications made it possible for the different area around the world as well as rural people to access the internet and cellular phones as well other applications. However, this growth came with many inherent issues. Some of which embody poor signal send and receiving in varied weather conditions. Figure 1 world s whole different climate zones classified in a range of various ways that however unremarkably used weather sectors include polar, arid, humid dry, tropical, and deserts. I. INTRODUCTION In recent years, there has been increased interest in the effects of rain and dust on millimeter wave communications and severe of signal propagation impairment. Over the few earlier decades, radio communication services and applications underwent broad development. The demands that a radio spectrum has to satisfy are larger through the day [1-4]. One of the most important characteristics of the propagation of signal are an environment of the path (attenuation) loss and an accurate prediction of the attenuation caused by rain or dust storm are essential for the design of communication systems. Moreover, a right estimation of the propagation losses provides a functional basis for an accurate choice of station locations and a correct determination of the frequency spectrum. In

2 Figure 1 world s different climates zone. Microwave and Millimeter wave bands links in space and terrestrial are prone to variations in attenuation over a large range due to rain and dust storm in links located in a raid and a tropical region, in some bad cases become more than 3 db/km. traditionally, it has been considered to be part of meteorological conditions, unavoidable variations. Propagation conditions vary from month to month and from year to year, and the probability of occurrence of these conditions may vary by as much as several orders of magnitude. Some phenomena occur only rarely, requiring many years of observation to make any conclusions. For instance, elevated ducting may occur only several times per year in some locations, and in many areas, rain intense enough to affect propagation paths occurs for less than 1 percent of a year. The outstanding sections of this paper are as follows: Effect of SDS to the higher frequency band and Microwave links is described in Section II. In Section III, we present the attenuation and development of prediction models and discussions. Finally, we end this study in Section IV with conclusion and recommendation and future work. II. RELEVANT STUDIES Rain attenuation is the dominant propagation impairment at frequencies above about 10 GHz. In addition, other impairments such as gaseous absorption, cloud attenuation, melting layer attenuation, and a tropospheric refractive effect becomes increasingly important with increasing operating frequency. A number of prediction models are available for the estimation of individual components [1]-[4]. However, methodologies that attempt to combine them in a cohesive manner are not widely available. This is partly due to the paucity of reliably measured data required to compare and verify such approaches. Three main categories of reduced visibility are often used to describe the severity of dust storms:1) Blowing Dust the horizontal visibility (Aeolian dust), is less than 11 km. 2) Dust Storm the horizontal visibility is less than 1000 m. 3)Severe Dust Storm the horizontal visibility is less than 200 m. The frequency of dust emission (FDE) is calculated using the present weather codes of SYNOP reports. FDE is highest at stations in Sudan and overall peaks in spring north of 23 N. Estimates statistics of various propagation effects that should be considered in the design of earth-space links. Propagation effects such as absorption, scattering and depolarization by hydrometeors, absorption due to atmospheric gasses, multipath effects and ionosphere effects (typically only notable below 1GHz) can cause signal fade and need to be considered when implementing a satellite system to maintain a quality of service. Statistics for propagation effects provide an attenuation cumulative distribution function (CDF), which can be combined with further ITU-R recommendations to create an overall average annual attenuation CDF. Other ITU-R recommendations include rainfall rate, P.837-5, rain attenuation, P.838-3, cloud attenuation, P.840, and gas attenuation, P.676. Significant of study it is the presence of the various forms of precipitation such as rain, snow, dust and haze in a radio wave or microwave path are always capable of producing major impairments to terrestrial communications. It is, therefore, necessary to identify and predict the overall impact of every significant attenuation effect along any given path. A procedure for predicting the combined effect of meteorological factors attenuation and several other propagation impairments along earth or space links is presented. There are several types of weather forecasts made in relation to time: Short-term activities is a weather observation made for a time period up to 48 hours. Medium - term activities are for a period extending from about 3-7 days. Long-term activities are for a period greater than seven days in advance but there are no absolute limits to the period. Table: Summary of Relevant Studies state in worst month Years References Locations Highlights Bashir[1] Yagasena[1] 2010 and after Emiliani[2] Owolawi[3] Villiers[4] CHEN Crane[8] Musa [5]Ting[9] Thorvaldsen[6] Malaysia(1) Colombia(1) South Africa(1) Abu Dhabi(1) U.S.A(1) Taiwan (1) Sudan(1) Norway(1) Malaysia(1) Nigeria(1) Iran(1) Iraq(1) Syria(1) To study monthly and seasonal rainfall and visibility, worst month and average worst month statistics of rain, and dust storm for the site under study. worst month distribution estimation problem because distributions may be predicted for each of the months in a year and the worst one is then readily apparent The worst month concept is relevant to the higher grade telecommunications services under which propagation engineers must design radio systems which ultimately determine link availability.

3 Meteorological data are used to give an indication of the statistical rate of dust storm occurrence of a given of rain and dust are serious influences the performance of a communications satellite and ground link. The concept of worst-month plays an important role in space and earth links design where there is a need to know the design margin that must be met in any particular month of the year. Worst-month statistics can be applied to quantities such as dust storm attenuation, visibility and polarization planes. The worst month statistics for a given link are obtained by compiling a composite curve using the highest exceedance probability obtained in any calendar month at each threshold level. From the literature reviewed considerable studies have been indicated around the world in relation to the spatial and worst month of rainfall and dust storm including extreme events overall contents. The worst-month and annual statistics is related by the following ratio: Q = X/Y (1) Where X is the average worst-month probability and Y is the average annual probability for the same threshold. Q is a function of the occurrence month level and the climatic region. Similar climatic regions will have similar values of Q in percentage. From a long term ( ) dust storm data more than twenty-five years data, we could find the average of the highest frequency of occurrence of dust storm events happens in the June in Capital Khartoum. Hence, June is the worst time statistics. Optical visibility is directly related to the severity of dust-storm. Low visibility implies high number concentration of particles while low number concentration of particles represents high visibility the following model [1]-[8-9] gives the monthly probability of visibility to be less than a distance (li): P (V<li) = [(ST t li)] / (30 24 L)] 100% (2) Where, P is the probability of exceedance in %, t is average dust-storm duration in hours, V is optical visibility in meters, T is an average number of storms per worst-month, and L is the maximum length of visibility considered in meters. We have been able to trace from data obtained from the Sudan Meteorological Authority (SMA -2014) that the visibility ranges from 0.1 km to 5 km and more in very few cases. The maximum length of visibility (L) considered in this case 1000 m. The table below depicts the visibility and its corresponding probability statistic results of the worst-month. We used the same model for rainfall worst-month statistics are useful in designing high quality MW communication links since the maximum occurrence of events that leads to outages is expected to be higher in the worst-month. Therefore, to meet the performance criteria of any month, communication systems designers need a model for the worst-month or worst 30-day statistics. The model is given as follow: P(V<l_i )=[(ST τr' l_i)/(30 24 L)] 100% (3) In which, P= probability of exceedance of rain, τr= rainfall duration in hours, V= rain visibility in km, ST= average number of storms per worst-month, L= maximum length of visibility in km, l_i = range of visibility. Worst-Month Visibility Statistics Worst-month statistics are of interest to those faced with designing a system to meet performance criteria expressed in terms of a percentage of any calendar month, or of any continuous 30-day period. The system designer in many cases needs to find the percentage of time that some threshold value of attenuation or rain rate and dust storm will be exceeded within a given month. For every threshold value, there corresponds a month of the year having the highest percentage of time exceeding the threshold (i.e., the percentage exceedance). This is designated the "worst-month" for that threshold. The percentage exceedance in this month, to be expected once every year or every given number of years, is of most interest. For high rain rates, the worst-month would probably correspond to the period of highest thunderstorm intensity or frequency whereas the worst-month for lower rain rates might be when most rainfall is of the steady, stratiform variety. An exponential model has been devised [9-10] for estimating the ratio of the percentage exceedance for a given threshold value in the worst-month to the average annual percentage exceedance for the same threshold. This exponential relationship is expected for statistics, the worst-month is defined as the month with the highest probability of exceeding the threshold level. Optical visibility is directly related to the severity of dust-storm. Low visibility implies high number concentration of particles while low number concentration of particles represents high visibility. III. METHODS The methodology of your paper should be described in this section. The fading and enhancement distributions for the average worst month obtained from the methods can be converted to distributions for the average year by employing the following procedure: Step 1: Calculate the percentage of time PW fade depth A is exceeded in the large tail of the distribution for the average worst month as appropriate, follow the method where A 0.01 is now the fade depth exceeded for 0.01% of the time in the average year. Obtain first pw by inverting and using p 0.01%. Then obtain fade depth A 0.01 exceeded for 0.01% of the time in the average year by inverting as appropriate, and using p in place of pw. The meteorological data that used in this study has been brought from SMA, Sudan for 10 years from 2004 to 2014 for 5 meteorological stations over the country with 1756 total number of examples. Figure 2 shown general method the used in the study. In addition to, figure 3 shown

4 Figure 2: Flow chart of Methodology of study Millimeter wave frequencies for ground or satellite links are prone to variations in attenuation over a large range due to rain and dust storm, more than 30 db some time. Traditionally, it has been considered to be part of meteorological conditions, unavoidable variations. A mathematical model is often an approximate representation of a more complex system. It is important to consider many technical issues before going to design and establish an expensive Wireless system. It is necessary to see the mathematical predictions or calculations of different parameters before going to design such type of systems. A mathematical model is often an approximate representation of a more complex system. In modeling complex systems, model parameters often abound in number. The value of these parameters may be approximately determined through the fitting of model predictions with calibration data obtained from laboratory experiments, first principle arguments, ab initio calculations, more refined models, etc. Unfortunately, repeating a given experiment multiple times may yield different results that are suitably described by a statistical distribution. Figure 3 five 5 meteorological stations of study in Sudan Consequently, model parameters obtained from calibration data sources are themselves often described statistically. This statistical model parameter uncertainty then potentially affects all future predictions that make use of this model. The major task at hand is to propagate this model parameter uncertainty throughout subsequent calculations, thus quantifying the statistical behavior of derived outputs. We refer to this task as uncertainty quantification from statistical model parameter uncertainty. Table 1 Total annual precipitation in Sudan last 10 years

5 Station _ID Station Long Lat Elev (m) Rec Yrs. Prep (mm) Dongola Port Sudan Khartoum ElFashir Wad Medani Sudan is the second largest country in the Middle East and Africa, with an area of about 1.3 m km2. It extends from the Sahara region of Africa to near the Equator. The population of Sudan is about 26 million and the main water resource is the surface water of the Nile and its tributaries. Rainfall amounts decrease from the south to the north and sand and dust storm form the north to the south. The number of stations studied was five. The area covered whole areas of the study area with a station density of approximately one station per km2. In fact, the length of the rainy and dusty season different zone in Sudan varies with latitude ranging from about one month in the northern extreme to five months in the southern areas. Throughout the region, August is the month of peak rainfall and in all but in the northern Sudan rainfall is also relatively high in July and September. Figure 4 shows the highest concentration of rainfall in August during the whole of the study period except for The results illustrated in the above-mentioned figure indicate that the perspective in mm and the rainy months, during each year, may represent the length of the rainy season at wad Madina in the Central Sudan during the ten years. For example, in 2004 the duration of this season is about 65 days. Figure 4 Average Annual Dust and Precipitation in Sudan IV. RESULTS AND DISCUSSION Currently, fast increasing in radio communication marketplace stronger and stronger demands on the design of a radio system. A suitable system proposal requires accurate and reliable radio links model, among which the arena asset statistical prediction models are significant of worst month in super high frequency. Although many researchers have been working hard during the past few decades in the area of field strength prediction in rain, there are still many problems to be solved. Empirical prediction models often provide a fair description of the fading process for distances and frequencies that lie within the data range for which measurements have actually been collected. For the design of communication system, a required availability, statistical knowledge of climatic propagation effects is essential. Predicting attenuation due to dust storm is an important but complex problem, and a variety of models have been developed [2]. Propagation conditions vary from month to month and from year to year, and the probability of occurrence of these conditions may vary by as much as several orders of magnitude. Some phenomena occur only rarely, requiring many years of observation to make any conclusions. For instance, elevated ducting may occur only several times per year in some locations, and in many locations, rain intense enough to affect propagation paths occurs for less than 1 percent of a year. In parts of Sudan at locations where the rain and dust occur some years but not others, several significant climate events may occur one year but not the following year. Worst-month statistics is of interest to those faced with designing a system to meet performance criteria expressed in terms of a percentage of any calendar month, or of any continuous 30-day period. The system designer, in this case, needs to find the percentage of time that some threshold value of attenuation or rain rate and dust storm will be exceeded within a given month. For every threshold value, there corresponds a month of the year having the highest percentage of time exceeding the threshold (i.e., the percentage exceedance). This is designated the "worst-month" for that threshold. The percentage exceedance in this month, to be expected once every year or every given number of years, is of most interest. For high rain rates or high visibility in a dust storm, the worst-month would probably correspond to the period of highest thunderstorm intensity or frequency whereas the worst-month for lower rain rates might be when most rainfall is of the steady, stratiform variety. Long-term surface observations are recorded as SYNOP reports. The annual, monthly, and daily data was supplied by the Sudan Meteorological Authority Khartoum, Sudan. The latter extends the records up to and including the year 2014 in contrast to the monthly data set which extends, for most of the stations. The percentage of data available covers more than 96.5% of the entire records. Compared with that of other Sudan stations, it can be concluded that the records of the Sudanese stations are more complete. V. CONCLUSION Implications of the results and future research directions are also presented. The main findings from this work can be summarized as follows: various prediction models developed for rain or sand and dust storm are comprehensively described. Secondly, determination of the attenuation statistics is indispensable in planning and designing of wireless communications systems and links. Worst-month statistics of total attenuation were studied by using data collected from five stations in the Sudan area at 4 different zones. A new and better approach to getting the results is to use both the rain and dust storm the distribution along term with one values to describe in one model are presented. Fast increasing in radio communication marketplace stronger and stronger demands on the design of a radio system. A suitable system proposal requires accurate and reliable radio links model, among which

6 the arena asset statistical prediction models are significant of worst month in super high frequency. Although many researchers have been working hard during the past few decades in the area of field strength prediction in rain, there are still many problems to be solved. The need to design radio communication systems to meet performance and availability objectives in any month is a task that must be accomplished. Since propagation conditions vary considerably from month to month and the monthly variability can change significantly from year to year, the propagation community came up with the concept of the worst month. ACKNOWLEDGMENT We wish to thank Razak School of Engineering and Advanced Technology, (UTM), Malaysia for providing the facilities, tools, and equipment for the radio signal propagation research. REFERENCES [1] S. O. Bashir, Statistical Modelling of Propagation Parameters through Sand/Dust Storms at Microwave Frequencies, IEEE Intern Conf on Antennas, Propagation and Systems, Johor Bahru Malaysia, 3-5, [2] J. A. Fernando, AndradeI, A. R. Luiz da Silva MelloI, Marlene S. PontesI, Marcio E. C. Rodrigues Statistical modeling of rain attenuation in tropical terrestrial links J. Microw. Optoelectron. Electromagn. Appl. vol.11 no.2 São Caetano do Sul Dec [3] Yagasena and S. I. S. Hassan, "Worst-month rain attenuation statistics for satellite-earth link design at Ku-band in Malaysia," TENCON Proceedings, Kuala Lumpur, 2000, pp vol.1. [4] Emiliani, L. D. "Worst-Month Rain Rate and Rain Attenuation Statistics for Satellite System Design in a Mountainous Region in Colombia," in IEEE Antennas and Wireless Propagation Letters, vol. 5, no. 1, pp , Dec [5] P.A. Owolawi and T.J. Afullo, Rainfall rate modelling and worst month statistics for millimetric line-of sight radio links in South Africa, Radio Sci., vol. 42, [6] A. Musa, S. O. Bashir, Md Rafiqul Islam and Othman O. Khalifa Dust- Storm Induced Cross-Polarization at MMW Bands in Northern Nigeria International Conference on Computer and Communication Engineering (ICCCE 2012), 3-5 July 2012, [7] CHEN C, CHEN Y, LIU C, LIN P and CHEN W Statistics of heavy rainfall occurrences in Taiwan. Weather Forecasting Vol 22 (5) [8] M. P. de Villiers and J. van Heerden Dust storms and dust at Abu Dhabi international airport Royal Meteorological Society, 2007 DOI: /wea.42 [9] Teh Sieh Ting and J.S. Mandeep Analysis of Worst-month Relationship with Annual Rain Attenuation in Malaysia Research Journal of Applied Sciences, Engineering, and Technology 7(7): , 2014 [10] Per Thorvaldsen and Ingvar Henne Long-term propagation measurements on a line-of-sight over-water radio link in Norway [11] R. Kumar, M. C. Barth, G. G. Pfister, M. Naja, and G. P. Brasseur WRF-Chem simulations of a typical pre-monsoon dust storm in northern India: influences on aerosol optical properties and radiation budget Atmos. Chem. Phys., 14, , [12] R. K. Crane, "A local model for the prediction of rain-rate statistics for rain attenuation models," in IEEE Transactions on Antennas and Propagation, vol. 51, no. 9, pp , [13] Ismail, Ahmad Fadzil and Abd Rashid, Nur Emeilin and Din, Jafri and Abdul Rahman, Tharek and Islam, Md. Rafiqul (2012) Analyses of worst-month rain fade statistics for Ku-band (26GHz) in the tropics. In: th International Conference on Telecommunication Systems, Services, and Applications (TSSA), October 2012, Bali, Indonesia. [14] M. F. Camara, S. O. Bashir, F. N. M. Isa and A. Musa, "Prediction of Rain-Induced Cross Polarization at Millimeter Wave Bands in Guinea," Computer and Communication Engineering (ICCCE), 2014 International Conference on, Kuala Lumpur, pp , 2014.