Performance Enhancement of LMDS Systems Using Adaptive Coded Modulation and SC Diversity under the Impact of Rain Attenuation in Indonesia

Transcription

1 Performance Enhancement of LMDS Systems Using Adaptive Coded Modulation and SC Diversity under the Impact of Rain Attenuation in Indonesia Suwadi #, G. Hendrantoro *, Wirawan #3 # Department of Electrical Engineering, Institut Teknologi Sepuluh opember, Indonesia Kampus ITS, Keputih-Sukolilo, Surabaya, 60 suwadi@ee.its.ac.id gamantyo@ee.its.ac.id 3 wirawan@ee.its.ac.id Abstract This paper presents the effects of rain on the performance of local multipoint distribution service (LMDS) in Indonesia. The significantly propagation factor effecting the link availability and channel capacity of the LMDS system operating at 30 GHz frequencies is rain attenuation. Adaptive coded modulation (ACM) and selection combining (SC) are methods to increase link availability and channel capacity for the system. The results of simulation study by using ACM and SC show that performance of the LMDS system is guaranteed to have signal reception quality with maximum BER 0 - in km link length. The performance also shows that the link availability is 99.99%. At 4 km link length, ACM and SC can increase 0.45% if they are compared with fixed transmission mode (for example, 64 QAM). The results also show that the channel capacity is bps/hz for the ACM and SC. Furthermore, the simulation study shows that the ACM and SC can enhance the LMDS performance. Index Terms LMDS, ACM, SC diversity, rain attenuation I. ITRODUCTIO The demand for broadband communication for high speed, high quality, and multimedia transmission is driving the use of the higher radio frequency spectrum. A local multipoint distribution service (LMDS) [] is a line of sight (LoS) pointto-multipoint wireless access system operating at the Ka-band millimeter wave frequency. They are designed to deliver broadcast services (multimedia, video, internet, etc) from a central transmitter to individual subscriber within its cell size. The frequency bands allocated by ITU R and CEPT are usually above 0 GHz. In this band, rain attenuation is the most influential propagation factor to determine the system availability [], [3]. At high rainfall intensities, the occurrence of which is of special interest in system requiring highreliability links, the horizontal structure of rain is highly variable []. It is frequently observed that during a shower, high intensity rain is localized in a very small area surrounded by a region of more uniform, low intensity rain. Hence, in a cellular system under rain with very localized storm, there is potential that a subscriber terminal receiving a heavily attenuated signal from one hub can obtain less attenuated, acceptable reception from another hub. This makes cell-site diversity appear as a very promising method for improving link reliability and area coverage. The cell-site diversity technique has been proposed in order to reduce the outage time due to the above reason [4], [5]. In many tropical countries, as Indonesia, its rainfall intensities are very high that cause significant rain attenuation in Ka Band communication channel [6]. Analysis of diversity gain using various combining techniques, i.e. selection combining (SC), equal gain combining (EGC), and maximalratio combining (MRC), in dual link millimeter wave communication systems has been proposed by Wijayanti et al. [7]. There is several rain fading compensation techniques, i.e. power control, adaptive coded modulation (ACM), diversity and relay. In this research, adaptive code modulation and SC diversity have been applied. This paper presents the results of simulation study about link availability evaluation, channel capacity and diversity gain in LMDS system at 30 GHz frequency using ACM and SC diversity. The evaluation is done based on rainfall intensities measurement in Surabaya Indonesia and adopts ITU-R P.530 recommendation for calculating rain attenuation [8]. This simulation uses identical dual-link length with line of sight condition. Lengths of links are 4 km with separation angle of 45 o (east with north-east), 90 o (north), 35 o (east with northwest), 80 o (west). The paper begins with a brief description of research background in introduction section. Section II describes ACM system model and SC diversity. Section III explains measurement set-up and rain rate and attenuation measurement results. An evaluation of link availability and channel capacity is reported in section IV. Finally, a summary is given in the last section. II. SYSTEM MODEL LMDS system model in this paper uses adaptive coded modulation to increase maximally channel capacity and link availability. The system uses M-QAM with various modulation levels, i.e. 4, 6, 64 QAM. Its coding uses code rate variation of convolution code CC(k/n) and Reed Solomon code RS(n,k). The system has to be maintained so that performance of the system has high signal reception quality with maximum bit error rate BER 0-6 and 0 -. Adaptive 50-

2 input Transmitter Tx ACM system / A n[k ] Channel Receiver Demodulation and decoding output Transmitter Tx ACM system / A n[k ] Channel Selection Combining Channel estimation delay Fig. Adaptive coded modulation system and selection combining diversity TABLE I M-QAM ADAPTIVE SCEARIO FOR MAXIMAL BER 0-6 Modulation Mode γ Interval (db) o transmission γ < QAM 3.54< γ <0.4 6 QAM 0.4< γ < QAM γ >6.56 coded modulation system and selection combining diversity can be shown in Fig..In this paper, the techniques are compared with M-QAM adaptive system. Probability of error of the M-QAM system in clear sky condition can be obtained as follow: P M 3T 0 B S = erf () b log M M ( M ) S Where M and are modulation level and signal to-noise ratio. The T 0 B is assumed by one. Rain attenuation A k (db) fades signal quality in the channel. SC diversity in the receiver receives two signals from transmitter or transmitter to be selected best signal. Bit error rate calculation of the system uses adaptive code modulation with CC (k/n) and RS (n,k) code in clear sky condition can be obtained as follow: [9] m m m j j P = j Pe Pe () log M B m j t cc cc j + where m, M and Pe cc are any bit in a symbol, modulation level and symbol error probability after it uses convolution code [9], [0], respectively. By using equation () and (), it can be obtained thresholds of signal to noise ratio γ reception for determining modulation level. Adaptation scenario based on γ for maximum bit rate 0-6 for M-QAM adaptive and ACM can be shown on Table I and Table II. By using same metod, it can be obtained thresholds of γ for maximum BER 0 -. TABLE II ACM SCEARIO FOR MAXIMAL BER 0-6 Modulation Mode γ Interval (db) o transmission γ <.8 4 QAM+RS(63,3)+CC(/3).8< γ <.6 6 QAM+RS(63,5)+CC(/).6< γ < QAM+RS(63,59)+CC(/3) γ >3.54 Channel capacity is an important factor for evaluating the performance of the LMDS system in Indonesia s rain environment. For our purpose, the channel-capacity, as defined by Shannon [0], is a good measure of the merit of any communication system, since it gives the maximum rate of transmission signal over the channel. If the channel is subject to fading, its capacity varies with the changes in the propagation medium, meaning that the channel capacity inherits the one-way stochastic properties of the fading process. Channel capacity of M-QAM system can be calculated as follow: C B = i= log (3) ( M i ). P( M i ) where M i, P(M i ), are i-modulation level, i-modulation level probability and data number, respectively. In this paper, ACM system using convolutiol code rate CC(k/n) and Reed-Solomon code rate RS(n,k). Channel capacity of the system can be calculated as follow: C B where kcc k i RSi = log( M i ). P( M i ).. (4) n n k n i= RSi RSi krsi, are number of information bit to code word n ratio of convolutional code and Reed Solomon code on the i- modulation level. 50-

3 0 Disdrometer 0 0 * C Prob[Rain-rate>abscissa](%) * B Rain-rate (mm/hr) Fig. 3 CCDF s of rain rate measurement result per year in Surabaya, Indonesia * A Fig. Measurement map in ITS campus, Indonesia III. MEASUREMET SET-UP AD RESULTS A. Rain Rate For rain measurement, two kinds of instruments were used to record, i.e. rain gauges and disdrometer. They are placed at the campus of Institut Technologi Sepuluh opember (ITS) Surabaya, Indonesia since 007 as shown at fig.. Rain gauge A is placed at medical centre, disdrometer B is placed at department of electrical engineering and rain gauge C is placed at ITS library. Rain rate has been measured until now. They were used to record rain rate data and then stored in a personal computer (PC) through RS3 and data sampling at every 0 s. In this research, it uses rain rate measurement data for years. The R 0.0 means that the rain rate at 0.0% of time in measured period that the rain-rate value is exceeded. Fig. 3 shows that Surabaya Indonesia s R 0.0 is 40 mm/h, larger than that of 00 mm/h suggested by ITU-R. B. Rain Attenuation Rain attenuation was estimated by using synthetic storm technique (SST). The SST was successfully tested for predicting conventional long term, long term statistics of fade duration, and long term relative to contiguous periods of the day of four hours []. The results of complementary cumulative distribution function of rain attenuation estimation in Surabaya Indonesia can be seen in Fig.4. Prob [Rain attenation > abscissa](%) East-link orth west-link orth east-link orth-link Attenuation(dB) Fig. 4 CCDF s of rain attenuation for single link at various link directions, in which the link length is 4 km The rain attenuations in Surabaya Indonesia at 4 km link length for outage probability 0.0% are 86, 93, 97 and 0 db with links direction are east (0 o ), north west (35 o ), north east (45 O ) and north (90 o ), respectively. The attenuation causes significantly signal impairment which it decreases performance of the system. Rain attenuation estimation results show that orth direction link has the largest attenuation. This is be caused by many wind directions from west and east. In this study, LMDS system operating at 30 GHz is evaluated its performance using parameters from ew Bridge etworks Corporation []. 50-3

4 TABLE III SIMULATIO RESULS OF LIK AVAILABILITY (%) FOR IDETICAL DUAL-LIK LEGTHS WITH SEPARATIO AGLE 90 0 I THE PRESECE OF RAI FADIG Link length Transmission Mode km km 3 km 4 km M-QAM Adaptive QAM QAM QAM ACM + SC Diversity 99,99 99,99 99,99 99,99 99,97 99,97 99,95 99,95 4 QAM+RS(63,3)+CC(/3) 99,99 99,99 99,99 99,99 99,97 99,97 99,95 99,95 6 QAM+RS(63,5)+CC(/) 99,99 99,99 99,98 99,98 99,96 99,95 99,93 99,93 64 QAM+RS(63,59)+CC(/3) 99,99 99,99 99,96 99,95 99,93 99,9 99,90 99,88 TABLE IV SIMULATIO RESULS OF CHAEL CAPACITY (bps/hz) FOR IDETICAL DUAL-LIK LEGTHS WITH SEPARATIO AGLE 90 0 Link length Transmission Mode km km 3 km 4 km M-QAM Adaptive QAM QAM QAM ACM + SC Diversity QAM+RS(63,3)+CC(/3) QAM+RS(63,5)+CC(/) QAM+RS(63,59)+CC(/3) IV. SIMULATIO RESULTS AD DISCUSSIO A. Link Availability in rain fading One of the LMDS system performances is link availability. The results of simulation are shown in Table III. Adaptive transmission method of LMDS system uses adaptive code modulation and selection combining in the presence of rain attenuation. System configuration is evaluated in identical duallink lengths with separation angle 90 0 (north and east). The system is guaranteed with maximum BER 0-6 and it is also evaluated at maximum BER 0 -. Base on simulation results, M-QAM adaptive technique in the LMDS system has 99.99% link availability for km link length at maximum BER condition 0-6 and 0 -. It also has 99.83% link availability for 4 km link length for maximum BER 0 -. For the fixed transmission method, i.e. the 64 QAM modulation has 99.50% link availability. Furthermore, LMDS system that it uses adaptive coded modulation and selection combining diversity is evaluated its link availability. At the condition, LMDS system has 99.99% link availability for km link length. It also has 99.95% link availability for 4 km link length for maximum BER 0 -. Additionally, LMDS system can be increased its link availability with using ACM and SC diversity. Simulation results show that ACM and SC diversity can enhance 0.45% link availability for maximum BER 0 - at 4 km link. B. Channel Capacity in rain fading Channel capacity is another important factor of the LMDS system performance. Table IV shows the simulation results of the system channel capacity that it uses M-QAM adaptive or ACM and SC diversity. By using M-QAM adaptive, the system produces channel capacity about 5.98 bps/hz for maximum BER 0 -. For the 4 QAM modulation, it has channel capacity about.860 bps/hz. Furthermore, LMDS system using adaptive transmission method (ACM and SC) is evaluated. The system produces channel capacity about bps/hz. If the system uses non adaptive transmission, for example 4 QAM, it produces channel capacity about.86 bps/hz and 0.39 bps/hz for 4- QAM+RS(63,3)+CC(/3). Channel capacity of 6 QAM is almost same with channel capacity of ACM and SC diversity, i.e. about 3.5 bps/hz, but LMDS using ACM and SC has highest link availability. Adaptive transmission using ACM and SC diversity technique can enhance relatively its performance. The performances are link availability and channel capacity. The technique is better than using fixed transmission mode, for example: 4 QAM or 4 QAM+RS(63,3)+CC(/3). 50-4

5 VI. COCLUSIO Based on the results of simulation study, it can be concluded that LMDS system using 64-QAM transmission mode without coding has 99.50% link availability. By using ACM and SC diversity can enhance 0.45% link availability for maximum BER 0 -. They give performance of the LMDS which is guaranteed to have maximum BER 0 -, 99.99% link availability, bps/hz maximum channel capacity at km link length. Additionally, the system produces 99.95% link availability, bps/hz channel capacity at 4 km link length. Finally, performance of the LMDS system can be enhanced with using adaptive transmission technique, i.e. adaptive coded modulation (ACM) and selection combining (SC). ACKOWLEDGEMET The reported work is supported in part by JICA through the PREDICT-ITS Project. REFERECES [] C. Smith, LMDS, ew York: McGraw-Hill, 000. [] R.K. Crane, Electromagnetic Wave Propagation Through Rain: ew York, Wiley, 996. [3] A. Paraboni, G. Masini, and A. Elia, The effect of precipitation on microwave LMDS networks-performance analysis using a physical rain cell model, IEEE J. Selected Areas Comm,. Vol. 0, pp , April 00. [4] G. Hendrantoro, R.J.C. Bultitude, and D.D. Falconer, Use of cell-site diversity in millimeter-wave fixed cellular systems to combat the effects of rain attenuation, IEEE J. Select. Areas Commun., vol. 0, pp , Apr. 00. [5] A.D. Panagopoulos and J.D. Kanellopulos, Cell-site diversity performance of millimeter-wave fixed cellular systems operating at frequencies above 0 GHz, IEEE Antennas and Wireless prop. Letter, vol., 00. [6] M. Salehudin, B. Hanantasena, L.J.M. Wijdemans, Ka-band line of sight radio propagation experiment in Surabaya Indonesia, Fifth Ka-band Utilization Conf., Oct [7] A. Wijayanti, H. Mahmudah, G. Hendrantoro, Cell-site diversity gain using various combining technique in dual-link millimeterwave communication system under impact of rain attenuation, Fifth International Conference on Wireless and Optical Communication etworks, May, 008. [8] ITU-R Rec. P.530-0, Propagation data and prediction methods required for the design of terrestrial line of sight systems, 00. [9] B. Sklar. Digital Communication, Prentice Hall, ew Jersey, 994. [0] Wicker, S.B, Error Control Systems for Digital Communication and Storage, Prentice Hall, 998. [] E. Matriani, Service oriented statistics of interruption time due to rainfall in earth-space communication systems, IEEE Transaction on antennas and propagation, vol. 5, o. 8 August, 004. [] Chu, C. Y., Chen, K. S. Effects of Rain Fading on the Efficiency of the Ka-Band LMDS System in the Taiwan Area, IEEE Transactions on Vehicular Technology, vol. 54, no., Januari