A Physical Layer Simulation for WiMAX MIMO-OFDM System

A Physical Layer Simulation for WiMAX MIMO-OFDM System Throughput Comparison Between 2x2 STBC and 2x2 V-BLAST in Rayleigh Fading Channel Hadj Zerrouki* Mohammed Feham STTC Laboratory Department of Electronics and Electrical Engineering Faculty of Technology Tlemcen University Algeria. zerrouki. hadj@gmail. com* feham _ m@yahoo. fr Abstract- WiMAX is a broadband wireless technology based on IEEE802.16 standards family which defines the physical (PHY) and medium access control (MAC) layers and makes several possible configurations available along with non-mandatory options. WiMAX is a new OFDM based technology and promises to combine high data rate services with wide area coverage. In this paper the performance of Wi MAX PHY layer is investigated for two MIMO (Multiple-Input Multiple-Output) PHY layer modifications (Space-Time Block Codes STBC) and (Spatial multiplexing SM) to provide high suppression against multipath fading provide high bandwidth efficiency and high throughput with high data rates. This work incorporates the model building of the WiMAX Physical layer using MA TLAB simulation. The results obtained for these modifications show that these mechanisms enhance the performance of the WiMAX PHY layer in fixed environments with high spectral efficiency. Keywords-component; WiMAX; IEEE 802.16; OFDM; MIMO; STBC; V-BLAST; SNR; Physical Layer (PHy); I. INTRODUCTION Worldwide Interoperability for Microwave Access (W imax) is introduced by the Institute of Electrical and Electronic Engineers (IEEE) which is standard designated 802. 16 d-2004 [1 ] (used in fixed wireless applications) and 802. 16 e-2005 [2] (mobile wireless) to provide a worldwide interoperability for microwave access. The IEEE 802. 16 d-2004 air interface standard is basically based on technology namely Orthogonal Frequency Division Multiplexing (OFDM) that has been regarded as an efficient way to combat the Inter Symbol Interference (lsi) for its performance over frequency selective channels for the broadband wireless networks. In an OFDM system the data is divided into multiple parallel sub-streams at a reduced data rate and each is modulated and transmitted on a separate orthogonal subcarrier. This increases symbol duration and improves system robustness. OFDM is achieved by providing multiplexing on users' data streams on both Uplink and Downlink transmissions. A valuable solution consists of the introduction of advanced digital signal processing techniques based on Multiple Input Multiple Output (MIMO) concept. The key feature of MIMO is the capability to increase channel capacity without increasing transmitted power and RF bandwidth [3 ]. Nowadays MIMO techniques present some well-promising applications in wireless standards like IEEE 802. 11n and IEEE 802. 16 x (W imax). Different space-time processing techniques have been proposed in literature in order to fully exploit potentialities of MIMO systems. The most popular one is Space-Time Coding [4] in which the time dimension is complemented with the spatial dimension inherent to the use of multiple spatially-distributed antennas Commonly used Space Time coding schemes are Space Time-trellis codes and Space-Time Block Codes (STBC). A well-known example of conceptually simple computationally efficient and mathematically elegant STBC scheme has been proposed by Alamouti in [5]. Substantially Alamouti's coding is an orthogonal space-time block code where two successive symbols are encoded in an orthogonal 2x2 matrix. The columns of the matrix are transmitted in successive symbol periods but the upper and the lower symbols in a given column are sent simultaneously through the first and the second transmit antenna respectively. The alternative solution to space-time coding is represented by Spatial Multiplexing (SM) [6 ]. Spatial multiplexing is a space-time modulation technique whose core idea is to send independent data stream from each transmit antenna. This is motivated by the spatially white property of the distribution which achieves capacity in MIMO i. i. d. Rayleigh matrix channels [7 ]. SM is addressed to push up link capacity rather than to exploit spatial diversity. Although various implementtation architectures for MIMO systems have been introduced since the BLAST (Bell Laboratories Layered Space-Time) system was proposed in [6 ] and [8] a variation of such system V-BLAST still emerges as a promising architecture due to lower receiver complexity (V-BLAST receiver algorithm) and higher data rates in the case of large number of antennas. 978-1-4799-3824-7/14/$31.00 2014 IEEE

The paper is organized as follows: Model of WiMAX PRY layer is explained in section 2. Explanation of the results obtained via simulation is done in Section 3. At the end conclusion is given in section 4. II. MODEL OF WIMAX PRY LAYER Physical layer set up the connection between the communicating devices and is responsible for transmitting the bit sequence. It also defines the ty pe of modulation and demodulation as well as transmission power. WiMAX physical layer is based on the orthogonal frequency division multiplexing (OFDM). OFDM is a good choice of high speed data transmission multimedia communication and digital video services. [t even can maintain very fast data rate in a non line of sight condition and multipath environment. [n the following subsection we provide a detailed description of the OFDM. The role of the PRY -layer is to encode the binary digits that represent MAC frames into signals and to transmit and receive these signals across the communication media. The proposed block diagram of WiMAX-MIMO-OFDM PRY system is given in Fig. I. : m - ' < m ' ' --------- --------- -' MIMO En coder STB C / SM... ' : z ' "" s: : :3: ga : The random binary signal information is first generated and grouped in symbols then coded for error correction. Digital modulation system is used because different modulation scheme is needed for different data rates. Then multiple antennas are used the M[MO space-time diversity encoder is implemented. After that guard band is inserted and the Inverse Fast Fourier Transform (IFFT) block transforms the data sequence into time domain. Then a cyclic prefix is used which is chosen larger than the expected delay spread to avoid intersymbol and inter-carrier interferences (lsi and ICI). The channel is considered to be a multipath fading channel followed by addition of white Gaussian noise. At the receiver the FFT is used to transform the data back to frequency domain. Adaptive filtering technique is used for channel estimation. The FFT is taken in each of receives antenna. Each antenna receives a different noisy superimposetion of the faded versions of the transmitted signals. Lastly the binary information data is obtained back after the demodulation and channel decoding. In following sections we will take each block one by one in detail. A. Channel Encoding The radio linle is a quickly vary ing linle often suffering from great interference. Channel coding whose main tasks are to prevent and to correct the transmission errors of wireless systems must have a very good performance in order to maintain high data rates. The 802. 16 channel coding chain is composed of three steps: Randomizer Forward Error Correction (FEC) and [nterleaving. They are applied in this order at transmission. 1) Randomization: The Randomizer performs randomization of input data on each burst on each allocation to avoid long sequence of continuous ones and zeros. This is implemented with a Pseudo Random Binary Sequence (PRBS) generator which uses a 15 stage shift register with a generator polynomial of 1 + X l4 + X is with XOR gates in feedback configuration as shown in Fig. 2. Figure 1. '.... _... -- - --------...... Block diagram of WiMAX-MIMO-OFDM PHY system D Data O u t D _"t_"_l n ---I ------- Figure 2. Randomization generator. Forward Error Correction (FEC) codes: The bits issued from the randomizer are then applied to the FEC encoder. FEC techniques ty pically use error-correcting codes that can detect with high probability the error location. These channel codes improve the bit error rate performance by adding redundant bits in the transmitted bit stream that are employed by the receiver to correct errors introduced by the channel.

The FEC is achieved using Convolutional Codes (correct independent bit errors) and Reed Solomon codes (correct burst errors at byte level) to provide the additional coding gain which measures the amount of additional SNR that would be required to provide the same BER performance for an uncoded message signal in the same channel conditions. The Reed Solomon (RS) codes are mandatory codes on both sides i. e. Uplink and Downlink. These are non-binary cyclic codes that add redundancy to the data that improves the probability of block errors. The outer RS encoded block is fed to inner binary convolutional encoder (see Fig. 3). The implemented encoder has native rate of 112 a constraint length of 7 (m = 7). The generator polynomials used to derive its two output code bits denoted X and Y are specified in the following expressions: G 1 = 1330cT for X and G2 = 17cT for Y. X=1330CT Figure 3. Convolutional encoder (rate = 112 m = 7). Coding rate is defined as the ratio of the input bits to the output bits. Higher rates like 2/3 and 3/4 are derived from it by employing "puncturing. " Puncturing is a procedure that involves omitting of some of the encoded bits in the transmitter thus reducing the number of transmitted bits and hence increasing the coding rate of the convolutional code and inserting a dumm y "zero" metric into the convolution Viterbi decoder on the receive side of WiMAX Physical layer in place of the omitted bits. For decoding the Viterbi algorithm is used at the receiver side of the PHY layer. To describe a convolution code one need to characterize the encoding function (m) so that given an input sequence m one can readily compute the output sequence U. 2) Interleaving: Interleaving is used to protect the transmission against long sequences of consecutive errors which are very difficult to correct. These long sequences of error may affect a lot of bits in a row and can then cause many transmitted burst losses. Interleaving by including some diversity can facilitate error correction. The encoded data bits are interleaved by a block inter-leaver with a block size corresponding to the number of coded bits per allocated sub-channels per OFDM symbol [1 ]. The interleaver is made of two steps: Distribute the coded bits over subcarriers. A first permutation ensures that adjacent coded bits are mapped on to nonadjacent subcarriers. The second permutation insures that adjacent coded bits are mapped alternatively on to less or more significant bits of the constellation thus avoiding long runs of bits of low reliability. B. Digital Modulation After channel coding data bits are mapped and modulated onto the allocated subcarriers. We passed the random values through the adaptive modulation schemes according to the constellation mapped. The data was modulated depending their size and on the basis of different modulation schemes like BPSK Gray-mapped BPSK QPSK 16 -QAM and 64-QAM modulation. Inverse process called demodulation is done by the receiver to recover the transmitted digital information. C. MIMO Encoder The proposed system consists of 2 transmit and 2 receive antennae. Our MIMO codes use two ty pes of encoders STBC and SM. The MIMO STBC encoder system is implemented using Alamouti 2x2 STBC. In case of spatial multiplexing (SM) technique MIMO encoder includes V-BLAST technology that is used to improve the spectral efficiency of the system with two transmits and two receive antennas. 1) Space-Time Block Codes: Space-Time Block Codes (STBCs) are the simplest ty pes of spatial temporal codes that exploit the diversity offered in systems with several transmit antennas. The transmit diversity technique proposed by Alamouti was the first STBC [9]. The encoding and decoding operation is carried out in sets of two modulated symbols. Therefore let us denote by S] and S2 the two modulated symbols that enter the space-time encoder. In the Alamouti scheme during the first time instance t] the symbols S] and S2 are transmitted by the first and the second antenna element respectively. During the second time instance t2 the negative of the conjugate of the second symbol i. e. -S2 ' is sent to the first antenna while the conjugate of the first constellation point. i. e. S/ is transmitted from the second antenna. The transmission rate is equal to the transmission rate of a SISO system. The space-time encoding mapping of Alamouti 2x2 can be represented by the coding matrix: -S;] (1) S' 1 The received signals at the time t and t + T can then be expressed as: r1 = r1(t) = hl1s1 + hz1sz + n1 rz = r1(t + T) = -hl1s; + hz1s; + nz r3 = rz(t) = h12s1 + hzzsz + n3 r4 = rz(t + T) = -h12s; + hzzs; + n4 where r] r3 are the received signals at time t and r] r4 are the received signals at time t + T n] n2 n3 and n4 are complex random variables representing receiver noise and interference. This can be written in matrix form as: (2) r = HS +n (3)

where H is the complex channel vector and n is the noise vector at the receiver. Figure 4. MIMO channel model (2x2). The estimate of the transmitted symbol using Zero Forcing (ZF) decoder is: 2) Spatial Multiplexing: In spatial multiplexing a signal is divided into different streams and each stream is transmitted from a different transmit antenna in the same frequency channel. If these signals arrive at the receiver antenna array with sufficiently different spatial signatures the receiver can separate these streams creating parallel channels for free. Spatial multiplexing is very powerful technique for increasing channel capacity at higher Signal to Noise Ratio (SNR). It can be used with or without transmit channel knowledge. This technique includes V-BLAST technology. Consider that we have a transmission sequence for example S] S200. Sn. For 2 transmit antennas we group the symbols into groups of two. In the first time slot send S] and S2 from the first and second antenna. In second time slot send S3 and S4 from the first and second antenna. Notice that as we are grouping two symbols and sending them in one time slot we need only n/2 time slots to complete the transmission so data is doubled. The V -BLAST transmission for 2 x2 MIMO system can be represented in matrix notation as follows: Where r] r2 are the received symbol on the first and second antenna respectively hii is the channel from /h transmit antenna to /h receive antenna S] S2 are the transmitted symbols that use first and second constellation mapped respectively and n] n2 is the noise on \ st 2n d receive antennas. The decoding is done using ZF technique which generates an estimate of the transmitted matrix as: D. OFDM System OFDM technique is a bandwidth efficient multicarrier technique which splits the system bandwidth into orthogonal (4) (5) (6) sub channels each of which occupies only a narrow bandwidth and a separate sub carrier is assigned to each. By means of guard interval and cyclic prefix an OFDM system also achieves good resistance against multipath fading. 1) Inverse Fast Fourier Transform (IFFT): This block implements the OFDM modulation and is preceded by a serial to parallel converter and at its output another block converts the data back to its serial format. The two converters are practically incorporated by the IFFT transform as it is implemented in Matlab. Hence An IFFT converts the input data stream from frequency domain to time domain representing OFDM Subcarrier as the channel is basically in time domain. IFFT is useful for OFDM system as it generates samples of a waveform with frequency components satisfy ing the orthogonality condition such that no interference occurs in the subcarriers. The mathematical model of OFDM symbol defined by IFFT which would be transmitted during our simulation as given bellow: N-l 1 "'\' j2rrnk xci n) = IFFTN[X(i k)] = N L XCi k) e- N- (7) k=o where X(ik) is the transmitted data symbol at the kth subcarrier of the ith OFDM symbol N is the FFT size. Similarly FFT converts the time domain to frequency domain as basically we have to work in frequency domain [\ 0]. By calculating the outputs simultaneously and taking advantage of the cyclic properties of the multipliers FFT techniques reduce the number of computations to the order of N log(n). The FFT is most efficient when N is a power of two. 2) Cyclic Prefix Insertion: After performing Inverse Fast Fourier Transform (IFFT) the cyclic prefix (CP) will be add with each OFDM symbol. The CP consists in a copy of the last samples composing the OFDM symbol added in front of it (see Fig. 5). This function is built according to IEEE 802.\6 specifications which define 4 possible values for the ratio between the duration of the cyclic prefix and the duration of the useful OFDM symbol. This ratio can be equal to 1/4 1/8 \/16 and 1/32. E. Communication Channel Figure 5. Cyclic Prefix insertion. Communication channels are kind of medium of communication between transmitter and receiver. The channel adds a white noise n of a certain variance to a flat faded variant of the useful signal S: r = Ray.S + n (8)

where Ray represents a Rayleigh random variable and n is the Additive White Gaussian Noise (A WGN). We selected the Rayleigh model for the channel to simulate a Non Line Of Sight (NLOS) communication. III. SIMULATION RESULTS The performance of MIMO-OFDM PHY layer of WiMAX system for different MIMO configurations (STBC V-BLAST) is analyzed for different values of FFT size Bandwidth cyclic prefix factor and modulation techniques. A comparative performance analysis with FEC encoder based WiMAX MIMO-OFDM PHY layer has also been approved. The simulated parameters used in the present study are shown in Table I. The WiMAX simulator presented in this paper allows a better understanding of the processes involved at the PHY layer level. Furthermore quantitative results may be provided by computing of WiMAX Downlink throughput in fading Rayleigh channel. TABLE I. Parameter Carrier Frequency SIMULA non PARAMETERS Value 3.5 GHz improves WiMAX throughput by 6.5 Mbps compared to SISO system. In contrast 2x2 V-BLAST mode is poorer than the SISO scheme. However for a SNR up to 13 db 2x2 V-BLAST system performs better then SISO and 2x2 STBC systems it can be observed that V-BLAST system has a throughput factor of 2 at SNR of db compared to the others systems. We also note that when SNR great than 20 db both SISO and STBC systems remain almost same throughput. So STBC system produces the best performance at low and medium values of SNR (from 1 to 13 db) due to their robustness in poor channel conditions. On the other hand at high SNR (up to 13 db) the increased error-free data rate makes V-BLAST the best choice. 18-------- SISO 16-2x2 MIMO SlBC - 2x2 MIMO-V-BLAST 14 12 "' "- ;[ - Channel Model MIMO Modulation Non-LOS 2x2 STBC 2x2 SM BPSK QPSK 16QAM 64QAM 4 OFDM subcarriers (IFFT/FFT) 12825651224 Channel Bandwidth Frame Duration 1.25 2.5 5 MHz 5 ms Number of Frames (per sec) 200 Guard Interval 1/41/81/161/32 EncoderlDecoder Fading Channel Noise A. MIMO Techniques CCNiterbi Rayleigh Fading Channel AWGN In this section we demonstrate the enhanced performance of the proposed detection scheme through simulation results. We consider a MIMO-OFDM system with two modes 2x2 STBC (Alamouti) and 2x2 SM (V-BLAST) compared to a traditional communications SISO. Moreover the number of subcarrier equals 512 convolutional encoder with constraint length of 7 and code rate 112 has been used with 16 QAM modulation and CP factor of 11 4. We assume that the channel is a frequency flat fading during two OFDM symbol periods. Moreover we suppose that Channel State Information (CSI) is known to receiver perfectly. Fig. 6 shows the comparison of WiMAX Downlink throughputs of various modes. It is seen the interest of spatial diversity. Throughput gains are highly significant for 2x2 V-BLAST. Thus at SNR of 7 db 2x2 MIMO STBC system 2- O('l ("1 ----{'>_... 1_; } -4 -;0"- = = -- -- -- -- :_... -5 0 5 15 20 25 30 Figure 6. Performance comparison for 2x2 STBC 2x2 V-BLAST and SISO systems. B. Effect a/fee Encoder In this section of our research work we represent various throughput vs. SNR plots to evaluate the effect of FEC encoding technique. The tow WiMAX MIMO systems (STBC and V-BLAST) performances are evaluated by convolutional encoding with three code rate cases (1/2 3/4 and 1= without FEC) the FFT size is fixed to 512 with CP factor of 1/4 and 16 QAM modulation is employed. On comparison From Fig 7 it is observable that the two systems throughput shows comparatively much better performance without FEC encoding at high SNR compared to coded data. On the other hand the systems performance with FEC is satisfactory at low SNR although the use of FEC causes redundant bits in the transmitted bit stream. When FEC is applied error reduces to a considerable label. For a typical SNR values respectively of db and 15 db the STBC and V-BLAST systems performances are improved by 4. 65 Mbps and 4. 76 Mbps for the case of 112 rated FEC encoded and 6.27 Mbps and 13 Mbps for FEC = 3/4. We can also see the benefit of using a Viterbi decoding to correct errors introduced by the channel at low SNR. The

counterpart is a loss throughput (a factor of code rate) and an additional time due to the interleaver. 30 - '2x2 STBC 1/2 2x2 STBC 3/4 2x2 STBC 1 2x2 V-BLAST 1/2 25 - - - '2x2 V-BLAST 3/4 5-2x2 V-BLAST 1 Ot c.' ':;' I '::J ( -5 0 5 Figure 7. 'c C. Effect a/constellation Size 15 20 L-- - r 25 30 Effect of FEC rate on throughput for 2x2 STBC and 2x2 V-BLAST systems. To verify the STBC and V-BLAST systems throughput performances an OFDM system with 512 subcarriers is simulated with 3/4 FEC convolutional code rate and CP factor of 1/4. Four different digital modulation schemes namely BPSK QPSK 16 QAM and 64QAM are used in our simulation system based on the observed channel conditions. In modulation bits are transmitted in symbols form not as they are. The number of bits included in each symbol denotes the constellation size. Therefore more this size will be large; throughput will be high and vice versa. Fig. 8 shows the effect of constellation size on the throughput as function of the SNR for 2x2 STBC and 2x2 V-BLAST systems. It is evident that the throughput increase when constellation size increase. Hence the higher throughput is obtained by 64QAM by transmit more bits per symbol and the lower throughput when using BPSK. Since increased constellation size implies shorter distance between neighboring symbols the received data is more susceptible to errors at higher rates when the channel is weak. In this way there is a balance between obtaining the higher throughput and maintaining an acceptable bit error rate for any radio communications system. As the SNR decrease so it required switching from higher modulation level to lower modulation level. The transmitter will choose the appropriate modulation scheme depend upon the SNR threshold value. D. Effect 0/ FFT Size This parameter specifies the FFT size used in our simulation. Four FFT sizes are supported here: 128 256 512 and 24. The FFT size determines the number of available subcarriers and OFOM symbol duration. In general for a given bandwidth a larger FFT size results in a greater number of available subcarriers and a longer OFOM symbol duration. Fig. 9 shows the throughput comparison of the two modes STBC and V-BLAST for 128 256 512 and 24 FFT size with 1/2 FEC CC rate 16 QAM modulation and 11 4 CP factor. It is clear from below figure that 2x2 STBC system throughput remains constant for any FFT size. ST6C128 -- - ST6C 256 - - ST6C512. - ST6C 24 V-BLAST 128 - V-BLAST 256 40 'STBC BPSK _ ' STBC QPSK STBC 16QAM - J _ 'STBC 64QAM 30 - V-BLAST BPSK 25.c a. 20.c OJ :J E'.c 15- f- 5 V-BLAST QPSK V-BLAST 16QAM V-BLAST 64QAM Figure 8. Effect of constellation size on throughput for 2x2 STBC and 2x2 V-BLAST systems. Figure 9. 15 20 25 30 Effect of FFT size on throughput for 2x2 STBC and 2x2 V-BLAST systems. The 2x2 V-BLAST system throughput is gradually enhanced by increasing the FFT size; this rate remains insufficient given the high complexity of the system due to this increase. An exception for the passage of a FFT size from 256 to 512 subcarriers the gain in throughput is significant. Thus for an SNR of 15dB we gain about 5. 7 Mbps. E. Effect a/cyclic Prefix Factor This parameter specifies the ratio of useful symbol time to cyclic prefix time. Four ratios are supported here: 1/4 1/8 1/16 and 1/32. We have compared the effect on throughput of

2x2 STBC and 2x2 V-BLAST systems. Let's check effect on the throughput. From Fig. it can be seen that throughput increases appreciably by decries the values of cyclic prefix factor. At the lower SNR fading is more and signal strength is going low as the distance increases. To defeat this problem higher value of CP need to be chosen. Large value of the CP means large time gap between two frames which mean extra time to receive signal from mu1tipath signals. Although the large values of the CP reduces throughput however it increases coverage up to large distance. Thus chosen the appropriate value of the CP gives the desired distance that need to cover by the signal [11]. Hence the throughput is reduced by a factor of 4/ 5 8/ 9 16/17 and 32/33 depending on the cyclic prefix configuration (1/4 1/8 1116 and 1/32) to extract the real useful bits. It was found that more the cyclic prefix duration is higher; more the resistance to OFDM Inter-Carriers Interference is effective. However throughput is then lower. and CP factor of 1/4. We can see that the throughput of the two systems is affected by change in bandwidth. It achieves maximum values for 24 FFT size and MHz bandwidth of both systems at SNR of25 db; these values are 13.3 Mbps and 34. 5 Mbps respectively for 2x2 STBC and 2x2 V-BLAST. Hence there is a fixed relationship between the occupied bandwidth and the OFDM symbol sample rate. The implementation of a bandwidth-scalable air interface makes the subcarrier separation and symbol duration remain invariant as the deployment bandwidth changes. 1.----- ---- ---- ---- ---- = 8TBC 128/1.25MHz 8TBC 256/2.5MHz 30 8TBC 512/5MHz 25 20 " a..c '" 15 f- 8TBC 24/MHz V BLAST 128/1.25MHz -c-- V BLAST 256/2.5MHz V BLAST 512/5MHz -- V BLAST24/MHz - </ - -;>_-_-=D=- =====---- 30 25 - "' Q. 20 " g 15 - e.c f- 5- STBC 1/ 16 STBC 1/32 V BLAST 1/4 --"- V BLAST 1/8 - V BLAST 1/ 16 V BLAST 1/32 o 0 o 25 30 Figure. Eflect of CP factor on throughput for 2x2 STBC and 2x2 V-BLAST systems. F. Effect of System Bandwidth WiMAX has a scalable physical-layer architecture that allows for the throughput to scale easily with available channel bandwidth. This scalability is supported in the OFDMA (Orthogonal Frequency Division Multiple Access) mode where the FFT size may be scaled based on the available channel bandwidth. In our simulation the WiMAX system may use 128 256 512 or 24 FFTs corresponding to the transmission channel bandwidths of l. 25MHz 2. 5MHz 5MHz or lomhz respectively. This scaling may be done dynamically to support user roaming across different networks that may have different bandwidth allocations [12]. Fig. 11 shows the effect of variation of bandwidth on throughput of 2x2 STBC and 2x2 V-BLAST systems 3/4 rate binary convolutional code is used with 16 QAM modulation O. :::l r: - I -r:' 5 0 Figure 11. Effect of system bandwidth on throughput for 2x2 STBC and 2x2 V-BLAST systems. The ability to scale system bandwidth while maintaining constant symbol duration provides greater commonality in equipment components and offers the operator the advantage of being able to deploy today and grow their future system bandwidth at lower cost and reduced network impact. IV. 15 CONCLUSION In this paper we proposed a downlink physical layer simulator for WiMAX-MIMO-OFDM in flat Rayleigh fading channel. The WiMAX simulator allows a better understanding of the signal processing steps taking place at the PHY layer corresponding to the IEEE 802. 16 specifications. A comparison for the Throughput performance at the downlink of 2x2 STBC and 2x2 V-BLAST systems are done using simulation. It is shown that the V-BLAST system has a better performance than STBC and SISO at high SNR range. On the other hand at low to medium values of SNR STBC produces the best performance due to its robustness in poor channel conditions and increase the diversity. The simulation results also indicate the effects of each bloc of MIMO-OFDM physical layer in WiMAX system namely Forward Error Correction (FEC) rate modulation constellation size OFDM IFF TIFFT size Cyclic Prefix (CP) factor and finally system bandwidth. 20 25

ACKNOWLEDGMENT The research reported in this paper was developed in the framework of research activities of ICS (Information and Communication Systems) team in Systems and Technologies of Information and Communication (STlC) laboratory at Tlemcen University Algeria. [I] REFERENCES IEEE Standard for Local and Metropolitan area networks Part 16 The Institute of Electrical and Electronics Engineering Inc. Std. IEEE 802. 16d-2004. [2] IEEE Standard for Local and Metropolitan area networks Part 16 The Institute of Electrical and Electronics Engineering Inc. Std. IEEE 802. I 6e-2005. [3] A1. Paulraj and C.B. Papadias "Space-Time Processing for Wireless Communications'" IEEE Sig.Process. Mag. pp. 49-83 Nov. 1997. [4] D. Gesbert M. Shafi et. ai. "From Theory to Practice: An Overview of MIMO Space-Time Coded Wireless Systems'" IEEE 1. Sel. Areas in Comm. vol. 21 no. 3 pp. 281-301 Apr. 2003. [5] S.M. Alamouti "A simple transmit diversity tecnique for wireless communications" IEEE J. Sel. Areas in Comm. vol.16 no. 8 pp. 1451-1458 Oct. 1998. [6] G.J. Foschini "Layered Space-Time Architecture for Wireless Communication in a Fading Environment when using Multi-Element Antennas'" Bell Labs Tech. Jour. vol. I no. 2 pp. 41-591996. [7] R. V. Heath and AJ. Paulraj "Switching between Diversity and Multiplexing in MIMO Systems'" IEEE Trans. on Comm. vol.53 no.6 pp. 962-968 June 2005. [8] P. W. Wolniansky et ai "V -BLAST: An architecture for realizing very high data rates over the rich-scattering wireless channel" Proc. ISSSE Conference Pisa Italy September 1998. [9] A Slaney and Y. Sun "Space-time coding for wireless communications: an overview" lee Proc.-Commun. vol. 153 no. 4 August 2006. [] M. Wang "Wi MAX Physical Layer: Specifications Overview and Performance Evaluation"' 2nd IEEE CCNC Research Student Workshop pp -122011. [II] L. Yang S. Cheng H. Wang "Effects of cyclic prefix on OFDM systems over time-varying channels... in IEEE 16th International Symposium on Personal Indoor and Mobile Radio Communications vol. 2no. 11-14 pp. 750-753 Sept. 2005. [12] J. G. Andrews A Ghosh and R. Muhamed. "Fundamentals of WiMAX Understanding Broadband Wireless Networking" Pearson Education Inc. 2007.