Implementation of channel estimation algorithms in ofdm for 64 subcarriers

Implementation of channel estimation algorithms in ofdm for 64 subcarriers

• 1.
  International Journal ofElectronics and Communication Engineering & Technology (IJECET), ISSN 0976 – 6464(Print), ISSN 0976 – 6472(Online) Volume 4, Issue 4, July-August (2013), © IAEME 42 IMPLEMENTATION OF CHANNEL ESTIMATION ALGORITHMS IN OFDM FOR 64 SUBCARRIERS Navdeep Bansal1 , Sukhjeet Singh2 , Pardeep Kumar Jindal3 1 ECE Department, GTBKIET Chhapianwali 2 ECE Department, GTBKIET Chhapianwali 3 ECE Department, GGSCET Talwandi Sabo ABSTRACT Main objectives of this paper is to design the PSK and QAM system for Symbol Error Rate (SER) performance analysis and to estimate the channels in OFDM. PSK & QAM syatems are designed for 64 Sbcarriers. In this paper we will compare the SER for both techniques for same modulation rate and same number of subcarrier. In this paper we will show how Symbol error rate is reduced as modulation rate increases for PSK & QAM. We will use three algotihms LS, LMMSE & Modified MMSE to compare the result parameters. Modified MMSE gives better results than LS & LMSE but computational complexity will be increased that is its major drawback. Keywords: OFDM, PSK, QAM, SER, LS, LMMSE, Modidied MMSE 1. INTRODUCTION In this paper we will design the system for PSK & QAM modulation techniques for 64 subcarriers for the channel estimation in OFDM. We will implement the LS and LMMSE algorithms to estimate the channels and compare the symbol error rate for the PSK & QAM Modulation technique. OFDM(Orthognal Frequency Division Modulation) is a multichannel modulation that divides a given channel into many parallel sub-channels or subcarriers, so that multiple symbols are sent in parallel. It is a block transmission technique.The transmitted OFDM signal multiplexes several low- rate data streams-each data stream is associated with a given subcarrier. The main advantage of this concept in a radio environment is that each of the data streams experiences an almost flat fading channel. In slowly fading channels, the inter-symbol interference (ISI) and inter-carrier interference (ICI) within an OFDM symbol can be avoided with a small loss of transmission energy using the concept of a cyclic prefix [4]. INTERNATIONAL JOURNAL OF ELECTRONICS AND COMMUNICATION ENGINEERING & TECHNOLOGY (IJECET) ISSN 0976 – 6464(Print) ISSN 0976 – 6472(Online) Volume 4, Issue 4, July-August, 2013, pp. 42-50 © IAEME: www.iaeme.com/ijecet.asp Journal Impact Factor (2013): 5.8896 (Calculated by GISI) www.jifactor.com IJECET © I A E M E
• 2.
  International Journal ofElectronics and Communication Engineering & Technology (IJECET), ISSN 0976 – 6464(Print), ISSN 0976 – 6472(Online) Volume 4, Issue 4, July-August (2013), © IAEME 43 The channel estimation can be performed by following pilot patterns. • Block type Pilot arrangement. • Comb type Pilot arrangement. Block Pilot Type: In block type, pilot tones are inserted in all subcarriers of an OFDM symbols periodically This type is suitable for slow-fading channels where channel characteristics are assumed stationary during one OFDM data block. For block type arrangements, channel at pilot tones can be estimated by using LS based or LMMSE based estimation, and assumes that channel remains the same for the entire block. So in block type estimation, we first estimate the channel, and then use the same estimates within the entire block [1]. Comb Type Pilot Type: The comb type pilot arrangement is generally based on inserting pilot tones in each individual OFDM data block as shown in figure 3.5.The channel is estimated in all OFDM symbols. The concept is to introduce some of the sub carriers as pilot carriers in each OFDM symbol. Comb type pilot tone estimation, has been introduced to satisfy the need for equalizing when the channel changes even in one OFDM block. The comb-type pilot channel ssestimation consists of algorithms to estimate the channel at pilot frequencies and to interpolate the channel. In a fast fading channel, the characteristics of a radio channel are changing within an OFDM block. Therefore, channel transfer function should be estimated in each OFDM symbol of a data block. [1] Channel Estimation Methods: Channel can be estimated at pilot frequencies by the following methods: 1. Least Square based Channel Estimation Method 2. Linear Minimum Mean Square Error based Estimation Method 2. BLOCK DIAGRAM Fig2.1: Channel Estimation using LS/MMSE algorithm In Block type pilot based channel estimation , each subcarrier in OFDM symbol is used in such a way that all subcarriers are used as pilots. The estimation of the channel is then done using Least square estimator amd Minimum mean square error estimator. [9], [13],[14].
• 3.
  International Journal ofElectronics and Communication Engineering & Technology (IJECET), ISSN 0976 – 6464(Print), ISSN 0976 – 6472(Online) Volume 4, Issue 4, July-August (2013), © IAEME 44 3. SIMULATION PARAMETERS Table No. 1 PARAMETER SPECIFICATION Number of Subcarriers N=64 FFT size 64 Length of Guard Interval L=4, L= 32 & L= 64 samples Modulation type Different types of QAM/PSK Pilot Type Block type arrangements Channel Model Rayleigh fading Channel Bandwidth 1 MHz Maximum Delay time 12 microseconds. Doppler frequency shift 100-250 Hz For block-type pilot channel estimation, it is assumed that each block consists of a fixed number of OFDM symbols. The each OFDM symbols consists of 64 subcarriers. Pilots are sent at all subcarriers of the first symbol of each block and channel estimation is performed by using LS, LMMSE. Channel estimated at the beginning of the block is used for all of the following symbols of the block. 4. IMPLEMENTATION AND RESULT DISCUSSION Figure 4.1: Performance comparison of Figure 4.2: Performance comparison of QAM symbol error rate 4QAM symbol error rate (N= 64, M=4, L=4, S1=32) 10 20 30 40 50 60 70 80 90 100 10 -1.6 10 -1.5 10 -1.4 10 -1.3 SNR in dB SymbolErrorRate SNR V/S Symbol Error Rate in OFDM SYSTEM LSE MMSE Modified MMSE(M) 10 20 30 40 50 60 70 80 90 100 10 -2 10 -1 10 0 SNR in dB SymbolErrorRate SNR V/S Symbol Error Rate in OFDM SYSTEM LSE MMSE Modified MMSE(M)
• 4.
  International Journal ofElectronics and Communication Engineering & Technology (IJECET), ISSN 0976 – 6464(Print), ISSN 0976 – 6472(Online) Volume 4, Issue 4, July-August (2013), © IAEME 45 Figure 4.3: Performance comparison of Figure 4.4 Performance comparison of 8QAM symbol error rate 16QAM symbol error rate (N= 64, M=8, L=4, S1=32) (N= 64, M=16, L=4, S1=32) Figure 4.5: Performance comparison of Figure 4.6: Performance comparison of 32QAM symbol error rate 64QAM symbol error rate (N= 64, M=32, L=4, S1=32) (N= 64, M=64, L=4, S1=32) Figure 4.7: Performance comparison Figure 4.8: Performance comparison of of BPSK symbol error rate QPSK symbol error rate (N= 64, M=2, L=4, S1=25) (N= 64, M=4, L=4, S1=25) 10 20 30 40 50 60 70 80 90 100 10 -2 10 -1 10 0 SNR in dB SymbolErrorRate SNR V/S Symbol Error Rate in OFDM SYSTEM LSE MMSE Modified MMSE(M) 10 20 30 40 50 60 70 80 90 100 10 -2 10 -1 10 0 SNR in dB SymbolErrorRate SNR V/S Symbol Error Rate in OFDM SYSTEM LSE MMSE Modified MMSE(M) 10 20 30 40 50 60 70 80 90 100 10 -0.9 10 -0.8 10 -0.7 10 -0.6 SNR in dB SymbolErrorRate SNR V/S Symbol Error Rate in OFDM SYSTEM LSE MMSE Modified MMSE(M) 10 20 30 40 50 60 70 80 90 100 10 -0.9 10 -0.8 10 -0.7 10 -0.6 10 -0.5 10 -0.4 SNR in dB SymbolErrorRate SNR V/S Symbol Error Rate in OFDM SYSTEM LSE MMSE Modified MMSE(M) 10 20 30 40 50 60 70 80 90 100 10 -1.6 10 -1.5 10 -1.4 10 -1.3 SNR in dB SymbolErrorRate SNR V/S Symbol Error Rate in OFDM SYSTEM LSE MMSE Modified MMSE(M) 10 20 30 40 50 60 70 80 90 100 10 -2 10 -1 10 0 SNR in dB SymbolErrorRate SNR V/S Symbol Error Rate in OFDM SYSTEM LSE MMSE Modified MMSE(M)
• 5.
  International Journal ofElectronics and Communication Engineering & Technology (IJECET), ISSN 0976 – 6464(Print), ISSN 0976 – 6472(Online) Volume 4, Issue 4, July-August (2013), © IAEME 46 Figure 4.9: Performance comparison of Figure 4.10: Performance comparison of 8QPSK symbol error rate 16QPSK symbol error rate (N= 64, M=8, L=4, S1=25) (N= 64, M=16, L=4, S1=25) Figures 4.1 to 4.6 and from Table-2, the results of computations of symbol error rates of different methods of channel estimation for different types of Quadrature Amplitude Modulations (QAM) have been calculated and compared under the conditions as per Table 2 parameters for 64 subcarriers. The channel correlation matrix RHH for LMMSE method consists of 64 coefficients and the modified MMSE method considered 32 coefficients in the matrix have been considered. It has been observed that the modified MMSE estimator has smaller MSE than that of LMMSE estimator and much smaller LS estimator Figure 4.11: Performance comparison of Figure 4.12: Performance comparison of 32 QPSK symbol error rate 64QPSK symbol error rate (N= 64, M=32, L=4, S1=25) (N= 64, M=64, L=4, S1=25) 10 20 30 40 50 60 70 80 90 100 10 -2 10 -1 10 0 SNR in dB SymbolErrorRate SNR V/S Symbol Error Rate in OFDM SYSTEM LSE MMSE Modified MMSE(M) 10 20 30 40 50 60 70 80 90 100 10 -0.7 10 -0.6 10 -0.5 10 -0.4 10 -0.3 SNR in dB SymbolErrorRate SNR V/S Symbol Error Rate in OFDM SYSTEM LSE MMSE Modified MMSE(M) 10 20 30 40 50 60 70 80 90 100 10 -0.5 10 -0.4 10 -0.3 10 -0.2 SNR in dB SymbolErrorRate SNR V/S Symbol Error Rate in OFDM SYSTEM LSE MMSE Modified MMSE(M) 10 20 30 40 50 60 70 80 90 100 10 -0.2 10 -0.1 SNR in dB SymbolErrorRate SNR V/S Symbol Error Rate in OFDM SYSTEM LSE MMSE Modified MMSE(M)
• 6.
  International Journal ofElectronics and Communication Engineering & Technology (IJECET), ISSN 0976 – 6464(Print), ISSN 0976 – 6472(Online) Volume 4, Issue 4, July-August (2013), © IAEME 47 TABLE-2: Symbol error rate performance comparison of different QAM (N=64) In Figures 4.7 to 4.12 and from Table-3, the results of computations of symbol error rates of different methods of channel estimation for different types of Phase Shift Keying (PSK) have been calculated and compared under the conditions as per Table 01 parameters for 64 subcarriers. The channel correlation matrix RHH for LMMSE method consists of 64 coefficients and the modified MMSE method considered 25 coefficients in the matrix have been considered. It has been observed that the modified MMSE estimator has similar MSE than LMMSE estimator and much smaller than LS estimator. Hence: On considering the optimal number of coefficients 25 out of 64, the performance of modified MMSE method is better than LMMSE and much better than the LS methods. Also it reduces the computational complexity of MMSE method by considering only the significant coefficients 25 out of 64 for different types of PSK modulations. Modulation Type Estimation Method SNR (10 dB) SNR (20 dB) SNR (30 dB) SNR (40 dB) SNR (50 dB) SNR (60 dB) SNR (70 dB) SNR (80 dB) QAM LS 0.0514 0.0376 0.0390 0.0394 0.0388 0.0391 0.0394 0.0386 LMMSE 0.0520 0.0309 0.0244 0.0234 0.0238 0.0233 0.0232 0.0239 Modified MMSE 0.0425 0.0267 0.0231 0.0229 0.0238 0.0234 0.0232 0.0240 4QAM LS 0.1019 0.0989 0.0992 0.0924 0.0902 0.0899 0.0900 0.0897 LMMSE 0.0950 0.0770 0.0738 0.0743 0.0744 0.0744 0.0742 .0742 Mod MMSE 0.0926 0.0783 0.0703 0.0639 0.0592 0.0589 0.0586 0.0586 8QAM LS 0.1106 0.0792 0.0729 0.0728 0.0714 0.0735 0.0713 0.0741 LMMSE 0.1335 0.1186 0.1224 0.1251 0.1258 0.1259 0.1250 0.1284 Mod MMSE 0.0914 0.0599 0.0512 0.0495 0.0481 0.0480 0.0480 0.0502 16QAM LS 0.1257 0.1065 0.1033 0.1041 0.1037 0.1043 0.1049 0.1047 LMMSE 0.1874 0.1694 0.1674 0.1658 0.1635 0.1640 0.1632 0.1629 Mod MMSE 0.1104 0.0893 0.0891 0.0884 0.0880 0.0888 0.0892 0.0891 32QAM LS 0.1321 0.1123 0.1139 0.1136 0.1151 0.1143 0.1155 0.1123 LMMSE 0.2733 0.2556 0.2602 0.2581 0.2598 0.2582 0.2580 0.2562 Mod MMSE 0.1715 0.1294 0.1182 0.1126 0.1134 0.1143 0.1121 0.1147 64QAM LS 0.1732 0.1504 0.1471 0.1478 0.1447 0.1471 0.1451 0.1485 LMMSE 0.4210 0.4106 0.4093 0.4127 0.4125 0.4119 0.4101 0.4144 Mod MMSE 0.1715 0.1294 0.1182 0.1126 0.1134 0.1143 0.1121 0.1147
• 7.
  International Journal ofElectronics and Communication Engineering & Technology (IJECET), ISSN 0976 – 6464(Print), ISSN 0976 – 6472(Online) Volume 4, Issue 4, July-August (2013), © IAEME 48 Table 3: Symbol error rate performance Comparision of Diff. PSK (N=64) Result Comparision TABLE- 4: Performance comparison of different QAM & PSK modulations ‘ Modulation Type Estimation Method SNR (10 dB) SNR (20 dB) SNR (30 dB) SNR (40 dB) SNR (50 dB) SNR (60 dB) SNR (70 dB) SNR (80 dB) BPSK LS 0.0535 0.0396 0.0409 0.0392 0.0398 0.0393 0.0383 0.0390 LMMSE 0.0554 0.0412 0.0409 0.0389 0.0388 0.0385 0.0396 0.0392 Modified MMSE 0.0389 0.0257 0.0238 0.0236 0.0234 0.0228 0.0236 0.0235 QPSK LS 0.1476 0.1250 0.1134 0.1115 0.1100 0.1054 0.1053 0.1057 LMMSE 0.1023 0.0664 0.0490 0.0432 0.0428 0.0432 0.0430 0.0432 Mod MMSE 0.0944 0.0601 0.0460 0.0423 0.0424 0.0432 0.0434 0.0432 8PSK LS 0.2343 0.1797 0.1661 0.1601 0.1563 0.1543 0.1543 0.1542 LMMSE 0.2207 0.1433 0.1464 0.1537 0.1541 0.1542 0.1543 0.1545 Mod MMSE 0.1697 0.0760 0.0583 0.0589 0.0607 0.0608 0.0606 0.0605 16PSK LS 0.5085 0.4217 0.4123 0.3947 0.3894 0.3896 0.3900 0.3898 LMMSE 0.4348 0.3166 0.2939 0.2846 0.2819 0.2803 0.2806 0.2801 Mod MMSE 0.3718 0.1924 0.1720 0.1788 0.1863 0.1863 0.1864 0.1864 32PSK LS 0.7160 0.6603 0.6785 0.6753 0.6715 0.6713 0.6714 0.6715 LMMSE 0.6876 0.5290 0.4844 0.4737 0.4707 0.4683 0.4684 0.4683 Mod MMSE 0.6406 0.3989 0.3212 0.2943 0.2785 0.2683 0.2652 0.2652 64 PSK LS 0.8657 0.8496 0.8522 0.8546 0.8590 0.8590 0.8592 0.8591 LMMSE 0.8339 0.8107 0.8318 0.8357 0.8339 0.8295 0.8281 0.8277 Mod MMSE 0.8025 0.6484 0.6226 0.6115 0.6008 0.5953 0.5933 0.5934 Modulation Type N=64 Subcarrier LS LMMSE Modified MMSE QAM 0.0514 0.0520 0.0425 4QAM 0.1019 0.0950 0.0926 8QAM 0.1106 0.1335 0.0914 16QAM 0.1257 0.1874 0.1104 32QAM 0.1321 0.2733 0.1266 64QAM 0.1732 0.4210 0.1715 BPSK 0.0535 0.0554 0.0389 QPSK 0.1476 0.1023 0.0944 8PSK 0.2343 0.2207 0.1697 16PSK 0.5085 0.4348 0.3718 32PSK 0.7160 0.6876 0.6406 64PSK 0.8657 0.8339 0.8025
• 8.
  International Journal ofElectronics and Communication Engineering & Technology (IJECET), ISSN 0976 – 6464(Print), ISSN 0976 – 6472(Online) Volume 4, Issue 4, July-August (2013), © IAEME 49 The results obtained from simulations showed that the LMMSE method performs significantly better than the LS estimator, but having drawback of more computational complexity. The results of the modified MMSE channel estimator are better than LMMSE, which is based on the rank-reduction of the correlation matrix with almost the same performance as the full-rank LMMSE method, while significantly reducing the computational complexity. Further, the simulations results showed that the symbol error rate increasing as the modulation type is increased for different QAMs and PSKs REFERENCES [1] Athaudage C.R.N.; Jayalath A.D.S., ‘Low-complexity Channel Estimation for Wireless OFDM Systems’, IEEE Proceedings on Indoor 7 mobile communications, 2003, Publication Year: 2003, Page(s): 521 - 525 Vol.1 [2] Chi-Hsiao Yih, “Effects of Channel Estimation Error in the Presenceof CFO on OFDM BER in Frequency-Selective Rayleigh Fading Channels,” JOURNAL OF COMMUNICATIONS, VOL. 3, NO. 3, JULY 2008 [3] J. Torrance and L, Hanzo, Multicarrier Modulation for Data Transmission: An idea whose time has come IEEE comun. Magazine, pp.5-14, May 1990. [4] J. J. van de Beek, P. Ödling. S.K. Wilso, P.O. Börjesson, ‘Orthogonal Frequency-Division Multiplexing’, International Union of Radio Science, Sweden. Oxford University Press, 1999, in press [5] J.V. de Beek, O. Edfors, M. Sandell, S.K.Wilson and P.O Borjesson, “On Channel Estimation In OFDM” Vehicular Technology Conference, vol. 2 pp. 815-819, Chicago, USA, September 1995
• 9.
  International Journal ofElectronics and Communication Engineering & Technology (IJECET), ISSN 0976 – 6464(Print), ISSN 0976 – 6472(Online) Volume 4, Issue 4, July-August (2013), © IAEME 50 [6] Jan-Jaap van de Beek, Ove Edfors, Magnus Sandell, Sarah Kate Wilson and Per Ola B.rjesson” On Channel Estimation In OFDM Systems”, In Proceedings of Vehicular Technology Conference (VTC Ô95), vol. 2, pp. 815-819, Chicago, USA, September 1995. [7] K. Fazel and G. Fettwis, “Performance of an Efficient Parallel Data Transmission System,” IEEE Trans. Commun. Tech., pp. 805-813, December1967. [8] M. Okada, S. Hara and N. Morinaga, “Bit Error Performances of Orthogonal Multicarrier modulation radio transmission schemes.” IEICE Trans. Commun, Vol.E76-B, pp. 113-119, Fed. 1993 [9] Sajjad Ahmed Ghauri, Sheraz Alam, M. Farhan Sohail, Asad Ali, Faizan Saleem, implementation of OFDM and channel estimation using LS and MMSE Estimators. International Journal of Computer and Electronics Research [Volume 2, Issue 1, February 2013]. [10] Sinem Coleri, Mustafa Ergen, AnujPuri, and Ahmad Bahai, “Channel Estimation Techniques Based on Pilot Arrangement in OFDM Systems,” IEEE transactions on Broadcasting, Vol. 48, No. 3, September 2002. [11] Sinem Coleri, Mustafa Ergen, Anuj Puri, and Ahmad Bahai “Channel Estimation Techniques Based on PilotArrangement in OFDM Systems” IEEE TRANSACTIONS ON BROADCASTING, VOL. 48, NO. 3, SEPTEMBER 2002. [12] Yuping Zhao, Aiping Huang, “A novel channel estimation method for OFDM mobile communication systems based on pilot signals and transform-domain processing ,” IEEE VTC , Vol. 3, May 1997. [13] Yuping Zhao, Aiping Huang, “A novel channel estimation method for OFDM mobile communication systems based on pilot signals and transform-domain processing ,” IEEE VTC , Vol. 3, May 1997. [14] Sinem Coleri, Mustafa Ergen, Anuj Puri, and Ahmad Bahai “Channel Estimation Techniques Based on PilotArrangement in OFDM Ssystems” IEEE VTC , Vol. 3, May 1997. [15] Haritha.Thotakura, Dr. Sri Gowri .Sajja and Dr. Elizabeth Rani.D, “Performance of Coherent OFDM Systems Against Frequency Offset Estimation under Different Fading Channels”, International Journal of Electronics and Communication Engineering & Technology (IJECET), Volume 3, Issue 1, 2012, pp. 244 - 251, ISSN Print: 0976- 6464, ISSN Online: 0976 –6472. [16] Nidhi Mehta, Mandeep Kaur Sekhon and Gurpadam Singh, “Performance Evaluation of Stbc Codes with Channel Estimation by Optimized Pilot Bits”, International Journal of Electronics and Communication Engineering & Technology (IJECET), Volume 3, Issue 1, 2012, pp. 54 - 61, ISSN Print: 0976- 6464, ISSN Online: 0976 –6472. [17] Jaimin K. Raval, Prof. Vijay K. Patel and Dr. D. J. Shah, “Research on Pilot Based Channel Estimation for Lte Downlink using LS and LMMSE Technique”, International Journal of Electronics and Communication Engineering & Technology (IJECET), Volume 4, Issue 3, 2013, pp. 70 - 82, ISSN Print: 0976- 6464, ISSN Online: 0976 –6472.