
Research Article


FEC and DWT Guided Multiband Orthogonal Frequency Division Multiplexing (MBOFDM)Enriched Quality 

J. Avila
and
K. Thenmozhi



ABSTRACT

For high data rate wireless applications over short distances,
multiband OFDM proves to be the right contender. The concatenation of error
control code for multiband OFDM cognitive radio system is analysed. Reed Solomon
code and convolutional code are taken in to account for analysis. From the simulation
results it can be inferred that because of the improved coding gain concatenated
codes prove to be superior to single codes. As a supplementary, the fast Fourier
transform and Inverse fast Fourier transform portion of the multiband OFDM systems
are exchanged with the discrete wavelet transform and inverse discrete wavelet
transform blocks respectively thus reducing the interferences drastically and
as a result the overall bit error rate (BER) gets improved.





Received: March 05, 2013;
Accepted: April 12, 2013;
Published: April 18, 2014


INTRODUCTION The Federal Communications Commission announced a frequency spectrum as unlicensed spectrum called the Ultra Wide Band which is a wide band ranging from 3.1 to 10.6 GHz in the year 2002. Handling such a huge spectrum brought many difficulties and so a unique candidate named Multiband Orthogonal Frequency Division Multiplexing (MBOFDM) came into being (Balakrishnan et al., 2003).It’s important features include high data rates, feeding little amount of power and silicon area. These features led to its popularity.
The frequency split up of UWB is as follows: there are five major divisions known as master groups. Each of the master groups possesses 3 sections and the final one has 2 sections. Totally there are 14 sections and 128 subcarriers. Of the 128 subcarriers, 100 is allocated for transmission of data, 12 are for piloting purpose, 10 subcarriers are put to use as protecting tones and 6 are treated as unused tones. Spacing between the subcarrier is 4.125 MHZ. Multiband OFDM supports a high data rates ranging from 53 to 480 Mbps.
Figure 1 gives the block diagram of multiband OFDM system. In multiband OFDM system the input data is first randomized by the scrambler and a channel coder is used to encode the information sequence. The multipath fading that occurs in the communication channel is combated by error control codes (Praveenkumar et al., 2012). Compensation of heavy noise conditions can be done by concatenation of error controls which leads to improvement of coding gain (Avila et al., 2012b; Thenmozhi et al., 2012).
 Fig. 1:  Block diagram of MBOFDM 
The encoded bits are made to pass through an interleaver to confirm strength against burst error. A mapper is used to map into constellation symbols. This output is passed to IFFT block where OFDM symbols are generated (Avila et al., 2012c). The noise expected here is AWGN (Additive White Gaussian Noise). At the receiver the signal is made to pass through FFT block, demodulator and decoding take place to reclaim the sequence of information (Batra et al., 2006; Avila et al., 2012a).
This study aims at enhancing the performance of multiband OFDM system by combating the channel noise. Channel noise can be effectively mitigated by the use of error control codes. To offer better results in high noise conditions concatenation of error control codes is preferred. Here Reed Solomon code is concatenated with convolutional code and their performance is analysed.
Since Fast Fourier Transform is not suitable for non stationary signals, it is substituted with Wavelet transform in the multiband OFDM system. Four wavelet families namely Haar, Daubechie, Coiflet and Symlet which satisfy orthogonality property similar to FFT has been chosen.
LITERATURE The channel coding is a process in which superfluous information is added to the data and this improves the capacity of the channel in which the data is transmitted. The forward error control codes implemented in the multiband OFDM of the proposed work are Reed Solomon code and convolutional code. Consisting of m binary bits, Reed Solomon Code is a technique where m can be any number exceeding 2. Hence for a known m, the range of the complete ReedSolomon codes consisting of mbit symbols is 2^{m} – 1 = 255. Reed Solomon code has two parts: a data part and a parity part. In a code which contains n symbols, the data part is the initial k symbols. This is to improve security and safeguard information from intruders (Nyirongo et al., 2006). The rest (nk) are parity symbols calculated with respect to the data part. This type of Reed Solomon Code is titled as an (n,k) code of RS(n,k) code. In this case, the (nk) parity symbol is an even number and referred to as 2t. Therefore, an RS Code of 2t parity symbols has the ability to correct up to‘t’ error symbols.
Convolutional code is the mechanism of adding surplus bits to the data to safeguard it. A k/n rate convolutional encoder, using binary shift registers, can work on an input of k bit information. This encoder processes the contents of the shift registers and computes on n bit output symbol. At the receiver side the Viterbi algorithm computes the path metrics (accumulated distances) from the input symbols to the possible transmitted output with the help of Trellis diagram (Viraktamath and Attimarad, 2010). Memory problems and complexity is caused when the stacked number of path metrics advances exponentially with the number of stages in the trellis. The number of paths in competition to have the smallest distance is limited to the number of states in one column of the trellis and this becomes a major advantage for the Viterbi algorithm. The modulation scheme proposed in this study are Binary phase shift keying (BPSK) and Quadrature phase shift keying (QPSK).Binary phase shift keying is the technique in which phase of the carrier varies by 180 degree (Sanjeev and Swati, 2010). The probability of error of BPSK is given by:
Where:
D = Energy
J_{0} = noise QPSK supports high data rates because each symbol has two bits. The phase of the carrier is 90 degrees (Bernard and Ray, 2001). The probability of error of QPSK is given by:
Where:
D = Energy
J_{0} = noise
PROPOSED METHODOLOGY To achieve high data rates and to make the multiband OFDM system error free Reed Solomon codes are combined with convolutional code with the former as outer code and the later as inner code. Here the rate of the convolutional code is fixed as ½. To boot, the IDWT and DWT blocks are used instead of IFFT and FFT blocks in the multiband OFDM system (Das et al., 2011). We use wavelet transform here because it offers lot of advantages. Wavelet transform gives accurate results both in the domains. Local information is not available in FFT (Zheng et al., 2005). Multiresolution capability is also present. Thus wavelet transform is popularly used in image processing, information hiding, signal processing etc. (Al Wadi et al., 2011).
Filtering and scaling are the two basic steps in DWT. By means of passing the input signal into low pass and high pass filter we split it into two components. Of the two high pass filter gives accurate results whereas low pass filter gives approximate results. Scaling is improvised by passing through a decimator. To reach the target the process is repeated. Result is obtained by adding the coefficients. With many wavelet families existing, four wavelet transforms namely Haar, Daubechie, Symlet and Coiflet are implemented. Haar wavelet with its simplest nature, gains fame because of its rapid response (Idi and Kamarudin, 2012). But at the same time it is not suited for audio applications. The reason is since it is two elements wide the processing capacity is not good.
On the other hand Daubechie wavelet is complex and costly, has both linear and non linear phase response and also has large number of vanishing moments (Mahmoud et al., 2007). D2D20 is the even index it has where index is nothing but the number of coefficients. Initially the inner product of data and coefficients are obtained and then with its help the scaling and wavelet coefficients are obtained. A better version of Daubechie is the Symlet wavelet which is symmetric and has even index number. It has minimal phase (Chavan et al., 2011). The most symmetric wavelet is the coiflet wavelet and it has other properties similar to that of Daubechie wavelet.
RESULTS AND DISCUSSION All graphs are obtained using MATLAB tool. The discussion is based on the bit error rate versus Eb/N_{0} curve obtained for the multiband OFDM system. Figure 2 gives the evaluation amid multiband OFDM system without error control code and with error control code. Here Reed Solomon code is utilized. From the figure it is evident that coded system overrules the system without error control code.
Figure 3 gives the assessment between Reed Solomon code, Convolutional code and the concatenated output of both.
 Fig. 2:  Multiband OFDM system with and without FEC 
 Fig. 3: 
RSCC combination with Haar wavelet and BPSK modulation scheme 
The figure clearly shows that concatenated code yields better result than single codes. They have improved coding gain than single codes. The figure is plotted using Haar wavelet transform and with BPSK modulation scheme. BPSK is less prone to error because the distance between signal points is greater in this case. But since either binary one or binary zero is transmitted it is not suitable for high data rate scenarios.
Figure 4 gives the output of Reed Solomon code, convolutional code and the concatenation of the two with Daubechie as the wavelet transform and QPSK as the modulation scheme. For a given BER of 10^{3} there is improvement in Eb/N_{0} for the concatenated output than single codes. Figure 5 gives the output for various orders of Symlet wavelet. As the order increases better results could be achieved. Figure 6 gives the output for various orders of Coiflet wavelet transform.
 Fig. 4:  CCRS combination with Daubechie wavelet and QPSK modulation scheme 
 Fig. 5:  Performance of RSCC code for various order of Symlet 
 Fig. 6:  Performance for various order of Coiflet transform 
With the increase in order more data is available for processing which in turn improves the accuracy of the system.
In the study proposed by Sato et al. (2007) Reed Solomon is concatenated with convolutional code in multiband OFDM system. In the study proposed by (Hussain et al., 2011) concatenated of Reed Solomon codes is done with convolutional code in OFDM system. (Jatoor et al., 2008) proposed the concatenation of convolutional with Alamouti code to achieve diversity. In the study proposed by Ahmed et al. (2009) the performance of OFDM is analysed in the presence of convolutional code. In the study proposed by (LeijaHernandez et al., 2009) analysis of convolutional code in CDMA system is carried out.
CONCLUSION In this research contribution the error analysis of the M ultiband OFDM system under concatenated coding scheme having convolutional code as inner code and Reed Solomon code as outer code is evaluated and it is found that concatenated code has better BER performance than single CC or single RS. In addition to enhance the system performance discrete wavelet transform is used as a substitute for FFT. Simulation is also carried out for various wavelet families and various modulation schemes. The results are also obtained for various orders of wavelet families. All this simulation proves that RSCC combination has good error correcting capability which makes the multiband OFDM system a more powerful candidate.

REFERENCES 
Batra, A., S. Lingam and J. Balakrishnan, 2006. Multiband OFDM: A cognitive radio for UWB. Proceedings of the International Symposium on Circuits and System, May 2124, 2006, Island of Kos, pp: 40944097 CrossRef 
Avila, J., E. Praveen and B.V. Rajan, 2012. ANN assistedaugmentation of AIC for MIMO multiband OFDM system. Proceedings of the IEEE International Conference on Advances in Engineering, Science and Management, March 3031, 2012, India, pp: 228232
Avila, J., E. Vinoth and K. Thenmozhi, 2012. Antiinterference for anticorroupt 3G spectruma multi carrier approach. Proceedings of the IEEE International Conference of Computer Communication and Informatics, January 1012, 2012, India, pp: 478481
Avila, J., E. Vinoth, E. Praveen and K. Thenmozhi, 2012. Performance augment of Multiband OFDM with ANN assisted concatenated codes. Proceedings of the IEEE International Conference on Computing, Electronics and Electrical Technologies, March 2122, 2012, Kumaracoil, India, pp: 7680 CrossRef 
Idi, B.Y. and M.N. Kamarudin, 2012. Interpretation of ground penetrating radar image using digital wavelet transform. Asian J. Applied Sci., 5: 174182. CrossRef 
Bernard, S. and P.K. Ray, 2001. Digital CommunicationFundamentals and Applications. 2nd Edn., Pearson Education, USA
Das, B., S. Tiwari and S. Das, 2011. Performance study of discrete wavelet packet based MBOFDM system for short range indoor wireless environment. Proceedings of the IEEE International Conference on Devices and Communication, Febuary 2425, 2011, Mesra, pp: 15 CrossRef 
Hussain, G.A., M.B. Mokhtar and R.S.A.B. Raja, 2011. Concatenated RSconvolutional codes for MIMOOFDM system. Asian J. Applied Sci., 4: 720727. CrossRef  Direct Link 
Ahmed, K.M., S.P. Majumder and B.K.M.M. Rahman, 2009. Performance analysis of an OFDM wireless communication system with Convolutional Coding. Proceedings of the 11th International Conference on Advanced Communication Technology, February 1518, 2009, Phoenix Park, pp: 10781083
Balakrishnan, J., A. Batra and A. Dabak, 2003. A multiband OFDM system for UWB communication. Proceedings of the IEEE Conference on Ultrawide Band Systems and Technology, November 1619, 2003, Dallas, TX, USA., pp: 354358 CrossRef 
Jatoor, J., J.R. Zeidler and J.G. Proakis, 2008. On the performance of concatenated convolutional code and Alamouti space time code with noisy channel estimates and finite depth interleaving. IEEE Trans. Communi., 56: 412423. CrossRef 
LeijaHernandez, G., M. Badaoui and A. IturriHinojosa, 2009. Performance analysis of convolutional coding in CDMA communications systems. J. Vectorial Relativity, 4: 4956. Direct Link 
Chavan, M.S., N. Mastorakis, M.N. Chavan and M.S. Gaikwad, 2011. Implementation of SYMLET wavelets to removal of gaussian additive noise from speech signal. Proceedings of the 10th International conference on Recent Researches in Communications, Automation, Signal Processing, Nanotechnology, Astronomy and Nuclear Physics, February 2022, 2011, India, pp: 3741
Mahmoud, I.M., M.I.M. Dessouky, S. Deyab and F.H. Elfouly, 2007. Comparison between haar and daubechies wavelet transformions on FPGA technology. Proc. World Acad. Sci. Eng. Technol., 20: 6872. Direct Link 
Nyirongo, N., W.Q. Malik and D.J. Edwards, 2006. Concatenated RSconvolutional codes for ultrawideband multibandOFDM. Proceedings of the IEEE International Conference on UltraWideband, September 2427, 2006, Waltham, MA., USA., pp: 137142 CrossRef 
Praveenkumar, P., R. Amirtharajan, K. Thenmozhi and J.B.B. Rayappan, 2012. Regulated OFDMrole of ECC and ANN: A review. J. Applied Sci., 12: 301314. CrossRef  Direct Link 
Al Wadi, S., M.T. Ismail and S.A.A. Karim, 2011. Discovering structure breaks in amman stocks market. J. Applied Sci., 11: 12731278. CrossRef 
Kumar, S. and S. Sharma, 2007. Error probability of different modulation schemes for OFDM based WLAN standard IEEE 802.11a. Int. J. Eng., 4: 262267. Direct Link 
Sato, H., T. Ohtsuki, T. Kashima and S. Jarot, 2007. MultibandOFDM system with concatenated coding scheme. Proceedings of the 65th Conference on Vehicular Technology, April 2225, Dublin, pp: 21652169 CrossRef  Direct Link 
Thenmozhi, K., P. Praveenkumar, R. Amirtharajan, V. Prithiviraj, R. Varadarajan and J.B.B. Rayappan, 2012. OFDM+CDMA+Stego = Secure communication: A review. Res. J. Inform. Technol., 4: 3146. CrossRef  Direct Link 
Viraktamath, S.V. and G.V. Attimarad, 2010. Impact of constraint length on performance of convolutional CODEC in AWGN channel for image application. Int. J. Eng. Sci. Technol., 2: 46964700.
Zheng, X., X. Huang and M. Wang, 2005. Object edge smoothing by wavelet transformation. Inform. Technol. J., 4: 451455. CrossRef  Direct Link 



