A RobustJitter Noise Power Reduction in Ultra-Speed Optical OFDM Systems

Transcription

1 A RobustJitter oise Power Reduction in Ultra-Speed Optical OFDM Systems GottemukkalaTherisa 1, Y Venkata Adi Satyanarayana ¹PG Scholar in DECS, Dr Samuel George Institute of Engineering and Technology, Markapur, Andhra Pradesh, India ²Professor and HOD, Department of Electronics and Communication Engineering, Dr Samuel George Institute of Engineering and Technology, Markapur, Andhra Pradesh, India Abstract Recently, there is a tremendous growth in wireless communication networks for achieving ultra-high-speed data transmission to enhance the performance of optical communication system. In such systems, jitter is a serious drawback and will leads to impairments in transceivers that causes overall mitigation on the system performance. To resolve this issue, we utilized a robust jitter noise power reduction approach to mitigate the jitter effect under several fading channel conditions such as additive white gaussian noise (AWG), Rayleigh and Rician distributions. Extensive simulation analysis shown that the proposed approach mitigates the average jitter noise under considered channel circumstances. We also compared the channel performances in terms of sub carrier index versus power. Keywords OFDM;onlinearities; PAPR; High power amplifiers;partial transmit sequence;precoding matrix and CCDF I. ITRODUCTIO ow a day every field needs higher speed data transmission over a communication network. Orthogonal frequency division multiplexing (OFDM) is a such network that allows high speed data transmission at higher data rates without degrading the quality of input signals. Recent years optical systems utilize OFDM for further enhancement of data rates from high to ultra-speed (see [8] and the references therein) since, the fiber optics data rates are much higher than the Radio Frequency (RF) wireless systems. Timing jitter is a vital and serious problem which causes impairments in transceivers at ultra-data rates that leads to poor system efficiency and performance degradation of the system. jitter is the deviation from true periodicity of a presumably periodic signal, often in relation to a reference clock signal. In clock recovery applications it is called timing jitter. Jitter is a significant, and usually undesired, factor in the design of almost all communications links. Sampling clock is the major source of jitter in the ultra-high-speed analog-to-digital converters (ADCs) which are required in these systems. It causes many serious issues in optical-ofdm radios which uses high frequency band pass sampling [9]. In [10] and [11-15], the jitter mitigation methodologies have been discussed. These papers focus on the colored low pass timing jitter which is typical of systems using phase lock loops (PLL). However, the conventional schemes have been tested under AWG channel environment with less number of symbols. Here, we utilized a robust jitter noise power mitigation scheme under various channel distributions instead of only AWG channel. The channels considered for simulation are fading channels. II. SYSTEM MODEL The ultra-speed optical OFDM block diagram has been demonstrated in figure 1, where the symbol mapper for mapping of input symbols, serial-to-parallel converter (S/P), inverse discrete Fourier transform (IDFT) or inverse fast Fourier transform (IFFT), guard internal (GI), digital-to-analog converter (D/A) with low pass filter (LPF) and radio frequency (RF)- to-optical up converter included in the transmitter section. Optical link has been provided as a Page 51

2 media which consists of fading channels such as AWG, Rayleigh and Rician. The receiver section contains the reverse operated blocks presented in transmitter section like de-mapper, DFT or FFT, analog-to-digital (A/D) and LPF, parallel-to-serial converter (P/S) and opticalto-rf. Generally, the jitter can be occurred at several points in practical optical-ofdm systems. But, we considered here that the jitter introduction will be at the receiver s ADC sampler block. Figure demonstrates that the definition of jitter. Ideally the OFDM signal is sampled at uniform intervals of T. In upper one of figure the uniform sampling intervals has been represented by dashed lines, where the actual sampling times represented by solid lines. The deviation between the actual sampling times and uniform sampling times is caused by the effect of timing jitter, it denoted byτ n. Discrete timing jitter τ n example is shown in down of fig.. Fig. 1 Block diagram of ultra-speed optical-ofdm communication system Fig. Definition of jitter Y = WHX T + (1) Where, Y= received signal X= transmitted signal = Additive white Gaussian signal H is channel response matrix and W is the timing jitter matrix Y = Y +1 Y 0 Y T X T = X +1 X 0 X T H = diag H +1 H 0 H Page 5

3 And () Timing jitter causes an added noise like component in the received signal. Y = HX T + W I HX T + (3) where I is the identity matrix. In eq. (3), the first term represents wanted term while the second term indicates jitter noise. In it was shown that the elements of the timing jitter matrix W are given by w l,k = 1 e jπk τ n π j k l n T e n= +1 (4) where n is the time index, k is the transmitted subcarrier index and l is the received subcarrier index. III. IHERET OVERSAMPLIG We now analyse theinherent oversampling (IOS) and show that it will mitigate the impairments caused by the timing jitter. The received signal will be sampled with M/T to achieve the IOS, where M is an integer. The band width of the signal is /T, when all sub carriers are modulated. So as shown in fig. sampling at interval of yquist sampling T/. The bandwidth of the signal is L+ U, If instead, only sub carriers with indices between T L and U are non-zero. In this case the sampling interval is above yquist rate. In general, where both inherent and fractional oversampling is applied, the signal samples after the ADC receiver are given by, j πk n M T y nm = y n M T = 1 U n H M k X k e T M + η M T k= L (5) M Where n M = the time index of discrete over sampling and ηis the AWG. The -point FFT in the receiver is replaced by an oversized M-point FFT in proposed algorithm. The output of this FFT is a vector of length M with elements Y lm = 1 1 M y M nm e j π n M l M M n M = M (6) +1 Where l M is the M-point FFT output index. Then combining the equations (4), (5) and (6), the modified coefficients for the proposed case is obtained by, w lm,k = 1 M e jπk τ n M π M j T e M k l M n M n M = M +1 (7) We now calculate each sub carrier average jitter noise power for the case of white jitter. From (3) and (6), Y lm = H lm X lm + U k= L w lm,k I lm,k H k X k + l (8) Where the second term represents the jitter noise. In the following, we consider a flat channel with, H k = 1, and consider that the power of transmitted signal has been distributed equally Page 53

4 across the used subcarriers so that for each used subcarrier E X k = σ s. Then the average jitter noise power, to received signal power of l th subcarrier is given by, σ = E w u k= l,k I l,k X k l s σ = E w l,k I U l,k s k= L (9) For a system in which ther-e are equal number of unused subcarriers at each band edge, so U = v, L = v + 1 Where v is number of used subcarriers. Then the eq. (9) becomes σ = E w l,k I v l,k s k= v +1 (10) Reducing the numbers of subcarriers will results in reduction of overall data, the transmitted power and even bandwidth when the symbol period in unchanged. To make a fair comparison the symbol period is also reduced so thatt v = v T, where T v is v used sub carriers symbol period and T is the symbol period of used subcarriers. 1 σ = v πk s k= v +1 M v T E τn (11) v Using k = 1 k= v +1 1 v + v, we obtain σ = E τ s 3M n v π 1 T + v E τ 3M n 3 v π T v Since the first term in eq. (1) is very small, so it can be ignored. Then the equation will be re written as σ s = π 3M E τ n v T (13) If we consider that there is no oversampling then M=1 and v =, therefore σ s = π (1) 3 E τ n T (14) Comparing the eq. (13) and (14), the combination of inherent and fractional oversampling reduces the jitter noise power by a factor of v M. IV. SIMULATIORESULTS Here in this section we demonstrated the extensive simulation analysis of proposed jitter noise power reduction model, which have been tested in MATLAB 014a version. Here we had considered 000 OFDM symbols, = 51 subcarriers and E τ n = 0.3T jitter variance, which will not vary when we apply oversampling. Fig.3 shows that the simulation and theoretical results of average jitter noise power as a function of the IOS factor under AWG, Rayleigh and Rician channel environments. Page 54

5 Fig.3 Performance of proposed model under AWG channel environment Fig. 4 Received power vs subcarrier index with AWG channel Page 55

6 Fig. 5 Rayleigh channel environment with power vs subcarrier index When you observe the graph, we can conclude that for every doubling of sampling rate, that will mitigate the jitter noise power by 3dB.Thevariance of the noise due to the jitter as a function of index of received subcarriers when band-edge subcarriers are unused, which shows that the index of subcarrier and that removing the band-edge subcarriers reduces the noise equally across the all subcarriers has been shown in figure 4, 5 and 6 respectively. As said above, figure 5 and 6 describes the performance of received power with the subcarrier index, where the power has distributed equally among the subcarriers. Figure 7 show that how the bit error rate (BER) varies according to the signal-to-noise ratio (SR) with the increment in the IOS factor, the BER is getting reduced while the IOS factor is increasing and at the same time the SR values also get improving. This leads that the improvement in system efficiency as well as the higher data rates. Fig. 6 Performance of received power vs subcarrier index under Rician channel Page 56

7 Fig. 7 Performance of BER vs SR with varying oversampling factor V. COCLUSIOS It has been shown that the proposed algorithm for jitter noise power reduction shows better performance and it reduces the impairments caused by the jitter noise in ultra-speed optical-ofdm systems. In this, we investigated inherent oversampling mathematically by deriving the equations to them in the proposed system model, which was implemented by receiver sampling rate increment. Finally, it shows that the proposed algorithm has given reduced noise power of 3dB for every doubling of sampling rate under various channel conditions such as AWG, Rayleigh and Rician. REFERECES 1. Wu Y., W. Y. Zou, Orthogonal frequency division multiplexing: A multi-carrier modulation scheme, IEEE Transactions on Consumer Electronics, vol. 41, no. 3, pp , Aug Van ee R., Prasad R., OFDM for wireless Multimedia Communications, Artech House, ETSI, Radio broadcasting systems; Digital Audio Broadcasting (DAB) to mobile, portable and fixed receivers, European Telecommunication Standard, Standard E , May Hiperlan, Broadband Radio Access etworks (BRA), HIPERLA Type ; Physical (PHY) layer, ETSI, Tech. Rep., ETSI, Digital Video Broadcasting (DVB); Framing structure, channel coding and modulation for digital terrestrial television, European Telecommunication Standard, Standard E , Part 16: Air Interface for Fixed Broadband Wireless Access Systems Amendment : Medium Access Control Modifications and Additional Physical Layer Specifications for -11 GHz, IEEE, Standard IEEE Std a- 003, Part 16: Air interface for fixed broadband wireless access systems. Amendment for physical and medium access control layers for combined fixed and mobile operation in licensed bands," IEEE, Standard IEEE Std 80.16e/D1, October J. Armstrong, OFDM for optical communications, J. Lightwave Technol., vol. 7, no. 1, pp , Feb V. Syrjala and M. Valkama, Jitter mitigation in high-frequency band pass sampling OFDM radios, in Proc. WCC 01, pp K.. Manoj and G. Thiagarajan, The effect of sampling jitter in OFDM systems, in Proc. IEEE Int. Conf. Commun., vol. 3, pp , May Page 57

8 11. U. Onunkwo, Y. Li, and A. Swami, Effect of timing jitter on OFDM based UWB systems, IEEE J. Sel. Areas Commun., vol. 4, pp , L. Yang, P. Fitzpatrick, and J. Armstrong, The Effect of timing jitter on high-speed OFDM systems, in Proc. AusCTW 009, pp L. Sumanen, M. Waltari, and K. A. I. Halonen, A 10-bit 00-MS/s CMOS parallel pipeline A/D converter, IEEE J. Solid-State Circuits, vol. 36, pp , Ch Pavan kumar and Poornima Padaraju Jitter oise Power Reduction in OFDM by Oversampling, International Journal of Engineering Research and Applications (IJERA), Vol., o. 4, pp , Monika Tiwari, Kanwar Preet Kaur, Analysis on Lowering the Effect of Timing Jitter in OFDM System using Oversampling, International Journal of Engineering and Advanced Technology (IJEAT), Vol. 3, o. 5, Page 58