Performance Evaluation of SNR in Radio Over Fiber ... - IEEE Xplore

International Conference on Microelectronics, Communication and Renewable Energy (ICMiCR-2013)

Performance Evaluation of SNR in Radio Over Fiber Systems for Different Modulation Schemes in AWGN Channel and Fading Channels Asha R. S

Jayasree V.K

Department of Electronics Engineering Model Engineering College, Thrikkakara Ernakulam, Kerala, India [email protected]

Department of Electronics Engineering Model Engineering College, Thrikkakara Ernakulum, Kerala, India [email protected]

Abstract— This paper discusses the performance of radio-overfiber systems in different channels for distributing gigabit-persecond wireless signals at millimetre wave frequencies. Nonlinear distortions (NLD) in the channel reduce the dynamic range (distortion-free range for the input signal) of radio-overfibre (ROF) links. Performance comparison of RoF in different channels shows that the dynamic range is highest in AWGN channel (highest SNR) and lesser in Rayleigh channel than Rician channel. An Orthogonal Frequency Modulated(OFDM) signal is Mach Zehnder Modulated (MZM), the generated optical signal is amplified by an Erbium Doped Fiber Amplifier (EDFA), noises are filtered, transmitted to a Radio Access Point (RAP) through a single-mode optical fiber and reached at the end user passing AWGN,Rayleigh and Rician channels.

cellular” for the radio cells which are smaller than 10 m. So the multi-gigabit-per-second wireless networking at mmwaves needs a high-capacity feeder network to interconnect different radio access points. Radio-over-Fiber (RoF) technology can provide the required feeder network and it is best suited with the demands of small-cell wireless networks. The major advantage of a fiber-based Distributed Antenna System (DAS) is its ability to support multiple diverse wireless applications and services on the same infrastructure. But the performance requirements for RoF links employed for low-frequency wireless systems differing from those required for mm-wave systems [2]. This paper explores the performance comparison of RoF system that uses different modulation schemes employed in the distribution of multigigabit-per-second in different wireless signals. Here the technical challenges faced by RoF systems in different wireless channels are discussed and their performances are compared.

Keywords— Radio Over Fibre, Orthogonal Frequency Division Multiplexing, Mach Zehnder Modulator, Erbium Doped Fibre Amplifier, Radio Access Point

I.

INTRODUCTION

Voice and low speed data services commonly use the conventional wireless communication. But they have been adequately provided by the existing wireless systems having data speeds up to a few Mbps. By the advent of popular High-Definition (HD) Video and high-speed Internet, future wireless systems must offer data speeds exceeding Gbps. Because of limited frequency spectra at low frequencies, and the congestion caused by the large number of consumer products sharing the frequency spectra, it will be necessary to use higher carrier frequencies in the future, including mm (millimeter)-waves, for much faster wireless communication at multi-gigabit-per-second speeds [1].Higher frequency bands are available at mm-wave frequencies. Even though the mm-waves provide the required bandwidth for ultra-fast wireless communication, they make wireless networking technically more challenging. Few technical challenges are the high carrier frequencies, channel nonlinearities and the wide channel bandwidths. They have higher air-link loss (e.g. about 30 dB higher at 60 GHz than at low GHz range), and low device performance. The wide channel bandwidth means higher noise power and reduced SNR (Signal to noise ratio). These factors make wireless networking at mm-waves “Pico-

978-1-4673-5149-2/13/$31.00 ©2013 IEEE

II.

ROF FOR FIBER AIDED WIRELESS (FI-WI )SYSTEMS

The fiber-wireless architecture for cellular networks is shown in Fig. 1(downlink). This architecture can increase the frequency reuse and broadband access by its micro/Pico cells for cellular radio networks. The micro/Pico cells can use with Radio Access Points (RAP) as in Fig. 1 These low power RAPs can provide inexpensive wireless access instead of conventional base stations because it is low complexity and cost in large-scale deployment [3]. The bigger cells can easily be split into smaller cells by dispersing RAPs throughout. These RAPs are connected to the central base station via the ROF links. In the RF communication system, the baseband signal is modulated to a convenient carrier frequency. The modulation scheme and the carrier frequency are predetermined. The aim of the ROF link is to provide a transparent, low distortion communication for the radio signal [4]. Here consider fiber-based wireless system architecture as shown in Fig. 1, where the radio signal s (t) from a central node is transmitted through an analog fiber optic link. The fiber terminates at the radio access point (RAP) where the

International Conference on Microelectronics, Communication and Renewable Energy (ICMiCR-2013) optical signal is converted to electrical signal u(t), amplified and transmitted to the air(different channels). The RAP is located as an interface between the fiber and the radio links [5].

SYSTEM MODELING

III.

The OFDM signal s (t) is transmitted through an optical link as in Fig. 3. The electrical signal i.e. OFDM signal is used to modulate the intensity of light in the fiber channel. This (E/O) conversion can be possible in two ways: by directly modulating the intensity of the light source or using a constant-intensity source followed by an external modulator. Consider a single OFDM symbol ∑

/

∆

(1)

Where N is the number of subcarriers separated by Δ and .Where and are the magnitude and centered at phase of n subcarriers. Fig .1 Radio over Fiber System

Fig.2 shows the gains and losses that occur in the optical and wireless portions. The input RF power has loss at the electro-optic (E/O) conversion stage and has a linear loss as a function of fiber length and again a loss due to O/E conversion [3]. All these three losses (plus any optical connector/splitter/splicing loss) are referred to as Lop. The optical link noise nop is added at the optical receiver which defines OSNR. This optical noise and signal are amplified by the optical amplifier with gain Gop then, goes through the wireless channel (AWGN, Rayleigh and Rician channel), experiencing a loss of Lwl [6].

Fig .3 System model

The output power after the signal has travelled through a of fiber is length {

/c)}

(2)

Where D is a dispersion parameter normally expressed in units of ps/nm/km, λ0 is the wavelength of the optical source and c is the speed of light in free space. In this paper, only power gain Gfiber is assumed for the fiber without other fiber distortions and therefore at the input of the Photo detector the optical power becomes, P0, r(t) = GfiberP0, t(t)

(3)

Where P0,t(t) is the optical output of the Mach-Zehnder modulator. Neglecting all the noises in the link, the received signal of the photo detector (PD) driving a load of RL ohms is Fig. 2

Losses, amplifications, and added noise to the radio signal

For multipath channels, an additional margin is to be given for fading. More noises will be added at the interface of air and portable receiver front end. These noises are cumulatively referred to as nwl. The SNR is the ratio between the signal and appropriately weighted noise power from all the sources. Like other wireless communication network, transmission medium consist two important problems in Fiber aided Wireless communication system. These problems are: AWGN noise & Rayleigh and Rician Fading.

ν ′r (t) = RL(RPD GfiberP0,t(t) + N r(t))

(4)

where RPD is the responsivity of the PD expressed in amps/watt (A/W), Nr(t) is the noise in the received optical signal from PD. PD noise sources are thermal noise, shot (or quantum) noise, dark current and surface leakage current. Thus s (t) is transmitted to the analog RoF link with MachZehnder modulator and finally expanding in to Taylor series, the electrical output of the optical link is = (Π 2 / νπ ) + (5) = (t) + + (6)


Π

Π

2 νπ

2 νπ

3ι.

Π

7ι. … . .

1

2 νπ

5ι.

Π

2 νπ

2n

1 ι.

.

7

is the nonlinear distortion that is K2 = (2 ) Π /νπ and obtained by all of the terms in the Taylor expansion of order greater than unity. IV.

MODULATION SCHEMES

∑

A. OFDM modulated RoF System RoF systems can handle wireless signals with different characteristics for achieving multi-standard system operation. Therefore, RoF systems can support single-carrier wireless signals having multilevel modulation formats and wireless signals having multicarrier modulation schemes, such as OFDM. In these systems, the sub-carriers themselves employ multi-level signal modulation. Channel uniformity is very critical for single-carrier systems [7]. However the employed RoF links are as simple as possible to reduce cost, and should provide the needed performance. Let N denote the number of sub-carriers used for parallel 0 1 denote information transmission and let the kth complex modulated symbol in a block of N information symbols. The outputs of the N -point Inverse are the OFDM signal Fast Fourier Transform (IFFT) of samples in one symbol interval, or mathematically √

∑

(8)

B. DPSK modulated RoF System Differential phase shift keying (DPSK) is a common form of phase modulation conveys data by changing the phase of carrier wave. Differential Shift Keying is a modulation technique that converts information by using the phase difference between two neighboring symbols. At the transmitter, each symbol is modulated relative to the previous symbol. At the receiver, the current symbol is demodulated using the previous symbol as a reference. The previous symbol acts as an estimate of the channel. A no change condition makes the modulated signal to remain at the same 0 or 1 state of the previous symbol. In this paper, we choose 8DPSK scheme to analyze the SNR in different fading channels. V.

B. Rayleigh Fading Channel When no Line Of Sight (LOS) path lies between transmitter and receiver, only have indirect path than the resultant signal received at the receiver, then the received signal is the sum of all the reflected and scattered waves. In wireless telecommunications, the multipath creates constructive and destructive interference, and phase shifting of the signal. This causes Rayleigh fading. The standard statistical model of this gives a distribution known as the Rayleigh distribution [8]. Mathematically, the multipath Rayleigh fading wireless channels modeled by the channel impulse response (CIR)

CHANNEL MODEL

A. AWGN noise channel AWGN is a noise that affects the transmitted signal when it is passing through a channel. It has a uniform continuous frequency spectrum over a particular frequency band.

(9)

Where, Lp is the number of channel paths, an are the complex value and delay of path l, respectively. The paths are assumed to be statistically independent and having normalized average power. The channel is time variant with respect to the motion of the mobile terminal, but we will assume that the CIR is constant during one OFDM symbol. C. Rician Fading Channel A Rician model is obtained in a system with LOS propagation and scattering effect[9]. The model is specified by the Rician factor, denoted by k and defined as the ratio of the line of sight and the scatter power components. The pdf of a Rician random variable x is given by 2 1

, x≥0

(10)

Where k= /2 and D2 and 2 are the powers of the LOS and scattered components, respectively [11]. D. Wiener Filter for Additive Noise Reduction Consider a signal x (t) considered in a broadband additive noise n (m), and represented as y(t) = x(t) + n(t) (11) Consider that the signal and the noise are uncorrelated, and the autocorrelation matrix of the noisy signal is the sum of the autocorrelation matrix of the signal x(t) and the noise n(t).i.e., (12) Ryy = Rxx + Rnn Where Ryy, Rxx and Rnn are the autocorrelation matrices of the noisy signal, the noise-free signal and the noise respectively. Substituting equations (11) and (12) in the Wiener filter Equation ,we get w = (Rxx + Rnn )-1 rxy

(13)

Equation (11) is the optimal linear filter for the removal of additive noise [11]. Where rxy is the cross correlation vector of the noisy signal and the noise-free signal. The frequency response of the Wiener filter provides useful information about the operation of the Wiener filter. In the frequency domain, the noisy signal Y (f) is given by

International Conference on Microelectronics, Communication and Renewable Energy (ICMiCR-2013) Y (f )= X (f ) +N (f )

(14)

Where X (f) and N (f) are the signal and noise spectra respectively. For a signal having additive random noise, the frequency-domain Wiener filter can be represented as 15 and

w (f) =

(16)

Where and are the power spectra of the signal and noise. Where SNR is the signal-to-noise ratio. The SNR (f) is expressed in terms of the power-spectral ratio. For additive noise, the Wiener filter attenuates each frequency component in proportion to an estimate of the signal to noise ratio. The Wiener filter can be considered as one of the most fundamental noise reduction techniques. E. Signal to Noise Ratio (SNR) SNR is the ratio of the received signal power over the noise power in the frequency range of the operation. It is an important parameter of the Local Area Wireless Network (LAWN). BER is inversely propotional to SNR. High BER causes large amount of packet loss, high delay and low throughput. It is tedious to find an exact relation between the SNR and the BER in the multi channel environment. Signal to noise ratio (SNR) is used to find out the quality of a communication link and measured in Decibels and it is represented by equation (17).

Fig. 4

Fig. 5

Block Diagram of Remote Access Point

SNR = 10 log10 (Signal Power / Noise Power) dB. VI.

(17)

SIMULATION SETUP OF RADIO OVER FIBER SYSTEM USING OFDM

The simulation set-up of the RoF system is shown in Fig. 2.The signal is generated from a PRBS source and then modulated using QAM. The resulting signal is OFDM modulated using a MATLAB program, and then upconverting the OFDM signal to 25 GHz [10]. Two OFDM sidebands having center frequency of 25GHz, with a combined bandwidth of 2 GHz is formed. Both side bands are transmitted through the RoF system.Because the Subcarriers are transmitted independently, and are demodulated independently at the receiver, the total bit-rate of the 2 GHz-wide OFDM signal is double that of the original OFDM signal generate by the MATLAB program. The 2 GHz OFDM signal at 25 GHz is amplified and combined with a 35.5 GHz Local Oscillator signal generated by a signal generator, as shown in Fig. 2. The composite signal was used to drive a single-electrode Mach Zehnder Modulator (MZM) located at the head-end unit (HEU).The HEU system consisted of an optical transmitter of 193.1 THz [12]. The optical signal was transmitted into the MZM, and modulated by the resultant OFDM and local oscillator signals as shown in Fig.4. The MZM modulator is biased at the point of minimum transmission in order to suppress the optical carrier.

Block diagram of RoF system

Fig. 6

Bock diagram of Receiver unit


The modulated signal (optical) was amplified by an Erbium Doped Fiber Amplifier (EDFA) of core radius 2.2 micrometer is filtered by a rectangular filter of 10GHz, and transmitted to a Radio Access Point (RAP) connected by a single-mode optical fiber of 50 Kilometers in multiple channels like AWGN, Rayleigh channel and Rician channel using OFDM and DPSK modulation as shown in Fig. 5&6. Then the Weiner filter is used to reduce the channel noises in DPSK and OFDM modulation . VII.

RESULTS

By using the proposed method, the RoF system was implemented for multiuser modulation scheme (OFDM) and single user modulation scheme (DPSK) in AWGN and fading channels. The amplitude spectrum and power spectrum were simulated (Fig. 7- Fig. 12). Results shows that the SNR is very high in Rayleigh channel compare to Rician and AWGN channel.

Fig. 9 Amplitude spectrum of ROF in Rician channel

Fig. 10 power spectrum of ROF in AWGN channel

Fig.7 Amplitude spectrum of ROF in AWGN channel

In the single user DPSK modulation, the noises are very high compared to multiuser OFDM modulation (Table 1&2). By using Weiner filter, noises can be reduced and SNR can be improved for both DPSK and OFDM (Table 3&4). For noises, the Wiener filter attenuates each frequency component in proportion to an estimate of the signal to noise ratio.

Fig. 11 Power spectrum of ROF in Rayleigh channel

Fig. 12 Fig. 8

Amplitude spectrum of ROF in Rayleigh channel

Power spectrum of ROF in Rician channel

International Conference on Microelectronics, Communication and Renewable Energy (ICMiCR-2013) TABLE 1. COMPARISON OF RESULTS( OFDM MODULATION)

Total Power

AWGN 66.522

CHANNELS RAYLEIGH RICIAN 56.732 56.732

Signal Power

69.126

0.077

2.201

Noise Power

56.732

56.732

56.732

Parameters

1.218

SNR (dB) TABLE 2.

0.00135

0.039

COMPARISON OF RESULTS( DPSK MODULATION) CHANNELS

Parameters Total Power Signal Power Noise Power SNR(dB) TABLE 3.

AWGN 4.62KW (66.64dBm)

RAYLEIGH 471.25W (56.73dBm)

RICIAN 471.26W (56.73dBm)

4.146KW (66.176dBm) 471.26W (56.73dBm) 1.166

0.325E-24W (-100dBm) 471.26W (56.73dBm) -1.76

1.34E-24W (-100dBm) 471.26W (56.73dBm) -1.54

COMPARISON OF RESULTS( OFDM MODULATION USING WEINER FILTER)

Parameters Total Power(dBm)

AWGN 69.29

CHANNELS RAYLEIGH RICIAN 66.56 66.614

Signal Power(dBm)

69.044

66.088

66.144

Noise Power(dBm)

56.732

56.732

56.732

SNR (dB)

1.217

1.218

1.166

TABLE 4.

Param eters Total Power Signal Power Noise Power SNR (dB)

COMPARISON OF RESULTS( DPSK MODULATION USING WEINER FILTER) AWGN

CHANNELS RAYLEIGH RICIAN

69.3

66.578

66.608

69.044

66.103

66.135

56.732

56.73

56.73

1.217

1.165

1.1657

VIII CONCLUSION

From the simulation results, the SNR of a RoF communication system is an important figure of merit used to quantify the integrity of

data transmitted through the system. The SNR in differenent channels were compared by using DPSK and OFDM modulation schemes. It has been found that the SNR is maximum for AWGN and for Rayleigh channel it is lower than that of Rician Channel. And also, if weiner filter is used to reduce the noises, the SNR can be improved effectively in Radio Over Fiber Communication.

REFERENCES [1]

Xavier N. Fernando and Abu B. Sesay, “Adaptive Asymmetric Linearization of Radio Over Fiber Links for Wireless Access,” IEEE transactions on vehicular technology, vol.51,pp.6, November 2002 [2] Reza Abdolee, Razali Ngah, Vida Vakilian and Tharek A.Rahman,“Application of Radio-Over-Fiber (ROF) in mobile communication”, IEEE proceedings on Applied Electromagnetics, pp. 1 – 5, December 2007 [3] Muhammad Haider Raza, Kamran Zaidi, S.M.H Zaidi, “A Review of Full-Duplex WDM RoF Architectures”, Proc. of SPIE Vol. 6784, pp. 6784-6789,November 2007 [4] X. Fernando and A. Anpalagan, “On the Design of Optical Fiber based Wireless Access Systems,” in Proceedings of the IEEE International Conference on Communications, vol. 6, pp.35503555,June 2004 [5] Christina Lim, Ampalavanapillai Nirmalathas,Masuduzzaman Bakaul, Prasanna Gamage, Ka-Lun Lee,Yizhuo Yang, Dalma Novak and Rod Waterhouse, “Fiber-Wireless Networks and Subsystem Technologies”, Journal of lightwave technology, vol.28,pp.4,February 2010 [6] A. Sudhir Babu, Dr. K.V Sambasiva Rao, “Evaluation of BER for AWGN, Rayleigh and Rician Fading Channels under Various Modulation Schemes”, International Journal Of Computer Applications (0975 – 8887)vol. 26,pp.9, July 2011 [7] AHM Razibul Islam, Rishad Ahmed Shafik1, Md. Shahriar Rahman1, Ju Bin Song, “ On the Nonlinear Distortion Effects in an OFDM-RoF Link”,in Proceedings of 2nd International Conference on Emerging Technologies.pp.20-26,November 2006 [8] Mohammaed Slim Alouini and Andrea J. Goldsmith “Capacity of Rayleigh fading channels under different Adaptive Transmission and Diversity combining Techniques”, IEEE Transactions on Vehicular Technology,Vol. 48, No. 4, July 1999 [9] Nuzhat Tasneem Awon, Md. Mizanur Rahman, A.Z.M. Touhidul Islam, “Effect of AWGN & Fading (Raleigh & Rician) channels on BER performance of a WiMAX communication System,” (IJCSIS) International Journal of Computer Science and Information Security,Vol. 10, pp. 8, August 2012 [10] Fumiyaki Adachi, “Error Rate Analysis of Differentially Encoded and detected 16-APSK under Rician fading”,IEEE Transactions on VehicularTechnology, Vol. 45, No.1, February 1996 [11] Advanced Digital Signal Processing and Noise Reduction, Second Edition. Saeed V. Vaseghi Copyright © 2000 John Wiley & Sons Ltd. [12] Anthony Ng’oma and Mike Sauer, “Radio-over-Fiber Systems for Multi-Gbps Wireless Communication”, in Proceedings of SPIE-OSAIEEE Asia Communications and Photonics, SPIE vol. 7632,pp.1-10 200