Implementation and simulation of basic structure of the radio over fiber link

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.



Wireless communication has gone through enormous growth in the past ten years. Less than a percent of world population had access to cellular technology before early nineties, and by the start of this millennium approximately every one in a five people has a mobile phone. In the same period different countries across the globe have increase the mobile network technology over ninety percent and future forecast says that by the end of 2010 there will be more than 1700 million mobiles users across the world. [1][2]

Apart from cellular technology WLANs has also seen phenomenal growth during the past ten years. These WLAN hotspots can be used in public places such as airports, cafes, hotels and restaurant etc.





Bit-Rate (MAX)


IEEE 802.11

2.4 GHz

Frequency Hopping and

Direct Spread Spectrum

2 Mbps



2.4 GHz

Wideband Frequency


1.6 Mbps


IEEE 802.11b

2.4 GHz

Direct Sequence Spread


11 Mbps


IEEE 802.11a

5 GHz


54 Mbps



5 GHz

OFDM Connection-Oriented

54 Mbps


IEEE 802.11g

2.4 GHz

OFDM compatible with


54 Mbps

Table 1.1 Evolutions of WLAN Standards [3]

The rapid growth in wireless communication achieved more fame due the ease of installation as compared to the fixed network. The first generation (1G) mobile system were analogue, discovered in 1980s. The second generation (2G) known as global system for mobile communication (GSM) came on the scene in 1990s, which has been very successful and has achieved extreme success across the globe. GSM is currently the major mobile communication system which is used worldwide. [1]

The graph above shows the relationship between coverage and capacity of communication systems. By looking at the graph the cell size of WPAN is of few meters but there transmission rate may go upto 10 Mbps. While considering 2G and 3G systems, there cell sizes may vary upto several kilometres but that are limited to less than 2Mbps. WiMAX technology can provide high bit rate mobile services using frequency span between 2 - 11 GHz. [6]



2 GHz

UMTS/ 3G Systems

2.4 GHz

IEEE 802.11 b/g WLAN

5 GHz

IEEE 802.11 a WLAN

2-11 GHz

IEEE 802.16 WiMAX

17/19 GHz

Indoor Wireless (radio) LANs

28 GHz

Fixed Wireless Access - Local point to multi point (LMD)

38 GHz

Fixed Wireless Access - Picocellular

58 GHz

Indoor Wireless LANs

57-64 GHz

IEEE 802.15 WPAN

10-66 GHz

IEEE 802.16 WiMAX

Table 1.2 Frequencies for Wireless Communication Systems [2]-[5]


Wireless networks can be categorized into different groups depending on the area they are applied to. As a result high numbers of standards have been making to public for the development of new techniques in order to increase the spectrum efficiency and perfect utilization of spectrum, which is scarce natural resource.

Wireless networks can be divided into three classes;

1.2.1 Wireless Private Area Network (WPAN)

Devices of such networks can communicate in the range of tens of metres. Infrared (IR) and Bluetooth are the two implementation of this principle.

1.2.2 Wireless Local Area Network (WLAN)

It is computer network that connects devices which are distributed over a local area (e.g office, house, mall, and airport). IEEE 802.11 which is commonly known as Wi-Fi, is an example of WLAN.

1.2.3 Wireless Metropolitan Area Network (WMAN)

Such a network covers a geographic area such as city or village. IEEE 802.16 which is commonly known as WiMAX, is an example of WMAN.

Depending upon the application, there are licensed and unlicensed frequency bands in which wireless systems can operate.


Now we will discuss wireless standards along with the overview of their applications:

1.3.1 Bluetooth - WPAN

Bluetooth is a radio standard, which operates in the unlicensed Industrial Scientific and Medical (ISM) band at 2.4 - 2.485 GHz. Frequency Hopping Spread Spectrum (FHSS) is used in order to minimize interference and fading. In order to make the transceiver architecture as simple as possible, binary modulation is used. The bit rate is up to 3 Mb/s. The benefits of Bluetooth include low power consumption and low cost, therefore they are used in devices such as laptops, mobile phones and PDAs. [7]

Power Class

Maximum Output Power

Minimum Output Power










Table 1.3 Bluetooth classes and power levels [7]

1.3.2 Wi - Fi - WLAN

The Wi-Fi alliance, the Institute of Electrical and Electronics Engineers (IEEE) and the European telecommunications standard Institute (ETSI) are the three organizations which influenced the standardization of WLAN. The IEEE WLAN standard is referred as 802.11. At the moment, the most used techniques are defined by the IEEE 802.11a, b and g standards. [8]


Release date

Operating frequency

Maximum Data Rate



5.15 - 5.35 GHz

5.725 - 5.825 GHz

54 Mbps



2.4 - 2.5 GHz

11 Mbps



2.4 - 2.5 GHz

54 Mbps

Table 1.4 IEEE 802.11a, b and g standards [8]

1.3.3 WiMAX - WMAN

WiMAX is an abbreviation for Worldwide Interoperability for Microwave Access. The WiMAX Forum is a non profit association. The aim and objective of the WiMAX technology is to provide fixed, portable or mobile connectivity to the users even if they located up to 6 miles away from base station and it is not necessary to be in line of sight. WiMAX can operate on any frequency below 66 GHz, as operating frequency may change for different countries depending on local regulation. It is possible replacement for mobile/cellular technologies such as GSM and CDMA. It has been considered to be the wireless backhaul technology for 2G, 3G and 4G networks. The limitations associated with WiMAX is that it can either provide high data rates or it can transmit data over longer distances but not both simultaneously. [9]

1.3.4 Distributed Antenna Systems and Radio Over Fiber

Distributed Antennas Systems (DAS) are used for several applications in the mobiles and wireless communications. It can be installing over indoor and outdoor sites. DAS can be implemented on those areas where there is lack of signals such as tunnels, underground stations etc. in order to extend the coverage of mobile network.

Radio over fibre consists of remote unit and central unit. Remote unit is kept very simple since it only consists of devices for reception of radio frequency signals and optoelectronic conversion. All expensive and complex equipments are located at central unit and functions such as modulation and up/down conversion etc. are done. This resulted in increase in efficiency and maintenance cost because as compared to central units, remote units are numerically high in numbers and often remote units are located in sites that are not easy to get in touch with. [10]



The aim of the dissertation is to implement and simulate the basic structure of the radio over fiber link using OFDM transceiver with the help of MATLAB/SIMULINK. The MATLAB version 7.8.0 (R2009a) is used for model implementation. Basically two models are designed: model number 1 consists of OFDM transceiver linked with a gain which represents the length of the fiber channel. Actually it is based on the theoretical fact that fiber has 0.2db loss per kilometre. For example 25km length fiber will be represented as - 5 dB(-ve sign to show loss). Later on simulations are carried out by varying the length of fiber and results are deduced. Model 2 consist of OFDM transceiver as well but linked with laser diode model, fiber channel model and photodiode model as these are the fundamental components of RoF link. Some additional parameters of measuring the transmitted and received power and bit error rate calculation are also introduced to enhance the diversity of the project.


The dissertation consists of six chapters:

Chapter 1 is the introduction chapter in which wireless communication systems and wireless applications have been discussed briefly.

Chapter 2 consist of the theory of radio over fiber which includes the need of RoF system, what RoF technology is, advantages and disadvantages of RoF system and applications of RoF technology.

Chapter 3 purely consist of theory related to OFDM technology. Sub topics include in this chapter are principles of OFDM, history, advantages and disadvantages and applications of OFDM. Fourier transform is also discussed in this particular chapter.

Chapter 4 consist of methodology of the dissertation. It consists of the models implemented using MATLAB/SIMULINK and the brief study of the essential blocks used in the models.

Chapter 5 is the chapter of simulations and results.

Chapter 6 includes the conclusion and future work regarding radio over fiber and OFDM.



Radio-over-fiber (RoF) is a communication technology for delivering broadband applications to wireless users such as satellite communications, mobile-radio communications, broadband access radio, multipoint video distribution and broadband mobile services. RoF technologies make use of optical and radio communication media for providing above mentioned broadband services. The optical part is used to transmit microwave signals between a central radio base station and a remote radio antenna and on the other hand radio part provides coverage to wireless users. In RoF system radio frequency (RF) signal is transmitted through an optical network in an easier way by directly modulating the intensity of the light source with the RF signal to be transmitted and on the receiving end direct detection of the signal at photo detector. The modulating of the laser-diode light intensity with electrical signals at multiple frequencies causes a number of problems such as relative intensity, noise chirp and inter modulation distortion. The main sources of non-linearity in a system are the laser-diode light source, the optical fiber and the photo detector. [27]


For the future prerequisite multimedia services and broadband over wireless media, some distinctive characteristics are needed such as cell size reduction in order to accommodate more users and to operate in the millimetre wave (mm-wave) frequency bands to overcome spectral clogging. Such a system would demands a large number of base stations to cover large geographical coverage area and base station should be cost effective as well, then only such a system would be successful in market. In such a competitive market, this necessity has led to the evolution of system architecture where microwave functions such as signal processing, signal routing, handover, modulation, protocols setting and frequency allocation etc. are performed at central control station (CS) rather than at remote station or base station (BS). This type of centralized arrangement allows complex, sensitive and expensive equipments to be positioned in safer environment and shared among several BSs or RSs (Remote Stations). Now the question arises how to link the central station (CS) with BS. In such type of radio network, the use of optical fiber is the most suitable choice for the linking of CS with BSs, as fiber is cheaper in cost, has low loss, immune to Electromagnetic Inter Modulation (EMI) and provides wider bandwidth. By keeping the BSs as simple as possible and by sharing the resources provided by CS among several BSs, can effectively minimizes the cost of entire network and thus maintenance cost. Modulation of RF sub carriers onto an optical carrier over an fiber is known as Radio over Fiber (RoF) technology. Typically RoF network consist of central CS, where functions like switching, routing, medium access control (MAC) and frequency management takes place whereas at BSs functions like optical to electrical and vice versa are performed. [32]


Radio over fiber system consists of a Radio Base Station (RBS) and Radio Access Point (RAP) which are connected by an optical fiber link. Optical fiber link is used to distribute RF signals from a RBS to RAP. RAP only contains optoelectronic conversion devices and amplifiers. In GSM technology RBS could be referred as Mobile Switching Centre (MSC) and RAP as Base Station (BS). The frequency used by the RoF systems usually lies under GHz region depending on the nature of application. Basically RoF systems were used to transmit microwave signals and to achieve mobility functions in RBS. Therefore modulated microwave signals had to be available at the input end of the system, which are then delivered to the RAP as optical signals. Signals at RAP are re-generated and radiated by antennas. Due to the advancement of technology, RoF systems are designed to perform added radio system functionalities other then transportation and mobility functions. The functions include are data modulation, signal processing and frequency conversion (up and down). The electrical signal at the input of the multifunctional RoF system may be baseband data, modulate IF or actual modulated RF signal for distribution. The modulated optical signal is carried over the optical fiber link to the remote station. At the receiving end, demodulation of the signal is carried out by the photo detector and the optical signal is converted back to electrical signal. [12] [13]


2.4.1 Low Attenuation

It is observed that high frequency signals when transmitted in free space or through transmission lines are expensive and sometimes due to different reasons challenging as well. In free space, losses are directly proportional to frequency due to absorption and reflection. Increase in frequency also gives rise in impedance when signal is delivered through transmission line. Therefore in order to overcome these issues, expensive signal regenerating equipment is required to distribute radio signal electrically over long distances. The cheaper solution is to use optical fibers which offer lower losses. Single Mode Fiber (SMF) made from glass (silica) has attenuation losses below 0.2dB/km and 0.5dB/km in the 1.5um and 1.3um windows respectively. [11]

2.4.2 Larger Potential Bandwidth

Larger bandwidth is being offered by optical fibers. Larger bandwidth provides high capacity for transmitting high frequency signals and also enables high speed signal processing which is difficult to achieve in electronics systems. Basically there are three main transmission windows, namely 850nm, 1310nm, and 1550nm wavelengths, which offer low attenuation. Anyhow optical system has to combine with electronic system in order to perform different tasks. But bandwidth mismatch of the systems create problem which is known as 'electronic bottleneck'. The solution to this problem is the use of effective multiplexing techniques such as OFDM, DWDM and SCM. [11]

2.4.3 Easy Installation And Maintenance

The plus point of RoF system is the Switching Centre (SC), which are less in numerical quantity because one SC is shared by several Remote stations (RSs), which are equipped with all the expensive and complex equipments and RSs are kept simpler which includes only photo detector, amplifier and an antenna, thus reducing system installation and maintenance cost. [11]

2.4.4 Reduced Power Consumption

As discussed earlier centralized SCs are equipped with complex equipment and RSs are kept simpler with less equipments thus resulting in reduced power consumption. Thus RSs can be operated in passive mode. [11]

2.4.5 Immune To Interference And Crosstalk

As we know that optical fibers form a dielectric waveguide therefore there are no concepts as electromagnetic interference (EMI), radio frequency interference (RFI), or switching transients giving electromagnetic pulses (EMP). In fact it doesn't require shielding form EMI. Hence optical signal can be transmitted through electrically noisy environment unaffectedly. The optical fiber can be used underground or overhead as it is not disposed to lightening strike. [11]

2.4.6 Signal Security

In RoF system, optical signals are transmitted in the form of light, which doesn't radiate drastically, thus providing high degree of signal security. Therefore it is widely used in military, banking and general data transmission applications. [11]


RoF systems can be called as analog communication system. Therefore signal impairments such as noise and distortion are worth considering in RoF. These impairments tend to limit Noise Figure (NF) and Dynamic Range (DR) of the RoF links. Chromatic dispersion may limit fiber link length when considering SMFs RoF. Modal dispersion can limit the available link bandwidth and distance when considering MMFs RoF system. Relative Intensity Noise(RIN), lasers phase noise, photodiode's shot noise, amplifier's thermal noise and fibre's dispersion are few examples of noise sources in analog optical fibre links.[10]


Listed below are the few applications regarding RoF:

2.6.1 Mobile Communication Network

A mobile network is a useful application of RoF technology. In the past decade the numbers of mobile subscribers coupled with the increasing demand of broadband service have been keeping massive pressure on the mobile service provider to provide vast capacity to the end user. [11]

2.6.2 Video Distribution Systems (VDS)

VDS is one of the major applications of RoF systems. In this case the Multipoint Video Distribution Service (MVDS) is used for mobile terrestrial transmission. In MVDS the transmitter serves the coverage area based on tall building. Gunn oscillators and heat pipes are used for frequency stabilization while a fiber link can be used for feeding a TWT or solid state amplifiers. This system provides reduction in weight and wind loading of transmitter. [11]

2.6.3 Cellular Broadband Services

Due to the very high bit rates of nearly 155 Mbps, carrier frequency is pushed into mm-waves. For this purpose frequency band in 66 GHz frequency band have been allocated. The 62-66 GHz band is used for downlink while 65-66 GHz frequency band can be used for uplink transmission. [11]

2.6.4 Vehicle Control And Communication

For vehicle communication and system the frequency band between 63 - 64 GHz and 76-77 GHz frequency band has been allocated. They are used to provide continuous mobile communication coverage in major areas for the purpose of intelligent transport systems which includes road to vehicle communication (RVC) and inter vehicle communication (IVC). These can be made simple and cost effective by feeding them through RoF system. [11]



Orthogonal frequency division multiplexing is a multi carrier technique which divides the bandwidth into several carriers. Each carrier is modulated by a low rate data stream. OFDM has the ability to use the spectrum efficiently by spacing the channels close to each other. Closeness of the channels can result in the interference therefore to prevent interference all carriers are orthogonal to each other which means all carriers are independent to each other. [14]

In FDMA a single channel is allocated to each user to transmit information. The bandwidth of each channel is about 10 kHz-30 kHz for voice communications. In order to prevent channels from interfering with one another, the allocated bandwidth is made wider than the minimum amount required. This extra bandwidth or spacing between channels is wasting about 50% of the total spectrum. As the channel bandwidth becomes narrower the problem becomes worst. [14]

In TDMA multiple users access the same channel or utilized the full bandwidth in different time slots. Many low data rates users can be combined to transmit in a single channel thus bandwidth or spectrum can be used efficiently. There are two problems associated with TDMA. Firstly the symbol rate of each channel is high resulting in multipath delay spread. Secondly at the start time of each user to use bandwidth for data transmission, a change over time has to be allocated in order to prevent from propagation delay variations and synchronization errors. This change over time is a loss, limiting the number of users that can be accommodated efficiently in each channel. [14]

OFDM is solution to both the problems occurring in FDMA and TDMA. Actually OFDM splits the available bandwidth into many narrow sub channels. As the carriers are orthogonal to each other which means they are purely independent of each other therefore they can be spaced very close to each other. Any time full utilization of bandwidth is possible in OFDM, therefore there is no need for users to be time multiplex and no more switching of the users for bandwidth. Users can send and receive data at any time unlike TDMA. [14]


The concept of OFDM was first developed in 1950's. A US copyright was issued in January 1970. The evolution of OFDM took place in order to use the available bandwidth or spectrum more efficiently. [15][16]

OFDM was first implemented in military communications just like CDMA. KINIPLEX [17] and ANDEFT [18] are two examples of OFDM application in high frequency military system. AN/GSC-10(KATHRYN) variable rate data modem was the early application of OFDM which was built for high frequency radio.

In 1980's, OFDM had been studied for high speed modems, digital mobile communications and high density recording. OFDM techniques for multiplexed QAM using DFT was discover by Hirosaki [19]. He has also designed 19.2 kbps voice band data modem which uses QAM modulation.

In 1990's, OFDM has been exploited for data communication over mobile radio FM channels, high bit rate digital subscribers lines(HDSL), very high speed digital subscriber lines(VHDSL), digital audio broadcasting(DAB), digital television, HDTV terrestrial broadcasting and asymmetric digital subscriber lines(ADSL).[14]

OFDM has been considered more towards mobile communication due to its robustness to multipath propagation. Recently OFDM has been put into practice in audio broadcasting applications such as DAB and DVB. And it has been successfully implemented in wireless LAN applications as well. [14]


The application of OFDM was not very practical in 1960's. Quite a few numbers of oscillators were needed to generate the carrier frequencies for sub channel transmission. At that time it was a bit difficult to make it practical, that is why OFDM scheme was said to be impracticable.

Complexity of the OFDM scheme was eliminated with the evolution of Fourier Transform where harmonically related frequencies are generated by Fourier and Inverse Fourier Transforms used to implement OFDM systems. Fourier Transform can be used in linear systems analysis, antenna studies, optics, random process modelling, probability theory, quantum physics and boundary-value problems.


In the last 10 years, the usage of OFDM has increased to enormous extent. It has been proposed for radio broadcasting such as EUREKA 147 standard and Digital Radio Mondiale (DRM). Some of the useful parameters are listed below: [20]

· Data rate: 6Mbps to 48 Mbps

· Modulation: BPSK, QPSK, 16-QAM and 64 QAM

· Coding: Convolutional concatenated with Reed Solomon

· FFT size: 64 with 52 sub-carriers uses, 48 for data and 4 for pilots

· Sub carrier Frequency Spacing: 200 MHz divided by 64 carrier or 0.3125 MHz

· FFT Period / Spacing Period: 3.2usec

· Guard Duration: One quarter of symbol time, 0.8usec

· Symbol time: 4usec


· Overlapping is used for efficient use of spectrum.

· OFDM systems are more often reluctant to freq selective fading by dividing the channel into narrowband sub channels.

· Cyclic prefix is used to discard ISI and IFI.

· The symbols lost due to selective fading can easily be recovered by using channel coding and interleaving.

· The use of single carrier systems makes channel equalization simpler by using adaptive equalization techniques.

· With reasonable complexity max likelihood decoding is possible.

· FFT techniques allow OFDM to be computationally efficient to the functions of modulation and demodulation.

· It can also be used for DAB systems and partial algorithms can be used for program selection.

· A channel estimator can easily be discarded with the use of differential modulation.

· As compared to single carrier systems OFDM is less sensitive to sample timing offset.

· OFDM gives extra protection concerning parasitic noise and co channel interference.

· In severe multipath orthogonality is preserved.

· OFDM is used in high speed applications and dynamic packet access is also supported.

· Transmitting and receiving diversity are supported. On the other hand OFDM also supports adaptive antenna arrays, space time coding and power allocation.


· The OFDM signal has a noise like amplitude with a very large dynamic range, therefore it requires RF power amplifiers with a high peak to average power ratio.

· It is more sensitive to carrier frequency offset and drift than single carrier systems.


3.7.1 Peak To Average Ratio

PAR is an important OFDM parameter which is defined as the ratio of peak instantaneous value to average time. It can also determine parameters such as current, voltage, phase and power of the signal. Since OFDM is a summation of several carrier signals therefore results in high PAR. The RF power needs to be increased to overcome the problem of efficiency in PAR. In order to increase the radio frequency power an amplifier is needed which can increase the cost of the system as it is expensive equipment.

In order to solve the problems created by PAR, different encoding schemes should be used before the modulation. Also the improvement in the amplification stage of transmitter is needed such as post processing the time domain signal to reduce the peak to mean signal ratio. [21][22]

3.7.2 Synchronization

The performance of OFDM link can be optimized by using two kinds of synchronizations between transmitter and the receiver.

· Timing Synchronization: The timing offset of the symbol is not need to be determined and then the optimal timing instants.

· Frequency Synchronization: The carrier frequency of the received signal must be aligned at the receiving end.

Timing sync can easily be achieved because the degree of sync error in OFDM structure is more severe. The sync techniques can be achieved by using known pilot tones that are embedded in OFDM signal or by using guard interval. [21][22]

3.7.3 Co-Channel Interference

In mobile communications co channel interference can be overcome by combining techniques related to adaptive antenna systems. Receiver antenna beam can be focused by beam steering while co channel interferer's are attenuated. This is useful as OFDM is sensitive to co- channel interference. [21][22]


· High frequency modems used for military

· Voice band modems




· Terrestrial Digital Video Broadcasting (DVB-T)

· Power line communication systems


· Cable modems

· Wavelength Division Multiplexing



This chapter includes the in depth study of the models built on MATLAB/SIMULINK. MATLAB version 7.8.0 (R2009a) is used for the modelling. Basically two models are designed which will be discussed in detail later. Each model consists of OFDM transceiver which is designed by using essential blocks which are already included in the MATLAB/SIMULINK library browser. One of the two model consist of optical components as well such as laser diode, fiber channel and the photo diode as these are the fundamental components of radio over fiber link. In the following section we will discuss briefly about the blocks used in the model.


The models can be divided into two sections as following

· OFDM Transceiver Design

· Optical Link Design

4.2.1 OFDM Transceiver Design

OFDM Transceiver as its name consists of both transmitter and receiver. Now we will briefly discuss the simulation blocks used in the model and set parameters are discussed in appendix. Bernoulli Random Binary Generator

The block generates random binary numbers by using Bernoulli distribution. The output signal of a generator can be a frame based matrix, sample based one dimensional array or a sample based row or column vector [23]. The pictorial representation is as under. Convolutional Encoder

The function of this block is to encode a series of binary input vectors in order to produce a series of binary output vectors. It can process multiple symbols at a time. The input can be a sample based vector or frame based column vector. [23] Matrix Interleaver

The function of this block is to perform interleaving by filling a matrix with the input symbols row by row and then matrix contents are send to the output port column by column. The block can accept the data type's int8, uint8, int16, uint16, int32, uint32, Boolean, single, double, and fixed-point. The output signal data type is same as that of input signal. [23] Rectangular QAM modulator Baseband

The modulator block modulates the signal using M-ary quadrature amplitude modulation with a constellation on a rectangular lattice. The output signal is a baseband version of the modulated signal. [23] Normalize

The function of this block is to normalize a vector. It accepts an input vector of any size and output represents the unit vector parallel to it. [23] OFDM Transmitter

The figure above is the pictorial representation of the OFDM trnasmitter and the figure below is the internal schematic of the OFDM transmitter which includes PN Sequence Generator, unipolar to Bipolar Convertor, Multiport Selector, Matrix Concatenation, Zero Pad, IFFT and add cyclic prefix block. The function block parameters of individual blocks are all given in the appendix. TX Spectrum Scope

TX Spectrum scope is designed to study the waveform of the transmitted signal after simulation. Above figure is the pictorial representation of the TX spectrum scope and the figure under is the internal schematic of the spectrum scope. The display following the spectrum scope in the model is used to show the transmitted power in decibels. OFDM Receiver

The received signal from the AWGN channel is fed to OFDM receiver which consists of remove cyclic prefix, FFT, Frame Conversion, remove zero padding and remove pilots. Basically most of the blocks are the inverse of the transmitter blocks. The figure below is the internal schematic of the OFDM receiver. Denormalize

The basic function of this block is to denormalize the filter coefficient and gain changes caused by normalization. [23] Rectangular QAM Demodulator Baseband

The demodulator block as its name demodulates the modulated signal with a constellation on a rectangular lattice. The impedance of 1 ohm is considered for all values of power. [23] Matrix Deinterleaver

The function of this block is the inverse of interleaving performing deinterleaving where input symbols are filled column by column and then are sends to the output port row by row. Input vector length must be number of rows times number of columns. It can allow the data type's int8, uint8, int16, uint16, int32, uint32, boolean, single, double, and fixed-point. The data type of output signal will be the same as that of the input signal. [23] Unipolar to Bipolar

The basic function of this convertor is to convert the unipolar signal to a bipolar output signal. [23] Viterbi Decoder

Decoding is inverse of encoding. The function of decoder block is to decode the input symbols to produce binary output symbols. Several symbols can be processed in order to achieve faster performance. [23] Error Rate Calculation

Error rate calculation block performs comparison of input data from a transmitter with input data from a receiver. It calculates the error rate by dividing the total number of unequal pairs of data elements by the total number of input data elements from one source [23]. It also displays the number of errors and number of bits transmitted. RX Spectrum Scope

RX Spectrum Scope is placed just after the OFDM receiver in order to study the spectrum scope of the received waveform. Following the spectrum is the display which shows the received power in decibels. The figure below is the internal schematic of the spectrum scope and the circuitry to calculate the power.

4.2.2 Optical Link Design

The main components of the optical link consist of laser diode, fiber channel and the photodiode. With the help of MATLAB/SIMULINK, these optical components are designed whose brief description is as under. Laser Diode

Laser shows a non-linear behaviour with the memory which known as weak-linearity. The diode input/output characteristic can be modelled by using Volterra series. Its input/output relationship can be modelled by a Volterra series of order 3 when the laser diode is driven its threshold current. When the essential parts of the Volterra series are taken as Dirac delta functions, the system may be modelled without memory. Power series of order 3 can be used to model the non-linear behaviour in order to simplify the analysis as such simpler models can be readily be used in the study of wideband systems such as OFDM.[10][28][29]

For an ideal linear characteristic, the laser-diode input output relationship is expressed as (1). [2]

Popt(t) = r(I(t) - Ith) ..............................eq(1)

Where I(t) is the input current of the microwave signal, I(th) is the diode threshold current, r is the P-I slope and Popt is the optical power of the laser diode.

For the non linear case, Popt is given by following

Popt(t) = a + b(I(t)-Ith) + c(I(t)-Ith)2 + d(I(t)-Ith)3

Where a,b,c and d are constants.

For the bias current of 25.5mA and a threshold current I(th) of 19.5mA, then the expression for input current will be stated as

I(t) - I = Bcos(4π10e9t) + 0.006 Amps ...........................eq(3)

Where B is the amplitude and the carrier frequency is 2 GHz. The instantaneous output electrical power is given by

Pe,out(t) = koPopt2(t) ....................................eq(4)

Where Ko represents the cascaded gains and losses within the system

With the help of curve fitting techniques, we can fit a polynomial of degree 3 as following

Py = 2.3*105Px3 - 670Px2 + 0.95Px + 1.4*10-6 ....................eq(5)

Where Py and Px are the average input and output powers respectively

If the overall amplification in the electrical front end is assumed to be 19dB and Px = 25B2, then the constant in equation (2) can be given as

[a b c d] = [ -0.0045 0.32 147.05 -12033] .........................eq(6)

By substituting these values for a, b, c and d in eq(2), we get

P(opt) = - 0.0045 + 0.32(I(t)-I(th))+ 147.05(I(t)-I(th))2- 12033(I(t)-I(th))3........eq(7)

So equation(7) is the final expression for the relationship between input current and output optical power of the laser diode which is only valid for the power range over which measurements were taken that is -20 to 0 dBm. Therefore with the help of eq(7), the laser diode modelling is done in MATLAB/SIMULINK, whose schematic diagram is as under. Fiber Channel

The schematic shown in the figure below is the linear model for Single Mode Fiber(SMF). Due to the time restriction, non linearities of fiber such as dispersion, self phase modulation, material non-linear refractive index etc are not considered. More detail of fiber model can be read from the reference number [30]. Photodiode

The purpose of the photodiode is to capture the transmitted signal which is in the form of light. The optical signal might be weak and thus needs amplification. Therefore following the photodiode circuitry, there is an amplifier, as shown in the figure below is placed to make sure that best possible power signal to noise ratio (SNR) is obtained.

S / N = Isig2 / Inoise2 ...............................eq(1)

Where Isig is the photo current and Inoise2 is the mean squared noise contributions from the photo detector.

As photodiode current needs to be amplified in order for the correct recovery of the data, therefore a gain related to an amplifier is placed with a noise figure (NF) of 0.5dB. This amplifier has a 33dB gain which translates to a power gain of 44.67. Therefore the following schematic has been developed by the addition of the photodiode noise to the photocurrent, the amplifier gain and amplifier noise.[31]

Gaussian Noise Generator is placed in the photodiode model in order to generate discrete time white Gaussian noise. Gaussian Noise generator uses a random number to add noise to the input signal which is basically initializes by the initial seed parameter. As recommended by the Matlab help that initial seed should be a prime number greater than 30, therefore we have set to 41.The other block parameter consists of mean value and variance. Mean value and variance can be either scalar or vectors. Couple of unit delays are placed in the model which delays its input by the specified sample period. It accepts continuous signals which can be real or complex [23]. More detail on photodiode model can be read from reference number [30].



The chapter comprises of the models designed by the help of MATLAB/SIMULINK. Basically two models has been designed which will be discussed in detail later in the chapter.

5.2 MODEL1

The model 1 represents the simple digital transmission in which data is entered in binary form. The Bernoulli Generator transmits the randon stream of data which is encoded by Convolutional encoder. The signal is then modulated with the help rectangular quadrature amplitude modulator. When the data enters the OFDM block, the bits are sequenced with the help of pseudo noise generator. The orthogonal frequency modulation is carried out by IFFT. Now the data is transmitted in form of OFDM signal. After the transmission of data the signal enters the optical fiber channel (represents as dB gain block).

The dB gain block is in this particular model represents the fiber channel. The purpose of using dB gain block is related to the theory of 0.2dB/km signal attenuation in single mode fiber. As we know that when signal propagates in optical fiber it faces a loss of 0.2dB/km therefore by varying the gain of the block, we can easily represents the required length of optical fiber for example 1km of fiber is represented as 0.2dB. now the signal enters the AWGN where additive white guassian noise is added to show the factor of distortion in wireless channel.

After passing through the fiber channel and AWGN channel the signal enters the OFDM receiver where it undergoes fast fourier transform. The signal is now demodulated in the original form with the help of QAM demodulator. The demodulated signal is travelled through the viterbi decoder where data is decoded and redundancy bits are removed and the original baseband signal is recovered.

In the final stage the transmitted and received signals are compared to calculate the bit error rate, the number of bits transmitted and the number of errors. The transmitted and received power can also be shown after the transmission and reception of signal with the help of spectrum scope.

5.2.1 Simulation 1.1

In the first simulation optical fiber of 5 km is used to represents the communication of OFDM signal via optical fiber. As stated before the gain of -0.2dB is set to represents 1 km, therefore 5 km means -1dB. The simulation test is run and the result can be concluded with the help of graph shown below:

The graph shown in figure 5.2 represents the constellation diagram of 16-QAM modulation. The QAM modulation can be used both as analog and digital modulation scheme in which two digital data bit stream can be represented by changing the amplitude of the carrier wave. These data streams are 90 degree out of phase and are thus called quadrature components or carriers. The graph above is potrait of a perfect transmitted 16 QAM modulated signal. When the signal travels along the optical fiber and passes through the AWGN channel, it undergoes certain attenuation on the receiver side, the signal can be shown as a graph below. The constellation diagram below represents the unmodulated distorted received signal in which constellation dots are scattered which shows the loss in signal power.

Comparing the above two graphs, we can analyze that there is no such difference between the transmitted and received signals. The structure of constellation dots in the received signal graph shows that due to high SNR and small length of fiber, a small amount of signal has been distorted

The transmitted signal shows the typical OFDM signal graph when considering the frequency and power of the signal. In figure 5.5 we can see that there is a lot of variation in the received signal when it is decoded at the receiving side.

5.2.2 Simulation 1.2

Now the gain has set to be -5 dB to represents fiber length of 25 km.

Considering figure 5.9 we can analyze that there is vast difference in the received constellation diagram as compare to simulation 1. Since we are using 25 km length fiber therefore the losses has been increased which can be shown as scattering of constellation dots.

In figure 5.11 the variation in graph can be observe graph by measuring the power loss on the y-axis. In this particular graph the peaks are touching -25 dBm which is more than earlier deduced simulation results where peaks were at -20 dBm.

Comparing figure 5.13 and 5.14 we can conclude that there is a certain amount of variation in the baseband transmitted and received signal. Other than the central peak the graph shows extreme variations as the frequency keeps increasing. This is observed due to the significant increase in length of optical fiber. As length keeps increasing there will be a gradual increase in the signal which can be proved in the next simulation.

5.2.3 Simulation 1.3

Now the gain has set to be -10dB in order to represent fiber length of 50 km.

In figure 5.17 the peaks of the received signal graph are showing steep rises and sudden falls by which we can analyze the loss of signal power and the great amount of attenuation the signal has gone through. In this graph the -45 dBm range proves that there has been a drastic change in the signal behaviour as compared to the earlier result. This is due to the 50 km length of fiber used in this simulation.

In figure 5.20 which is a baseband received signal of third simulation. We can see that signal changes its behaviour on regular intervals but it almost same at the center frequency. The difference in transmitted and received signal can be prove initially when there is a change of signal shown as -3dB which eventually drops to -7dB in the received signal.

5.3 MODEL 2

Radio over fiber optical link components such as laser diode, fiber channel and photodiode are interconnected in the model with OFDM transceiver. The overall working will be same as that of model 1. There are quite few spectrum scopes that have been placed in order to study the behaviour of the signal in the form of waveform. The essential blocks used in the model are already discussed in previous chapter. Following figure is the pictorial representation of the model.

5.3.1 Simulation 2.1

Comparing figure 5.22 and 5.23 we can observe that high amount of signal variation when comparing basic model 1 and model 2 which comprises of laser diode and photodiode. Due to the presence of these two components the constellation dots are completely scattered and enormous amount of attenuation is observed in the signal strength. This is due to the internal noises and impedances that are present in the structure of laser diode and photodiode. Since laser diode is a non linear device and Gaussian noise generator block in the photodiode model add up to show a vast difference in signal strength.

Comparing result of figure 5.27 and figure 5.28 it can be observed that other than the signal peaks the signal is varied at regular interval. The kind of variation observed here is different to earlier cases as there is a huge amount of attenuation in the signal strength as compared to the earlier result. Also in the received signal graph the wavelength of the signal is increased as there is a decrease in signal frequency. We are not using different length of fibers as we can analyzed that due to the presence of laser diode and photodiode that the signal strength is so much weaken that the rest of the results can't be obtained easily and which can't be shown graphically.



Transmitted power(dB)

Received power(dB)

Number of bits transmitted

Number of errors

Bit Error Rate

Simulation 1.1







Simulation 1.2







Simulation 1.3







Simulation 2.1







Table 5.1 Key Parameters of the Simulations



The idea behind choosing the concept of RoF technology as my final year dissertation came in my mind considering the development of 3G mobile communication, IMT-2000 which can envision a global village where people could transmit and receive anytime, anything and anywhere. RoF systems have many advantages such as lower cost, lower power, enhanced micro cellular services, higher capacity and easier installation. Therefore this technology is suitably applicable in areas such as airports, shopping centres, underground tunnels and dead zones such as highways also.

The concept of introducing radio over fiber with OFDM is to demonstrate the use of discrete fourier transform to perform baseband modulation and demodulation. DFT increases the efficiency of modulation and use of guard space and raised cosine filtering encounters the problem of inter symbol interference to a great extend. The fundamental system of the OFDM system is to decompose the high data rate into lower data rate stream and then transmit them simultaneously.

We also introduce the concept of using laser diode and photdiode to give some diversity to the project. But one of the major drawbacks in RoF system is the laser diode non linearity which gives rise to inter modulation distortion. This can be easily proved if we go through the results and figures obtained in the simulation test of model 2 of the project.

In radio over fiber transmission technology the length of the optical fiber, the wavelength of the laser diode and RF signal frequency should be paid great attention. The responsibility of the optical fiber is to show chromatic dispersion while on the other hand robustness is adequate for outdoor applications. As compared to conventional network structure, the RoF network structure is more flexible and easy to expand. Nowadays still various problems are being faced by this particular technology but it can become a core technology for telecommunication in the 21st century which can introduce a new style in mobile and ITS communication.[10]


1) World Telecommunication Development Report 2002: Reinventing Telecom, March, 2002, available online:

2) B. J. Jeong, J. Chung, C-S. Hwang, J. S. Ryu, Y. Kim, K-H. Kim and Y. K. Kim, "Beyond 3G: Vision, Requirements, and Enabling Technologies", March 2003

3) The Industrial Wireless Book, "De-mystifying IEEE 802.11 for Industrial Wireless LANs", available online at, May 2005.

4) S. Ohmori, "The Future Generations of Mobile Communications Based on Broadband Access Technologies", December, 2000

5) D. Novak, "Fiber Optics in Wireless Applications"

6) IEEE 802.16 Working Group, "Developing the IEEE 802.16 Wireless MAN® Standard for Wireless Metropolitan Area Networks", available online at , May 2005.

7) Specifications of the Bluetooth System:

8) Roshan, P. and J. Leary, "802.11 Wireless LAN Fundamentals", Cisco Press, 2003

9) WiMAX Specifications. Available from:

10) Al-Raweshidy, H. And Komaki,S., "Radio over Fiber Technologies for Mobile Communications Networks", Universal personal communications series, Artech House, Boston, London, 2002

11) A. Ng'oma, document, "Deliverable D6.1:Design of a radio over fibre system for wireless LAN's".

12) Abdullah Saad Mohammed Al-Ahmadi, thesis," Front-End Design of Low Power Radio Access Point for Radio over Fiber Technology", Universiti Teknologi Malaysia, May 2007.

13) Xavier N. Fernando and Stephen Z. Pinter ,"Fiber-Wireless Solution for Broadband Multimedia Access", Ryerson University, Toronto.

14) Teddy Purnamirza, thesis, "The performance of OFDM in mobile radio channel", Universiti Teknologi Malaysia, APRIL 2005.

15) R.W. Chang, "Synthesis of band -limited orthogonal signals for multichannel data transmission system", December 1996.

16) Saltzberg, B.R., " Performance of an efficient parallel data transmission system,", December 1967.

17) R.G. Clabaugh and R.R. Mosier, " Kineplex, a bandwidth efficient binary transmission system", January 1958.

18) G.C. Porter, "Error distribution and diversity performance of a frequency differential PSK HF modem", IEEE Trans Commun Technol, August 1968

19) S. B Hasegawa, A Sabato and S. Hirosaki, " Advanced groupband data modem using orthogonally multiplexed QAM technique", IEEE Transaction Commun,, June 1986



22) R.V. Nee and R.Prasad, "OFDM for wireless multimedia communications", Universal Personnal Communications, Artech house publishers, Boston, London.

23) MATLAB HELP version 7.8.0 (R2009a)




27) Ali Akdagli and M. Emin Yuksel, "Application Of Differential Evolution Algorithm To The Modelling Of Laser Diode Nonlinearity In A Radio Over Fiber Network", Erciyes University, 38039, Kayseri, Turkey, November 2005.

28) J.R. Cavallaro and Y.Guo, "A Novel Adaptive Pre-Distortion Using LS Estimation of SSPA Nonlinearity in the Mobile OFDM Systems " , IEEE Symposium, May 2002

29) G. Baghersalimi, V. Postoyalko, T.O'Farrell, "Modelling Laser-Diode Non-linearity in a Radio-over-Fibre Link", School of Electronic and Electrical Engineering University of Leeds, Leeds, LS2 9JT

30) LN Binh and B. Laville, "Simulink Models For Advanced Optical Communications: Part IV- DQPSK Modulation Format", Department of Electrical and Computer Systems Engineering, Monash University, Clayton, Melbourne, Victoria 3168, Australia.

31) G. Keiser, "Optical Fiber Communication" , New York, McGraw Hill, 1991

32) Ankush Kumar, thesis, "Studies On Optical Components and Radio Over Fiber Systems", National Institute Of Technology, Rourkela, Orissa-769008, 2009