This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.
We are aware of that previously there was Advance Mobile Phone System (AMPS) which is analogue mobile system. Disadvantage of this system is that it does not use the frequency spectrum efficiently so companies like AT&T and Verizon did not support this service.
So there was a demand for the digital systems which gave rise to 2G digital systems like GSM and CDMA. Advantages of these systems are conversation between the users is digitally encrypted so privacy is achieved more over these systems make use of spectrum very efficiently. And one more feature provided by these systems is short message service.
But as time progress the demand for the spectrum goes on increasing so there was need to manage available spectrum efficiently so that users will get satisfactory data rates and quality so people started looking for the new ways for the next generation wireless systems (4G) so as to increase throughput, efficiency.
And one of the ways was to carry Radio over Fiber Link Design.
Radio over Fiber Link is a technique which is used to distribute wireless signals and one of the main applications is Distributed Antenna System (DAS).
DAS basically consists of Central Unit (CU), Base Station (BS) and Remote Antenna Units (RAU). There are RoF transmission links between Central Unit (CU) containing the radio base stations to a number of remote antenna units (RAUs) in order to provide excellent coverage and dedicated capacity for short range communication.
LTE (Long Term Evolution) is 4G system specified by3rd Generation Partnership Project (3GPP). According to specifications LTE will have peak downlink user rates of around 100 Mbps with a channel bandwidth of up to 20 MHz, modulation complexity up to 64-QAM, 2048 subcarrier OFDM and MIMO with up to 4*4 antennas.
So according to FUTON (an EU Framework 7 project) to achieve throughput up to 1Gbps channel bandwidth up to 100 MHz and modulation level up to 256-QAM. And also OFDMA with up to 2048 subcarriers and various MIMO configurations can be used.
This paper is focused on the performance of the RoF links in the light of above requirements. Also covers major issues affecting the performance and design of RoF links like Carrier Frequency Radio Channel Bandwidth, Number of Channels, OFDM, Modulation Complexity, and MIMO. It also shows that noise introduced by the RoF links does not have a significant impact on wireless range and finally compare the cost and performance of RoF Links for this application with other types of links.
RoF transmission is classified into two characteristics:
RF over fiber architecture: In this architecture RF signal (beyond 10 GHz) is used to impose on the light signal before it is carried over the optic link.
IF over fiber architecture: In this architecture RF signal is first down converted to the intermediate frequency and then is used to impose on the light signal
Advantages of RoF: Low attenuation, Low complexity, Low cost, Future proof
Radio over Fiber Link Design Issues:
Requirements of next wireless communications systems are very high so while designing Radio over fiber links we have consider some major factors that can affect the performance of RoF links.
Some of them are as follows- Carrier Frequency, Radio Channel Bandwidth, Number of Channels, OFDM, Modulation Complexity, and MIMO.
Carrier Frequency: This is the important factor for RoF links that do not consider the frequency translation that means they transmit data at radio frequency and do not convert that frequency to intermediate frequency. And moreover radio frequency on which we are transmitting decide the type of optoelectronic components that can be used. Because we can get lasers with upper modulation frequency limit of 3 GHz at low cost beyond that semiconductor diodes are no longer cheaper but their price increases. (Up to 20 GHz) Beyond 20 GHz lasers are not commercially available because of this we have to use external optical modulators but those are expensive.
It is predicted that next wireless systems are going to use bands below 6 GHz also higher frequencies can be used in some scenarios due to greater spectrum availability. But problem with higher frequencies is fiber dispersion. Also we have to use advance modulation and transmission schemes because of which the cost increases unnecessarily. So we have to find some solution for that and the best solution for this is to use intermediate frequency for transmission. This is called as frequency translation.
There are so many advantages of using IF transmission:
As IF frequency is not higher frequency we can use low cost components (Lasers). Cost effective.
We can use simple intensity modulation direct detection scheme and direct modulation of semiconductor diodes
We can use MIMO channels
Frequency Translation is easy to implement and cost effective but we have to send low frequency tone over RoF link for the perfect locking at the receiver side.
Radio Channel Bandwidth: By considering the requirements of users the estimated bandwidth for the next generation wireless system is approximately 100 MHz and there is direct relation between system bandwidth and the noise that means as bandwidth increases noise introduced in the system also increases. Because of noise receiver sensitivity decreases and so radio range. But RoF link should support the required CNR in presence of higher channel noise.
Number of Channels: There is limit on the maximum composite input power of the RoF links and this power is to be shared by all the channels. So if the number of channels increases per channel input power decreases. As the input power per channel decreases required CNR at the receiver also decreases. So receiver cannot demodulate received signal properly. In downlink direction we can increase the amplifier gain to increase the input power but maximum amplifier gain is also limited by some factors.
Maximum transmitted noise and spurious emissions requirements of the wireless system specification
Stability requirements in the remote unit, since high gain for uplink and downlink directions may lead to oscillations ----------------------------- (1)
Peak to average power ratio of OFDM signal is generally high and it's due to the relative phase in subcarriers that combine to give the composite signal. Naturally when there is constructive interference we will be getting maximum peak power and it is possible when all the subcarrier will have same phase. Now big problem is that how to accommodate the signals with high PAPR, simple solution is to reduce the input power by factor of PAPR and we can do it by introducing a back-off factor. Back-off factor reduces the nonlinear distortion but because of high back-off factor the output power and output CNR of the RoF link reduces.
We have to maintain certain minimum level of CNR at the output of the RoF link for the perfect demodulation of the signal. And modulation complexity and required CNR are directly proportional to each other that mean as modulation complexity increases minimum requirement of CNR also increases.
MIMO (Multiple Input Multiple Outputs):
Using MIMO we can transport many radio channels between the central unit and remote units at the same frequency.
Separate RoF links on separate optical fibers ( space division multiplexing,SDM)
Separate RoF links on the same fiber but on separate optical wavelengths ( Wavelength Division Multiplexing)
Same RoF link where each channel is frequency translated to the separate IF ( Subcarrier multiplexing)
We have to choose appropriate option depending upon many factors such as availability and cost of the fiber optic cable. And most importantly where we have to deploy this fiber optic network. Because in metropolitan areas it is very expensive to deploy optic network. SCM which is described above has ability to reduce the number of fibers in the optical network. Implementation of the SCM is simple and cost effective. It's cheaper than the WDM. In SCM we downscale the original RF frequency so we can use low cost optoelectronic components.
EXAMPLE RADIO OVER LINK DESIGN
Link Architecture: Considers the MIMO channels. Uses SCM scheme that minimizes the need of optical fibers and optical wavelengths. Low cost semiconductor lasers and photo diodes are used. External modulation is not preferred because of high cost. Frequency translation is relatively simple and cost efficient and we can do it using cost effective oscillators and mixer. But in this case we have to send a low frequency reference to the receiver for the perfect frequency locking.
Untitled.pngSimplified DBWS link design. S1 and S2 are the two sectors; Tx1, Tx2, Rx1 and Rx2 are the 2*2 MIMO channels.
Above figure shows the MIMO configuration with two transmission antennas and two receiving antennas and the RAU supports 2 sectors. These two things together means 4 radio channels per link direction.
Frequency plan for SCM transmission system.
Base station supports digital in phase and quadrature inputs and outputs.
Incoming digital I/Q signals are converted to analog form using D to A converter and then modulated onto individual subcarriers using IQ modulators. Then all the modulated signals are combined and composite signal that we get is used to modulate a laser diode for optical transmission to RAU. I f several RAUs are to be supported from single CU the WDM can be used. In RAU photodiode is used to covert a composite signal from optical domain to the electrical domain. And then splitter is used to split the composite signal into appropriate subcarriers. IF signals are the converted to RF signals for transmission over air interface.
Reverse process will take place.
There are some other channels are also there that should be transmitted between CU and RAU those are for control and monitoring purpose.
Link Measurements: The main things that are considered in the link measurements are gain, equivalent input noise, and maximum input power. And for calculating these parameters we can use a simple link consisting of laser, an optical attenuator and a photodiode. We can determine maximum input power by measuring error vector magnitude as a function of RF input power.
EVM and CNR are related to each other by formula: CNR= -(r+20 log [EVM/ (100%)]
Where r is the peak to average energy ratio of the QAM constellationâ€¦â€¦.. (1)
We cannot use built in vector signal generator and vector signal analyzer settings for the measurement of error vector magnitude (EVM) because we have to make it user friendly so that user can enter different combinations of QAM and OFDM subcarrier.
User defined setup for calculating the EVM is shown in the fig below:
User generates OFDM signals using Matlab/Simulink offline. And then this OFDM is downloaded to the Agilent VSG to produce real experimental signal at an intermediate frequency of 1.8 GHz. Then this signal is fed to the link consisting of laser, optical attenuator (10 dB) and a photodiode. The job of the Agilent VSA at the receiver is to create a data file from incoming signal and fed this to the matlab for the offline processing. We can do manual timing synchronization, OFDM demodulation; blind frequency offset (FO) correction and blind equalization in the offline OFDM signal processing.
Frequency Offset: Frequency offset occurs when there is no perfect frequency locking. That means there is difference between the frequencies generated by the local oscillator at transmitter side and the receiver side; they are not exactly the same.
Verification of the user defined OFDM technique was carried through the comparison of the experimental demodulation results for a user defined OFDM signalâ€¦â€¦â€¦.. (1)
Experimentally we can show that EVM fairly remains the same for the same drive input power for different QAM levels which proves that the statistical PAPR is independent of the QAM level.
Wireless Range Calculations:
Downlink Direction: In downlink direction the amplifier gain depends on the two factors output power of the RoF link and the required transmit power at the antenna connector.
Uplink Direction: For reducing the noise figure we have to increase amplifier gain but it has some limit depending upon the power of input signal (For example mobile station which is very close to base station).
So now we can calculate the signal power and CNR levels for both uplink and downlink directions, also we can calculate maximum path loss. And finally we can now calculate the wireless range using particular environment.
In this system the limiting factor in determining the wireless range is uplink direction because the transmission power of the mobile station is limited by the battery.
Maximum path loss can be calculated by subtracting receiver sensitivity from the transmitted power. And the receiver sensitivity is given by
S(rx) = -174+B+CNR+N*F(t)+L(impl)
S(rx): Receiver Sensitivity
-174: Thermal Noise Power
CNR: Minimum required signal to noise ratio for the successful demodulation of the signal
N*f (t): Total noise figure
L (impl): Implementation loss of receiverâ€¦â€¦â€¦â€¦â€¦â€¦â€¦â€¦â€¦â€¦â€¦â€¦â€¦â€¦â€¦â€¦ (1)
Using above formula we can compare the wireless ranges with no RoF link with those having RoF link. And analysis shows that for few Kilometers distance both the system will have almost same characteristics. But for larger distances performance of the RoF link decreases considerably because of fiber dispersion and fiber non-linearity
There are two ways to implement the RoF links. Those are as follows:
In first approach communication between CU and RAU is in digital format and uses TDM and in second approach communication between CU and RAU is in analog form and uses SCM.
And it can be shown that the cost of implantation for first approach is higher than the second one.
As we are inserting radio link between CU and RAU, RAU is responsible for both fiber and wireless communication and therefore there are also changes in protocols of protocol stack. The main problem with inserting optical fiber between CU and RAU, it introduces propagation delay. In cellular systems we have to see that the propagation delay does not exceed the certain limit. In case of wireless local area networks and fixed broadband wireless access we can use centralized MAC scheme to relax fiber length constraints.
Applications of Radio over Fiber Link:
Access to dead zones:
It provides the coverage where wireless coverage is not possible. For example tunnels, places behind the buildings, areas like mountains, jungle
FTTA (Fiber to the Antenna)â€¦â€¦â€¦â€¦â€¦â€¦â€¦â€¦â€¦â€¦â€¦â€¦â€¦â€¦â€¦â€¦â€¦â€¦â€¦â€¦. (2)
Here we described Radio over Fiber Link Design for Next Generation Wireless Systems. And this system consists of wide radio channel bandwidth, complex modulation scheme, a high number of OFDM subcarriers and several MIMO channels. And this system uses SCM in order to conserve need of optical fibers and optical wavelengths. Also we have seen that for larger distances the performance of the RoF decreases considerably. Transmitting digital signal over fiber optic cable is much more expensive than transmitting analog signal.