Excerpt
Contents
Abstract
1 Introduction
2 OFDM adjacent channel interference
3 Conventional Methods
3.1 Windowing
3.2 Filtering
3.3 Guard-band and virtual sub-carrier
3.4 Forward Error Correction coding
3.5 Adaptive modulation
4 Combinations of modulation schemes
4.1 Performance of OFDM modulation schemes
4.2 Proposal
4.3 Simulation model and results
4.3.1 Simulations of the proposed method
4.3.2 Optimization
5 Conclusion
Acknowledgments
References
List of Figures
illustration not visible in this excerpt
List of Tables
illustration not visible in this excerpt
Abstract
Non-orthogonality among adjacent OFDM channels creates OFDM adjacent channel interference and it heavily affects the entire system’s performance. Conventional methods to avoid OFDM adjacent channel interference are not only insufficient but also are wasting a lot of frequency resources. In this research, a method using combinations of modulation schemes is proposed to avoid effects of OFDM adjacent channel interference. It can be obtained by modulating the sub-carriers at the outer sides of an OFDM channel with lower order modulation schemes (such as BPSK or QPSK), while modulating the sub-carriers at the inner side of the OFDM channel with higher order modulation schemes (such as 16QAM or 64QAM). Intensive simulations have been carried out to evaluate the performance of the proposed method. The simulation results have shown an increase in the OFDM system’s resistance against adjacent channel interference while still maintain the bandwidth efficiency.
1.Introduction
The growth of demand on wireless mobile multimedia services has made OFDM technology a very popular modulation scheme for high-speed communication systems. OFDM has been applied in almost all kinds of communication media such as wireless, copper wires, power-line or fiber optic [7]... It can be defined as either a modulation or a multiplexing technique. It has been used in many applications such as Digital Terrestrial Television Broadcasting, Digital Audio Broadcasting, wireless networking and broadband internet access. IEEE 802.11 standard extension targets a range of data rate from 6 up to 54 Mbps using OFDM in the 5 GHz band making OFDM effectively a world-wide standard for this band [1].
One of the main reasons to use OFDM is because it can handle and have the potential to handle efficiently the multipath fading and interference problems in a wireless communica-tion channel [2]. In a single carrier system, if the transmitted signal has a greater bandwidth than the bandwidth of the multipath fading channel and the interference, the received signal will be distorted and interfered [3]. As a result, a single fade or interference can cause the entire link to fail. However, if the transmitter has a narrower bandwidth as compared to the multipath channel and interference, the received signal will not be distorted in time domain. Since OFDM is a multi carrier transmission technique, its transmitting bandwidth is divided into subcarriers each has a much smaller bandwidth. Only a small percentage of the sub carriers will be affected by the multipath fading and interference [17]. Error detection and correction coding can then be used to make necessary corrections to the few erroneous sub carriers. OFDM is also an efficient way to deal with multipath propagation; for a given delay spread, the implementation complexity is significantly lower than that of a single carrier system with an equalizer [18].
In a classical multi carrier system, the total single frequency band is divided into N nonoverlapping frequency subchannels. All sub channels are then modulated and frequency multiplexed. Guard bands must be used to avoid spectral overlap of the frequency sub channels in order to eliminate inter-channel interference. However, it leads to an inefficient usage of the available spectrum [16].
OFDM is a special case of multi carrier transmission, where a single data stream is transmitted over a number of lower rate sub carriers. The main difference between OFDM and a classical multi carrier system is that the sub carriers in OFDM are overlapped. To realize the overlapping multi carrier technique, however we need to reduce crosstalk among sub carriers, which means that we want orthogonality among the different sub carriers [7].
Orthogonality means that there is a precise mathematical relationship between the signals of the sub carriers in an OFDM system. In a normal frequency-division multiplexing system, many carriers are spaced apart in such a way that the signals can be received using conventional filters and demodulators. In such receivers, guard-bands are introduced between the different carriers and in the frequency domain which results in a lowering of spectrum efficiency. In an OFDM system, the receiver acts as a bank of demodulators, translating sub carriers down to DC, with the resulting signal integrated over a symbol period to recover the raw data. If the other sub carriers all beat down the frequencies that, in the time domain, have a whole number of cycles in the symbol period T, then the integration process results in zeros contribution from all other sub carriers. Thus, the sub carriers are orthogonal if the sub-carrier spacing is a multiple of 1/T[1].
To eliminate the banks of sub carriers oscillators and coherent demodulators required by frequency-division multiplex, completely digital implementations can be built around the fast Fourier transform (FFT). Using this method, both transmitter and receiver are implemented using an efficient FFT technique [8].
The orthogonality of sub carriers in OFDM can be maintained, and individual sub carriers can be completely separated by using an FFT circuit at the receiver when there are no inter-symbol interference (ISI) and inter-carrier interference (ICI) introduced by transmission channel distortion. In practice, however, these conditions can not be obtained [19]. To reduce the distortion, a simple solution is to increase the symbol duration or the number of sub carriers. However, this method may be difficult to implement in terms of carrier stability against Doppler frequency and FFT size [1]. Each sub-carrier can be modulated with a different modulation scheme. In WLAN environment, differential encoding and detection-based modulation schemes, such as D8PSK, are used. However, according to several standardization committees, the use of a broadband data terminal is possible not only in an indoor environment but also in an outdoor micro-cellular environment. The common modulation schemes used in such kinds of OFDM systems are BPSK, QPSK, 16QAM, and 64QAM [1].
The main motivation and contribution of this research are to find combinations of modulation schemes (which are commonly used in OFDM systems) in order to increase the OFDM system’s resistance against Adjacent Channel Interference while still maintain its bandwidth efficiency.
The thesis is organized as followed: Chapter 2 will introduce the background knowledge of adjacent channel interference and its effects on OFDM system’s performance. In Chapter 3, main conventional methods will be introduced to see their advantages and disadvantages. The proposal will then be explained in detail in Chapter 4, followed by intensive simulations and simulation results. In Chapter 5, some conclusions will be derived from the simulation results.
2. OFDM adjacent channel interference
OFDM operators are assigned different channels to provide their services. For example, the 2.4GHz frequency band is available for license-exempt use in Europe, the United State and Japan. Table 2.1 lists available frequency bands and restrictions to devices which use this band for communications.
Table. 2.1 International 2.4 GHz ISM bands [1] illustration not visible in this excerpt
In Japan, equipment manufacturers, service providers and the Ministry of Post and Telecommunications are cooperating in a Multimedia Mobile Access Communication (MMAC) project to define new wireless standards similar to those of IEEE 802.11. Additionally, MMAC is also looking into the possibility for ultra-high-speed wireless indoor LANs supporting large-volume of data transmission at speeds up to 156 Mbps using frequencies in the 30 to 300 GHz band [1].1 illustration not visible in this excerpt
Fig. 2.1 Channelization in lower and middle UNII band.
Figure 2.1 shows the channelization for the lower and middle UNII (Unlicensed National Information Infrastructure) bands (in GHz). Eight channels are available with a channel spacing of 20 MHz and a guard spacing of 30 MHz at the band edges in order to meet FCC restricted band spectral density requirements. FCC also defined an upper UNII band from 5.725 to 5.825 GHz, which carries four other OFDM channels. For this upper band, the guard spacing (guard band) from the band edges is only 20 MHz, as the out-ofband spectral requirements for the upper band are less severe than those of the lower and middle UNII bands. In Europe, a total of 455 MHz is available in two bands, one from 5.15 to 5.35 GHz and another from 5.470 to 5.725 GHz. In Japan, a 100 MHz wide band is available from 5.15 to 5.25 GHz, carrying four OFDM channels.
Obviously, the signals generated by commercially available OFDM wireless equipments are by no means perfect. Indeed, these signals generate some amount of energy outside of their approved spectrum bands. This is called side-band emissions [6]. This also is true in other wireless devices using the same OFDM technology, such as Bluetooth, cordless telephones, Digital Terrestrial Television Broadcasting, Digital Audio Broadcasting... Although band-pass filters are usually applied to minimize the RF interference and the side-band emission from adjacent channels, they still generate side lobe energy that falls into the pass-band of other adjacent channels’ signals. This kind of interference is called Adjacent Channel Interference (ACI). If the total ACI from adjacent channels is much stronger than the wanted signal from the wanted channel, side band energy from the ACI can dominate the channel’s pass-band. This is shown in Figure
illustration not visible in this excerpt
Fig. 2.2 Adjacent channel interference.
The side-band emission is characterized by the adjacent channel leakage ratio (ACLR), which is defined as the ratio of the transmitted power to the power measured in the adjacent channel. Receiver imperfections also contribute to ACI. The receiver filter aims to reject all signal energy that lies outside the wanted band; but in practice, a certain amount of this energy is passed and added to the wanted signal. A parameter, known as the adjacent channel selectivity (ACS), has been defined, which is the ratio of the receiver filter attenuation on the assigned channel to the receiver filter attenuation on the adjacent channel.
As the effects of ACLR and ACS work together to generate ACI, they are combined into a single parameter known as the adjacent channel protection (ACP). This is defined as the ratio of the total power transmitted from a source to the total interference power affecting a victim receiver, and is calculated using Equation 2.0.1. illustration not visible in this excerpt
Figure 2.3 illustrates the near-far problem in wireless communication. Near-far problem in wireless communication systems also contributes to ACI. The near-far problem happens when a Mobile Station (MS) of one operator moves away from it own Base Station (BS) toward a BS of another operator operating in an adjacent channel. In this case, the wanted signal that the MS receives from its own BS decreases while the unwanted ACI received from the BS of the adjacent channel increases. In this case, the ACI power may dominate the pass band of the wanted signal and heavily damage the system’s performance.
For example, if a mobile station is 20 times as closer to a base station as another mobile station and has energy spill out of its pass-band, the signal to noise ratio (SNR) of the illustration not visible in this excerpt
Fig. 2.3 Near-far problem in wireless communication systems. weak mobile (before receiver filtering) is approximately: illustration not visible in this excerpt
From Equation 2.0.2, we can calculate that, for a path-loss exponent n=4, the SIR is equal to -52 dB. If the band-pass filter at the base station receiver has a slop of 20 dB/octave, then an adjacent channel interference must be moved by at least six times the pass-band bandwidth from the center of the receiver frequency pass-band to achieve 52 dB attenuation. This means that, a frequency spacing of greater than six times the channel bandwidth is needed to bring the adjacent channel interference to an acceptable level, or tighter filters are needed when close-in and distant users share the same cell [3]. Near-far problem can be effectively simulated by changing the relative transmitting powers of wanted and interfering channels.
Adjacent channel interference mainly affects sub-carriers at two sides of a frequency channel using OFDM technology. According to that fact, in the IEEE 802.11a standard, the sub-carriers at two sides of the frequency band are left unmodulated and unused to protect the OFDM signal from the adjacent channel interference.
3.Conventional Methods
To separate adjacent OFDM channels and to avoid adjacent channel interference, several conventional methods have been applied. The most common methods are: Windowing, Filtering, Guard-band, Virtual Sub-carrier, Forward Error Coding, and Adaptive Modulation. In this chapter, those conventional methods are explained and analyzed to see their advantages and limitations.
3.1 Windowing
Essentially, an OFDM signal consists of a number of unfiltered sub carriers. As a result, the out-of-band spectrum decreases rather slowly, according to a sinc function. For a larger number of sub-carriers, the spectrum goes down more rapidly in the beginning. It is caused by the fact that, the side lobes are closer together. However, the spectrum of 256 sub carriers has a relatively large -40dB bandwidth that is almost four times the -3dB bandwidth. This causes serious adjacent channel interference to the adjacent OFDM channels.
To make the spectrum go down more rapidly, windowing or pulse shaping techniques must be applied to individual OFDM symbols. When rectangular pulses are passed through a band limited channel, the pulse will spread in time, and the pulse for each symbol will spread into the time intervals of succeeding symbols. This causes ISI and leads to an increased probability of the receiver making an error in detecting a symbol. Out-of-band radiation in the adjacent channel should generally be 40dB to 80dB below that in the desired pass-band.
There are a number of well known pulse shaping techniques which are used to simultaneously reduce the ISI effects and the adjacent channel interference for an OFDM system. The ISI could be completely nullified if the overall response of the communication system (including transmitter, channel and receiver) is designed so that in every sampling instant at the receiver, the response due to all symbols except the current symbol equal to zero [3]. Expressed in terms of the channel response, if the channel is ideal, then the shaping filters at both the transmitter and receiver must produce an channel response that satisfies the Nyquist ISI criterion: illustration not visible in this excerpt
WhereinTs is the symbol interval.
Filters which satisfy the Nyquist ISI criterion in Equation 3.1.1 are called Nyquist filters. Windowing an OFDM symbol by using Nyquist filter makes the amplitude go smoothly to zero at the symbol boundaries. A commonly used Nyquist window type is the raised cosine window [3], which is defined as: illustration not visible in this excerpt
Here,Ts is the symbol interval, which is shorter than the total symbol duration because we allow adjacent symbols to partially overlap in the roll-off region. αis the roll-off factor of the raised-cosine window.
In practice, the OFDM signal is generated as follow: first,
Ncinput modulated values are padded with zeros to get N input samples that are used to calculate an IFFT. Then, lastTpref ixsamples of the IFFT output are inserted at the beginning of the OFDM symbol, and the firstTpref ixsamples are appended at the end. The OFDM symbol is then multiplied by a raised cosine window W(t) to more quickly reduce the power of out-of-band sub-carriers. The OFDM symbol is then added to the output of the previous OFDM symbol with a delay ofTs such that there is an overlap region ofα.Ts. illustration not visible in this excerpt
Fig. 3.1 Magnitude transfer function of a raised-cosine window.
Figure 3.1 shows the magnitude transfer function of a raised cosine window with different values of the roll-off factorα. Smaller roll-off factors improve the spectrum further in fre-quency domain, at the cost, however, of a decreased delay spread tolerance in time domain. There is a trade-off between a roll-off factor ofαwith the effective guard time.
Because of this trade-off, raised-cosine window should only been carefully used with a suitable roll-off factor. Moreover, the spectral efficiency offered by a raised-cosine window only occurs if the exact pulse shape is preserved at the carrier. This becomes difficult if nonlinear RF amplifiers are used. Small distortions in the base band pulse shape can dramatically change the spectral occupancy of the transmitted signal. If not properly controlled, this can make adjacent channel interference more serious. These problems lead to the fact that, using windowing in an OFDM system is inadequate against adjacent channel interference in OFDM systems.
[...]
1EIRP: Effective Isotropic Radiated Power.