The concept of cooperative communication and its numerous advantages has brought a new dimension to mobile communication networks. With improvement in spectral efficiency without the need for increasing available power or bandwidth makes it even more attractive. Orthogonal Frequency Division Multiplexing (OFDM) is acknowledged as the technique for future generation wireless systems and many schemes have been proposed for incorporating cooperative diversity in OFDM based mobile networks, but these different approaches result in a huge amount of feedback being generated from the relay channel to the source. The OFDM cooperative network model of interest, uses asymmetric time duration and non-pairing of sub-channels to limit the amount of overhead information in the network.
Mobile Communication is simply defined as "Communication on the Go". It enables users make phone calls from anywhere conveniently without being restricted to a particular area. People make calls from airplanes, in the parks, cars, while jogging, etc.
The exciting world of mobile communication since its inception in 1946[1], is continually being driven to overcome difficult limitations, in a bid to attain the incessant demands for perfection from its users. The demands for higher data rates, better quality and higher network capacity with respect to very limited available resources has become a challenge to network designers. In view of limited power and bandwidth resources, there is need for effective and efficient resource allocation. The problems posed by the nature of the wireless network environment in the form of fading, shadowing, multipath propagation, etc have led to the discovery of a viable solution in antenna diversity or the use of multiple antennas at the transmitter and receiver parts of the network. Although this implementation is not attainable with mobile phones, a possible means is suggested via cooperative communication. OFDM technology has proved itself a more effective means of improving data rate and network capacity. Its alliance with a form of antenna diversity referred to as Cooperative communication (Antenna Diversity for Mobile phones) has opened doors to more possibilities and new dimensions of operation for the wireless mobile phone system.
Review of Recent Research In OFDM Based Cooperative Networks
In OFDM transmission, the total bandwidth is divided into a number of subcarriers, over which multiple data symbols are transmitted in parallel. Therefore the symbol duration of an OFDM signal increases with the number of subcarriers. From the viewpoint of space-time cooperation, the longer symbol duration along with the use of cyclic prefix makes the system robust to timing errors [1] . According to [2] OFDM is a mature technique to mitigate the problems of frequency selectivity and inter-symbol interferences.
A lot of research has been carried out in recent times. Some of the works that led to the realization of this project's design approach are listed below -
Optimizing power allocation for nonregenerative OFDM relay links[3]
A chunk based OFDM amplify-and-forward relaying scheme for 4G mobile radio systems[4]
Optimally joint subcarrier matching and power allocation for wireless relay networks[5]
The optimization of subcarrier assignment for different users offers substantial gains to the performance of a cooperative network.
In [6], multiple-access channels with cooperative diversity are special cases of multiple-access channels with generalized feedback, a model originally developed by Carleial[7] and also studied by Willems[8]. The generalized feedback allows the sources to act as relays for one another. In [9], multiple-access channel with relaying and fading is considered.
Relaying in OFDM based network was considered theoretical in [10] but the author proposed an adaptive relaying scheme that takes channel state information (CSI) at the relay station into account. In [11], the feasibility of a multiuser OFDM network in cooperative diversity was studied and in [12] selective relaying in OFDM network was discussed.
In [13] and [14] power is allocated uniformly across all sub-channels. Power optimization to improve SNR is considered in [15] and [16].
In [3], a means of maximizing available power for effective throughput was addressed. The proposed relaying scheme in [4] reorders subcarrier signals at the relay station such that an optimal pairing of sub-channels at the source and relay which maximizes network capacity is made possible.
In [17], the maximum outage for cooperative networking with DF (decode and forward) relay is analyzed. In [18] a feasible scheme that minimizes the outage probability is considered for DF and the results approximated for AF.
In [5], an optimally joint subcarrier matching and power allocation scheme is proposed whereby a system model that incorporates optimal sub-channel pairing and optimal power allocation is considered. The outcome outperforms that of [3] and [4] that involve one optimization per model.
The research in [19] gave the capacity bounds of cooperative systems where transmission periods at the source and relay are asymmetric. However, it was a single-carrier based system.
The chapter is divided into sections that give a brief summary of how the different aspects of the project define it as a whole.
1. Overview of the Mobile Communication Network
The basic model of a mobile phone network consists of a geographic area that is divided up into cells which is why the devices are sometimes called cell phones. Each cell is under the central control of a Base-station. The fundamental function of the base-station is to manage the calls to/from a mobile phone in its cell to/from other mobile phones in its cell or other cells efficiently with the available power and bandwidth. Also it performs handoff operations when a mobile phone leaves its cell for another cell in order not to interrupt an ongoing call. Each cell uses a set of frequencies not used by neighboring cells. The key idea that gives cellular systems far more capacity than previous systems is the use of relatively small cells and the reuse of transmission frequencies in nearby but not adjacent cells[20]. The capacity of the network increases as the cells sizes decrease. A smaller cell size requires less power and cheaper antennas for both transmitters and receivers. All base-stations are connected to a single device called MTSO (Mobile Telephone Switching Office) which is the second tier of interconnection. As the network grows, more MTSOs are required and interconnected to form a second-level MTSO. More tiers are formed as the network gains more geographical coverage. The MTSOs communicate with the base-station, each other and the PSTN as well.
Figure 1: Diagram of the cellular distribution of the mobile wireless network [37]
1.1. Characteristics of The Mobile Network
Unlike the wireline networks in which the signals are transmitted via cables or guided media, the wireless networks are characterized by RF (Radio Frequency) waves that are scattered in space by the presence of obstacles (buildings, cars and moving objects) in the signal path. This results in a particular signal following different paths to reach its destination as shown in figure 2 below. The signal strength, quality and originality of information are distorted and even more so when these various copies or delayed versions of it are combined destructively at the receiver. This effect constitutes the two major problems that limit the ultimate gains of wireless networks. First is the phenomenon of fading: the time variation of the channel strengths due to the small scale effect of multipath fading, as well as larger scale effects such as path loss via distance attenuation and shadowing by obstacles. Secondly, the nature of the wireless network channel (over the air) makes it prone to interference between signals. The interference can be between transmitters communicating with a common receiver, between signals from a single transmitter to multiple receivers or between transmitter-receiver pairs [21]. The problem is further complicated if the channel is slowly changing, i.e. people and objects are moving through the environment. The changing nature of the channel may even be such that the direct line of sight signal is completely obstructed[22].
Figure 2: Radio propagation environment [24]
The aforementioned problems hinder the network from realizing high potential with data rates and coverage. The following set of diagrams illustrates that fact. Consider a signal that follows the line-of-sight path and another signal that is slightly reflected. At the receiver, the two signals form a constructive composite signal additionally. The received signal is not affected adversely.
Figure 3: diagram showing the positive effect of multipath propagation [22]
The next set of diagrams shows what happens if the reflected signal is completely out of phase with the line-of-sight signal. The resultant signal is combined destructively to form a weak signal. The receiver may not be able to track the signal and hence goes into outage/no reception mode for the duration of the weak signal.
Figure 4: Diagram showing the negative effect of multipath propagation[22]
2. Overview of Antenna Diversity
Fortunately research found a solution in using this diversity created by multipath to enhance performance. This diversity enhancement scheme is termed spatial diversity. Spatial diversity is simply the use of multiple antennas to convey copies of the same signal to a receiver. Via coherent or selective combination techniques at the receiver, a version of the signal which is extremely better in quality and with less variation than either of the individually transmitted copies of the same signal is obtained. According to [23], the basic principle of antenna diversity is that multiple antenna outputs experience different signals due to different channel conditions and these signals are only partially correlated. Thus it is likely that if one antenna is in a deep fade, then the other one is not and provides sufficient signal.
Diversity provides two major benefits. First relaiability is improved in multipath channels and secondly the overall average received signal power is increased [23].
2.1 Antenna Diversity Techniques
It is well known in signal processing that there are actually not one, but five different types of diversity that can be used to increase signal reception: spatial, temporal(time), polarization, frequency, and pattern (angle). Of these, only spatial, polarization and pattern make for a practical implementation in WLAN antenna systems [24].
Spatial Diversity - this method employs multiple antennas that are physically separated from one another[25]. Spatial diversity mitigates this problem by using two similar receive antennas separated by a fixed number of wavelengths. Given that the multi-path interference is localized to a specific location (such as antenna 1), antenna 2 will not suffer the same degradation. The separation spacing is chosen specifically to maximize the reception of antenna 2 when antenna 1 is at minimum. The receiver simply switches to whichever antenna is currently receiving the strongest signal[24]. This is especially beneficial for mobile communication[25].
Pattern Diversity - pattern diversity consists of two or more co-located antennas with different radiation patterns. This type makes use of directive antennas that are usually separated by a short distance[25].
Polarization diversity - polarization diversity combines pairs of antennas with orthogonal polarizations. Reflected signals can undergo polarization changes depending on the media. By pairing two complementary polarizations, this scheme can immunize a system from polarization mismatches that would otherwise cause a signal fade. Additionally such diversity has proven valuable at radio and mobile communication base-stations since it is less susceptible to the near random orientation transmitting antennas[25].
Transmit/Receive Diversity - Transmit/Receive diversity uses two separate, collocated antennas for transmit and receive functions. Such a configuration eliminates the need for a duplexer and can protect sensitive receiver components from the high power used in transmit[25].
Adaptive Arrays- Adaptive arrays can be a single antenna with active elements or an array of similar antennas with ability to change their combined radiation pattern as different persists. It is especially beneficial for radar applications[25].
The technique of interest is the spatial diversity method.
2.2 Spatial Diversity
The first form of spatial transmit diversity is the delay diversity scheme proposed by Wittenberg where a signal is transmitted from the second antenna then delayed one time slot and transmitted from the first antenna. Signal processing is used at the receiver to decode the superposition of the original and delayed codewords[26]. More advanced methods have been devised since then. The early 1990s witnessed new proposals for using antenna arrays to increase the capacity of wireless links, creating enormous opportunities beyond just diversity. It turned out that diversity was only a first step to mitigate multipath propagation. With the emergence of MIMO systems, multipath was effectively converted into a benefit for the communication system [27].
Spatial diversity as stated above utilizes multiple antennas at one end of the transmission path which is the transmitter to perform interference cancellation and to realize diversity and array gain through selective superposition of the signals. In contrast, MIMO employs the use of multiple antennas at both sides of the link and offers additional fundamental gain such as spatial multiplexing, which results in spectral efficiency [28].
The advantages of MIMO have been recognized as a breakthrough in wireless communication to the point that certain diversity methods have been incorporated into wireless standards. Although clearly advantageous on a cellular base station, it may not be practical for other scenarios. Specifically due to size, cost or hardware limitations, a wireless agent (mobile handset) may not be able to support multiple transmit antennas. A technique was recently developed to enable the application of MIMO to single-antenna mobile phones. The basic idea is that single-antenna mobiles in a multi-user scenario can share their antennas in a manner that creates a virtual MIMO system [29] or cooperative network.
3. Overview of Cooperative Communication Network
Cooperative communication networks also referred to as Virtual MIMO systems is designed in such a way that each mobile device in multi-user network increases the quality of its transmitted signal via mutual partnership with a second mobile device. According to [cooperative communication in resource constrained wireless networks], cooperative communication refers to a system where users share and coordinate their resources to enhance the transmission quality. The diagram below illustrates how it works.
Figure 5 Diagram showing a three-node cooperative network model [22]
From the diagram above, if the two users are viewed from the point that they both transmit individually over independent fading channels, at times when the SNR is below the threshold (region marked out to identify good signal quality) due to the channel experiencing deep fade as indicated by the dark areas, the transmission fails. If the users cooperate with each other and share resources, then transmission failure will only occur if both users experience poor channel conditions simultaneously.
The foundation, on which the idea of cooperative communication is built, is attributed to the work of Cover and El Gamal on the information theoretic properties of the relay channel[30]. This work analyzed the capacity of a three node network consisting of a source, relay and a destination. Although the analysis was in view of their current understanding of the challenge at hand, its principle of incorporating a relay channel in a network model has been adopted but in a different way to suit the goals of a cooperative network. The relay channel is provided by a mobile device which also is an information source unlike the relay channel proposed by Cover and El Gamal whose sole purpose is to help the main channel [29] or information source. In a cooperative mobile network, all users can be a source, relay or destination node depending on its functional status at any time i.e. sender, relay or receiver.
Most cooperation strategies involve two phases: the coordination phase and the cooperative transmission phase [cooperative comm. In resource constrained network]. Coordination phase defines how feedback is distributed. Feedback is necessary in a cooperative network because of the interdependency factor amongst users. The cooperative transmission phase defines how the users compute and transmit messages based on the feedback information received.
3.1 Cooperative Signaling Methods
The source, relay and destination channels of a cooperative network can be designed to perform in certain ways. A cooperative network is classified based on the choice of performance method. A summary of the different methods of implementation are given below:
Decode And Forward - this method is closest to the idea of a traditional relay. In this method a user attempts to detect the partner's bits and then retransmits the detected bits. The partners may be assigned mutually by the base station or via some other techniques [29]. It is also called the regenerative method. This method is considered simplified and easily adaptable to changing channel conditions. Adversely, it is prone to loss of signal detection by partnering mobile phone and the source needs to know the error probability of the channels for optimum decoding. According to J Laneman[31], a hybrid method of decode and forward method is suggested to minimize the effect of loss of signal detection. In the case of high SNR, cooperation by detect and forward is implemented but engages non-cooperative if otherwise.
Figure 6. Illustration of Decode and Forward method [29]
Amplify And Forward - this method allows the relay node to amplify and not merely repeat the noisy signal from the source. The base station via superposition of the signals from the relay and source obtains the transmitted information signal. this method, according to Laneman[31], produces the best outcome at high SNR. It is a non-regenerative method.
Figure 7 Illustration of Amplify and Forward method [29]
Coded Cooperation - this method integrates cooperation into channel coding by sending different portions of the user's codeword via two independent fading paths. The basic idea is that each tries to transmit incremental redundancy to its partner. Whenever that is not possible, the users automatically revert to a non-cooperative mode.
Figure 8: Illustration of Coded Cooperation Method [29]
Among these strategies, DF and AF are the most popular ones due to their simplicity and intuitive designs[22].
4. Multiple Access Techniques for Wireless Networks
In order to further improve network capacity and efficiently utilize limited resources, access technologies were developed. These access technologies make it possible for many calls to share a set of frequencies and not just a single user occupying the entire bandwidth. These multiplexing techniques are FDMA (Frequency Division Multiple Access), TDMA (Time Division Multiple Access) and CDMA (Code Division Multiple Access).
FDMA - It is a technology based on FDM. The frequency spectrum is divided into frequency bands with each user having exclusive possession of some band[20]. Used in first generation mobile networks.
TDMA - This developed from TDM. The users take turns (in a round- robin fashion), each one periodically getting the entire bandwidth for a little burst of time [20].
CDMA - allows each station to transmit over the entire frequency all the time. Multiple simultaneous transmissions are separated using unique codes[20]. It is based on spread spectrum.
Figure 9: Diagram showing the different technologies in frequency and time domain[37]
Cooperative communication assumes that the destination can separately receive the original and relayed transmissions. This is attained by transmitting the two parts orthogonally so that they can be separated[29]. Orthogonality is best achieved using OFDM[2][3][4][5], owing to the fact that the relay node retransmits the same signal from the source after a certain time, Time Division Duplex (TDD) is incorporated.
Over the years, FDM has developed into OFDM. OFDM was born out of a need to solve the mitigation problems associated with wireless transmissions. It is a type of FDM that uses orthogonal carriers.
4.1 Overview of OFDM Technology
FDM splits the available bandwidth into multiple channels. These channels have different frequency ranges which all add up to the total bandwidth. FDM systems have guard bands (inter-space) between channels to prevent one channel from interfering with another. These guard bands lower the data information rate[32]. Also these channels are carriers
Figure 10: Diagram showing OFDM subcarriers in Frequency Domain[33]
The basic principle of OFDM is this; the data stream is split into N parallel streams of reduced data rate and each is transmitted on a separate subcarrier. In practice it means that bits are transmitted in parallel over a number of flat fading channels[34].OFDM communication systems utilize more effectively and efficiently the frequency spectrum via overlapping sub-carriers. This is possible because of the orthogonality characteristic it possesses. Orthogonality is defined when signals are uncorrelated with each other. Their dot product results is zero[35]. Hence they partially overlap without interfering with adjacent subcarriers because the maximum power of each subcarrier corresponds directly with the minimum power of each adjacent subcarrier.
The OFDM signal requires less bandwidth as the number of subcarriers is increased due to its orthogonality characteristics or spectral overlapping.
Further Work
The challenge still remains on how partners are assigned and managed in multi-user networks. Systems such as the cellular, in which the users communicate with a central base-station, offer the possibility of a centralized mechanism. Assuming the base station has some knowledge of all the channels between users partners could be assigned based on some performance criterion[29].
Also the cost function, how the network operator charges users based on data rates shared or the duration of cooperation.
CHAPTER 2
1. Methodology
1.1 Aims and Objectives
The aim of this project is to propose a model of an OFDM based cooperative network in which sub-pairing of subcarriers is unnecessary which means that the amount of feedback generated is independent of the number of sub-channels. This results in a dramatic overhead reduction in comparison to previous works[36].
The design also distinguishes itself from other designs by using asymmetric timing slots for its TDD ( Time Division Duplex)in transmitting signals from source and relay nodes to the destination node.
The project uses Matlab simulations to show graphically the analytical comparison of other models with the proposed scheme.
In order to have a balanced analysis of the different models, each will be briefly discussed and their major characteristics pointed out.
2. Design Models
2.1 Optimal Power Allocation for Non-Regenerative OFDM Relay Links
This model discussed in [3] approaches its design from the view point of maximizing the available power resources. In earlier designs, the power allocation between source and relay are discussed fior the case that both share a total amount of transmit power over the two time slots required for relaying but this model shows that the instantaneous rate of the link is attainable via allocating more power to certain channels based on a channel attenuation factor.
The model focuses on a two-hop AF (amplify and forward) relay link using OFDM modulation. The transmitted signals are subject to frequency-selective fading channels.
Joint optimization of the source and the relay PA with joint transmit power constraints at source and relay would certainly provide a higher rate. This optimization is capable of responding more efficiently to the relay and destination. If attenuation between source and relay is much smaller than between relay and destination, this optimization would give a higher fraction of the overall transmit power to the relay, according to their average channel attenuation.
A higher performance in terms of rate can be achieved if the subcarrier of both channels; source to relay and relay to destination are paired according to their actual strength. In other words the best source to relay channel is paired with the best relay to destination channel.
Such optimization is only a reasonable approach in low mobility wireless networks, where the channel does not vary fast over time. In other practical systems, the signaling overhead due to joint optimization seems to be prohibitive. The solution was to calculate their respective power constraints according to the average channel attenuation.
Yet the fact that the relay needs information from all subcarrier pairing and the continuous calculation of the separate channel power constraints per transmission constitutes a lot of feedback information.
2.2 A Chunk Based OFDM Amplify and Forward Relaying Scheme for 4G Mobile Radio Systems
This method was proposed in [4]. Communication between the source and relay covers two time slots. The base-station (BS) transmits an OFDM packet during the first time slot and the relay receives and stores the signal. During a second time slot, the relay retransmits a processed version of the stored signal towards the mobile station.
It assumes that the OFDM sub-channels be arranged on a chunk basis in the sense that a number of subchannels are grouped together and taken as a single sub-channel. This is done for BS to RS (relay station) and for RS to MS (mobile station). The average sub-channel SNR (signal-noise ratio) within chunk is calculated for the first and second hop channels respectively.
An adaptive OFDM AF (amplify and forward) relaying scheme makes use of the estimated transfer functions of the first and second hop channels to optimize relaying to the MS. This means that the strongest received sub-channel signal has to be coupled into the strongest sub-channel of the second hop. The sub-channel reordering that takes place at the relay also affects the sub-channel numbering at the destination. Hence the destination learns the current sub-channel mapping function in order to reconstruct the originally transmitted OFDM packet.
Also like the previous method, this approach accumulates a lot of feedback information to the relay and destination nodes. The large overhead information lowers the channel capacity of the system.
2.3 Optimally Joint Subcarrier Matching and Power Allocation in OFDM Multihop System
This design approach considered in [5], proposes that subcarrier matching be carried out based on the order of the channel power gains and power allocation to these paired subcarriers by water filling method. The scenario is based on a three terminal network; the source, relay and destination nodes. The source is assumed to be a huge distance away from the destination and the relay is the only link between them. Communication between them covers two time slots. The source transmits an OFDM symbol over the first time slot and at the same time, the relay decodes the symbol. During the second time slot, the relay re-encodes the signal with the same codebook as the one used at the source and transmits it towards the destination over the relay - destination channel. The destination decodes the relay based on the received signal form the relay.
The sub-channels at the source and relay nodes are arranged in ascending order based on the channel power gains. Power is allocated to these sub-channels based on the equivalent power gain of the paired sub-channels, with the most power given to the pair with the highest power gain, the next channel with the second highest gain is given most of the power too but not as much as the first and so forth. Hence based on the power gain ratio, the channel power is apportioned amongst the channels.
2.4 PROPOSED MODEL - Dynamic Resource Allocation with Limited Feedback
The project's model is designed to limit feedback information from relay to source[36]. It is different from other previous models in the following aspects -
Sub-channel pairing is not required as in past works. The amount of feedback information is independent of the number of sub-channels leading to a dramatic overhead reduction[36].
The transmission duration for the first time slot and second time slot from BS to RS and RS to MS respectively are designed to be asymmetric rather than symmetric as in past works, which enhances the degree of freedom for the transmission[36].
3. Future Work
For the next stage of the project, a detailed analysis of the various design models considered above will be carried out.
An in-depth performance analysis will be carried out for each model. Their performances will be compared with the project's model based on the amount of feedback information generated in their respective systems. The comparison results will be shown via Matlab simulation. The results will attest to the fact that the model of interest provides a higher system capacity than the other conventional works.
Finally, if time permits, a more efficient model will be recommended based on findings.
CHAPTER 3
Probable Challenges
The most obvious challenge to be faced is getting to use Matlab to model the different system designs. Each system uses OFDM TDD technology, a technology that uses FFT (Fast Fourier Transform) and IFFT (Inverse Fast Fourier Transform) to convert symbols from time domain to frequency domain and vice versa. The FFT and IFFT are identified as advanced complex mathematics. In order to simulate any of the designs aforementioned, a deep understanding of FFT and IFFT are required as well as how OFDM implements it to attain orthogonality.
The practical aspect of the project is most likely going to uncover inherent issues that cannot be foreseen at the moment.
Conclusion
Cooperative communication is an ingenious technique that is being exploited in depth, to fully realize its potentials. Merging it with OFDM system is just one of various exploitations that seem to be yielding substantial results in the field of mobile telecommunications. So far it has been proved beyond doubt that the collaboration improves signal-to-noise ratio (SNR) and network capacity by a considerable amount in comparison to non-cooperative OFDM systems.
Gantt Chart
