Research On Key Technologies Of Cooperative Wireless Communications

Tremendous improvements have been achieved by the evolution from the first-generation wireless telecommunication technology to the fourth-generation wireless telecommunication technology, and evolve from the voice-only services to a high speed broadband system with 1Gbps of downlink rate and 500 Mbps of uplink rate. In the next several years can be expected, with the wide use of video stream, audio stream as well as mobile game we still need a novel generation of wireless technology that can support much higher data rate. As is known to all, relay has been regarded as a promising technology that can readily enhance the reliability of wireless communication and simultaneously enlarge the cellular coverage. Consequently, relay related topic have become the hot topic during the past decade, and widely discussed in the industrial community with basic applications. This thesis will focus relay application in a more general circumstance, and attempt to solve problems in practical networks. Besides, device to device(D2D) communication will surly become possible in the future. Nevertheless, the secrecy issue brought by D2 D communication should be paid more attention by researchers as well as engineers.In this thesis, we first discuss the scenario with application of two-way relays. Compared with traditional research of two-way relay, we will study a more general model that incorporates multiple relays and multiple users. Specifically, we take the multiple-input multiple-output assumption into account. Most of the existing research concerning mean squared error(MSE) optimization can be viewed as a special case of our studies. Besides, we proposed two kinds of optimizing solutions, which aim at optimizing users’ sum MSE and maximum user’s MSE, respectively. With this purpose, we need to carefully design the user’s receiver, relay and user’s precoder, respectively. Further, we notice that simultaneous optimization the above three variables is a difficult work since the problem is non-convex. However, we also see that if we fix two variables and attempt to optimize the third one, such a sub-problem is convex and can be readily solved by convex software.Therefore, we adopt the iterative strategy. Finally, simulation demonstrates the our proposed two algorithm will perform better the existing algorithm, and the algorithm based on maximum user’s MSE can achieve better BER performance than the algorithm based on users’ sum MSE. Since the BER performance will largely depend on the maximum user’s MSE, and the former one can average users’ MSE.Second, we also discuss the problem that might be encountered in the practical scenario. To start with, we study the case with existence of channel estimation error obeying Gaussian distribution. With the purpose of eliminating the channel uncertainties, we attempts to calculate the expectation of MSE and relays’ power consumption over channel estimation errors. Afterwards, we utilize iterative algorithm that iteratively optimize users’ receiver, relay precoders and user precoders, until that converges. Similarly, we also give the robust algorithms that can optimizing users’ sum MSE and maximum user’s MSE, respectively. In the end, from simulation results it can be seen that our proposed robust algorithm indeed outperform their non-robust versions in terms of bit-error-rate(BER) performance, and will avoid the performance degradation in high signal-to-noise-ratio(SNR) region. Furthermore, we also discuss the norm-bounded error model which corresponds to the scenario with limited feedback strategy. However, it turns out that the whole optimization problem is non-convex due to the existence of norm-bounded error. Therefore,we refer to optimizing the worst case of users’ signal-to-interference-plus-noise-ratio(SINR) and relays’ power. Finally, simulation results verify that our robust algorithms can better confine the eavesdropper’s SINR compared with non-robust scheme.Third, we study the topic of physical layer security in cooperative communication scenario.Firstly, we assume a relay-aided case with existence of one eavesdropper where the legal user’s desired data can be obtained only with the help of relay. Nevertheless, in the second timeslot eavesdropper might attempt to wiretap the relay’s transmitting message. Considering the special identity of eavesdroppers, we assume the channel between eavesdropper and relay experiences norm-bounded error. We provide iterative-based algorithm that can alternatively optimize relay precoder and base station precoder, respectively, and simulation verify the effectiveness of our proposed algorithm. Then, we discuss another cooperative scenario with existence of multiple base stations and one eavesdropper. With the assumption of with/without pre-known wiretapping knowledge, we provide their corresponding algorithms, respectively. Similarly, we also study the case with assumption of norm-bounded error between base stations and eavesdropper. We give our robust strategy and simulation results verify that our proposed two algorithms outperform the traditional signal-to-leakage-plus-noise-ratio. Besides, under the imperfect channel state information case, compared with non-robust scheme our robust algorithm can better restrict the eavesdropper’s wiretapping ability.Finally, we summarize the research topics studied in this thesis, and give the promising research orientation in the future.