Font Size: a A A

Research On Optimization Approach For Multi-user Wireless Transmission Based On Interference Alignment

Posted on:2016-07-05Degree:DoctorType:Dissertation
Country:ChinaCandidate:L H JiangFull Text:PDF
GTID:1108330503469775Subject:Information and Communication Engineering
Abstract/Summary:PDF Full Text Request
Multi-user interference has become an emerging problem of wireless transmission, and how to suppress the interference has become a key technology of increasing the throughput of the networks. Within the existed interference suppression methods, interference alignment(IA) can cast the desired signal and interference into different signal subspaces, and thus can suppress the interference and increase the sum rate effectively. Therefore, the multi-user wireless transmission based on IA has promising prospect. In spite of this, it still faces many challenges, such as how to suppress the interference leakage effectively, how to increase the signal-to-interference plus noise ratio(SINR), how to reduce the computational complexity, how to deal with the strong interference when the number of users exceeds the IA feasibility constraint, and how to optimize the modulation scheme. Therefore, the thesis focuses on the optimization approaches of the multi-user wireless transmission based on IA, including the low complexity interference leakage suppressing algorithm, the low complexity maximizing SINR algorithm, the user grouping approach, and the modulation scheme.Firstly, the low complexity interference leakage suppressing algorithm is investigated. To mitigate the high computational cost of the traditional minimizing interference leakage(Min IL) algorithm for IA, a low complexity directional quartic optimal(DQO) algorithm is proposed. IA must be able to suppress the interference leakage effectively. Although the Min IL algorithm can achieve low interference leakage, its computational complexity increases dramatically with the number of users and antennas, posing limit to its applications to the practical networks. The proposed DQO algorithm employs the line search strategy and obtains the optimal step size by optimizing a quartic function. Theoretical analysis and simulations demonstrate that DQO algorithm can achieve rapider convergence rate and lower computational cost than the Min IL algorithm. Under the same interference leakage, the proposed algorithm requires much less iterations.Secondly, the algorithm for increasing the received SINR is studied. To mitigate the high complexity of the traditional Max-SINR algorithm for IA, a principle direction search(PDS) algorithm is proposed. IA has to consider not only the interference leakage, but also the received SINR. Although the traditional Max-SINR algorithm can increase the SINR of each data stream and the sum rate effectively, the iterations and computational time increase dramatically when the number of users approaches the IA feasibility bound. Therefore, the properties of the Max-SINR algorithm are studied, and the concept of principle direction and the associated PSD algorithm are developed. Theoretical analysis and simulation results demonstrate that the proposed PDS algorithm can achieve high sum rate with higher convergent rate and lower computational cost than the traditional Max-SINR algorithm.Thirdly, the user grouping approach is studied, and a joint spatial-code clustered(JSCC)-IA algorithm is proposed for the case when the number of users exceeds the IA feasibility constraint. The number of users in the traditional IA network is constrained by the IA feasibility condition and the number of antennas has to become larger when the network scale increases. When the number of users exceeds that IA can afford, the traditional IA method will fail to separate the desired signal and interference, the quality of server will degrade significantly, and the communication might even be terminated. The proposed JSCC-IA algorithm can mitigate the interference and increase the number of users that the network can accomodate, when the the IA feasibility constraint cannot be met. The analytical expressions of the bit error rate(BER) of the JSCC-IA algorithm are formulated for the symmetric networks. In addition, a random grouping selection algorithm is also developed, which can select better grouping combinations and further improve the performance of the JSCC-IA algorithm in the asymmetric networks.Finally, the modulation scheme in the transmission is studied, and an IA algorithm is proposed based on continuous phase modulation(CPM). The CPM-IA scheme can reduce the bandwidth, increase the spectrum efficiency, and reduce the BER effectively. Different from the traditional binary phase shift keying(BPSK) modulation, the phase of the CPM signal varies continuously, leading to much lower sidelobe and rapider spectrum rolloff. In addition, the continuity and memory inherent in the phase can be further explored by the detector, leading to lower BER. Moreover, to mitigate the high complexity of the CPM-IA recever, a low-complexity pulse amplitude modulated(PAM) decomposition algorithm is applied to CPM-IA scheme, which can reduce the number of front end correlators and the phase states of the Viterbi decoder. Finally, a novel spatial-frequency domain(SFD)-CPM-IA scheme is proposed based on the advantage of narrow bandwidth and high sprectrum efficiency of CPM-IA. The SFD-CPM-IA can mitigate the interference that traditional IA cannot handle when the IA feasibility condition cannot be satisfied, and can accommodate many more users to communicate simultaneously.
Keywords/Search Tags:Interference Alignment, Directional Quartic Optimal, Principle Direction Search, Joint Spatial-Code Clustered, Continuous Phase Modulation
PDF Full Text Request
Related items