Research On Resource Optimization And Receiver Detection Algorithms For Ambient Backscatter Communication

The Internet of Things?IoT?is the core driver of a new round of scientific and technological revolution,and an important force leading the development of society.Radio frequency identification?RFID?is a typical backscatter communication technology.It enables communication equipments to get rid of the shackles of fixed batteries,by harvesting energy from radio-frequency signals,and solves the energy problem of IoT sensors.However,current RFID systems need a reader to transmit specified radio-frequency exciting signals to activate the tags,which limits the communication distance and deployment flexibility of RFID systems,being difficult to satisfy the requirement of future IoT communications.Ambient backscatter communication?AmBC?technology utilizes existing RF signals in the environment to communicate,without the need for a special excitation source and radio spectrum,which brings the possibility of low-power and high-speed IoT communication.Most of the current research works on AmBC mainly focuses on the AmBC system of single backscatter device?BD?,and lack the research on the AmBC system with multiple BDs.Therefore,this thesis studies the resource optimization problem for a multiple-BD full-duplex AmBC network?F-ABCN?based on OFDM signals,and the detection algorithm design for a F-ABCN with multiple-BD parallel transmission.First,aiming to maximize the minimum throughput among all BDs by jointly optimizing the backscatter time coefficients and power reflection coefficients of BDs,as well as full-duplex access point's subcarrier power allocation coefficients,ensure optimization of system throughput performance under the premise of fairness.For the special case of a F-ABCN with a single BD,a closed-form solution of joint resource allocation is obtained.For the general multiple-BDs case,an efficient iterative algorithm based on the block coordinate descent and successive convex optimization techniques are proposed to solve the complex non-convex optimization problem.The convergence of the proposed algorithm is proved.Furthermore,the concept of throughput region is introduced to characterize throughput trade-offs among all BDs.The simulation results show that the F-ABCN joint resource optimization proposed in this paper can achieve higher system throughput,compared to F-ABCN equal resource allocation and joint resource optimization of half-duplex backscatter communication system.Also,numerical results validate the effects of legacy-user throughput requirement constraints,the BD's harvested energy constraints,and access point subcarrier peak power constraints on the F-ABCN's throughput performance.In addition,this paper studies the receiver algorithm of an F-ABCN system in which multiple BDs reflect incident signal in parallel.A parallel-transmission detection algorithm based on expectation-maximization?EM?clustering is proposed.After obtaining an appropriate initial value by the pre-clustering of the received signal's constellation points through the K-means algorithm,EM algorithm is utilized to perform fine constellation clusterings.Then,the level-flipping characteristic of Bi-Phase space coding?i.e.,FM0 coding?is used to obtain the mapping relationship between BD's bits and the clusters without requiring channel state information and cluster-mapping pilots,and thus recoves the information bits of multiple parallel-transmission BDs.Finally,compared to the benchmark of parallel transmission detection algorithm based on density-based spatial clustering of applications with noise,simulation results show that the proposed parallel-transmission detection algorithm can decrease the bie-error-rate?BER?performance.Specifically,for the case of two-BD paralle transmission and two-antenna receiving,under the reliable-detection requirement of BER 10^-2,the proposed scheme achieves 7.5 dB signal-to-noise-ratio improvement compared to the benmark.