Beamforming For Joint Communication And Radar Sensing Techniques In Autonomous Vehicular Networks

With the rapid development of technology,there are increasing demands of autonomous vehicular networks.The next generation vehicular networks require high efficiency and flexibility to guarantee safety.In such networks,vehicles need to sensing the surround-ing area to retrieve the information of other objects,such as the position,velocity and the shape of the target.At the same time,vehicles need to communicate with other nodes to share and update the latest events.In such a context,joint communication and radar(radio)sensing(JCAS,also known as Radar-Communications)technology is particularly promising in autonomous vehicular networks,for its appealing capability of operating both communication and radar sensing functions.JCAS systems have appealing features such as low-cost,resource-saving(for both frequency spectrum and hardware),reduced size and weight,and mutual sharing of information for improved communication and sens-ing performance.Millimeter wave(mm Wave,defined at 30-300 GHz),as a traditional waveband for radar applications,is particularly promising for JCAS due to its very wide bandwidth and hence,fine resolution capability.Due to the small wavelength,the path loss,attenuation,and blockage of electromagnetic waves can become severe in mm Wave.Therefore,beamforming(BF)technology,which can produce definable high-directional spacial beams and hence efficiently transmit/receive the signals,is becoming a demanding feature in JCAS.In this dissertation,we investigate the BF techniques for JCAS in autonomous vehic-ular networks,addressing the following two issues:1.Communication and sensing often have different requirements for BF.Radio sens-ing often requires time-varying directional scanning beams,while a stable and accurately-pointing beam is usually expected.In most of the current JCAS schemes,a single beam is used for both communication and radar sensing,which is highly restricted for both functions.Multibeam technology enables the use of more than one mainlobe(called as subbeams hereafter)in JCAS systems,to meet different requirements of beamwidth and pointing directions.Regarding this issue,this dissertation mainly study two classes of multibeam generation methods for the transmitter: 1)the optimal combination of two pre-generated subbeams,and their BF vectors,using a combining phase coefficient? 2)global optimization methods which directly find optimal solutions for a single BF vector.Con-sidering different requirements for communication and radar sensing,a series of different BF algorithms are proposed,analyzed,and compared.These works can be regarded as a good reference for the BF applications in JCAS systems.2.This dissertation also studies the quantization of the BF weight vector with the use of phase shifters,which can have a non-negligible influence on the performance of BF al-gorithms.This dissertation focus on the two-phase-shifter array,where two phase shifters are used to represent each BF weight.This dissertation proposes novel joint quantiza-tion methods by combining the codebooks of the two phase shifters,where a novel phase shift constant is proposed.The phase shift constant can largely influence the number and distribution of the quantization codes.Based on two typical joint codebooks,the quantiza-tion of BF vectors is studied.The improved golden section search-quantization(IGSS-Q)method is also proposed based on the one-dimensional search,which enables better scalar quantization by considering the property of vector quantization.Analytically,the mean squared quantization error(MSQE)is also derived for various quantization methods.Sim-ulation results validate the accuracy of the analytical results and the effectiveness of the joint quantization methods.3.In conventional BF algorithms,the antenna arrays are ideally modelled from the aspect of signal processing,ignoring the radiation characteristics of the antenna array.The array manifold mismatches between the ideal array model and the practical array can also deteriorate the BF performance.For the first time,this dissertation proposes ro-bust BF algorithms embedding the active radiation patterns of antennas.By combining the idea of classic robust BF algorithms with the radiation characteristics of the antenna array,the proposed algorithms can improve the performance of the classic robust BF algo-rithms considering the ideal array model.Besides,with the use of robust BF algorithms,the proposed algorithms show better tolerance to both engineering and electromagnetic mismatches caused by the modelling,manufacturing,aperture assembling,and channel debugging of the antenna array.