Research On Beamforming Design For Integrated Sensing And Communication In Massive MIMO Systems

The international mobile communication 2030(IMT-2030)has issued the White Paper on 6G Vision and Candidate Technologies,implying that the integrated sensing and communication(ISAC)with massive MIMO system will be the development trend in the 6G wireless communication systems.ISAC refers to the design methodology and the corresponding technologies that integrate sensing and communication functionalities to achieve efficient usage of wireless resources and to mutually benefit each other.For massive multiple input multiple output(MIMO)ISAC system,there are many factors such as the interference between sensing and communication services,the computational complexity and hardware consumption,the trade-off between sensing and communication,and the physical layer security et al,which determine the performance of the system simultaneously,bringing many new challenges into the beamforming design.To this end,this paper focuses on the energy-and compute-efficient beamforming design methods,aiming to improve the system performance in the following three folds:(1)reducing the computational complexity,(2)improving the sensing and communication quality of service,(3)enhancing the physical layer security.The main research contents of this paper are summarized as follows.1.In the massive MIMO ISAC scheme,the high computational complexity and poor real-time performance prohibit the application of the full digital beamforming design method.To this end,a low-complexity sub-optimal beamforming iterative algorithm is proposed,where the closed-form expressions for both radar signal covariance and communication beamforming are obtained by relaxing the original problem.The proposed algorithm can adjust the sensing and communication performance in terms of the different requirements,and significantly reduce the computational complexity and memory consumption at the price of sacrificing a little performance.2.The hybrid analog and digital beamforming(HBF)algorithm for massive MIMO systems is studied to significantly reduce the required number of radio frequency chain,thereby reducing the power consumption and hardware complexity.For the multi-user interference,the modified two-stage HBF algorithm is proposed,where the analog beamforming is obtained by quasi-Newton method to achieve beamforming gain and the minimize mean square error(MMSE)criterion is used to eliminate the multi-user interference in the second stage.Compared to the state-of-the-art HBF algorithms,the proposed algorithm can improve the system’s achievable sum rate effectively.Furthermore,the low-resolution(2-bit)PS-based HBF algorithm is proposed to compensate the quantization error caused by low-resolution phase shifters.The candidate set with much fewer elements than the original one is constructed,to reduce the computational complexity of the exhausting search method from exponential to quadratic power,while guaranteeing satisfactory system performance and energy efficiency.3.To further enhance the energy of the received sign al,an electromagnetic(EM)lens is put at the front end to achieve energy focusing gain.However,the inherent power leakage problem will degrade the sensing and communication performance.To this end,a joint highresolution channel or direction-of-arrival(DoA)estimation and beamforming algorithm is proposed to tackle the power leakage problem.Specifically,the criterion function is constructed based on DFT beam difference to infer the real position of arrived signal,followed by a low-dimensional beamspace multiple signal classification(MUSIC)to estimate the DoA for each selected antenna group.The proposed algorithm can tackle the power leakage problem effectively,hence improving the sensing and communication performance.Furthermore,when the targets are in the near-field region of the array,the response of the lens antenna array is a window effect function with respect to DoA and distance.The parameter estimation method including time-delay and Doppler frequency estimates through the match filtering for dual-functional radar and communication(DFRC)signal,and the DoA estimates based on sparse Bayesian inference are proposed for targets localization.Subsequently,a communication served by sensing scheme based on the Kalman filter technique is proposed,where the formulated beam is predicted by the position at the last epoch and is updated according to the parameter estimation results.The proposed algorithm can significantly reduce the pilot overhead by circumventing the frequent pilot training and channel estimation.4.The ISAC base station transmits DFRC signal to detect the targets,while the communication symbols are also transmitted to the targets at the same time.It will cause communication security problems when the targets are potential eavesdroppers.To this end,the physical security beamforming design is also considered to protect the transmission of confidential information.Specifically,the weighted sum of radar and communication signals is used instead of the traditional artificial noise method for inflicting interference upon the eavesdropper,while additionally increasing the degrees of freedom for target detection.To compensate for the performance loss caused by the semidefinite relaxation(SDR)technique,the global optimal solution is constructed,which implies the relaxation procedure of the proposed algorithm is tight.Moreover,the zero-forcing constraint is imposed to reduce the computational complexity,where the zero-forcing based low-complexity algorithm is no performance loss at the high signal-to-noise ratio regime.Finally,the trade-off between target detection,communication quality of service,and physical layer security with different thresholds are theoretically analyzed and validated by numerical simulations.