Signal Processing Methods For Integarted Sensing And Communications

Traditional multifunctional electronic devices have employed a superimposed system structure for a long time.Since then,radio sensing and communication systems,as the two fundamental functional modules,have developed parallelly and independently,leading to a separated radio sensing and communication pattern.As a result,the dedicated sensing and communication functionalities in a device are lack integration and coordination,and the global coordination from underlying mechanisms to high-layer applications is absent in a cross-function or cross-device manner.Furthermore,with the development of mobile communications,such as the exploration of new functions in the vertical scenario in the 5G and forthcoming 6G networks,the technological progress of radiofrequency electronic devices and the ultra-high frequency/very high-frequency microwave devices,the commercial applications of the largescale antenna system,sensing and communication modules tend to be similar to their hardware architectures,channel characteristics,and signal processing approach.As a result,the boundary between sensing and communication modules is gradually blurred,and the above similarities provide a clear opportunity for the integration and coordination of sensing and communication modules.The integrating sensing and communication(ISAC)systems are capable of bringing up two potential gains,i.e.,the integration gain and coordination gain.The former is attained by the integrated design of sharing frequency resources,hardware devices,and transmitting a joint radar-communication waveform for more spectral efficiency and hardware efficiency.The latter is attained from the shared information between communication and sensing modules and a joint signal processing scheme.However,in practice,the deployment and implementation of the ISAC system face several critical challenges.(1)As two dominant wireless devices,radar and communication systems coexist in the same frequency band and interfere,leading to severe inter-system interference.(2)The performance limits of communication and sensing from an informationtheoretical perspective are unclear,and there are still many gaps in the research.(3)When the communication network is dominant,the sensing performance of the transmitted signal cannot meet the basic requirements of sense,and the approaches to enhancing the sensing performance in this scenario still need to be explored.(4)In the realistic scenario,the information from the surrounding environment is beneficial to improve communication performance,but there are still defects in the signal design.The aforementioned challenges motivate the investigation of the IS AC system framework,performance limits,and relevant signal processing methods.To solve these challenges,we employ the interference alignment,Degrees of Freedom theory,ambiguity function analysis,and optimization theory in this paper to implement the interference management based on the standard in 5G New Radio(5G NR).And we perform the asymptotic analysis of performance limits and investigate the method of coordination in IS AC systems.The main contributions are listed in the following five aspects.(1)A multi-carrier interference cancellation scheme in radar-communication systems based on the joint design of the precoder and decoder was proposed.This paper extended the traditional multi-carrier radar signal model to the interference channel between radar and communications.Further,a joint design of a multi-user precoder-decoder design was proposed based on the maximum signal-to-noise ratio(SNR)criterion and interference alignment constraints.Compared with the null-space precoder,radar users can achieve better detection performance and diversity gain.Simulation results demonstrated the significant performance gain of the proposed scheme.(2)An interference management scheme of MIMO radar and communication systems based on the collaborative design of the transceiver was proposed.Aiming at the difficulties in multiple-antenna ISAC systems,we proposed a rank minimization rank constraint transceiver design method improve the sensing performance and diversity gain,which maximized the available signal space of the sensing and communication channel with the guarantee of system diversity gain.If the decoder matrix is unitary and the interference alignment is perfect,the detection performance of radar is nearly consistent with the expected diversity gain when without interference.Subsequently,an efficient algorithm based on the idea of alternating iteration was proposed to obtain the solution to joint transceiver design more efficiently.At last,the mutual information maximization criterion was utilized to optimize the capacity of sensing and communication systems simultaneously,and the sub-optimal solution was obtained.In the simulation,when the SNR is relatively high,the designed precoder and decoder can suppress the interference effectively.The sum of the mutual information increases linearly with the increase of SNR.(3)The degree of freedom(DoF)limit of sensing and communication channel was derived,and the performance of fuzzy function was analyzed with NR standardized waveform.We utilize the DoF to depict the performance of interfering channels in the ISAC system in high SNR and proposed a radar DoF definition method based on Renyi information dimension.Simultaneously,we defined the relation between entropy and information dimension.The radar DoF controls the radar diversity gain,and the maximum DoF can bring up maximum available diversity gain with the Neman-Pearson detection strategy.When the targets obey the exponential distribution,the radar DoF equals the diversity gain.Moreover,the self-fuzzy function and cross-fuzzy function are employed to analyze the sensing performance with the 5G NR standardized structure,and the half-frame level and frame-level system-level simulation of NR waveform are conducted.The simulation results show that the current NR waveform can be reused for sensing while the accuracy cannot meet the daily needs.(4)We proposed a waveform design scheme in sensing-assisted communication systems.By analyzing the lower limit of mean squared error(MSE)of the sensed target in downlink ISAC systems,a waveform enhancement method based on DoF completion was proposed,which introduced additional signal structure to improve the performance of parameter estimation and achieved the lower limit of MSE.Subsequently,a waveform design scheme aimed at energy efficiency was explored in ISAC systems.In order to guarantee the performance of target estimation and improve the energy efficiency,the energy efficiency maximizing problem was formulated with the constraints of the Cramer-Rao bound,the Dinkelbach method was employed to solve the fractional programming problem and the semi-definite relaxation method was utilized to solve its sub-problem.(5)A predictable beamforming scheme in sensing-assisted communication systems was proposed.In order to decrease the influence of non-linear factors and improve the performance of angle estimation,we first proposed a predictive beamforrming scheme for vehicle-road cooperation based on a particle filter.Further,we designed a predictable beamforming scheme based on machine learning.Compared with the existing beamforming scheme,the proposed deep learning method can accurately learn the complex relationships between unknown variables.Furthermore,our proposed method can achieve better estimation performance due to utilizing exact values instead of approximate values.