Research On Power Allocation Of Integrated Sensing And Communication Based On OTFS

Integrated sensing and communication(ISAC)technology can simultaneously realize radar detection and communication transmission on the basis of sharing hardware and software resources.Thus,the spectrum efficiency,hardware efficiency and information processing efficiency of the system are effectively improved..In recent years,orthogonal time frequency space(OTFS)modulation has attracted widespread attention in 6G wireless communications.Therefore,OTFS is considered to realize the full potential of the ISAC system.This thesis mainly studies power allocation technology based on OTFS-ISAC.The specific research content and innovation points are as follows:Based on the above inherent advantages of the new OTFS system,a new framework for the design of ISAC waveforms based on OTFS is proposed,and a system model for ISAC is established,which breaks through the principle that a signal cannot be accurately represented in both the time domain and the frequency domain at the same time.For the Doppler frequency shift and delay extension of high-speed targets,the signal waveform of ISAC is jointly designed directly in the delay-Doppler domain,which can realize the functions of sensing and communication for high-speed maneuvering targets at the same time.Furthermore,the waveform design of the OTFS-ISAC system is carried out,and the application scenario of communicating with a single user in the downlink and realizing the detection target is analyzed.On the basis of this model,a further proposed in the delay Doppler domain A new method of ISAC power distribution is proposed.The power allocation scheme proposed in this thesis is to solve the problem that the perceived waveform and the waveform used for communication conflict with each other in the delay-Doppler domain,and to minimize the Cramer-Rao Lower Bound(CRLB)of sensing parameter under the condition of satisfying the communication rate constraint.Since this problem is a non-convex problem,the block coordinate descent method is used to optimize the non-convex problem.The simulation results show that in the high-speed mobile scene,compared with the traditional OFDM-based power allocation scheme,the proposed power allocation scheme based on OTFS is better than that of OFDM.Finally,the OTFS-ISAC power allocation problem for sensing-assisted communication is studied.Considering that with the continuous movement of the user,the channel state information is constantly changing,and the sensing function can obtain information such as the user’s speed and distance as accurately as possible,thereby improving the communication quality.This thesis adopts a power allocation scheme based on deep reinforcement learning,establishes the relationship between communication power and sensing power,and reconstructs the original problem into a Markov decision process,which can better interact with the changing environment,so that Adjust the power distribution according to the current environment.The function of sensing-assisted communication is realized,and the final simulation results verify that the average reachable rate obtained by the power allocation scheme based on deep reinforcement learning proposed in this thesis has good convergence.