Abstract:We propose a novel deep learning based method to design a coded waveform for integrated sensing and communication (ISAC) system based on orthogonal frequency-division multiplexing (OFDM). Our ultimate goal is to design a coded waveform, which is capable of providing satisfactory sensing performance of the target while maintaining high communication quality measured in terms of the bit error rate (BER). The proposed LISAC provides an improved waveform design with the assistance of deep neural networks for the encoding and decoding of the information bits. In particular, the transmitter, parameterized by a recurrent neural network (RNN), encodes the input bit sequence into the transmitted waveform for both sensing and communications. The receiver employs a RNN-based decoder to decode the information bits while the transmitter senses the target via maximum likelihood detection. We optimize the system considering both the communication and sensing performance. Simulation results show that the proposed LISAC waveform achieves a better trade-off curve compared to existing alternatives.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is to design an integrated sensing and communication (ISAC) waveform that can simultaneously satisfy good sensing performance and high communication quality. Specifically, the authors propose a deep - learning - based method, called LISAC (Learned Coded Waveform Design for ISAC with OFDM), to optimize the design of orthogonal frequency - division multiplexing (OFDM) waveforms. The following is a detailed description of this problem: ### 1. **Problem Background** The integrated sensing and communication (ISAC) system aims to achieve communication and radar sensing functions through the same waveform, which has important application prospects in future 6G networks. However, traditional waveform design methods are difficult to optimize these two functions simultaneously because their requirements for waveforms are conflicting: - **Sensing Performance**: The ideal sensing waveform should have a constant amplitude to maximize the estimation accuracy. - **Communication Performance**: The ideal communication waveform should follow a random Gaussian distribution to improve the channel capacity. ### 2. **Research Objectives** The objective of this paper is to design a coded waveform that can provide satisfactory sensing performance while maintaining high communication quality (measured by bit error rate (BER)). To this end, the authors propose the LISAC model, which uses a deep neural network (DNN) to encode and decode information bits and optimize the waveform to balance communication and sensing performance. ### 3. **Specific Problems** - **Waveform Design Challenges**: How to design a waveform that can effectively sense targets and ensure high - quality communication? - **Performance Trade - off**: How to find the optimal trade - off point between sensing performance and communication performance? - **Deep Learning Applications**: How to use deep learning techniques (such as RNN) to design and optimize such waveforms? ### 4. **Solutions** The LISAC model solves the above problems in the following ways: - **Encoder Design**: Use a recurrent neural network (RNN) to encode the input bit sequence into a transmission waveform, which is used for sensing and communication. - **Decoder Design**: The receiving end also adopts an RNN decoder to decode the information bits, while the transmitting end senses the target through maximum - likelihood detection. - **Joint Optimization**: Optimize the performance of the entire system by alternately training the encoder and decoder and introducing loss functions (including the abnormal mean - square error (MSE) of sensing performance and the cross - entropy loss of communication performance). ### 5. **Experimental Verification** Through simulation experiments, the authors show that the LISAC waveform can achieve a better trade - off between sensing and communication performance under different parameter settings (such as different λ values, different block lengths C, etc.), and exhibits better performance compared to existing methods. In summary, this paper aims to design an ISAC waveform that can simultaneously optimize sensing and communication performance through deep learning techniques, thereby promoting the development of future 6G networks.

LISAC: Learned Coded Waveform Design for ISAC with OFDM

Short-Length Code Designs for Integrated Sensing and Communications Using Deep Learning

A Dual-Functional Sensing-Communication Waveform Design Based on OFDM

End-to-End Learning for SLP-Based ISAC Systems

Integrated Sensing and Communication Waveform Design With Sparse Vector Coding: Low Sidelobes and Ultra Reliability

A Power-Domain Non-Orthogonal Integrated Sensing and Communication Waveform Design Using OFDM

Dual-Functional Waveform Design with Local Sidelobe Suppression via OTFS Signaling

DL-based Joint Waveform and Beamforming Design for Integrated Sensing and Communication.

MIMO-OFDM ISAC Waveform Design for Range-Doppler Sidelobe Suppression

Waveform Design and Performance Analysis for Full-Duplex Integrated Sensing and Communication

Interference-Resilient OFDM Waveform Design with Subcarrier Interval Constraint for ISAC Systems

OFDM Waveform Design With Subcarrier Interval Constraint for Narrowband Interference Suppression in ISAC Systems

Low-Complexity Minimum BER Precoder Design for ISAC Systems: A Delay-Doppler Perspective

Coherent Compensation based ISAC Signal Processing for Long-range Sensing

Cross-Domain Multicarrier Waveform Design for Integrated Sensing and Communication

Triangular FM-OFDM Waveform Design for Integrated Sensing and Communication

Integrated Sensing and Communications Waveform Design for OTFS and FTN Fusion

Low-Range-Sidelobe Waveform Design for MIMO-OFDM ISAC Systems

Waveform Design for OFDM-based ISAC Systems Under Resource Occupancy Constraint

Sensing Integrated DFT-Spread OFDM Waveform and Deep Learning-Powered Receiver Design for Terahertz Integrated Sensing and Communication Systems

Waveform Design for Communication-Assisted Sensing in 6G Perceptive Networks