General Overview

This collection of preprints explores diverse advancements in signal processing, communication, and sensing, with a notable emphasis on distributed systems and intelligent surfaces. Several papers address challenges in distributed tracking and sensor fusion, where unknown correlations between tracks hinder traditional Kalman filtering. Sharma et al. introduce harmonic mean density (HMD) fusion in two separate preprints (Sharma et al., 2024a; Sharma et al., 2024b), demonstrating its efficacy in handling both Gaussian and Gaussian mixture densities and offering an alternative implementation based on importance sampling. They highlight HMD's ability to achieve faster convergence and consistency compared to existing methods like covariance intersection (CI) while also addressing the complexities of Gaussian mixture fusion. This work opens promising avenues for more robust and efficient track fusion in complex, real-world scenarios. Furthermore, Kost et al. (Kost et al., 2024) propose novel techniques for noise covariance matrix identification in time-varying state-space models, utilizing a generalized measurement difference method (MDM) with varying weighting schemes for improved estimation quality and computational efficiency.

Another prominent theme is the application of reconfigurable intelligent surfaces (RIS) in various communication systems. Hou et al. (Hou et al., 2024) tackle the challenge of dynamic channel prediction in massive MIMO systems with temporal non-stationarity, proposing a tensor-structured approach that leverages dual-timescale and cross-domain correlations. Sokal et al. (Sokal et al., 2024) address the complex channel estimation problem in beyond diagonal RIS (BD-RIS), introducing a decoupled channel estimation method based on Khatri-Rao Factorization (KRF) to improve accuracy. Li et al. (Li et al., 2024) propose RIS-aided distributed MIMO radar systems, leveraging amplitude-controlled beamforming for enhanced multi-target detection. Nwalozie et al. (Nwalozie et al., 2024) further explore double RIS-aided MIMO systems, proposing interference-free channel training and estimation protocols based on coupled tensor decomposition. These contributions collectively advance the understanding and practical implementation of RIS technology for improved communication and sensing capabilities.

Several papers focus on specific applications and methodologies within wireless communication and sensing. Kang et al. (Kang et al., 2024) investigate MIMO detection under hardware impairments, employing data augmentation with boosting to enhance likelihood function estimation. Park et al. (Park et al., 2024) introduce a ViT-based semantic communication system with importance-aware quantization for efficient image transmission. Chemingui et al. (Chemingui et al., 2024) propose an EMF-aware architecture for uplink communications in MU-MIMO networks, utilizing aerial RIS technology to minimize EMF exposure. John et al. (John et al., 2024) provide a comprehensive review of multisensor data fusion for wearable health monitoring, discussing various fusion techniques and their applications. These works demonstrate the breadth of research in applying advanced signal processing techniques to real-world problems.

Beyond communication and sensing, the collection also includes works on specific signal processing techniques and applications. Dunik et al. present two papers on state estimation (Dunik et al., 2024a; Dunik et al., 2024b), exploring data-augmented numerical integration and stochastic integration-based estimators, respectively. Vu Thanh (Vu Thanh, 2024) investigates low-rank matrix factorizations with volume-based constraints and regularizations for enhanced interpretability and uniqueness. Wang et al. (Wang et al., 2024) analyze the Cramér-Rao bound and propose beamforming designs for 3D extended target sensing in ISAC systems, employing both optimization and learning approaches. These contributions offer valuable insights into fundamental signal processing techniques and their applications in diverse fields.

Finally, several papers explore novel applications and methodologies in specific domains. Crilly (Crilly, 2024) presents a simplified method for signal discovery in interstellar communication using interferometer measurements. Giannakopoulos & Shaikh (Giannakopoulos & Shaikh, 2024) discuss the design of multiband monopole and microstrip patch antennas using HFSS simulation. Kong et al. (Kong et al., 2024) introduce a reconfigurable calibration-free digital-to-time converter based on a high-speed transceiver. These diverse contributions showcase the ongoing innovation in applying signal processing principles to a wide range of challenging problems. Overall, this collection of preprints highlights the significant progress being made in signal processing, communication, and sensing, with a focus on distributed systems, intelligent surfaces, and data-driven approaches.

Paper Highlights

Paving the Way for 6G: A Unified Channel Model

New Characteristics and Modeling of 6G Channels: A Unified Channel Model towards Standardization by Huiwen Gong, Jianhua Zhang, Yuxiang Zhang, Guangyi Liu https://arxiv.org/abs/2412.07336

Caption: The image presents the core equation of the Extended Geometry-Based Stochastic Model (E-GBSM) for 6G channel modeling, highlighting key parameters that capture the unique characteristics of 6G technologies like XL-MIMO, ISAC, and RIS. These parameters, including modifications to cluster numbers (N₁, N_s), SnS (S), near-field effects (A), intra-cluster K-factor, and RCS/radiation pattern (F^Tar), are incorporated to address the limitations of existing 5G models. The equation demonstrates how the channel is modeled as a combination of sub-channels, incorporating target/RIS characteristics and near-field effects for a comprehensive representation of the 6G channel.

The arrival of 6G heralds a new era in wireless communication, bringing with it groundbreaking technologies like Integrated Sensing and Communication (ISAC), Extremely Large MIMO (XL-MIMO), and Reconfigurable Intelligent Surfaces (RIS), along with the utilization of broader frequency bands. These advancements, while transformative, present significant challenges for channel modeling and standardization. Existing 5G models, primarily based on the Geometry-Based Stochastic Model (GBSM), struggle to encapsulate the unique characteristics of these emerging 6G technologies. This necessitates a unified and adaptable model for effective research, performance evaluation, and the seamless deployment of 6G.

This paper directly addresses this challenge by proposing an extended GBSM (E-GBSM) framework. This innovative model integrates the distinctive features of each 6G technology while maintaining backward compatibility with existing 5G models, ensuring a smooth transition between generations. The E-GBSM tackles key 6G channel characteristics, including the Radar Cross Section (RCS) of ISAC targets, channel sparsity in new mid-band and higher frequencies, near-field effects and spatial non-stationarity (SnS) in XL-MIMO, and multi-segment concatenated channels in RIS systems.

The E-GBSM achieves this comprehensive modeling by introducing several new parameters and modifying existing ones within the established GBSM framework. These include parameters for frequency-dependent cluster numbers (N₁), shared clusters in ISAC (N_s), SnS (S^Tx/Rx), near-field effects (A^Tx/Rx_n,m), intra-cluster K-factor, and RCS/radiation patterns. The channel between intermediate nodes (like an ISAC target or RIS) and the transmitter/receiver is modeled as:

h^Tx/Rx_p,x(in/out) = Σ^N₁+N_s_n Σ^M₁_m A^Tx/Rx_n,m h^Tx/Rx_n,m(τ) * F^Tar(n_out, n_in) dn_out dn_in

The overall Tx-Target (or RIS)-Rx channel is modeled as a convolution of the two sub-channels, incorporating the target/RIS characteristics. This detailed modeling approach allows for a more accurate representation of the complex interactions within 6G channels.

The implementation of the E-GBSM is thoroughly detailed, building upon the existing 3GPP channel model implementation process. Modifications include updates to existing steps, the addition of new steps (e.g., generating SnS and near-field parameters for XL-MIMO, configuring target RCS for ISAC), and parameter updates (e.g., frequency-dependent cluster numbers, intra-cluster K-factor). A simulation platform, BUPTCMCCCMG-IMT2023, was developed based on the E-GBSM, providing a valuable tool for researchers and engineers. Simulations using this platform have validated the effectiveness of the E-GBSM, demonstrating its ability to accurately capture the unique characteristics of 6G channels. For example, the model accurately captured the RCS characteristics of a UAV target at different frequencies and the SnS behavior of XL-MIMO channels, showing close alignment with ray-tracing simulations. The impact of RIS configuration on received SNR was also successfully demonstrated, with continuous phase shift configurations outperforming other configurations. Furthermore, the model accurately captured the increasing channel sparsity with higher frequencies, as measured by the Gini Index, with values at 6 GHz and 13 GHz aligning closely with measured data. These results highlight the E-GBSM’s potential to serve as a crucial tool in the development and standardization of 6G technologies.

Rydberg Atoms Tune In to the Future of Wireless

Rydberg Atomic Quantum Receivers for Classical Wireless Communications and Sensing: Their Models and Performance by Tierui Gong, Jiaming Sun, Chau Yuen, Guangwei Hu, Yufei Zhao, Yong Liang Guan, Chong Meng Samson See, Mérouane Debbah, Lajos Hanzo https://arxiv.org/abs/2412.05554

Caption: Comparison of Receive SNR for RAQRs and Conventional RF Receiver

Quantum sensing is revolutionizing measurement science, offering unprecedented precision in detecting physical quantities. Among the emerging technologies, Rydberg atomic quantum receivers (RAQRs) stand out for their exceptional sensitivity to electric fields, making them promising candidates for both classical wireless communications and sensing applications. This paper presents a comprehensive model for superheterodyne RAQRs, effectively bridging the gap between the intricate physics of quantum systems and the practical considerations of wireless system design. While previous experimental studies have showcased the potential of RAQRs, a robust and detailed signal model has been lacking, hindering theoretical analysis and system optimization.

This research develops a closed-form expression for the density coefficient related to the crucial |1⟩→|2⟩ transition in a four-level Rydberg atom system, based on realistic assumptions. This density coefficient is directly linked to the probe beam, enabling precise characterization of its amplitude and phase, and ultimately providing a detailed evaluation of the RAQR's transfer function. The model incorporates the effects of the probe, coupling, and local oscillator (LO) beams, including their detuning frequencies and powers, offering a comprehensive platform for joint optimization of these critical parameters. A detailed end-to-end reception scheme is presented, outlining each functional block and culminating in an equivalent baseband signal model. This model reveals that the RAQR applies a gain Φ and phase shift to the baseband transmit signal, aligning with conventional RF receiver models and facilitating seamless integration with established signal processing techniques.

Noise analysis is a fundamental aspect of receiver system design. This research considers both extrinsic noise sources, such as black-body radiation-induced thermal noise, and intrinsic noise mechanisms within the RAQR itself. These intrinsic noise sources include photodetector shot noise (PSN), intrinsic thermal noise (ITN), and quantum projection noise (QPN). The model reveals that PSN often dominates, while QPN can be mitigated by increasing the number of Rydberg atoms involved in the detection process. Two photodetection schemes are analyzed: direct incoherent optical detection (DIOD) and balanced coherent optical detection (BCOD). The receive signal-to-noise ratio (SNR) is derived for both schemes, showing that BCOD generally outperforms DIOD, particularly when detuning parameters are optimized. The SNR ratio for the PSN-dominated case is given by:

Ratio₁ = (κ₂(Ω_l)/κ₁(Ω_l))² = 1 + (ℛ{χ'(Ω_l)}/ℐ{χ'(Ω_l)})² > 1

where κ₁(Ω_l) and κ₂(Ω_l) are related to the DIOD and BCOD schemes respectively, Ω_l is the Rabi frequency of the LO, and χ'(Ω_l) is the derivative of the susceptibility of the atomic vapor medium. This detailed analysis provides valuable insights into the noise characteristics of RAQRs and guides the selection of optimal detection schemes.

Finally, the performance of RAQRs is compared against conventional RF receivers employing a half-wavelength dipole antenna. A lower bound condition for the RAQR gain ρ is derived to ensure superior performance. Simulations validate the accuracy of the proposed model and demonstrate that even a modestly configured RAQR can achieve significant SNR gains over traditional receivers. Specifically, gains of ~22.6 dB and ~33.5 dB were observed in standard and optimized configurations, respectively. These results highlight the transformative potential of RAQRs in the future of wireless communications and sensing.

Sending Secrets in Plain Sight: New Semantic Communication Approach Uses Images to Carry Digital Signals

Semantic Communications for Digital Signals via Carrier Images by Zhigang Yan, Dong Li https://arxiv.org/abs/2412.07173

Caption: This diagram illustrates a novel Semantic Communication (SemCom) framework that uses images as carriers for digital signals. Digital data is encoded into the masking pattern of an image, processed by a Masked Autoencoder (MAE), and transmitted as a latent representation along with a sparsely encoded mask matrix. The receiver reconstructs both the image and the embedded digital signal using the MAE decoder.

Semantic Communication (SemCom) has emerged as a promising paradigm for efficient information transmission by prioritizing the meaning of data over its raw form. However, current SemCom frameworks primarily focus on data types with inherent semantic features, such as text and images, leaving the transmission of semantically-bare digital signals largely unaddressed. This research introduces a novel SemCom approach that leverages images as carrier signals for digital data, effectively embedding information within the visual content.

The proposed method ingeniously integrates Masked Autoencoders (MAEs), a deep learning model known for its ability to reconstruct masked images, into the SemCom framework. Digital signals are first converted into binary streams, which then determine the masking pattern applied to an image. The MAE encoder processes the visible patches of the masked image, generating a compact latent representation while also identifying the indices of the masked patches. Critically, instead of transmitting the full set of mask tokens, which can incur significant overhead, a sparse encoding module compresses these indices into a sparse matrix, further enhancing transmission efficiency. At the receiver, the MAE decoder utilizes the received latent representation and the recovered sparse matrix to reconstruct both the original image and the embedded digital signal. This innovative approach effectively addresses the challenge of transmitting digital signals within a SemCom framework without requiring a complete system overhaul.

The researchers rigorously evaluated the performance of their approach using ImageNet-1K images and randomly generated binary signals, transmitting over both AWGN and Rayleigh fading channels. Image reconstruction quality was assessed using Peak Signal-to-Noise Ratio (PSNR) and Multi-Scale Structural Similarity (MS-SSIM), while Bit Error Rate (BER) measured the accuracy of digital signal transmission. The results demonstrated notable improvements in PSNR and MS-SSIM compared to baseline MAE models, especially in high-SNR regimes. Moreover, the proposed method exhibited robustness to varying lengths of transmitted binary data, indicating its adaptability to different data size requirements. The BER performance also improved with increasing SNR, showcasing the resilience of transmitting the sparse matrix over wireless channels.

A key advantage of this method lies in its efficiency. The transmission overhead, measured in bits, was significantly lower compared to traditional methods and existing SemCom frameworks. The number of transmitted bits for the proposed approach is given by: $\frac{W \times H}{P \times P} \times (1-M_r)L_e + L_m$, where H and W are the image dimensions, P is the patch size, Mr is the mask ratio, Le is the number of bits for the latent representation, and Lm is the number of bits for the sparse matrix. This formula highlights a counterintuitive yet powerful result: increasing the length of the binary data (and consequently the mask ratio) actually reduces the overall transmission overhead. This efficiency gain arises from the MAE's ability to effectively handle high mask ratios, combined with the sparse encoding of mask tokens. This innovative approach opens new avenues for SemCom by extending its application to a broader range of data types. The clever use of images as carriers, coupled with the efficient encoding of masking information, provides a robust and efficient solution for transmitting digital signals within the semantically-rich world of SemCom.

Conclusion

This newsletter highlights significant advancements across several key areas in signal processing, communication, and sensing. The development of a unified channel model (E-GBSM) marks a crucial step towards standardizing 6G, addressing the unique challenges posed by technologies like ISAC, XL-MIMO, and RIS. The E-GBSM's ability to incorporate parameters for near-field effects, spatial non-stationarity, and RCS, while maintaining backward compatibility with 5G, positions it as a foundational tool for future 6G research and deployment. Shifting from the macro to the quantum realm, the introduction of a comprehensive model for Rydberg atomic quantum receivers (RAQRs) opens exciting possibilities for ultra-sensitive wireless communication and sensing. The demonstrated SNR gains of RAQRs over conventional RF receivers, coupled with the detailed noise analysis and optimization strategies presented, underscore the potential of this technology to revolutionize receiver design. Finally, the novel SemCom approach utilizing images as carrier signals for digital data offers a clever solution for transmitting semantically-bare information within a meaning-centric framework. The integration of MAEs and sparse encoding techniques results in a highly efficient system capable of robustly transmitting both images and embedded digital signals. Collectively, these advancements showcase the ongoing innovation and convergence of diverse fields, driving progress towards more efficient, robust, and intelligent communication and sensing systems.