Audio and Speech Processing
Phase Aware Ear-Conditioned Learning for Multi-Channel Binaural Speaker Separation
Ruben Johnson Robert Jeremiah, Peyman Goli and Steven van de Par
Separating competing
speech in reverberant environments requires models that preserve spatial cues while maintaining separation efficiency. We present a Phase-aware Ear-conditioned speaker Separation network using eight microphones (PEASE-8) that consumes complex STFTs and directly introduces a raw-STFT input to the early decoder layer, bypassing the entire encoder pathway to improve reconstruction. The model is trained end-to-end with an SI-SDR-based objective against direct-path ear targets, jointly performing separation and dereverberation for two speakers in a fixed azimuth, eliminating the need for permutation invariant training. On spatialized two-speaker mixtures spanning anechoic, reverberant, and noisy conditions, PEASE-8 delivers strong separation and intelligibility. In reverberant environments, it achieves 12.37 dB SI-SDR, 0.87 STOI, and 1.86 PESQ at T60 = 0.6 s, while remaining competitive under anechoic conditions.
Read the full abstract ▾
Signal Processing
A Dynamic Watermarking Technique for Matching Communication Addresses with Cars in a Visual Field
Woo-Hyun Ko, Jaewon Kim, Tzu-Hsiang Lin et al. (+2 more)
We consider a problem faced by an intelligent roadside unit (RSU) monitoring a roadway by a
video camera. Suppose the RSU notices that a particular car in its visual field needs to execute a specific evasive maneuver to avoid danger. It would like to send a packet addressed to that particular car with this suggestion. The problem is that while all the cars are communicating with the RSU, the RSU does not know which car in the
video is associated with what IP address. So, it does not know which IP address to send the packet to. Indeed, the problem of matching addresses with cars in the visual field is a fundamental open problem. We provide an active solution employing dynamic watermarking that was originally developed for the security of cyber-physical systems. This technique calls for a car to superpose a small random excitation onto its actuation commands for steering angle or throttle/brake positions. The car sends this random waveform to the RSU in a packet containing its IP address. By signal processing of the
video stream of a car at the RSU it can verify whether it matches with the waveform in the packet and thereby associates that the IP address of the packet with that car in the visual field. The RSU thereby determines which IP address is associated with which car in its visual field. We present two demonstrations of performance. We demonstrate experimental results on a laboratory transportation automated vehicles, a
vision system, and a network, as well as on the field with two passenger sedans in practice. The results demonstrate that employing the dynamic watermarking method enables an RSU to distinguish the communication of a target vehicle from that of other IP addresses of nearby vehicles.
Read the full abstract ▾
All authors Woo-Hyun Ko, Jaewon Kim, Tzu-Hsiang Lin, Samin Moosavi, P. R. Kumar
Systems and Control · Cross-listed from math.OC
Exponential convergence of multiagent systems with lack of connection
Fabio Ancona, Mohamed Bentaibi and Francesco Rossi
Finding conditions ensuring consensus, i.e. convergence to a common value, for a networked system is of crucial interest, both for theoretical reasons and applications. This goal is harder to achieve when connections between
agents are temporarily lost. Here, we prove that known conditions (introduced by Moreau) ensure an exponential convergence to consensus, with explicit rate of convergence. The key result is related to the length of the graph (i.e. the number of connections to reach a common
agent): if this is large, then convergence is slow. This general result also provides conditions for convergence of second-order cooperative systems with lack of connections.
Read the full abstract ▾
Systems and Control
Efficient LLM Inference over Heterogeneous Edge Networks with Speculative Decoding
Bingjie Zhu, Zhixiong Chen, Liqiang Zhao et al. (+2 more)
Large
language model (LLM) inference at the network edge is a promising serving paradigm that leverages distributed edge resources to run inference near users and enhance
privacy. Existing edge-based LLM inference systems typically adopt autoregressive decoding (AD), which only generates one
token per forward pass. This iterative process, compounded by the limited computational resources of edge nodes, results in high serving latency and constrains the system's ability to support multiple users under growing demands.To address these challenges, we propose a speculative decoding (SD)-based LLM serving framework that deploys small and large models across heterogeneous edge nodes to collaboratively deliver inference services. Specifically, the small model rapidly generates draft
tokens that the large model verifies in parallel, enabling multi-
token generation per forward pass and thus reducing serving latency. To improve resource utilization of edge nodes, we incorporate pipeline parallelism to overlap drafting and verification across multiple inference tasks. Based on this framework, we analyze and derive a comprehensive latency model incorporating both communication and inference latency. Then, we formulate a joint optimization problem for speculation length, task batching, and wireless communication resource allocation to minimize total serving latency. To address this problem, we derive the closed-form solutions for wireless communication resource allocation, and develop a dynamic programming
algorithm for joint batching and speculation control strategies. Experimental results demonstrate that the proposed framework achieves lower serving latency compared to AD-based serving systems. In addition,the proposed joint optimization method delivers up to 44.9% latency reduction compared to
benchmark schemes.
Read the full abstract ▾
All authors Bingjie Zhu, Zhixiong Chen, Liqiang Zhao, Hyundong Shin, Arumugam Nallanathan
Audio and Speech Processing · Cross-listed from cs.SD
Diffusion-Link: Diffusion Probabilistic Model for Bridging the Audio-Text Modality Gap
KiHyun Nam, Jongmin Choi, Hyeongkeun Lee et al. (+2 more)
Contrastive audio-
language pretraining yields powerful joint representations, yet a persistent audio-text modality gap limits the benefits of coupling
multimodal encoders with large
language models (LLMs). We present
Diffusion-Link, a
diffusion-based modality-bridging module that generatively maps audio
embeddings into the text-
embedding distribution. The module is trained at the output
embedding from the frozen
multimodal encoder and implemented as a lightweight network with three residual MLP blocks. To assess the effect of
Diffusion-Link on
multimodal encoder-LLM coupling, we evaluate on Automatic Audio Captioning (AAC); to our knowledge, this is the first application of
diffusion-based modality bridging to AAC. We report two results. (1) Modality-gap analysis: on similarity and
geometric criteria,
Diffusion-Link reduces the modality gap the most among prior
diffusion-based methods and shows a collective migration of audio
embeddings toward the text distribution. (2) Downstream AAC: attaching
Diffusion-Link to the same
multimodal LLM baseline achieves state-of-the-art on AudioCaps in both zero-shot and fully supervised captioning without external knowledge, with relative gains up to 52.5% and 7.5%, respectively. These findings show that closing the modality gap is pivotal for effective coupling between
multimodal encoders and LLMs, and
diffusion-based modality bridging offers a promising direction beyond knowledge-
retrieval-centric designs. Code will be released upon acceptance https://github.com/DevKiHyun/
Diffusion-Link
Read the full abstract ▾
All authors KiHyun Nam, Jongmin Choi, Hyeongkeun Lee, Jungwoo Heo, Joon Son Chung
Systems and Control
pyspect: An Extensible Toolbox for Automatic Construction of Temporal Logic Trees via Reachability Analysis
Kaj Munhoz Arfvidsson, Loizos Hadjiloizou, Frank J. Jiang et al. (+2 more)
In this paper, we present pyspect, a Python toolbox that simplifies the use of reachability analysis for temporal logic problems. Currently, satisfying complex requirements in cyber-physical systems requires significant manual effort and domain expertise to develop the underlying reachability programs. This high development effort limits the broader adoption of reachability analysis for complex verification problems. To address this, pyspect provides a method-agnostic approach to performing reachability analysis for verifying a temporal logic specification via temporal logic trees (TLTs). It enables the specification of complex safety and liveness requirements using high-level logic formulations that are independent of any particular reachability technique or set representation. As a result, pyspect allows for the comparison of different reachability implementations, such as Hamilton-Jacobi and Hybrid Zonotope-based reachability analysis, for the same temporal logic specification. This design separates the concerns of implementation developers (who develop numerical procedures for reachability) and end-users (who write specifications). Through a simple vehicle example, we demonstrate how pyspect simplifies the synthesis of reachability programs, promotes specification reusability, and facilitates side-by-side comparisons of reachability techniques for complex tasks.
Read the full abstract ▾
All authors Kaj Munhoz Arfvidsson, Loizos Hadjiloizou, Frank J. Jiang, Karl H. Johansson, Jonas Mårtensson
Signal Processing
Channel-Aware Deep Learning for Superimposed Pilot Power Allocation and Receiver Design
Run Gu, Renjie Xie, Wei Xu et al. (+2 more)
Superimposed pilot (SIP) schemes face significant challenges in effectively superimposing and separating pilot and data signals, especially in multiuser mobility scenarios with rapidly varying channels. To address these challenges, we propose a novel channel-aware learning framework for SIP schemes, termed CaSIP, that jointly optimizes pilot-data power (PDP) allocation and a receiver network for pilot-data interference (PDI) elimination, by leveraging channel path gain information, a form of large-scale channel state information (CSI). The proposed framework identifies user-specific, resource element-wise PDP factors and develops a deep neural network-based SIP receiver comprising explicit channel estimation and data detection components. To properly leverage path gain data, we devise an
embedding generator that projects it into
embeddings, which are then fused with intermediate feature maps of the channel estimation network. Simulation results demonstrate that CaSIP efficiently outperforms traditional pilot schemes and state-of-the-art SIP schemes in terms of sum throughput and channel estimation accuracy, particularly under high-mobility and low signal-to-noise ratio (SNR) conditions.
Read the full abstract ▾
All authors Run Gu, Renjie Xie, Wei Xu, Zhaohui Yang, Kaibin Huang
Systems and Control
Edge-to-Cloud Computations-as-a-Service in Software-Defined Energy Networks for Smart Grids
Jack Jackman, David Ryan, Arun Narayanan et al. (+2 more)
Modern power grids face an acute mismatch between where data is generated and where it can be processed: protection relays, EV (Electric Vehicle) charging, and distributed renewables demand millisecond analytics at the edge, while energy-hungry workloads often sit in distant clouds leading to missed real-time deadlines and wasted power. We address this by proposing, to our knowledge, the first-ever SDEN (Software Defined Energy Network) for CaaS (Computations-as-a-Service) that unifies edge, fog, and cloud compute with 5G URLLC (Ultra-Reliable Low-Latency Communications), SDN (Software Defined Networking), and NFV (Network Functions Virtualization) to co-optimize energy, latency, and reliability end-to-end. Our contributions are threefold: (i) a joint task offloading formulation that couples computation placement with network capacity under explicit URLLC constraints; (ii) a feasibility preserving, lightweight greedy heuristic that scales while closely tracking optimal energy and latency trade-offs; and (iii) a tiered AI (Artificial Intelligence) pipeline-reactive at the edge, predictive in the fog, strategic in the cloud-featuring
privacy-preserving,
federated GNNs (Graph Neural Networks) for fault detection and microgrid coordination. Unlike prior edge-only or cloud-only schemes, SDEN turns fragmented grid compute into a single, programmable substrate that delivers dependable, energy-aware, real time analytics establishing a first-ever, software defined path to practical, grid-scale CaaS.
Read the full abstract ▾
All authors Jack Jackman, David Ryan, Arun Narayanan, Pedro Nardelli, Indrakshi Dey
Signal Processing
Two-Dimensional Graph Bi-Fractional Fourier Transform
Mingzhi Wang and Zhichao Zhang
Graph signal processing (GSP) advances spectral analysis on irregular domains. However, existing two-dimensional graph fractional
Fourier transform (2D-GFRFT) employs a single fractional order for both factor graphs, thereby limiting its adaptability to heterogeneous signals. We proposed the two-dimensional graph bi-fractional
Fourier transform (2D-GBFRFT), which assigns independent fractional orders to the factor graphs of a Cartesian product while preserving separability. We established invertibility, unitarity, and index additivity, and developed two filtering schemes: a Wiener-style design through grid search and a differentiable framework that jointly optimizes transform orders and diagonal spectral filters. We further introduced a hybrid interpolation with the joint time-vertex fractional
Fourier transform (JFRFT), controlled by a tunable parameter that balances the two methods. In the domains of synthetic Cartesian product graph signals, authentic temporal graph
datasets, and dynamic image deblurring, 2D-GBFRFT consistently surpasses 2D-GFRFT and enhances JFRFT. Experimental results confirmed the versatility and superior performance of 2D-GBFRFT for filtering in GSP.
Read the full abstract ▾
Signal Processing · Cross-listed from cs.LG
Learning the Structure of Connection Graphs
Leonardo Di Nino, Gabriele D'Acunto, Sergio Barbarossa et al. (+1 more)
Connection graphs (CGs) extend traditional graph models by coupling network
topology with orthogonal transformations, enabling the representation of global
geometric consistency. They play a key role in applications such as synchronization, Riemannian signal processing, and neural sheaf
diffusion. In this work, we address the inverse problem of learning CGs directly from observed signals. We propose a principled framework based on maximum pseudo-likelihood under a consistency assumption, which enforces spectral properties linking the connection
Laplacian to the underlying combinatorial
Laplacian. Based on this formulation, we introduce the Structured Connection Graph Learning (SCGL)
algorithm, a block-optimization procedure over Riemannian
manifolds that jointly infers network
topology, edge weights, and
geometric structure. Our experiments show that SCGL consistently outperforms existing baselines in both topological recovery and
geometric fidelity, while remaining computationally efficient.
Read the full abstract ▾
All authors Leonardo Di Nino, Gabriele D'Acunto, Sergio Barbarossa, Paolo Di Lorenzo
Signal Processing
Normalized Ambiguity Function Characteristics of OFDM, OTFS, AFDM, and CP-AFDM for ISAC
Hyeon Seok Rou and Giuseppe Thadeu Freitas de Abreu
This paper presents a unified and system-agnostic analysis of the ambiguity function (AF) characteristics of four representative multicarrier waveforms, orthogonal frequency division multiplexing (OFDM), orthogonal time frequency space (OTFS), affine frequency division multiplexing (AFDM), and chirp-permuted AFDM (CP-AFDM), which are considered as key candidates for enabling integrated sensing and communications (
ISAC) in future sixth generation (6G) networks. The AF of each waveform is obtained directly from its discrete-time definition and enhanced via ideal fractional interpolation, enabling precise characterization of its continuous-time delay-Doppler response. Two signaling modes are examined: a communication-oriented case with random information symbols suitable only for monostatic scenarios, and a sensing-oriented case with fixed unimodular symbols suitable for general multi-static scenarios. Furthermore, the AFs and the ambiguity metrics including the 3dB mainlobe width, peak-to-sidelobe ratio (PSLR), and integrated sidelobe ratio (ISLR), are evaluated in normalized delay-Doppler units, enabling direct translation to any physical system configuration defined by bandwidth, sampling frequency, or symbol duration, while ensuring straightforward and consistent comparison across waveforms. The results establish a consistent
benchmark for comparing waveform sensing capabilities in
ISAC design, consolidating known behaviors: OFDM exhibits excellent delay resolution and sidelobe behavior but poor Doppler response, whereas advanced waveforms achieve improved balance between delay and Doppler resolution with varying sidelobe characteristics. The simulation code of the smooth AFs, is openly shared to promote reproducibility and support future
ISAC waveform research.
Read the full abstract ▾
Signal Processing
CSI Prediction Using Diffusion Models
Mehdi Sattari, Javad Aliakbari, Alexandre Graell i Amat et al. (+1 more)
Acquiring accurate channel state information (CSI) is critical for reliable and efficient wireless communication, but challenges such as high pilot overhead and channel aging hinder timely and accurate CSI acquisition. CSI prediction, which forecasts future CSI from historical observations, offers a promising solution. Recent deep learning approaches, including recurrent neural networks and
Transformers, have achieved notable success but typically learn deterministic mappings, limiting their ability to capture the
stochastic and
multimodal nature of wireless channels. In this paper, we introduce a novel probabilistic framework for CSI prediction based on
diffusion models, offering a flexible design that supports integration of diverse prediction schemes. We decompose the CSI prediction task into two components: a temporal encoder, which extracts channel dynamics, and a
diffusion-based generator, which produces future CSI samples. We investigate two inference schemes-autoregressive and sequence-to-sequence- and explore multiple
diffusion backbones, including U-Net and
Transformer-based architectures. Furthermore, we examine a
diffusion-based approach without an explicit temporal encoder and utilize the DDIM scheduling to reduce model complexity. Extensive simulations demonstrate that our
diffusion-based models significantly outperform state-of-the-art baselines.
Read the full abstract ▾
All authors Mehdi Sattari, Javad Aliakbari, Alexandre Graell i Amat, Tommy Svensson
Image and Video Processing
Generalisation of automatic tumour segmentation in histopathological whole-slide images across multiple cancer types
Ole-Johan Skrede, Manohar Pradhan, Maria Xepapadakis Isaksen et al. (+20 more)
Deep learning is expected to aid pathologists by automating tasks such as tumour
segmentation. We aimed to develop one universal tumour
segmentation model for histopathological images and examine its performance in different cancer types. The model was developed using over 20 000 whole-slide images from over 4 000 patients with colorectal, endometrial, lung, or prostate carcinoma. Performance was validated in pre-planned analyses on external cohorts with over 3 000 patients across six cancer types. Exploratory analyses included over 1 500 additional patients from The Cancer Genome Atlas. Average Dice coefficient was over 80% in all validation cohorts with en bloc resection specimens and in The Cancer Genome Atlas cohorts. No loss of performance was observed when comparing the universal model with models specialised on single cancer types. In conclusion, extensive and rigorous evaluations demonstrate that generic tumour
segmentation by a single model is possible across cancer types, patient populations, sample preparations, and slide scanners.
Read the full abstract ▾
All authors Ole-Johan Skrede, Manohar Pradhan, Maria Xepapadakis Isaksen, Tarjei Sveinsgjerd Hveem, Ljiljana Vlatkovic, Arild Nesbakken, Kristina Lindemann, Gunnar B Kristensen, Jenneke Kasius, Alain G Zeimet, Odd Terje Brustugun, Lill-Tove Rasmussen Busund, Elin H Richardsen, Erik Skaaheim Haug, Bjørn Brennhovd, Emma Rewcastle, Melinda Lillesand, Vebjørn Kvikstad, Emiel Janssen, David J Kerr, Knut Liestøl, Fritz Albregtsen, Andreas Kleppe
Systems and Control
Utilizing Bayesian Optimization for Timetable-Independent Railway Junction Performance Determination
Tamme Emunds, Paul Brunzema, Sebastian Trimpe et al. (+1 more)
The efficiency of railway infrastructure is significantly influenced by the mix of trains that utilize it, as different service types have competing operational requirements. While freight services might require extended service times, passenger services demand more predictable schedules. Traditional methods for addressing long-term traffic assignment problems often rely on fixed-value capacity limitations, determined based on specific assumptions about traffic composition. This paper introduces a methodology for determining timetable-independent capacity within the traffic rate assignment problem, enabling the calculation of junction capacities under dynamic traffic distributions. We solve the underlying non-linear constrained optimization problem maximizing the traffic throughput using
Bayesian optimization (BO). This setting combines a known objective function with expensive- to-compute capacity constraints, motivating an adaption of standard BO problems, where objective functions are usually unknown. We tailor the acquisition process in BO to this specific setting and increase performance by incorporating prior knowledge about the shape of the constraint functions into the Gaussian process surrogate model. Our derived approaches are benchmarked on a railway junction near Paris, significantly outperforming fixed traffic composition models and highlighting the benefits of dynamic capacity allocation.
Read the full abstract ▾
All authors Tamme Emunds, Paul Brunzema, Sebastian Trimpe, Nils Nießen
Signal Processing
WiNPA: Wireless Neural Processing Architecture
Sai Xu and Yanan Du
This article presents a wireless neural processing architecture (WiNPA), providing a novel perspective for accelerating edge inference of deep neural network (DNN) workloads via joint optimization of wireless and computing resources. WiNPA enables fine-grained integration of wireless communication and edge computing, bridging the research gap between wireless and edge intelligence and significantly improving DNN inference performance. To fully realize its potential, we explore a set of fundamental research issues, including mathematical modeling, optimization, and unified hardware–software platforms. Additionally, key research directions are discussed to guide future development and practical implementation. A case study demonstrates WiNPA's workflow and effectiveness in accelerating DNN inference through simulations.
Read the full abstract ▾
Systems and Control · Cross-listed from cs.NI
Visible Light Communication for Vehicular Networks: A Tutorial
Pedro E. Gória Silva, Eduardo S. Lima, Jules M. Moualeu et al. (+2 more)
The advent of the fifth-generation technology promises to bring about more vertical applications and emerging services that include vehicular networks and intelligent transportation systems (ITSs). To achieve their
vision of real-time and safetyapplications, vehicular networks rely on short-range to medium-range communications. One emerging technology that aims to provide reliability and high-data rate in short-range communications is the visible light communications (VLC). Due to its remarkable advantages, some studies have recently investigated the integration of VLC in vehicular networks and ITSs. Despite their attractive features, such networks also face several implementation issues. This paper provides an extended tutorial on the implementation of VLC-based vehicular networks. To begin with, we present the implementation characteristics of these systems and discuss some related issues. The underlying system considers a general structure with transmitters, channels, and receivers based on photodetectors and cameras, as well as standardization efforts and types of
topologies. In addition, we discuss the impact of the sun and artificial light sources, flickering, dimming, throughput enhancement, uplink security, and mobility on practical implementation. Finally, we highlight some key challenges and potential solutions and provide some directions for future research investigations that could constitute an advancement toward the development of commercial VLC-based vehicular systems.
Read the full abstract ▾
All authors Pedro E. Gória Silva, Eduardo S. Lima, Jules M. Moualeu, Mohamed Korium, Pedro H. J. Nardelli
Signal Processing
Navigating the Dual-Use Nature and Security Implications of Reconfigurable Intelligent Surfaces in Next-Generation Wireless Systems
Hetong Wang, Tiejun Lv, Yashuai Cao et al. (+4 more)
Reconfigurable intelligent surface (RIS) technology offers significant promise in enhancing wireless communication systems, but its dual-use potential also introduces substantial security risks. This survey explores the security implications of RIS in next-generation wireless networks. We first highlight the dual-use nature of RIS, demonstrating how its communication-enhancing capabilities can be exploited by adversaries to compromise legitimate users. We identify a new class of security vulnerabilities termed “passive-active hybrid attacks,” where RIS, despite passively handling signals, can be reconfigured to actively engage in malicious activities, enabling various RIS-assisted attacks, such as eavesdropping, man-in-the-middle (MITM), replay, reflection jamming, and side-channel attacks. Furthermore, we reveal how adversaries can exploit the openness of wireless channels to introduce
adversarial perturbations in artificial intelligence-driven RIS networks, disrupting communication terminals and causing misclassifications or errors in RIS reflection predictions. Despite these risks, RIS technology also plays a critical role in enhancing security and
privacy across
radio frequency (RF) and visible light communication (VLC) systems. By synthesizing current insights and highlighting emerging threats, we provide actionable insights into cross-layer collaboration, advanced
adversarial defenses, and the balance between security and cost. This survey provides a comprehensive overview of RIS technology's security landscape and underscores the urgent need for robust security frameworks in the development of future wireless systems.
Read the full abstract ▾
All authors Hetong Wang, Tiejun Lv, Yashuai Cao, Weicai Li, Jie Zeng, Pingmu Huang, Muhammad Khurram Khan
Signal Processing
The Post-Electromagnetic Era: A Vision for Wireless Communication Beyond 6G
Shumaila Javaid and Nasir Saeed
Electromagnetic (EM) communication is nearing its physical and thermodynamic limits, where further performance gains through spectrum optimization alone have become increasingly unsustainable. Finite bandwidth, propagation loss at higher frequencies, and the inherent trade-offs between energy and information constrain the scalability of 6G and beyond systems. These limitations drive the search for alternative mechanisms for information transfer beyond conventional EM propagation. This work introduces a state-centric framework for post-6G communication, in which information is conveyed by manipulating physical, biological, and cognitive states rather than EM waves. It identifies ten foundational paradigms that define potential carriers and interaction mechanisms for the post-electromagnetic era and outlines a research roadmap toward self-organizing, cognitively integrated networks. Together, these developments envision a new class of communication systems that are energy-aware, adaptive, and capable of uniting matter, life, and intelligence within a single informational continuum. By establishing the conceptual basis for this transition, the work provides a foundation for future research aimed at realizing communication paradigms that transcend the limitations of spectrum-bound systems.
Read the full abstract ▾
Systems and Control
Establishing assembly-oriented modular product architectures through Design for Assembly enhanced Modular Function Deployment
Fabio Marco Monetti, Adam Lundström, Colin de Kwant et al. (+2 more)
Modular product design has become a strategic enabler for companies seeking to balance product variety, operational efficiency, and
market responsiveness, making the
alignment between modular architecture and manufacturing considerations increasingly critical. Modular Function Deployment (MFD) is a widely adopted method for defining modular product architectures, yet it lacks systematic support for assembly considerations during early concept and system-level development. This limitation increases the risk of delayed production ramp-up and lifecycle inefficiencies. This paper proposes a set of enhancements to MFD that integrate Design for Assembly (DFA) logic into architectural synthesis. The extended method introduces structured heuristics, assembly-oriented module drivers, a coded interface taxonomy, and quantitative metrics for assessing assembly feasibility and automation readiness. These additions preserve compatibility with standard MFD workflows while enriching decision-making with traceable, production-informed
reasoning. An illustrative case study involving a handheld leaf blower demonstrates the method's usability and effectiveness. The redesigned architecture shows reduced assembly effort, simplified interfaces, and increased automation potential. By supporting early-stage evaluation of architectural alternatives through an assembly lens, the method enables faster transition to efficient volume production and provides a foundation for continuous improvement throughout the product lifecycle.
Read the full abstract ▾
All authors Fabio Marco Monetti, Adam Lundström, Colin de Kwant, Magnus Gyllenskepp, Antonio Maffei
Systems and Control · Cross-listed from cs.RO
PhysHSI: Towards a Real-World Generalizable and Natural Humanoid-Scene Interaction System
Huayi Wang, Wentao Zhang, Runyi Yu et al. (+10 more)
Deploying
humanoid robots to interact with real-world environments–such as carrying objects or sitting on chairs–requires generalizable, lifelike motions and robust scene perception. Although prior approaches have advanced each capability individually, combining them in a unified system is still an ongoing challenge. In this work, we present a physical-world
humanoid-scene interaction system, PhysHSI, that enables
humanoids to autonomously perform diverse interaction tasks while maintaining natural and lifelike behaviors. PhysHSI comprises a simulation training pipeline and a real-world deployment system. In simulation, we adopt
adversarial motion prior-based policy learning to imitate natural
humanoid-scene interaction data across diverse scenarios, achieving both generalization and lifelike behaviors. For real-world deployment, we introduce a coarse-to-fine object localization module that combines
LiDAR and camera inputs to provide continuous and robust scene perception. We validate PhysHSI on four representative interactive tasks–box carrying, sitting, lying, and standing up–in both simulation and real-world settings, demonstrating consistently high success rates, strong generalization across diverse task goals, and natural motion patterns.
Read the full abstract ▾
All authors Huayi Wang, Wentao Zhang, Runyi Yu, Tao Huang, Junli Ren, Feiyu Jia, Zirui Wang, Xiaojie Niu, Xiao Chen, Jiahe Chen, Qifeng Chen, Jingbo Wang, Jiangmiao Pang
Audio and Speech Processing · Cross-listed from cs.LG
Efficient Test-Time Adaptation through Latent Subspace Coefficients Search
Xinyu Luo, Jie Liu, Kecheng Chen et al. (+6 more)
Real-world deployment often exposes
vision models to distribution shifts, making test-time adaptation (TTA) important for reliable prediction. Existing TTA methods can be difficult to deploy on edge devices because they require backpropagation, activation storage, or test-time batches, which increase computation,
memory, and state. We propose ELaTTA (Efficient Latent Test-Time Adaptation), a forward-only, gradient-free framework for single-instance TTA. ELaTTA freezes the pretrained model and adapts each test input by searching a low-dimensional coefficient vector in a principal latent subspace computed offline from a source-domain cache. The basis is stored once and source data are discarded before deployment. At test time, CMA-ES performs covariance-adaptive Gaussian search, ranks candidate corrections by prediction
entropy, and selects the best candidate within a fixed evaluation budget. On five visual
benchmarks spanning corruption, natural, rendition, sketch, and cross-domain shifts, ELaTTA provides strong accuracy under strict and continual single-instance protocols. In the strict single-instance comparison, MEMO requires \emph{64.7$\times$} the computation and \emph{12.4$\times$} the peak
memory of ELaTTA. A complementary keyword-spotting experiment shows that the same coefficient-search design transfers beyond
vision. We further demonstrate ViT-Base adaptation on a Raspberry Pi 5, confirming the feasibility of deploying ELaTTA on an edge device.
Read the full abstract ▾
All authors Xinyu Luo, Jie Liu, Kecheng Chen, Yihua Shao, Junyi Yang, Bo Ding, Nicu Sebe, Arindam Basu, Haoliang Li
Signal Processing · Cross-listed from physics.med-ph
Basis for a hands free blood flow measurement with automated vessel focus
Reinhard Fuchs, Nathalie Sumrah, Johannes Schwerdt et al. (+5 more)
Cardiopulmonary resuscitation (CPR) is an essential tool to ensure oxygen supply during cardiac arrest, yet not quantifiable to this day. Low-quality chest compressions or wrong pressure placement go unnoticed. This paper presents a solution for the quantification of blood flow to guide first responders in their efforts. An approach for automated vessel identification with three different steps was developed, featuring a new sensor probe for ultrasonic measurements with non-symmetrically angled piezo ceramics. The probe was used with prototype ultrasound hardware for Pulsed Wave Doppler (PW Doppler) in a phantom. Initial measurements evaluated sensor vessel
alignment at different sensor positions by examining Doppler results with a large sample volume during periodic flow. Afterward, an iterative mode was used for depth-dependent frequency measurements with score calculation of flow periodicity and power. The configuration with the best score was used for a prolonged monitoring mode. Initial mode and iterative mode aligned with ultrasound imaging regarding the best position and vessel depth. Simultaneous flow-sensor data and flow values of monitoring mode calculated via Doppler substitution showed a minimum correlation coefficient of 0.98, a minimum R2 of 0.96, and an average root mean square error of 3.84 ml/s. With the proposed hardware and software solutions, a basis for future developments was made, which could lead to a fully automated vessel identification and blood flow calculation during CPR. When used in emergencies, a miniaturized device could provide vital information about CPR efficiency that has yet to be included in the therapy of people during cardiac arrest.
Read the full abstract ▾
All authors Reinhard Fuchs, Nathalie Sumrah, Johannes Schwerdt, Michael Unger, Georg Stachel, Michael Schultz, Karsten Lenk, Thomas Neumuth
Signal Processing · Cross-listed from cs.LG
Robust Photoplethysmography Signal Denoising via Mamba Networks
I Chiu, Yu-Tung Liu, Kuan-Chen Wang et al. (+2 more)
Photoplethysmography (PPG) is widely used in wearable health monitoring, but its reliability is often degraded by noise and motion artifacts, limiting downstream applications such as heart rate (HR) estimation. This paper presents a deep learning framework for PPG denoising with an emphasis on preserving physiological information. In this framework, we propose DPNet, a Mamba-based denoising backbone designed for effective temporal modeling. To further enhance denoising performance, the framework also incorporates a scale-invariant signal-to-distortion ratio (SI-SDR) loss to promote waveform fidelity and an auxiliary HR predictor (HRP) that provides physiological consistency through HR-based supervision. Experiments on the BIDMC
dataset show that our method achieves strong robustness against both synthetic noise and real-world motion artifacts, outperforming conventional filtering and existing neural models. Our method can effectively restore PPG signals while maintaining HR accuracy, highlighting the complementary roles of SI-SDR loss and HR-guided supervision. These results demonstrate the potential of our approach for practical deployment in wearable healthcare systems.
Read the full abstract ▾
All authors I Chiu, Yu-Tung Liu, Kuan-Chen Wang, Hung-Yu Wei, Yu Tsao
Signal Processing · Cross-listed from cs.LG
Conformal Inference for Time Series over Graphs
Sonakshi Dua, Gonzalo Mateos and Sundeep Prabhakar Chepuri
Trustworthy decision making in networked, dynamic environments calls for innovative uncertainty quantification substrates in predictive models for graph time series. Existing conformal prediction (CP) methods have been applied separately to multivariate time series and static graphs, but they either ignore the underlying graph
topology or neglect temporal dynamics. To bridge this gap, here we develop a CP-based sequential prediction region framework tailored for graph time series. A key technical innovation is to leverage the graph structure and thus capture pairwise dependencies across nodes, while providing user-specified coverage guarantees on the predictive outcomes. We formally establish that our scheme yields an exponential shrinkage in the volume of the ellipsoidal prediction set relative to its graph-agnostic counterpart. Using real-world
datasets, we demonstrate that the novel uncertainty quantification framework maintains desired empirical coverage while achieving markedly smaller (up to 80% reduction) prediction regions than existing approaches.
Read the full abstract ▾
Signal Processing
Dual-Waveguide Pinching Antennas for PLS: Parallel Placement or Orthogonal Placement?
Yang Lu, Xinke Xie, Yanqing Xu et al. (+3 more)
Pinching
antennas (PAs) represent an emerging flexible-
antenna paradigm, where movable PAs arranged along dielectric waveguides enable on-demand customization of wireless channel characteristics. This paper exploits PAs to enhance physical-layer security (PLS) by increasing the channel diversity gap between legitimate users (LUs) and eavesdroppers (Eves). We concentrate on a dual-waveguide PA system, in which each waveguide is equipped with multiple PAs to serve multiple LUs under wiretapping from an Eve. Two waveguide placement schemes are studied: parallel placement and orthogonal placement. To capture practical waveguide propagation effects, we adopt two in-waveguide channel models: in-waveguide phase shifts, and in-waveguide phase shifts and attenuation. We formulate secure sum rate (SSR) and secrecy energy efficiency (SEE) maximization problems, and develop a two-stage optimization framework to tackle these non-convex problems. In the first stage, we design a feasibility-enhanced particle swarm optimization (PSO)
algorithm named FeaPSO to search for valid PA positions complying with minimum inter-PA spacing constraints. The second stage leverages successive convex approximation (SCA) to jointly optimize baseband beamforming and artificial noise vectors. Simulation results validate the effectiveness of the proposed two-stage
algorithm and confirm that PA adjustment yields substantial gains in both SSR and SEE. Moreover, the simulation outcomes demonstrate the performance benefits brought by orthogonal waveguide placement for PLS enhancement.
Read the full abstract ▾
All authors Yang Lu, Xinke Xie, Yanqing Xu, Bo Ai, Octavia A. Dobre, Arumugam Nallanathan
Image and Video Processing
Bit Allocation Transfer for Perceptual Quality Enhancement of VVC Intra Coding
Runyu Yang and Ivan V. Bajić
Mainstream image and
video coding standards – including state-of-the-art codecs like H.266/VVC, AVS3, and AV1 – adopt a block-based hybrid coding framework. While this framework facilitates straightforward optimization for Peak Signal-to-Noise Ratio (PSNR), it struggles to effectively optimize perceptually-aligned metrics such as Multi-Scale Structural Similarity (MS-SSIM). To address this challenge, this paper proposes a low-complexity method to enhance perceptual quality in VVC intra coding by transferring bit allocation knowledge from end-to-end image compression. We introduce a lightweight model trained with perceptual losses to generate a quantization step map. This map implicitly captures block-level perceptual importance, enabling efficient derivation of a QP map for VVC. Experiments on Kodak and CLIC
datasets demonstrate significant advantages, both in execution time and perceptual metric performance, with more than 11% BD-rate reduction in terms of MS-SSIM. Our scheme provides an efficient, practical pathway for perceptual enhancement of traditional codecs.
Read the full abstract ▾
Audio and Speech Processing · Cross-listed from cs.SD
Unify Variables in Neural Scaling Laws for General Audio Representations via Embedding Effective Rank
Xuyao Deng, Yanjie Sun, Yong Dou et al. (+1 more)
Scaling laws have profoundly shaped our understanding of model performance in computer
vision and natural
language processing, yet their application to general audio representation learning remains underexplored. A key challenge lies in the multifactorial nature of general audio representation-representation quality is jointly influenced by variables such as audio length,
embedding dimensionality, model depth, model architecture, data volume, etc., many of which are difficult to isolate or express analytically. In this work, we present a systematic study of scaling laws for general audio representations by utilizing
embedding effective rank (RankMe) as a unifying metric that encapsulates the impact of diverse variables on representation quality. RankMe enables a label-free, information-theoretic quantification of audio
embeddings, allowing us to examine scaling behaviors across a wide hyper-parameter space, including model size, training data volume, computational budget, architectural configurations, etc. Our empirical findings reveal a consistent power-law relationship between RankMe and representation quality, suggesting that
embedding effective rank serves as a reliable proxy for assessing and predicting model performance in audio representation learning. This work not only validates the applicability of classical scaling principles to the general audio domain but also offers a theoretically grounded and empirically robust framework for guiding future model scaling strategies in audio foundation models.
Read the full abstract ▾
All authors Xuyao Deng, Yanjie Sun, Yong Dou, Kele Xu
Signal Processing
Spatial Signal Focusing and Noise Suppression for Direction-of-Arrival Estimation in Large-Aperture 2D Arrays under Demanding Conditions
Xuyao Deng, Yong Dou and Kele Xu
Direction-of-Arrival (DOA) estimation in sensor arrays faces limitations under demanding conditions, including low signal-to-noise ratio, single-snapshot scenarios, coherent sources, and unknown source counts. Conventional beamforming suffers from sidelobe interference, adaptive methods (e.g., MVDR) and subspace
algorithms (e.g., MUSIC) degrade with limited snapshots or coherent signals, while sparse-recovery approaches (e.g., L1-SVD) incur high computational complexity for large arrays. In this article, we construct the concept of the optimal spatial filter to solve the DOA estimation problem under demanding conditions by utilizing the sparsity of spatial signals. By utilizing the concept of the optimal spatial filter, we have transformed the DOA estimation problem into a solution problem for the optimal spatial filter. We propose the Spatial Signal Focusing and Noise Suppression (SSFNS)
algorithm, which is a novel DOA estimation framework grounded in the theoretical existence of an optimal spatial filter, to solve for the optimal spatial filter and obtain DOA. Through experiments, it was found that the proposed
algorithm is suitable for large aperture two-dimensional arrays and experiments have shown that our proposed
algorithm performs better than other
algorithms in scenarios with few snapshots or even a single snapshot, low signal-to-noise ratio, coherent signals, and unknown signal numbers in two-dimensional large aperture arrays.
Read the full abstract ▾
Systems and Control
Optimal Multi-Modal Transportation and Electric Power Flow: The Value of Coordinated Dynamic Operation
Jiajie Qiu, Dakota Thompson, Kamal Youcef-Toumi et al. (+1 more)
The electrification of transportation represents a critical challenge in the global transition toward net-zero emissions, as the sector often accounts for more than one-quarter of national energy consumption. Achieving this transformation requires not only widespread adoption of electric vehicles (EVs) but also their seamless integration into interdependent infrastructure systems-specifically, the transportation-electricity nexus (TEN). This paper develops an optimal multi-modal transportation and electric power flow (OMTEPF) model to evaluate the benefits of coordinated, dynamic system operation. Building on recent advances in hetero-functional graph theory, the framework enables joint optimization of five key operational decisions in intelligent TEN management: vehicle dispatch, route choice, charging station queuing, coordinated charging, and vehicle-to-grid stabilization. The mesoscopic, dynamic model explicitly represents individual EVs and their state-of-charge trajectories, thereby extending beyond the prevailing literature's focus on static, macroscopic traffic assignment. It further captures the full scope of the TEN as a system-of-systems, incorporating five distinct charging modalities: private residential, private commercial, wired public commercial, inductive public, and discharging. On the power system side, an IV-ACOPF formulation ensures globally optimal solutions to the electrical subproblems. Comparative analysis demonstrates the substantial value of coordinated TEN operation relative to the status quo of siloed, uncoordinated infrastructure management. This work provides both a novel methodological contribution and actionable insights for the co-design and operation of next-generation sustainable mobility-energy systems.
Read the full abstract ▾
All authors Jiajie Qiu, Dakota Thompson, Kamal Youcef-Toumi, Amro M. Farid
Audio and Speech Processing · Cross-listed from cond-mat.mtrl-sci
Delayed 1T to 2H Phase Transition Upon Electrochemical Delithiation of LiMoS2
Yerin Hong, Juhwan Lim, Jinhong Min et al. (+4 more)
Molybdenum disulfide (MoS2) is a widely studied layered material for electronic, optical, and catalytic applications. It can host lithium ions between the van der Waals layers, which triggers a phase transition between the semiconducting 2H phase and metallic 1T phase. While lithium insertion triggers a phase transition to the 1T phase, the phase behavior upon electrochemical lithium removal is not resolved. In this work, we conduct single-flake electrochemical (de)lithiation of MoS2 using microelectrode arrays. Through both electrochemical voltage analysis and correlative Raman spectroscopy, we show that an electrochemically cycled and delithiated MoS2 flake initially remains in the 1T phase. However, over the course of several days, it transitions back into the thermodynamically stable 2H phase. This result resolves the phase transformation pathway upon delithiation and showcases the ability to electrochemically synthesize the metastable 1T-MoS2 phase.
Read the full abstract ▾
All authors Yerin Hong, Juhwan Lim, Jinhong Min, Nishkarsh Agarwal, Robert Hovden, Ageeth A. Bol, Yiyang Li