# Radar Published articles for Radar. This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles. ## 4D Point Splatting for mmWave Radar Micro-Doppler Differentiable Rendering DevFeed: [4D Point Splatting for mmWave Radar Micro-Doppler Differentiable Rendering]() Original publisher: [Read original article]() Author: Adnan Armouti, Guanyue Qian, Rajalakshmi Nandakumar Published: 2026-11-06T05:00:00Z Content type: article Language: en Sources: [arXiv: Machine Learning]() Topics: [3D rendering](), [Machine Perception](), [Artificial Intelligence](), [GPU rendering](), [data augmentation](), [3D](), [gaussian splatting]() Tags: [cs-cv](), [cs-gr](), [cs-lg](), [cuda](), [data-augmentation](), [dataset](), [datasets](), [doppler](), [eess-sp](), [features](), [gap](), [generation](), [mmwave](), [point](), [radar](), [range](), [raw](), [rendering](), [signal-processing](), [synthesis](), [synthetic]() ### AI overview 4D Point Splatting (4DPS) fits an explicit moving scene to real mmWave radar captures, aiming to narrow the sim2real gap that forward simulators, learned generators, and static-scene inverse renderers do not address. It initializes room and body geometry from synchronized camera data, represents the scene as oriented points with material, normal, and velocity, and splats them into range and Doppler bins to produce raw complex radar returns. The same signal processing can then create representations for downstream models. Fitting 235 HuPR sequences and 100 RT-Pose sequences yielded median correlations of 0.940/0.830 for range-azimuth, 0.829/0.695 for range-Doppler, and 0.988/0.966 for micro-Doppler. On RT-Pose, a pose model trained on 4DPS renders achieved 21.9 cm MPJPE on real data, compared with 16.7 cm for real-data training and 127.6 cm for the RF-Genesis baseline. A sequence fit took 5.7 minutes on HuPR and 12.9 minutes on RT-Pose using one H100. ### Source excerpt arXiv:2611.08342v1 Announce Type: cross Abstract: Millimeter-wave radar resolves the micrometer displacements of a moving body as micro-Doppler, enabling gait, gesture, activity, pose and vital-sign sensing. Each task relies on a downstream model trained on real data, yet radar datasets lag orders of magnitude behind vision and language. Brute-force collection will not close that gap, and existing synthesis methods (forward-only simulators, learned generators and static-scene inverse renderers) cannot fit a real capture of a moving body, leaving a sim2real gap. We present 4D Point Splatting (4DPS), a physics-based differentiable renderer that fits an explicit moving scene to a real capture. Using the synchronized camera to initialize the room and body, we represent the scene as oriented points carrying material, normal and velocity, and splat each point into range and Doppler bins. Our output is the raw complex return a real radar records, so we apply the same signal processing on the rendered data to obtain any representation for training any SOTA downstream model. We fit 235 sequences on HuPR and 100 on RT-Pose, two datasets with different radars. Our reconstructions set the state of the art, with median correlations of 0.940 and 0.830 on range-azimuth, 0.829 and 0.695 on range-Doppler, and 0.988 and 0.966 on micro-Doppler. Our renders also set the state of the art in downstream sim2real gap. On RT-Pose, a pose model trained on them reaches 21.9 cm MPJPE on real data, against 16.7 cm for real training and 127.6 cm for the RF-Genesis baseline, where both the rendered- and real-trained models learn nearly the same features. Our custom CUDA kernels fit a sequence in 5.7 minutes on HuPR and 12.9 on RT-Pose on one H100, making dataset-scale inverse rendering of mmWave radar possible for the first time. This work can enable synthetic data generation for any human-centered downstream task and model, and scene editing for data augmentation. ## Autonomous combat drone maker Shield AI buys radar startup Echodyne for $5.3B DevFeed: [Autonomous combat drone maker Shield AI buys radar startup Echodyne for $5.3B]() Original publisher: [Read original article]() Author: Mike Wheatley Published: 2026-11-06T01:13:14Z Content type: news Language: en Sources: [SiliconANGLE]() Topics: [autonomous vehicles](), [Artificial Intelligence]() Tags: [acquisition](), [advanced-radar](), [autonomous](), [autonomous-drones](), [autonomous-systems](), [autonomous-weapons](), [autonomy](), [defense](), [defense-technology](), [drone](), [drone-radar](), [drone-software](), [echodyne](), [emerging-tech](), [gary-steele](), [mesa-radar](), [news](), [radar](), [shield-ai](), [startup](), [the-latest](), [v-bat-drones](), [x-bat]() ### AI overview Shield AI agreed to acquire Echodyne for $5.3 billion to combine Echodyne's MESA solid-state radar, designed to detect small targets such as drones, with Shield AI's autonomous systems. The deal includes $4.8 billion at closing and $500 million tied to undisclosed performance targets; Shield AI says it will fund it with its own cash and proceeds from a $2.5 billion Series H round. Echodyne will remain a standalone unit in Shield AI's mission systems business, with CEO Eben Frankenberg staying on; the companies anticipate closer radar integration with Shield's drones over time. The article also notes Echodyne's recent $250 million U.S. defense contract for drone tracking. ### Source excerpt Defense technology startup Shield AI Inc. said today it has agreed to buy the Bill Gates-backed radar startup Echodyne Inc. in a massive deal valued at $5.3 billion. The acquisition is intended to enhance Shield's autonomous drone warfighting capabilities, it said. Echodyne is a 13-year-old startup that's best known for its Metamaterials Electronically Scanned Array [...] The post Autonomous combat drone maker Shield AI buys radar startup Echodyne for $5.3B appeared first on SiliconANGLE. ## Camera Radar LiDAR Fusion via Point-Cloud Projection for Multi-Task Learning DevFeed: [Camera Radar LiDAR Fusion via Point-Cloud Projection for Multi-Task Learning]() Original publisher: [Read original article]() Author: Kavin Chandrasekaran, Sorin Grigorescu Published: 2026-11-04T05:00:00Z Content type: article Language: en Sources: [arXiv: Artificial Intelligence](), [arXiv: Machine Learning]() Topics: [autonomous vehicles](), [Machine learning](), [AI agents in autonomous vehicles]() Tags: [autonomous](), [autonomous-vehicles](), [camera](), [computational-complexity](), [cs-ai](), [cs-cv](), [cs-lg](), [dataset](), [experiments](), [feature](), [free](), [fusion](), [learning](), [lidar](), [light](), [measurements](), [model](), [models](), [performance](), [perspective](), [point](), [projection](), [radar]() ### AI overview The study reports that camera images fused in front view with perspective-projected HD radar and LiDAR point clouds outperform recent standalone and fusion models on the RADIal dataset. Its early-fusion architecture jointly detects vehicles and segments free drivable space. The abstract motivates fusion by contrasting camera detail with radar reliability in poor conditions and LiDAR's precise distance measurements, while noting the sensors' respective limitations. It does not provide performance figures or further experimental details. ### Source excerpt arXiv:2611.03942v1 Announce Type: cross Abstract: Autonomous vehicles require robust perception systems to operate safely under diverse environmental conditions. While cameras provide high-resolution visual data, their performance significantly degrades in low-light and adverse weather. High Definition (HD) radar offers superior reliability and angular resolution under such conditions but lacks rich visual feature recognition. LiDAR, though expensive, delivers highly precise distance measurements and dense point clouds. Each sensor exhibits unique trade-offs, necessitating the development of advanced sensor fusion algorithms to harness complementary strengths while minimizing computational complexity and maintaining accuracy. In this study, the proposed fusion architecture performs early fusion by combining camera images with perspective-view point cloud data from HD radar and LiDAR sensors. The proposed model is trained to detect vehicles and segment free drivable space simultaneously. Experiments are performed on the RADIal dataset where we demonstrate that fusing point cloud data with camera images in front-view surpasses the performances of the most recent standalone and fusion models. ## A Universal Raw-IQ Radar Encoder for Cross-Sensor Human Sensing DevFeed: [A Universal Raw-IQ Radar Encoder for Cross-Sensor Human Sensing]() Original publisher: [Read original article]() Author: Jonas Leo Mueller, Dario Zanca, Sebastian Hoefler, Jan Petermann, Thomas Altstidl, Naga Venkata Sai Jitin Jami, Emmanuelle Salin, Bjoern M. Eskofier Published: 2026-11-04T05:00:00Z Content type: article Language: en Sources: [arXiv: Machine Learning]() Topics: [AI Foundation Models](), [Artificial Intelligence]() Tags: [abstract](), [all](), [angular](), [arxiv](), [based](), [benchmark](), [benchmarks](), [best](), [beta](), [carrier](), [choice](), [code](), [compare](), [costs](), [covering](), [cross](), [cs-cv](), [cs-lg](), [radar](), [raw](), [sensing](), [sensor]() ### AI overview The paper reports that a frozen encoder pretrained without labels on clutter-removed raw radar measurements transfers across sensors and human-sensing tasks. Its convolutional β-VAE uses measurements from four platforms spanning 60 and 77 GHz and 4-12 virtual antennas, aligned on a shared spectral interpolation grid; each benchmark trains only a small supervised temporal classifier on the encoder's embeddings. Across ten settings, it leads on all five benchmarks when test users and radar are represented in training, and on three of four for unseen users, by 4.8-23.2 points. For subject-disjoint gait recognition, it leads for each of four held-out subjects and averages 8.3 points above the strongest baseline, despite pretraining excluding that radar. Excluding the test radar from pretraining reduces performance by at most 5.3 points on any benchmark. These are abstract-level reported results; the supplied text does not include experimental details. ### Source excerpt arXiv:2611.04534v1 Announce Type: cross Abstract: Radar-based human sensing typically relies on end-to-end task-specific models trained on hand-chosen signal projections, which generalize poorly across unseen users, tasks, and hardware. We argue that the dominant signal processing choice, the micro-Doppler spectrogram, is a poor starting point for transfer because it keeps the Doppler axis, which carrier frequency rescales, and discards the angular axis, which carrier frequency leaves unchanged. We instead pretrain a convolutional beta-VAE without labels on clutter-removed raw radar measurements from four platforms spanning 60 and 77 GHz and 4 to 12 virtual antennas, aligned by a unified spectral interpolation grid. For every downstream benchmark the encoder stays frozen, and only a small supervised temporal classifier is trained on its embeddings. We compare against supervised models trained end to end on each benchmark and a pretrained radar foundation model across ten settings covering activity, gesture, sign-language, and gait recognition. When the test users and the radar are seen during training, our encoder is best on all five benchmarks. For users never seen in training, it leads on three of four benchmarks, by 4.8 to 23.2 points. On subject-disjoint gait recognition, it is best for each of the four held-out subjects and 8.3 points above the strongest baseline on average, even though its pretraining excluded every recording of this radar. Excluding the test radar from pretraining costs at most 5.3 points on any benchmark. Code is available at https://github.com/jonasmueler/UniversalRadarEncoder_WACV. ## Introducing Zero Wire, a hub for investigating critical Internet vulnerabilities DevFeed: [Introducing Zero Wire, a hub for investigating critical Internet vulnerabilities]() Original publisher: [Read original article]() Author: Celso Martinho Published: 2026-11-03T14:00:00Z Content type: release Language: en Sources: [Cloudflare Blog]() Topics: [cve remediation](), [Cloudflare](), [Security research](), [cybersecurity and infrastructure security agency]() Tags: [actions](), [active-exploitation](), [agent](), [ai](), [ai-agent](), [applications](), [assessment](), [attack](), [basic](), [blocking](), [cisa](), [cloudflare](), [critical](), [cve](), [cves](), [how-to](), [infrastructure](), [internet](), [product-news](), [radar](), [remediation](), [security](), [threat-intelligence](), [vulnerabilities](), [waf]() ### AI overview Zero Wire is a public vulnerability intelligence feed on Cloudflare Radar focused on significant critical CVEs in Internet-facing software. Its briefings explain impact, who should investigate, current exploitation evidence, remediation guidance, and relevant Cloudflare controls, with links to vendor advisories and other sources. Readers can use a supplied prompt for a read-only, evidence-based assessment of their own infrastructure. Briefings distinguish severity from observed exploitation, and traffic matching WAF rules is not proof of compromise or a count of successful exploits. A WAF rule, where available, is a compensating control requiring configuration review; it does not replace vendor patches or undo actions already taken at the origin. Status, last-checked dates, and update chronologies show how guidance changes, while "Reference" status does not mean every affected system is remediated. ### Source excerpt Zero Wire on Cloudflare Radar pairs plain-language briefings with global attack telemetry and remediation steps. Triage critical CVEs faster and decide how to protect your infrastructure. ## Communication-Efficient Model-Distributed Inference for CBRS Radar Detection at the Edge DevFeed: [Communication-Efficient Model-Distributed Inference for CBRS Radar Detection at the Edge]() Original publisher: [Read original article]() Author: Teng Li, Hulya Seferoglu Published: 2026-11-03T05:00:00Z Content type: article Language: en Sources: [arXiv: Distributed Computing]() Topics: [AI Infrastructure](), [Local AI](), [Data transmission](), [Jetson Orin](), [communications]() Tags: [5g](), [cbrs](), [communication](), [compression](), [cs-dc](), [data](), [detection](), [device](), [distributed](), [distributed-inference](), [edge](), [eess-sp](), [i-q](), [inference](), [jetson](), [jetson-orin](), [jetson-orin-nano](), [pipeline](), [quality](), [radar](), [radio](), [raw]() ### AI overview CEDAR reduces communication costs in edge-distributed CBRS radar detection by redesigning both the detector and model partitioning. CEDAR-YOLO removes YOLO's multi-scale neck and two detection scales, using a single-scale path with independently configured frequency and time resolutions. CEDAR-MDI adds lightweight activation compressors and chooses each compressor's partition to minimize the pipeline bottleneck. On an NVIDIA Jetson Orin Nano with real-world over-the-air data, CEDAR-YOLO reduced parameters by 35-57% and inference time by 16-33%; CEDAR-MDI achieved up to 3.27x the throughput of single-device inference. The abstract reports preserved detection quality for both, but provides no further evaluation details. ### Source excerpt arXiv:2611.00625v1 Announce Type: new Abstract: The Citizens Broadband Radio Service (CBRS) shares the 3550-3700 MHz band between incumbent radar systems and commercial LTE/5G users, so radar activity must be detected accurately and promptly. Spectrogram-based YOLO detectors can localize coexisting radar and cellular signals, but running them on resource-constrained edge devices is challenging. Model-distributed inference (MDI) can increase throughput by partitioning a model across edge devices. Using an end-to-end pipeline from raw I/Q samples to on-device detection, we show that partitioning recent YOLO generations incurs high activation-transfer costs due to their multi-scale structures. To reduce these costs, we propose CEDAR, a communication-efficient, edge-distributed architecture for radar detection. Its RF-aware detector, CEDAR-YOLO, removes YOLO's multi-scale neck and two of its three detection scales, leaving a single-scale path with independently configured frequency and time resolutions, which lowers the communication cost of partitioning. Its compression-aware MDI framework, CEDAR-MDI, integrates lightweight activation compressors and selects, for each compressor, the partition that minimizes the pipeline bottleneck. On an NVIDIA Jetson Orin Nano testbed with real-world over-the-air data, CEDAR-YOLO reduces the parameter count by 35-57% and model inference time by 16-33%, and CEDAR-MDI achieves up to $3.27\times$ the throughput of single-device inference, with detection quality preserved in both cases. ## MazeBolt launches native RADAR deployment for Akamai DDoS protection customers DevFeed: [MazeBolt launches native RADAR deployment for Akamai DDoS protection customers]() Original publisher: [Read original article]() Author: Duncan Riley Published: 2026-11-02T14:00:45Z Content type: news Language: en Sources: [SiliconANGLE]() Topics: [DDoS]() Tags: [ai](), [akamai](), [akamai-technologies](), [api-s](), [cloudflare](), [configuration-gaps](), [cybersecurity](), [ddos](), [deployment](), [distributed-denial-of-service](), [eyal-rahimi](), [f5](), [financial-services](), [john-grady](), [launches](), [matthew-andriani](), [mazebolt](), [mazebolt-technologies](), [native](), [news](), [omdia](), [radar](), [radware](), [security](), [security-validation](), [the-latest](), [vectorai]() ### AI overview MazeBolt's RADAR now offers native Layer 7 testing for Akamai DDoS protection customers. It runs simulated attacks from within a customer's environment without taking services offline, then provides remediation guidance tailored to the protection vendor and retests after fixes. The article says testing addresses gaps that can emerge as services, APIs, and regions are added; MazeBolt reports that about 37% of attack vectors passed existing protection in initial deployments. Its VectorAI module includes AI-orchestrated and AI-generated attack vectors. The Akamai support is available to all RADAR customers; the article also says RADAR tests Cloudflare, F5, and Radware products across scrubbing centers, on-premises equipment, and cloud services. These findings and product claims are attributed to MazeBolt and the excerpt does not provide independent validation. ### Source excerpt Distributed denial-of-service testing company MazeBolt Technologies Ltd. today launched a native deployment of its RADAR validation platform for customers of Akamai Technologies Inc. Enterprises using Akamai's DDoS protection can have continuous testing running within hours without taking services offline. The release targets the configuration gaps that open up as a company adds services, application programming interfaces [...] The post MazeBolt launches native RADAR deployment for Akamai DDoS protection customers appeared first on SiliconANGLE. ## Beamfocusing Beyond the Near Field: Frequency Diverse Arrays for OFDM-ISAC Systems DevFeed: [Beamfocusing Beyond the Near Field: Frequency Diverse Arrays for OFDM-ISAC Systems]() Original publisher: [Read original article]() Author: Arda Akyazi, Ahmed Hussain, Asmaa Abdallah, Ahmed M. Eltawil, Abdulkadir Celik Published: 2026-11-02T05:00:00Z Content type: article Language: en Sources: [arXiv: Networking]() Topics: [communications](), [synchronization]() Tags: [arrays](), [beam](), [coexistence](), [communication](), [cs-it](), [cs-ni](), [eess-sp](), [math-it](), [radar](), [sensing](), [users]() ### AI overview Frequency diverse arrays (FDAs) use offsets across antenna elements to form range-dependent beams whose finite depth can extend into the far field; unlike phased arrays, their range selectivity is not limited to near-field spherical wavefront curvature. The excerpt explains that integer-multiple offsets align with OFDM subcarriers and avoid offset-induced intercarrier interference under ideal synchronization, while still requiring suitable symbol mapping and receiver processing. It distinguishes FDA beamdepth from waveform-based radar range resolution. In the studied configurations, FDAs add no single-user spatial degrees of freedom; their reported benefit is location division multiple access (LDMA). Numerical studies using an idealized FDA channel model show higher sum-rates than phased-array schemes for far-field users with small angular separation, even after accounting for bandwidth reserved for offsets. ### Source excerpt arXiv:2610.41415v1 Announce Type: new Abstract: Traditional phased arrays (PAs) achieve near-field (NF) beamfocusing through spherical wavefront curvature, but lose this range selectivity in the far field (FF). Frequency diverse arrays (FDAs) instead introduce frequency offsets across antenna elements. These offsets synthesize a range-dependent beam whose beamdepth is governed by the aggregate frequency-offset span rather than the physical aperture, thereby extending finite beamdepth into the FF. Although FDAs have been extensively studied for radar, their broader potential for wireless communications remains largely unexplored. Accordingly, this article examines how this capability can support wireless communications and integrated sensing and communication (ISAC). We compare PA and FDA focusing mechanisms, explain FDA coupling and periodicities, and assess coexistence with orthogonal frequency-division multiplexing (OFDM). Integer-multiple offsets preserve alignment with the OFDM subcarrier grid and avoid offset-induced intercarrier interference under ideal synchronization, but require suitable symbol mapping and receiver processing. We also distinguish FDA beamdepth from waveform-based radar range resolution and discuss their complementary roles in sensing. For the configurations studied, FDAs offer no additional single-user spatial degrees of freedom. Interestingly, their principal benefit is multiuser separation through location division multiple access (LDMA). Numerical case studies under an idealized FDA channel model show that FDA-LDMA achieves higher sum-rates than PA-based schemes for users with small angular separation in the FF, even after accounting for the bandwidth reserved for frequency offsets. ## Build an Aliens style motion scanner with an ESP32-S3 and radar DevFeed: [Build an Aliens style motion scanner with an ESP32-S3 and radar]() Original publisher: [Read original article]() Author: Anne Barela Published: 2026-10-30T14:33:53Z Content type: tutorial Language: en Sources: [Adafruit Industries]() Topics: [ESP32-S3](), [animations]() Tags: [aliens](), [code](), [display](), [displays](), [electronics](), [esp32-s3](), [handheld](), [ld2450](), [microcontroller](), [motion](), [projects](), [radar](), [scanner](), [screen](), [sound](), [speaker](), [tracker]() ### AI overview NorthWorks Lab's handheld Aliens-style motion scanner uses an LD2450 radar module to report moving-target positions; an ESP32-S3 converts that data into an animated display sweep and beeps when contacts appear. The build also includes a small display and speaker, with three connections linking the radar, display, and speaker to the board. Free build files are available to copy or modify. ### Source excerpt Julian Horsey at Geeky Gadgets writes about a handheld radar built by NorthWorks Lab, modeled on the motion tracker from the movie Aliens. It pairs an ESP32-S3 microcontroller with an LD2450 radar module, a small display and a speaker. The LD2450 reports the position of moving targets, and the ESP32-S3 turns that data into an [...] ## PT$^3$: Prototype-Guided Self-Supervised Pretraining for Generalizable mmWave Human Point Cloud Representation DevFeed: [PT$^3$: Prototype-Guided Self-Supervised Pretraining for Generalizable mmWave Human Point Cloud Representation]() Original publisher: [Read original article]() Author: Ziyang Liu, Kieren Yu, Xiao Yang, Xin Yang, Kaishun Wu Published: 2026-10-30T04:00:00Z Content type: article Language: en Sources: [arXiv: Artificial Intelligence]() Topics: [Representation learning](), [Model Development]() Tags: [autoregressive](), [classification](), [consistency](), [cs-ai](), [datasets](), [evolution](), [framework](), [future](), [generation](), [local](), [mmwave](), [models](), [motion](), [multiple](), [performance](), [point](), [prediction](), [pretraining](), [prototype](), [prototypes](), [radar]() ### AI overview PT$^3$ pretrains a reusable mmWave human point cloud encoder from unlabeled sequences collected across multiple sources. It represents local spatial states as sparse mixtures of shared prototypes, predicts their temporal evolution autoregressively, and uses observation consistency to stabilize representations under radar perturbations. The same frozen encoder supports action classification, dense point generation, joint pose estimation, and future motion prediction on datasets not used for pretraining. The abstract reports better performance than compared pretraining baselines and competitive or superior results versus task-specific models; it provides no quantitative results or details of the comparisons. ### Source excerpt arXiv:2610.38551v1 Announce Type: new Abstract: Multi-source millimeter-wave (mmWave) observations offer an opportunity to learn human representations reusable across sensing tasks. However, sparse detections and multipath effects make these observations unstable, challenging existing point cloud pretraining methods. We propose PT$^3$, a prototype-guided self-supervised pretraining framework for mmWave human point clouds. PT$^3$ represents local states in spatial regions as sparse compositions of shared prototypes and learns their temporal evolution through autoregressive prediction. Observation consistency further stabilizes representations under radar perturbations. After pretraining on unlabeled sequences from multiple sources, the same frozen encoder supports action classification, dense point generation, joint pose estimation, and future motion prediction on datasets not used for pretraining. Across these tasks, PT$^3$ outperforms the compared pretraining baselines and achieves competitive or superior performance to task-specific models. ## Multistatic ISAC with Deep Learning-Based Clutter Suppression for 5G and Beyond DevFeed: [Multistatic ISAC with Deep Learning-Based Clutter Suppression for 5G and Beyond]() Original publisher: [Read original article]() Author: Yalin E. Sagduyu, Tugba Erpek, Sastry Kompella Published: 2026-10-30T04:00:00Z Content type: article Language: en Sources: [arXiv: Artificial Intelligence](), [arXiv: Machine Learning](), [arXiv: Networking]() Topics: [Deep learning](), [Artificial Intelligence](), [Neural Network](), [communications]() Tags: [5g](), [attention](), [cs-ai](), [cs-lg](), [cs-ni](), [deep-learning](), [distributed](), [doppler](), [eess-sp](), [framework](), [localization](), [neural](), [neural-network](), [position](), [processing](), [radar](), [range](), [robustness](), [sensing](), [signal](), [signal-processing](), [target](), [velocity](), [wireless]() ### AI overview The paper presents a multistatic ISAC framework aimed at improving target position and velocity estimates when dense, dynamic clutter corrupts bistatic range and Doppler measurements. It models clutter as randomly located scatterers with varying Doppler spreads and radar cross sections, then evaluates conventional signal-processing suppression baselines alongside residual and attention-enhanced neural networks that enhance target responses in clutter-contaminated observations. The abstract reports that conventional methods have limited robustness under strong clutter, while the deep learning methods substantially outperform them across diverse clutter conditions. The supplied text does not provide quantitative results or implementation details. ### Source excerpt arXiv:2610.40586v1 Announce Type: new Abstract: Integrated sensing and communications (ISAC) systems using 5G and beyond signals of opportunity enable joint wireless sensing and communication using shared infrastructure and spectrum resources. While multistatic ISAC can increase sensing coverage with spatially distributed receivers, dense and dynamic clutter can significantly degrade bistatic range and Doppler estimation, leading to errors in target localization and velocity estimation. This paper presents a multistatic ISAC framework with clutter suppression for high-fidelity estimation of target position and velocity. Distributed clutter is modeled with randomly located scatterers with varying Doppler spreads and radar cross sections to capture dense and dynamic clutter effects. A broad range of conventional signal processing baselines are evaluated for clutter suppression, showing limited robustness under strong clutter conditions. To address this limitation, deep learning-based clutter suppression methods are introduced to improve robustness against dense and dynamic clutter by employing residual and attention-enhanced deep neural network architectures that directly enhance target responses from clutter-contaminated observations. These methods substantially outperform conventional approaches in target position and velocity estimation across diverse clutter conditions in multistatic ISAC. ## Benchmark Radar: A Living Database and Search Engine for AI Benchmarks and Evaluation DevFeed: [Benchmark Radar: A Living Database and Search Engine for AI Benchmarks and Evaluation]() Original publisher: [Read original article]() Author: Koutian Wu, Wenjun Zhang, Junjie Zhou, Pengqian Han, Qiming Bao, Ergan Shang, Jiayu Wang, Junkai Wang, Wanghan Xu, Songyuanyi Lu, Yixuan Li, Rudy Mizrahi Celekli, Lazizbek Ravshanov, Yaoliang Bian, Xu Published: 2026-10-28T04:00:00Z Content type: article Language: en Sources: [arXiv: Artificial Intelligence]() Topics: [LLM evaluation / benchmarking](), [search-engines](), [data-engineering](), [Dataset development]() Tags: [abstract](), [adoption](), [agentic](), [ai](), [ai-benchmarks](), [ai-systems](), [analysis](), [arxiv](), [audit](), [benchmark](), [benchmarks](), [candidate](), [catalog](), [cli](), [cs-ai](), [cs-ir](), [database](), [engine](), [evaluation](), [radar](), [search](), [search-engine]() ### AI overview Benchmark Radar is a living database and search engine for finding AI benchmarks and inspecting the evidence behind evaluations. It discovers benchmark papers, repositories, datasets, and releases daily from 37 sources (13 direct connectors and 24 research and engineering feeds), and links catalog records to mentions in model cards and technical reports and to score histories. The catalog covers 1,283 source records and 12,916 numeric observations across 790 records. Its dashboard offers a leaderboard, score-versus-measured-use Pareto view, saturation and trend views, daily feeds, and downloadable evidence; a CLI supports offline queries. The authors describe a worked prior-art search and audit the catalog, while noting limitations in score comparisons. The supplied text does not detail those limitations. ### Source excerpt arXiv:2609.11115v4 Announce Type: replace Abstract: Benchmark researchers and developers of large language models (LLMs) and other AI systems need to find relevant evaluations, locate their benchmark datasets and code, and understand the settings behind reported scores. We present Benchmark Radar, a living database and search engine for retrieval and discovery of AI benchmarks, covering LLM evaluation, agentic and tool-use benchmarks, coding, reasoning, safety, and domain-specific evaluations. The system combines daily discovery of benchmark papers, repositories, datasets, and releases with a searchable benchmark catalog, mentions in model cards and technical reports, and score histories. It retains source identities and citations so readers can inspect candidate benchmarks and their evaluation evidence. Daily discovery draws on 37 sources: 13 direct connectors and 24 first-party research and engineering feeds. The catalog contains 1,283 source records drawn from 4 benchmark catalogs and 12,916 numeric observations on 790 records. We describe collection and retrieval, audit the full catalog, and examine benchmark saturation, adoption trends, and the limits of score comparisons. A worked example walks through a complete prior-art search, showing how to query the catalog and inspect benchmark evidence when designing a new evaluation. We release the web dashboard with a benchmark leaderboard, a Pareto frontier view of score against measured use, saturation and trend views, daily feeds, downloadable evidence, a command-line interface (CLI) for offline queries, and reproducible analysis. ## What took the Internet offline in Q3 2026 DevFeed: [What took the Internet offline in Q3 2026]() Original publisher: [Read original article]() Author: Lai Yi Ohlsen Published: 2026-10-27T13:00:00Z Content type: article Language: en Sources: [Cloudflare Blog]() Topics: [Internet Trends](), [Internet Traffic](), [Resilience](), [TCP/IP](), [Cloudflare]() Tags: [2026](), [cloudflare](), [connectivity](), [government](), [internet](), [internet-shutdown](), [internet-traffic](), [internet-trends](), [natural-disasters](), [outage](), [power](), [radar]() ### AI overview Cloudflare Radar recorded 49 Internet disruptions in Q3 2026. Government-directed shutdowns led with 19 incidents, including 18 scheduled exam shutdowns in Iraq; power outages accounted for 14 incidents across 10 countries, and natural disasters for five. Power failures affected connectivity through grids and facilities: a Honduras interconnection failure disrupted at least 11 providers and cut HTTP traffic 63% for about three hours, while outages in Sevastopol reduced traffic by up to 98%. Storms, earthquakes, conflict, and infrastructure damage also caused regional or localized outages. In Japan, the earthquake's connectivity impact appeared mainly in desktop traffic, with mobile traffic largely unaffected. Recovery varied: 14 disruptions returned to typical traffic levels in under an hour and 16 in under three hours. The account is based on Cloudflare's observations of traffic to its customers; Radar data alone could not establish whether Russia's cable damage was related to elections. ### Source excerpt An examination of global Internet disruptions observed by Cloudflare Radar during the third quarter of 2026. This review analyzes how power failures, natural disasters, and government shutdowns impacted network connectivity and traffic. ## Continuous accountability for Verified bots DevFeed: [Continuous accountability for Verified bots]() Original publisher: [Read original article]() Author: Chris Tsoukaladelis Published: 2026-10-16T13:00:00Z Content type: article Language: en Sources: [Cloudflare Blog]() Topics: [Bot Management](), [Bot](), [trust](), [Digital Trust](), [webservice](), [Cross-origin resource sharing (CORS)]() Tags: [ai](), [automated-traffic](), [bot-management](), [bots](), [cloudflare](), [http](), [internship-experience](), [radar](), [robots-txt](), [security]() ### AI overview Cloudflare is introducing a system to assess whether Verified bots consistently honor website preferences, including robots.txt instructions and HTTP 429 requests to slow down. The system evaluates patterns over time and aims to give site owners context for handling automated traffic, with plans to make it available to all Cloudflare customers. ### Source excerpt Identity alone doesn't guarantee a bot will behave responsibly. We are introducing a system to continuously evaluate whether Verified bots honor site preferences, like robots.txt and HTTP 429s, ensuring they remain accountable over time. ## Cloudflare analysis finds malicious bot traffic shifted providers after Yandex disruptions DevFeed: [Cloudflare analysis finds malicious bot traffic shifted providers after Yandex disruptions]() Original publisher: [Read original article]() Author: Lai Yi Ohlsen Published: 2026-10-14T20:24:19Z Content type: article Language: en Sources: [Cloudflare Blog]() Topics: [Threat Research](), [Bot Management](), [data centers](), [Threat monitoring](), [Internet Traffic]() Tags: [attack](), [bot-management](), [bot-traffic](), [cloud-services](), [cloudflare](), [crawler](), [data-centers](), [networks](), [outage](), [radar]() ### AI overview Cloudflare analyzed bot traffic around reported strikes on Yandex data centers. Its data shows that the decline in Yandex Search traffic was mostly verified search crawling, while malicious automation using Yandex.Cloud moved to other hosting providers. The analysis found no evidence that the disruption caused a broader interruption in malicious internet activity or social media manipulation. ### Source excerpt When strikes hit Yandex data centers, many assumed Russian bot operations went dark. Cloudflare data reveals that while search crawling stopped, malicious bots simply migrated to other networks. ## Bridging technical depth and usability: The story behind Radar's redesign DevFeed: [Bridging technical depth and usability: The story behind Radar's redesign]() Original publisher: [Read original article]() Author: Lai Yi Ohlsen Published: 2026-10-08T13:00:00Z Content type: article Language: en Sources: [Cloudflare Blog]() Topics: [GigaOm Radar](), [User interface design](), [Design system](), [Cloudflare]() Tags: [cloudflare](), [front-end](), [internship-experience](), [radar](), [redesign](), [usability]() ### AI overview Cloudflare redesigned Radar's landing page to make its real-time Internet traffic and outage data easier for a wider audience to explore. The update introduces a map-based overview and tabbed navigation, aiming to improve usability while retaining the depth of Radar's data. ### Source excerpt We redesigned Cloudflare Radar to make real-time global traffic and outage data accessible to a wider audience -- including journalists, less technical researchers, and everyday users. By introducing an interactive map and standardized components, we simplified navigation while preserving technical depth. ## AutoSens 2026: Regulation Drives Automotive Sensing Architectures DevFeed: [AutoSens 2026: Regulation Drives Automotive Sensing Architectures]() Original publisher: [Read original article]() Author: Pablo Valerio Published: 2026-10-02T15:58:50Z Content type: news Language: en Sources: [EE Times]() Topics: [Cybersecurity](), [Self-organizing Team](), [Post-quantum cryptography](), [scheduling](), [Linux Kernel]() Tags: [ai](), [analog](), [analysis](), [automotive](), [automotive-architecture](), [autosens](), [blog](), [business](), [centralization](), [chip](), [cnsa-2-0](), [community](), [compute](), [cybersecurity](), [data](), [designline](), [detection](), [digital](), [digital-products](), [digital-twin](), [education](), [ee-times](), [ee-times-university](), [eelife](), [eelive-electronic-design](), [eet](), [electrical-engineering](), [electronic-components](), [electronic-engineering-times](), [electronics](), [embedded-systems](), [euro-ncap-2026](), [europe](), [fundamentals-courses](), [ieee-802-15-4ab](), [incabin](), [learning](), [lidar](), [mixed-signal](), [mmwave](), [news](), [opinion](), [part-search](), [product-search](), [radar](), [regulation](), [semiconductors](), [sensor-technology](), [software](), [standards](), [supply-chain-distribution](), [tech-papers](), [technology](), [uwb](), [video]() ### AI overview At AutoSens Europe 2026, automotive sensing designs reflected tighter safety standards, advances in AI processing, and growing cybersecurity requirements. Companies demonstrated camera and radar systems for cabin monitoring, new image sensors and zonal edge-computing approaches, and hardware supporting post-quantum cryptography. The article reports these product demonstrations and industry developments; many claimed performance figures are attributed to the companies presenting them. ### Source excerpt At AutoSens Europe, automotive sensing designs reflected tighter safety standards, advances in AI processing, and growing cybersecurity requirements. The post AutoSens 2026: Regulation Drives Automotive Sensing Architectures appeared first on EE Times. ## Synthetic aperture radar drone gets interferometric imaging DevFeed: [Synthetic aperture radar drone gets interferometric imaging]() Original publisher: [Read original article]() Author: Anne Barela Published: 2026-09-29T19:42:14Z Content type: article Language: en Sources: [Adafruit Industries]() Topics: [ESP32-C3]() Tags: [drone](), [drones](), [imaging](), [radar](), [rtk](), [rtk-gps](), [sensors](), [synthetic-aperture-radar]() ### AI overview Henrik Forstén's homebuilt synthetic aperture radar drone now performs interferometric imaging using the same low-cost hardware. The update adds RTK GPS, including an F9P module, helical antenna, and base station. ### Source excerpt Henrik Forstén has posted a follow-up on hforsten.com to his homebuilt synthetic aperture radar drone, and it now does interferometric synthetic aperture radar (SAR) with the same cheap hardware. The key addition is RTK GPS: an F9P module with a helical antenna found on AliExpress for 144 EUR, plus a 180 EUR base station. Position [...] ## Bliki: Sensible Default DevFeed: [Bliki: Sensible Default]() Original publisher: [Read original article]() Author: Martin Fowler Published: 2026-09-29T13:43:00Z Content type: article Language: en Sources: [Martin Fowler]() Topics: [bestpractice](), [Self-organizing Team](), [CI/CD]() Tags: [bliki](), [first](), [practices](), [radar](), [software-development](), [thoughtworks]() ### AI overview A sensible default is a known-good software development practice to start with when no overriding constraints apply. Unlike a universal best practice, it should be reassessed and changed when context calls for it. The article describes Thoughtworks' use of defaults and regular review of them. ### Source excerpt A Sensible Default is a practice that, absent some overriding context, should be used when carrying out a certain kind of task. In software development such sensible defaults might include things like "use version control", "separate UI logic from domain logic", "automate deployment pipelines". The term "sensible default" is a deliberate contrast to "best practice". Folks dislike calling things "best practice" because that term implies a general presumption that the best practice is something we should always expect to do. A "sensible default", however, is something that should be reassessed in a new context, something that can (and should) be overridden when circumstances change. I first heard the term when it was popularized within Thoughtworks by Evan Bottcher. He got the name from talking to James Ross, and found the phrasing appealing as carried the nuance he was seeking, a known-good starting point. A sensible default is what we'd expect you to do, the practices to apply, if there are no hard constraints in the environment. Do these practices, or do better, and be prepared to explain why you've chosen some other way. -- Evan Bottcher Thoughtworks has since made much of this concept, including publishing a playbook of the ones we use. We expect people to be familiar with these defaults, ready to use them when starting any new piece of work. They are our defaults because we've used them in many situations and found them to be effective. But teams should also be familiar with their limitations, and able to judge whether they should be changed depending on the particular circumstances. As the twelfth agile principle says: "At regular intervals, the team reflects on how to become more effective, then tunes and adjusts its behavior accordingly." We also reassess these defaults regularly - this is the heart of the Thoughtworks Technology Radar. Searching on the web led me to a post from Steve Bennett on Sensible Defaults. The post was written at about the time Evan t ## Synthetic Aperture Radar Drone Gets Interferometric Imaging DevFeed: [Synthetic Aperture Radar Drone Gets Interferometric Imaging]() Original publisher: [Read original article]() Author: Aaron Beckendorf Published: 2026-09-28T20:00:17Z Content type: article Language: en Sources: [Hackaday]() Topics: [Medical imaging](), [3D]() Tags: [drone](), [fmcw-radar](), [fpga](), [gps](), [hardware](), [interferometry](), [radar](), [radio-hacks](), [radio-interferometry](), [rtk](), [rtk-gps](), [software](), [synthetic-aperture-radar]() ### AI overview The article describes improvements to a synthetic-aperture radar imaging drone, including autofocus, more accurate positioning, hardware changes, and a compression algorithm for recording data. It also explains how interferometric imaging uses flights at different altitudes to produce an elevation map. ### Source excerpt It's been more than a year since [Henrik Forstén] built the first iteration of his synthetic-aperture radar (SAR) imaging drone, and he's certainly been productive in the meantime. Not only ...read more ## IonQ | Quantum Change Detection on Satellite Radar Data DevFeed: [IonQ | Quantum Change Detection on Satellite Radar Data]() Original publisher: [Read original article]() Author: IonQ Staff Published: 2026-09-25T17:20:37Z Content type: article Language: en Sources: [IonQ]() Topics: [Quantum Computing](), [Machine learning](), [geospatial]() Tags: [change-detection](), [data](), [quantum](), [radar](), [satellite](), [space]() ### AI overview IonQ researchers evaluated a Quantum Circuit Born Machine on real SAR and InSAR satellite imagery to detect changes. The model achieved higher filtered F1 scores than a classical baseline on datasets with strongly skewed intensity distributions, while results were comparable on approximately Gaussian data. The article describes these as early results and says further testing is needed to determine where quantum methods add value. ### Source excerpt How IonQ used a Quantum Circuit Born Machine and a trapped-ion QPU to detect changes in SAR and InSAR imagery from Capella Space satellites. ## How 60 GHz Radar Improves Low-Power Presence Sensing In IoT Devices DevFeed: [How 60 GHz Radar Improves Low-Power Presence Sensing In IoT Devices]() Original publisher: [Read original article]() Author: Chris Wengerter Published: 2026-09-21T07:01:47Z Content type: article Language: en Sources: [Semiconductor Engineering]() Topics: [Internet of things](), [Hardware](), [systems](), [processing]() Tags: [antenna-in-package](), [blogs-security](), [devices](), [fmcw-radar](), [hardware](), [infineon](), [iot](), [iot-and-security](), [low-power](), [mcu](), [motion](), [presence-sensing](), [radar](), [secured-intelligent-mobility](), [sensors]() ### AI overview The article explains how 60 GHz radar enables low-power presence and motion sensing in battery-powered IoT devices. It describes duty-cycled sensing architectures and uses Infineon's XENSIV BGT60CUTR13AIP as an example of antenna-in-package integration, on-chip radar processing, and autonomous operation. It also compares radar with PIR, cameras, and ultrasonic sensing. ### Source excerpt Track occupancy and motion continuously without keeping the full sensing and processing chain active. The post How 60 GHz Radar Improves Low-Power Presence Sensing In IoT Devices appeared first on Semiconductor Engineering. ## Analyzing rising fraud attempts among travel and leisure businesses on Stripe DevFeed: [Analyzing rising fraud attempts among travel and leisure businesses on Stripe]() Original publisher: [Read original article]() Author: Andrew Beckmann; James Lemon Published: 2026-09-18T00:00:00Z Content type: article Language: en Sources: [Stripe]() Topics: [stripe](), [Risk](), [Security](), [Transactions](), [Artificial Intelligence]() Tags: [2023](), [2025](), [ai](), [apac](), [emea](), [fraud](), [payment](), [radar](), [security](), [stripe](), [transactions](), [travel]() ### AI overview Stripe analyzes rising fraud attempts affecting travel and leisure businesses, based on payment activity from more than 200,000 active businesses. The analysis describes increasing attempts, regional differences, and the effectiveness of Stripe Radar in blocking suspected fraud. ### Source excerpt Last year, Stripe data shows fraud attempts against travel and leisure businesses hit a four-year high. We analyzed payment activity from more than 200,000 active travel and leisure businesses on Stripe to understand where fraud is rising, how effectively it's being blocked, and what businesses can do in response. ## Total eclipse of the Internet: traffic impacts in Iceland, Spain, and Portugal DevFeed: [Total eclipse of the Internet: traffic impacts in Iceland, Spain, and Portugal]() Original publisher: [Read original article]() Author: Lai Yi Ohlsen Published: 2026-08-13T19:58:01Z Content type: article Language: en Sources: [Cloudflare Blog]() Topics: [Internet Traffic](), [Cloudflare](), [data]() Tags: [blog](), [cloudflare](), [data](), [internet](), [internet-traffic](), [internet-trends](), [radar]() ### AI overview Cloudflare Radar data shows that Internet traffic decreased as a total solar eclipse deepened across affected European countries, with the largest dips along the path of totality and traffic typically rebounding within minutes after maximum coverage. ### Source excerpt Cloudflare's data shows a clear impact on Internet traffic from Iceland to Spain and Portugal, following the path of totality of the total solar eclipse that occurred on August 12, 2026. [Next page]()