Digital Commons Network^™

Horizon‑Based Operating Point Selection For A System On A Chip Using Predicted Thermal Trajectories And Adaptive Cost Weighting, Mohsen Heidarinejad, Farid Kenarangi, Arpit Mittal

Defensive Publications Series

Thermal and power management on a system on a chip (SoC) typically responds to its present state, revising the operating point only after reaching a temperature or power limit.  This reactive approach often overlooks the delayed and cumulative heating produced by earlier execution, failing to effectively reconcile performance, power, and completion goals.  The disclosed methods forecast temperature ramp rates and trajectories from the present temperature and power of the SoC, form candidate operating points expected to meet a workload completion time, and limits them with caps from thermal and power controllers.  A cost function evaluated across a finite prediction horizon …

High-Capacity Server Ramp With Crank-Adjustable Height, N/A, N/A

Defensive Publications Series

The high-capacity server ramp with crank-adjustable height is a portable steel structure for unloading heavy server racks from shipping crates. The ramp may support a maximum load of 8,000 pounds. An operator adjusts the integrated landing gear assemblies of the ramp using a hand crank. The hand crank drives an internal gearbox and screw mechanism to extend or retract the legs of the ramp, enabling continuous height adjustment. This adjustment aligns the surface of the ramp with various shipping crate heights to create a consistent incline for moving equipment.

Keywords: server rack ramp, adjustable height, crank adjustment, internal gearbox, two-speed …

Warranted Extraction: A Verification Architecture For Machine-Read Coverage, James H. Rosing

Defensive Publications Series

Large language models can read a payer medical-policy document and produce structured coverage criteria that look right. Nothing in the model’s output distinguishes the criteria that are right from the criteria that are invented, and in clinical revenue-cycle use an invented criterion is acted on. We describe an architecture in which an LLM is treated as an untrusted proposer whose output becomes queryable only after a deterministic, separately built verifier has resolved every claim to a verbatim span in a hash-pinned rendering of the source document. Every served criterion carries that span, called a warrant, so any consumer can check …

Large Language Model (Llm) Powered Input Method Editor (Ime), Liyang Jiang, Heng Su, Gang Feng

Defensive Publications Series

Traditional Input Method Editors (IMEs) that provide next phrase prediction to ease data entry have limited context and lack deep semantic understanding. The prediction accuracy of such IMEs is suboptimal for long-tail phrases such as domain-specific vocabulary, acronyms, slang, etc. or in contexts where the text entry has a long semantic context. This document describes techniques that integrate a Large Language Model (LLM) into the text decoding and prediction path of an IME to reduce keystroke overhead and improve output quality by providing advanced state-dependent predictions. An asynchronous prediction pipeline is implemented where a client application (e.g., that operates as …

Device-Level Revocation Of Digital Asset Signing Capabilities, Paul J. Rohde

Defensive Publications Series

Attackers can circumvent content provenance standards by physically modifying a device to generate valid device attestation reports, allowing the attacker to sign fake digital assets such as images, video, or audio content as authentic assets captured using the device. For example, an attacker may physically interact with and voltage-glitch the device's random-access memory (RAM) to obtain runtime root access, disable security modules, and impersonate an authorized capturing process to generate valid attestation reports. This disclosure describes techniques for revoking certificates assigned to a device detected as having been used to produce a fraudulent digital asset. The techniques involve generating a …

Centrifugal Compressor With Pressure Responsive Flexible Seal Plate For Dynamic Diffuser Geometry Control

Defensive Publications Series

Traditional centrifugal compressors utilize fixed-geometry diffusers that are optimized for a single operational design point, leading to reduced aerodynamic efficiency under off-design flow conditions and varying pressure ratios. Mechanical variable-geometry alternatives introduce complex linkages, actuators, and sealing vulnerabilities. This proposed solution discloses a compressor configuration that utilizes controllable fluid pressure within an integrated coolant circuit to actively or passively deform a flexible seal plate. By dynamically modulating the coolant circuit pressure, the diffuser channel width and geometric profile can be altered seamlessly to maximize aerodynamic and thermal exchange efficiency without external mechanical moving parts.

Remote Display Spatial Pointing Using Processed Inertial Sensor Data From A Mobile Device, Chien-Feng Yu, Sophia Chen, Eusden Lin, Sara Lin, Dewita Sonya

Defensive Publications Series

A mobile computing device (e.g., a smartphone, smart watch, wearable device, augmented reality glasses, etc.) may function as a spatial pointer for a remote display by processing data from an on-board inertial measurement unit. An on-device spatial engine can translate the device's physical orientation and movement into two-dimensional screen coordinates. This translation process may, for example, involve techniques such as isotropic aspect ratio correction and a dynamic motion amplification layer that adjusts sensitivity based on angular velocity. The processed coordinates can then be streamed to the remote display over a low-latency network protocol. This approach may provide improved cursor control …

Fully Isolated Floating Nut Terminal Block For 3-Phase High Voltage Systems

Defensive Publications Series

In electric vehicle (EV) powertrains, establishing robust electromechanical connections between the inverter and the electric motor's stator busbars (U, V, W phases) is critical. Traditional terminal blocks utilize a fixed-type geometry that cannot account for manufacturing and assembly tolerance variations. This rigidity introduces severe mechanical stress into the busbars, potentially leading to material fatigue, weld failures, and fluid leakage at the inverter’s overmolded housing seals. This proposed solution discloses an innovative, fully isolated floating nut terminal block. By integrating floating threaded subassemblies within a polyphenylene sulfide (PPS) overmolded bracket containing specific hexagonal grooves, the system allows multi-axis movement to absorb …

Runtime Adaptation Of Controller-To-Operating-Point Mapping Policies For Thermal, Power, And Performance Coordination In A System On A Chip, Mohsen Heidarinejad, Farid Kenarangi, Arpit Mittal

Defensive Publications Series

A system on a chip (SoC) converts outputs from its temperature and power controllers into operating points through mapping policies set during nominal silicon characterization.  Because thermal and electrical behavior shifts with workload activity, silicon process variation, and aging, a fixed mapping policy can leave thermal and power budgets underused or request operating points that press measured temperature or power toward platform limits.  The disclosed system retains existing temperature and power control loops while adding a temperature impact block, a power impact block, and a performance impact block that evaluate measured SoC behavior and propose localized refinements to the mapping …

Macroscopic Momentum-Space Tomography: Utilizing High-Z Cryogenic Target Arrays As Resonant Phase-Retrieval Lenses For Substrate Interferometry, Christopher L. Eckes

Defensive Publications Series

This paper introduces a formal engineering architecture to transition the Dimensionally Extended Holographic Projection (DEHP) model from numerical simulation to empirical instrumentation. Integrating the spatial phase-retrieval mechanics of 3D Photoemission Orbital Tomography (3D-POT) with the high-density containment engineering of deep-underground liquid xenon detectors, we outline a method for a macroscopic momentum-space microscope. Rather than treating the September 1, 2026 LUX-ZEPLIN 248 keV anomaly as an exogenous particle collision, we demonstrate that a high-Z condensed target array can function as an active, resonant phase-retrieval lens. We provide the complete mathematical tensor frameworks, specific material specifications, and structural engineering steps required to …

Detecting Scan Flop State Corruption During A Functional-To-Scandump Mode Transition Using Parallel Seed Initialization And Comparison, Sunder Garg, Prakash Narayanan, Krunal Siddhapathak, Shivam Sharma, Mayank Parasrampuria

Defensive Publications Series

During silicon debug, a system-on-chip (SoC) captures its internal sequential state through a scan-based extraction step, referred to as Scandump.  Moving the design from functional operation into Scandump mode is an asynchronous event, where a glitch on a reset or clock signal can toggle sequential elements and corrupt the captured state.  This document describes a verification method that freezes the design with a global clock-stop signal, writes a known random seed in parallel into the scannable flip-flops (scan flops) as value A, moves the design into Scandump mode, and samples the resulting flop state in parallel as value B.  Comparing …

Machine‑Learning Estimation Of Static Ir Drop In System‑On‑Chip Power Grids Using Gradient‑Boosted Regression And Image‑To‑Image Translation, Prateek Pendyala

Defensive Publications Series

Designing the power delivery network of a system‑on‑chip (SoC) involves repeated changes to the layout and to the power grid, and engineers conventionally re‑evaluate each change with a commercial static current‑resistance (IR) drop analysis.  These analyses can run from tens of minutes to several hours across many compute workers, which lengthens each design iteration.  This disclosure describes a machine‑learning approach that estimates per‑node static IR drop for multi‑million‑node power grids from electrical and spatial features of a design database.  The approach provides two complimentary techniques: a gradient‑boosted decision‑tree regression that predicts incremental IR drop from extracted power and resistance features, …

Early Prediction Of Design‑Rule‑Check Violation Hotspots From Placement‑Stage Congestion Feature Maps Using A Fully Convolutional Network, Prateek Pendyala

Defensive Publications Series

In the physical design of a system‑on‑chip (SoC) layout, design‑rule‑check (DRC) violations often appear late, during detailed routing (a stage that can take days to complete and that may trigger repeated placement adjustments).  This document describes a method that estimates where DRC violation hotspots may gather earlier in the process, working from feature maps that become available after placement and global routing.  The method arranges these maps (macro regions, cell and pin densities, rectangular uniform wire density (RUDY) maps, and routing congestion) as multi‑channel two‑dimensional inputs to a fully convolutional network (FCN) built as a multi‑scale encoder‑decoder with skip connections …

A Place‑And‑Route‑Independent Solver For Early Voltage‑Drop Estimation Of System‑On‑Chip Power Grids, Prateek Pendyala

Defensive Publications Series

Designing the power grid of a system‑on‑chip (SoC) balances two competing goals: a denser grid maintains voltage drop within a target, while a sparser grid frees routing resources for timing closure and layout density.  Reaching a workable balance calls for repeated passes through grid design, electromigration and voltage‑drop analysis, floor-planning, and place‑and‑route.  Preparing those passes with separate commercial tool flows can take up to forty minutes for a single grid specification.  This document describes a standalone solver that reads a power‑grid specification, builds a resistor‑and‑via mesh model of the grid, and computes the direct‑current node‑voltage distribution by assembling and solving …

Predictive, Time‑Scaled Scheduling Of Concurrent Workloads For Power And Thermal Management In A System On A Chip, Mohsen Heidarinejad, Farid Kenarangi, Arpit Mittal, Amin Khajeh

Defensive Publications Series

A system on a chip (SoC) that runs several workloads at the same time can experience uneven power draw, localized heating, and reduced responsiveness when it relies on a fixed policy that treats the die as uniform.  This document describes a closed‑loop framework that observes on‑die temperature and power, estimates power for subsystems without direct sensors, and predicts near‑term spikes and longer‑term demand from queued work.  Using these estimates, the framework stretches or compresses task timelines and recommends per‑subsystem power caps, which a control layer applies through operating‑point adjustment, clock and power gating, and task migration.  A supervisory layer then …

When The Protocol Decides - Replay-Ready Infrastructure For Programmable Financial Institutions, Deepak Aggarwal

Defensive Publications Series

The rails are becoming intelligent. The governance frameworks developed to manage them remain as before. In the ASEAN+3 group, the financial infrastructure stack is undergoing redesign, starting from the message layer upward. Project Nexus connects national payment systems at near-real-time speeds. ISO 20022 incorporates compliance controls within the financial messages themselves. Project Agorá brings programmable settlement to a distributed ledger. AI is parsing liquidity, fraud, identity, and credit information even before humans can see the information.

This paper presents a simple yet powerful illustration of its main thesis: One decision, three jurisdictions, seven seconds. Decision chains start unfolding themselves at …

The Irrecoverable Institution - Why Replayability - Not Explainability - Is The Governance Standard For Agentic Banking, Deepak Aggarwal

Defensive Publications Series

Financial institutions have accelerated the deployment of Agentic AI systems. These systems have the capabilities to make decisions and are operating faster than any governance mechanism can respond. However, when such decisions go wrong, the biggest cost incurred is not due to their very first mistake, but because the financial institution lacks the capability of replaying and governing the decision path until other costs increase significantly.

The objective of this paper is to propose Decision Replay as the base governance framework for agentic financial institutions, which would be the capability to replay and govern decisions at any moment, via any …

Banking Rewired - Reimagining The Future Of Banking In A Protocol-Driven World, Deepak Aggarwal

Defensive Publications Series

The World is Changed. Has Your Bank?  

A fintech executive in New York, USA. An AI researcher in Amsterdam, Netherlands. An MSME owner in Bengaluru, India. Each of them interacts with money not by logging in but living. Money flows, credit adjusts, identity streams, and ethics calculate, all ambiently, intelligently, and consent driven. Not science fiction. This is the shape of banking in a  programmable world. Banks are at a crossroads. The question is no longer: "How digital are we?" It is: "How programmable, ethical, and modular is our institution?"

Four Strategic Shifts Redefining the Future Bank

1. From Platform …

System And Method For Elemental Decomposition-Based Provenance Attribution Of Ai-Generated Audio, Eric Zavesky

Defensive Publications Series

A system and method determine whether a candidate audio catalog contributed to the training of a generative audio model. An elemental encoder decomposes generated audio and catalog items into elemental embeddings corresponding to aesthetic elements including melodic contour, harmonic progression, rhythmic pattern, and timbral palette. A segmentation engine defines regions of the generated output for scoped analysis. Within each element independently, instance-level scores compare output regions against catalog items while distribution-level alignment compares the catalog's element distributions against distributions over model-generated outputs relative to a reference set. A multi-order analyzer scores recurrence patterns, including motif persistence. A provenance scorer combines …

Localized Generation Of Privacy-Preserving Ambient Chronicles, Karl Tao, Gaurav Arora

Defensive Publications Series

A system is disclosed for generating a localized activity chronicle within a private environment. Local cameras capture sparse static snapshots of room environments over a local network. A local smart hub platform receives each snapshot directly into an ephemeral, volatile memory buffer. An on-device vision model, such as a quantized vision-language model executed by a local hardware neural processing unit or other local processing resource, processes the snapshot to generate a time-stamped natural-language description of activities within the private environment. Immediately following text generation, the raw snapshot is purged from the volatile memory, preventing raw image data from being stored …

Wearable Apparatus Configured For Ambient Marker Sensing And Application Actions, Carlos Henrique Esteves Mendonca, Joost Korngold

Defensive Publications Series

Transactions involving physical objects often require manual interaction with a mobile device, such as opening a camera application to scan a visual marker, resulting in an interruptive and multi-step process. A wearable extended reality apparatus or device (e.g., smart glasses) is described that may be configured for hands-free processing of digital transactions. The device can use an ambient camera to sense visual markers in an environment. To establish user intent, a dwell mechanism may evaluate gaze duration before triggering a secondary image capture. The device is configured to decode a uniform resource identifier from the visual marker and link to …

Semantics-Aware And Controllable Hand-Object Grasp Generation, Alberto García, Dominik Thabo Beeler, Jesús Pérez, Vasileios Choutas, Hiroyasu Akada, Emre Aksan

Defensive Publications Series

Current systems configured to generate three-dimensional interactions between a hand and an object typically rely on fragmented computational pipelines that can fail to generalize to novel target objects and may offer restricted control. A system is disclosed for generating poses of a hand utilizing a hand diffusion transformer configured to process conditioning signals, such as target object geometry and a direction of a hand. The hand diffusion transformer may output a distance profile, which represents target distances between regions of a hand and surfaces of an object. A subsequent optimization layer can use the distance profile to refine candidate hand …

Ai-Driven Browser Automation System For Attribute-Based Resource Access, Benjamin Albert Azose, Akancha Murlidhar Gupta, Theodore Thomas Olsauskas-Warren

Defensive Publications Series

An AI-driven browser agent framework is disclosed to facilitate attribute-based access to resources by enabling resource access decisions based on user or account characteristics rather than transferable authorization mechanisms. With user permission, the framework allows a browser agent to communicate user credentials, selected eligibility-related information requested by a resource provider for evaluating predefined eligibility criteria associated with a requested resource. Based on the evaluation, the browser agent may be granted immediate access to a resource acquisition workflow, assigned access at a future date or time, placed into a waiting queue, or denied access. By supporting multiple communication mechanisms, coordinated admission …

Dynamic Binary Patching Of Sub-Processes In A Multi-Process Browser Architecture Without Application Relaunch, Kentaro Hara

Defensive Publications Series

A vulnerability window is created when software patches are committed to open source repositories before an application restart is executed. An approach is disclosed where a primary process remains active on an original binary while helper processes run on updated binary versions. These updates are restricted to changes that preserve inter-process communication boundaries and shared memory layouts. New rendering processes are initiated from the updated binary during web navigations. Graphics and network processes are restarted using the updated binary during unutilized periods. Security updates are deployed to a running application without a full restart or modifying live memory structures.

Real-Time Relighting Of Physical Hands In Mixed Reality Using 3d Skeletal Tracking And Skin Thickness Estimation, Ashish Singh, Ahmet Cengiz Öztireli

Defensive Publications Series

This publication describes a framework for seamlessly integrating a user’s physical hands into a Mixed Reality (MR) environment by replacing real-world lighting with virtual illumination in real-time. The technique utilizes 3D skeletal tracking to generate a simplified volumetric model of the hand, which is then used to estimate skin thickness and define material properties (albedo and surface texture). By incorporating 3D bone velocity data, rather than standard 2D pixel tracking, the system stabilizes the decomposition process during rapid movements. The resulting maps allow virtual lighting to be applied realistically, simulating effects like light translucency through thin tissue (e.g., finger webbing) …

Cross-Surface Ad Narrative Orchestration Using Llm Sequential Reasoning With Engagement-Responsive Story Branching, Tamar Yakar, Ilia Labzovsky, N/A

Defensive Publications Series

A system orchestrates stateful, multi-chapter advertising narratives across different digital platforms within real-time bidding environments. The system separates the ad-serving process into an asynchronous control plane for computationally intensive planning and a synchronous data plane for low-latency auction execution. The asynchronous control plane uses a large language model (LLM) to generate a narrative directed acyclic graph (NDAG) that defines a story arc with branching logic based on potential user engagement. The system then compiles this narrative graph into a deterministic state transition matrix for execution at the network edge. To track user progression, the synchronous data plane uses a compact …

Automating Recurring Tasks Using Ai-Driven Browser Automation System, Benjamin Albert Azose, Jonathan Jerome Azose

Defensive Publications Series

An AI-driven browser automation system collects browser activity information associated with a user, including webpage visits, timestamps, transaction-completion events, autofill activity, and other interaction data. The system analyzes the browser activity information to identify recurring activities, determine recurrence intervals, and classify and rank the recurring activities. Based on the analysis, the system generates recurring-task recommendations and presents the recommendations to a user for approval. Upon receiving approval, the system creates recurring tasks within a calendar system, reminder system, task-management system, or agentic browser workflow. In some cases, the system determines workflows associated with prior occurrences of recurring activities and utilizes …

Plastic Resins Or Adhesive Resins With Uv Fluorescent Additive As Tracers, Steven Crochiere, Michael Yi Zhao Yao, Anothai Kamphan, Randi Walz

Defensive Publications Series

UV fluorescent additives are added to primary sources of plastic resins or adhesive formulations to serve as tracers and contamination-detection aids. When a plastic or adhesive particle/debris is detected in failed hard disk drives (HDDs) or at other locations, the different tracers added in different plastic or adhesive items (such as trays, assembly tools, direct plastic components, or adhesives used on direct components or indirect materials in manufacturing processes) help identify and narrow down the source(s) of the plastic or adhesive particle/debris.

Deterministic Multi-Domain Computational Framework For Bi-Ionic Propulsion And Thermoelectric Resonance (Stalwart V18.1), Daniel Schramm

Defensive Publications Series

This paper details the mathematical and computational architecture of a multi-domain physics engine designed to validate interrelated aerospace systems, specifically bi-ionic vector thrust and solid-state thermoelectric energy recycling. To eliminate computational hallucination and enforce strict physical boundaries, the engine is constructed utilizing highly deterministic Python libraries: SciPy for immutable, universally accepted physical constants; SymPy for strict, symbolic equation rendering and mathematical proofs; and NumPy for continuous spatial array generation. This framework interlinks discrete physical phenomena (magnetic containment, plasma gyration, kinetic wave damping, and thermal energy harvesting) into a singular, codependent mathematical ecosystem. The novel contribution is the Master Resonance Equation …

Physical Oracles And The Instrumental Status Of The Terrestrial Biosphere, Volodymyr Kotegov

Defensive Publications Series

This document develops a general instrumental framework for evaluating physical systems that function as oracles — systems capable of generating high-dimensional observations on the actual physics of the universe, exhibiting substantial computational irreducibility, and producing data in measurement bases not yet fully formalized by the agent. Within this framework the terrestrial biosphere is analysed as the strongest currently known Slow Oracle.

Humanistic, ethical, and sentimental considerations are excluded. Evaluation proceeds solely in terms of expected utility, option value under residual uncertainty, basis diversity rent, thermodynamic opportunity cost, and resistance to model collapse. The central claim is that, while residual uncertainty …