The search results confirm that the USPTO Patent Public Search tool can be used to look up patent numbers. I will proceed with generating the defensive disclosure directly, using the provided patent text as the authoritative source, as I already have the full patent text and do not need to retrieve it from USPTO for this task. I have also verified that the patent number is 7051306 and not a similar number.

---

Defensive Disclosure Document for US Patent 7051306

Title: Advanced Dynamic Power Management for Modular Integrated Circuits Using Power Islands

Purpose: This defensive disclosure aims to document numerous technical advancements and alternative implementations related to dynamic power management using power islands on integrated circuits (ICs), building upon the concepts described in US Patent 7051306. The intent is to establish prior art, rendering obvious or non-novel future incremental improvements in this field by competitors. This document explores variations across materials, operational parameters, cross-domain applications, integration with emerging technologies, and failure/low-power modes.

---

Derivative Variations for Independent Claims 1, 16, and 37

The following derivatives expand upon the core concepts of US Patent 7051306, which generally describe methods, systems, and software products for managing power on an integrated circuit with a plurality of power islands, where power consumption is independently controlled within each island by a power manager determining a target power level, determining an action, and performing that action (e.g., selecting a frequency).

1. Material & Component Substitution

Derivative 1.1: Gallium Nitride (GaN) based Power Islands with Adaptive HEMT Switching

• Enabling Description: This derivative implements power islands within an integrated circuit fabricated using Gallium Nitride (GaN) semiconductor technology. Each power island utilizes GaN High Electron Mobility Transistors (HEMTs) as the primary power switching and regulation components within its local power control circuitry. The power manager (e.g., SPM, IPM, MPM) determines a target power level and dynamically controls the gate drive and switching frequency of the GaN HEMTs, or adjusts the bias voltage of enhancement-mode HEMTs, to achieve precise Vdd modification or power gating for the corresponding island. This provides superior power efficiency, higher switching speeds, and reduced form factor compared to traditional silicon-based power control, especially for high-frequency or high-power density applications. Signal isolation between GaN islands operating at different Vdd levels employs optical isolators integrated on-chip, leveraging GaN's transparent properties in certain wavelength ranges for improved noise immunity and breakdown voltage.

flowchart TD
    A[Application/OS] --> B(PMCL Firmware)
    B --> C(MPM/IPM)
    C --> D{SPM for GaN Power Island}
    D -- Target Vdd/Freq/GateBias --> E(GaN HEMT Control Logic)
    E --> F(GaN Power HEMT Array)
    F -- Regulated Power --> G(GaN Logic/IP Block)
    G -- Optical Feedback --> E
    H[External Voltage Input] --> F
    I[Optical Isolator] -- Isolated Signal --> G
    F -- Isolated Sense --> D

Derivative 1.2: Flexible Substrate Power Islands with Hybrid Integration

• Enabling Description: The integrated circuit is fabricated on a flexible polymer substrate, such as polyimide, incorporating power islands defined by areas of active circuitry. The power control circuitry for each island utilizes thin-film transistors (TFTs) made from metal oxides (e.g., IGZO) or organic semiconductors, allowing for flexible Vdd multiplexing and clock gating. The power manager, implemented as a separate rigid silicon die, is heterogeneously integrated onto the flexible substrate and communicates with the flexible power islands via embedded flexible interconnects (e.g., stretchable copper traces). Energy storage elements (e.g., thin-film supercapacitors) are co-integrated within each power island to buffer dynamic power demands. This enables power management in deformable or wearable electronic devices.

classDiagram
    class FlexibleSubstrate {
        +Polyimide Material
    }
    class PowerIsland_Flex {
        +Thin-Film Transistors
        +Metal Oxide/Organic Logic
        +Thin-Film Supercapacitor
    }
    class RigidSi_PowerManager {
        +MPM/IPM/SPM Logic
        +Communication Interface
    }
    class FlexibleInterconnect {
        +Stretchable Cu Traces
    }

    FlexibleSubstrate <|-- PowerIsland_Flex
    FlexibleSubstrate <|-- RigidSi_PowerManager
    RigidSi_PowerManager -- FlexibleInterconnect
    FlexibleInterconnect -- PowerIsland_Flex : Controls Power
    PowerIsland_Flex -- FlexibleInterconnect : Status Feedback

Derivative 1.3: Cryogenic Superconducting Power Islands with Quantum-Dot Switching

• Enabling Description: This derivative envisions an integrated circuit designed for cryogenic operation, where power islands are comprised of superconducting circuits (e.g., using NbN junctions) for ultra-low resistance power distribution paths. Power gating and Vdd adjustment within each island are achieved using quantum-dot single-electron transistors (SETs) or Josephson junction-based switches, which exhibit near-zero power dissipation in the superconducting state and highly precise switching characteristics at extremely low temperatures (e.g., 4K or below). The power manager's control signals are transmitted via superconducting transmission lines, and the target power levels are precisely managed by manipulating the quantum states of the switching elements, enabling unprecedented energy efficiency for quantum computing or deep-space applications.

stateDiagram-v2
    state Active_Superconducting {
        PowerOn: Power Island Active
        CriticalThreshold: Thermal Excursion
        PowerOn --> CriticalThreshold : T > T_critical
        CriticalThreshold --> Degraded_Superconducting : System Action
    }

    state Degraded_Superconducting {
        PowerOff: Power Island Power-Off
        Restored: Return to nominal temp
        Degraded_Superconducting --> Restored : T < T_critical
        Restored --> Active_Superconducting : Resume Operation
    }

    [*] --> Active_Superconducting
    Active_Superconducting --> PowerOff : Power_Off_Command
    Degraded_Superconducting --> PowerOff : Forced_Shutdown

    state Power_Manager {
        Control_Signals: Manage SETs/JJ switches
        Monitor_Temp: Monitor Cryo Temperature
    }
    Power_Manager --> Active_Superconducting
    Power_Manager --> Degraded_Superconducting

2. Operational Parameter Expansion

Derivative 2.1: Terahertz (THz) Frequency Power Islands with Dynamic Frequency Scaling

• Enabling Description: This implementation focuses on integrated circuits operating in the Terahertz (THz) frequency range, where each power island contains circuitry designed for THz signal processing (e.g., using resonant tunneling diodes or plasmonic interconnects). The power manager dynamically adjusts the operating frequency of individual THz power islands by controlling THz-source (e.g., photomixers or frequency multipliers) output power and resonant cavity parameters. The "target power level" for a THz island directly correlates with its required THz output power and processing throughput. Actions include scaling the THz clock signal frequency (e.g., 0.5 THz to 2 THz), modulating the amplitude of the THz carrier, or enabling/disabling THz-specific IP blocks. This allows for fine-grained power-performance trade-offs in next-generation high-bandwidth communication or imaging systems.

graph TD
    A[Application Request] --> B(PMCL)
    B --> C(MPM/IPM)
    C -- THz_Freq_Target, THz_Power_Target --> D{SPM for THz Island}
    D -- Control THz Source --> E(THz Clock & Power Generator)
    E --> F(THz Power Island Logic/IP)
    F -- THz_Freq_Actual, THz_Power_Actual --> D
    D -- Feedback --> C

Derivative 2.2: Extreme Temperature (Cryo/High-Temp) Tolerant Power Islands with Predictive Thermal Management

• Enabling Description: The integrated circuit is designed to operate under extreme temperature conditions, either cryogenically (e.g., -196°C) or at very high temperatures (e.g., +200°C for automotive under-hood applications). Each power island is equipped with integrated micro-thermocouples or resistance temperature detectors (RTDs). The power manager employs a predictive thermal model, fed by real-time sensor data from each island, to anticipate "hot spots" or cold-related performance degradation. The target power level is dynamically adjusted to maintain the island within its optimal operating temperature range or prevent thermal runaway/freeze-out. Actions include dynamically adjusting Vdd, clock frequency, or applying localized back-biasing to control leakage, or even pre-emptively powering down an island if thermal limits are projected to be exceeded, and restarting it after stabilization.

stateDiagram-v2
    state Normal_Op {
        entry / Monitor Temp Sensors
        exit / Update Thermal Model
        Normal_Op --> Overheat_Warning : Temp > Threshold_High_Warning
        Normal_Op --> Cold_Warning : Temp < Threshold_Low_Warning
    }
    state Overheat_Warning {
        entry / Reduce Power; Log Event
        exit / Re-evaluate Thermal State
        Overheat_Warning --> Thermal_Shutdown : Temp > Threshold_High_Critical
        Overheat_Warning --> Normal_Op : Temp < Threshold_High_Recover
    }
    state Thermal_Shutdown {
        entry / Power Off Island; Isolate
        Thermal_Shutdown --> Normal_Op : Temp < Threshold_High_Safe && Restart_Cmd
    }
    state Cold_Warning {
        entry / Increase Power; Log Event
        exit / Re-evaluate Thermal State
        Cold_Warning --> Cold_Shutdown : Temp < Threshold_Low_Critical
        Cold_Warning --> Normal_Op : Temp > Threshold_Low_Recover
    }
    state Cold_Shutdown {
        entry / Power Off Island; Maintain Standby Heat
        Cold_Shutdown --> Normal_Op : Temp > Threshold_Low_Safe && Restart_Cmd
    }

    [*] --> Normal_Op

Derivative 2.3: Ultra-Scale Data Center on Chip (DCoC) with Hierarchical Power Orchestration

• Enabling Description: This derivative scales the power island concept to an entire "Data Center on Chip" (DCoC) comprising hundreds to thousands of logically or physically segregated power islands. Each island represents a computational core, memory block, or I/O fabric portion. The power management hierarchy extends to multiple levels: Slave Power Managers (SPMs) at the individual core level, Intermediate Power Managers (IPMs) for racks/clusters of cores, and a Master Power Manager (MPM) orchestrating the entire DCoC. The MPM, often a dedicated power orchestration unit, dynamically allocates power budgets based on global workload scheduling and service level agreements (SLAs). Actions include migrating workloads between islands, throttling performance, or putting entire sections of the DCoC into deep sleep, optimizing for peak performance, energy cost, or cooling capacity across the entire system.

graph TD
    A[Global Workload Scheduler] --> B(MPM - DCoC Orchestrator)
    B -- Power Budget Allocation --> C1(IPM - Rack 1)
    B -- Power Budget Allocation --> C2(IPM - Rack 2)
    C1 -- Power Cmds --> D1(SPM - Core A)
    C1 -- Power Cmds --> D2(SPM - Core B)
    C2 -- Power Cmds --> D3(SPM - Core C)
    D1 -- Power/Freq Control --> E1[Power Island: Compute Core A]
    D2 -- Power/Freq Control --> E2[Power Island: Memory Block B]
    D3 -- Power/Freq Control --> E3[Power Island: I/O Fabric C]
    E1 -- Status/Telemetry --> D1
    E2 -- Status/Telemetry --> D2
    E3 -- Status/Telemetry --> D3
    D1 -- Aggregated Status --> C1
    D2 -- Aggregated Status --> C1
    D3 -- Aggregated Status --> C2
    C1 -- Global Status --> B
    C2 -- Global Status --> B

3. Cross-Domain Application

Derivative 3.1: Aerospace - Satellite Payload Power Management

• Enabling Description: In a satellite's onboard computer (OBC) or reconfigurable payload, power islands are defined for critical sub-systems such as communication transceivers, imaging sensors, attitude control processors, and telemetry units. The power manager, under control of the flight software, dynamically adjusts power to these islands based on orbital mechanics, mission phase (e.g., launch, orbit insertion, nominal operations, science data collection), and available solar array power. For instance, during eclipse, non-essential sensor islands might be powered down or placed in a low-frequency mode, while communication links are maintained at minimal power. Actions involve power cycling specific transponders, dynamically reconfiguring power converters (Vdd adjustment) for sensor arrays, or shifting clock frequencies for error-correction code (ECC) engines, optimizing for battery life and thermal stability in a radiation-hardened IC.

sequenceDiagram
    participant FS as Flight Software
    participant PM as Power Manager (MPM/IPM)
    participant SPM_Comm as SPM (Comm Transceiver)
    participant SPM_Sens as SPM (Imaging Sensor)
    participant SPM_Att as SPM (Attitude Control)

    FS->>PM: Mission_Phase_Update(Eclipse)
    PM->>SPM_Sens: Set_Rate(Low_Power_Mode)
    SPM_Sens->>SPM_Sens: Reduce_Vdd_Freq
    PM->>SPM_Comm: Set_Rate(Min_Comm_Power)
    SPM_Comm->>SPM_Comm: Reduce_Tx_Power_Freq
    PM->>SPM_Att: Maintain_Nominal_Power
    SPM_Att->>SPM_Att: No Change
    SPM_Sens->>PM: Status(Low_Power_Achieved)
    SPM_Comm->>PM: Status(Min_Power_Achieved)
    PM->>FS: Status(Power_Optimized_Eclipse)

    FS->>PM: Mission_Phase_Update(Science_Collection)
    PM->>SPM_Sens: Set_Rate(Full_Performance)
    SPM_Sens->>SPM_Sens: Increase_Vdd_Freq
    PM->>SPM_Comm: Set_Rate(High_Bandwidth)
    SPM_Comm->>SPM_Comm: Increase_Tx_Power_Freq
    PM->>FS: Status(Science_Ready)

Derivative 3.2: Automotive - Zonal ECU Power Optimization

• Enabling Description: In a modern vehicle's zonal electronic control unit (ECU) architecture, power islands are logically or physically segregated for functions like Advanced Driver-Assistance Systems (ADAS) processing (e.g., radar, lidar, camera fusion), infotainment, powertrain control, and body electronics. A central vehicle power manager (VPM), which acts as the MPM, dynamically controls the power levels of these zonal ECUs (IPMs/SPMs) based on driving conditions (e.g., highway cruising, city stop-and-go, parking), driver engagement, and active features. For example, during low-speed parking, high-performance ADAS processing for highway autonomy can be scaled down or powered off, while parking assist features are prioritized. Actions include enabling/disabling ADAS compute clusters (power islands), adjusting voltage/frequency for GPU cores in infotainment, or entering sleep modes for non-critical body control modules, all in real-time to optimize battery consumption and thermal dissipation.

graph LR
    A[Vehicle State/Driver Input] --> B(Vehicle Power Manager (MPM))
    B -- Power Policy --> C(ADAS Zonal ECU (IPM))
    B -- Power Policy --> D(Infotainment Zonal ECU (IPM))
    B -- Power Policy --> E(Powertrain Zonal ECU (IPM))

    C -- Control Signals --> C1[ADAS Sensor Fusion (SPM)]
    C -- Control Signals --> C2[ADAS Perception (SPM)]

    D -- Control Signals --> D1[Display Processor (SPM)]
    D -- Control Signals --> D2[Audio Processor (SPM)]

    C1 -- Vdd/Freq Adjust --> F1[Power Island: Radar/Lidar Processing]
    C2 -- Vdd/Freq Adjust --> F2[Power Island: Camera Vision Processing]
    D1 -- Vdd/Freq Adjust --> G1[Power Island: Graphics/UI]
    D2 -- Vdd/Freq Adjust --> G2[Power Island: Sound Codecs]

    F1 -- Status --> C1
    F2 -- Status --> C2
    G1 -- Status --> D1
    G2 -- Status --> D2
    C -- Status --> B
    D -- Status --> B
    E -- Status --> B

Derivative 3.3: Biomedical Implants - Energy Harvesting & Consumption Management

• Enabling Description: For long-term biomedical implants (e.g., neural interfaces, pacemakers, continuous glucose monitors), power islands are defined for sensing arrays, stimulation circuits, wireless communication modules, and data logging. The power manager intelligently balances energy consumption with often intermittent energy harvesting (e.g., thermoelectric, kinetic, RF-to-DC conversion) and limited battery capacity. It determines target power levels based on physiological activity, communication schedule, and battery state-of-charge. For example, during periods of low activity or sufficient battery charge, the implant might sample physiological data at a higher frequency, while during low battery or intense activity, non-critical sensing or communication is scaled down or temporarily disabled. Actions include reducing the sampling rate of bio-sensors, lowering the output power of telemetry radios, power-gating specific processing units, or adapting the duty cycle of stimulation pulses, all to extend the operational lifetime of the implant.

stateDiagram-v2
    state High_Battery {
        entry / Enable Full Functionality
        Normal_Op: High_Battery
        Normal_Op --> Low_Battery : Battery < Threshold_Low
    }
    state Low_Battery {
        entry / Enter Power_Saving Mode
        Reduced_Functionality: Low_Battery
        Reduced_Functionality --> Critical_Battery : Battery < Threshold_Critical
        Reduced_Functionality --> High_Battery : Battery > Threshold_Recharge
    }
    state Critical_Battery {
        entry / Minimal Functionality; Prioritize Life Support
        Shutdown_Imminent: Critical_Battery
        Shutdown_Imminent --> Low_Battery : Battery > Threshold_Emergency_Recharge
    }

    [*] --> High_Battery
    High_Battery --> PowerManager : Monitor_Energy_Harvesting
    Low_Battery --> PowerManager : Prioritize_Charging
    Critical_Battery --> PowerManager : Initiate_Emergency_Protocol

    PowerManager --> High_Battery
    PowerManager --> Low_Battery
    PowerManager --> Critical_Battery

    PowerManager -- Controls --> PowerIsland_Sensor
    PowerManager -- Controls --> PowerIsland_Comm
    PowerManager -- Controls --> PowerIsland_Stim

4. Integration with Emerging Tech

Derivative 4.1: AI-Driven Predictive Power Optimization

• Enabling Description: The power management system integrates a machine learning (ML) model (e.g., a Recurrent Neural Network or Transformer-based architecture) within the Power Management Control Layer (PMCL) and/or Master Power Manager (MPM). This ML model continuously analyzes historical workload patterns, real-time sensor data (temperature, current draw, voltage fluctuations from SPMs), and application-specific demands to predict future power requirements for each power island. Based on these predictions, the ML model proactively recommends or directly implements optimal power states (Vdd, frequency, Vt biasing, sleep modes) for individual power islands, thereby minimizing latency associated with reactive power scaling and preventing potential thermal excursions or brown-outs. Reinforcement learning can be employed for self-optimizing power policies over long operational periods.

flowchart TD
    A[Workload Data] --> B(Sensor Telemetry)
    C[Historical Data] --> B
    B --> D(Data Preprocessing)
    D --> E(ML Model Training - PMCL)
    E -- Trained Model --> F(ML Model Inference - MPM/IPM)
    F -- Predicted Optimal State --> G(Power Manager)
    G --> H(Power Island Control)
    H -- Actual Power State --> B

Derivative 4.2: IoT Sensor-Integrated Real-time Power Monitoring and Edge Control

• Enabling Description: Each power island is equipped with a micro-IoT sensor module, comprising ultra-low-power current, voltage, and temperature sensors, along with a tiny embedded microcontroller running an MQTT (Message Queuing Telemetry Transport) client. These IoT sensor modules, acting as distributed SPMs, publish real-time power consumption levels and environmental data from their respective power islands to a central MQTT broker. The power manager (MPM/IPM) subscribes to these topics, processes the aggregated sensor data at the edge, and issues optimized power commands (e.g., clock gating, dynamic Vdd scaling) back to the relevant IoT sensor modules/SPMs via MQTT, enabling highly granular, low-latency, and distributed power control across the IC, especially beneficial for large-scale SoCs or multi-chip modules.

graph TD
    A[Power Island 1] --> B1(IoT Sensor Module/SPM 1)
    A[Power Island N] --> B2(IoT Sensor Module/SPM N)
    B1 -- MQTT Publish (Telemetry) --> C(MQTT Broker)
    B2 -- MQTT Publish (Telemetry) --> C
    C -- MQTT Subscribe (Telemetry) --> D(Power Manager (MPM/IPM))
    D -- Decision Logic --> E(Power Policy Engine)
    E -- MQTT Publish (Commands) --> C
    C -- MQTT Subscribe (Commands) --> B1
    C -- MQTT Subscribe (Commands) --> B2
    B1 -- Power Control --> A
    B2 -- Power Control --> A

Derivative 4.3: Blockchain for Secure and Auditable Power Resource Allocation

• Enabling Description: For integrated circuits used in shared computing environments or critical infrastructure, a blockchain layer is introduced to manage and audit power resource allocation. Each power island is represented as a "node" on a permissioned blockchain. Power requests (e.g., "activate high-performance mode for Island X for Y duration") from applications or higher-level power management layers are recorded as transactions. The power manager acts as a smart contract executor, validating these requests against pre-defined power budgets and policies (e.g., maximum power, priority levels) before issuing actual power control commands. The power state changes (e.g., Vdd changed, frequency scaled) and actual power consumption reported by SPMs are cryptographically signed and stored on the blockchain, providing an immutable, transparent, and auditable log of power usage for billing, compliance, or forensic analysis.

sequenceDiagram
    participant App as Application/Client
    participant PMCL as PMCL Layer
    participant SmartContract as PowerAllocation Smart Contract
    participant MPM as Master Power Manager
    participant SPM as Slave Power Manager
    participant PI as Power Island

    App->>PMCL: PowerRequest(IslandID, TargetLevel, Priority)
    PMCL->>SmartContract: ProposePowerTx(IslandID, TargetLevel, Priority)
    SmartContract->>SmartContract: Validate (Budget, Policy, Permissions)
    alt Transaction Valid
        SmartContract-->>MPM: ExecutePowerCommand(IslandID, Action)
        MPM->>SPM: IssuePowerControl(Action)
        SPM->>PI: ApplyPowerChange(Action)
        PI->>SPM: ReportActualState(Power, Freq, Vdd)
        SPM->>MPM: SignedPowerReport(ActualState)
        MPM->>SmartContract: RecordPowerLog(SignedPowerReport)
        SmartContract->>SmartContract: UpdateBlockchain(PowerLog)
        SmartContract-->>PMCL: TxConfirmed(LogID)
        PMCL-->>App: PowerChangeAcknowledged
    else Transaction Invalid
        SmartContract-->>PMCL: TxRejected(Reason)
        PMCL-->>App: PowerRequestDenied
    end

5. The "Inverse" or Failure Mode

Derivative 5.1: Graceful Degradation and "Limp Home" Mode with Prioritized Island Operation

• Enabling Description: The integrated circuit incorporates a sophisticated "limp home" power management strategy. In response to a detected critical fault (e.g., external power supply failure, internal over-temperature event in a critical island, or significant power draw deviation) or a command to enter a low-power maintenance state, the power manager initiates a pre-programmed graceful degradation sequence. It identifies critical power islands (e.g., minimal CPU, emergency communication, safety monitoring) and prioritizes their operation, while non-essential islands are immediately powered off (hard-gated) or forced into an ultra-low-leakage sleep mode. The target power level for critical islands is automatically adjusted to the lowest functional setting required to maintain basic operation, ensuring system survivability or safe shutdown, effectively operating at a fraction of its nominal power budget and functionality.

stateDiagram-v2
    state Normal_Operation {
        entry / Full Functionality
        Normal_Operation --> Fault_Detected : Critical_Fault
        Normal_Operation --> Limp_Home_Command : User/System Cmd
    }
    state Fault_Detected {
        entry / Initiate Graceful Degradation
        Fault_Detected --> Limp_Home_Mode : Fault Assessment Complete
    }
    state Limp_Home_Command {
        entry / Prepare Limp Home Mode
        Limp_Home_Command --> Limp_Home_Mode : Configuration Loaded
    }
    state Limp_Home_Mode {
        entry / Prioritize Critical Islands; Power Off Non-Essential
        Limp_Home_Mode --> Recovery_Mode : Fault Cleared/Restore Cmd
        Limp_Home_Mode --> Emergency_Shutdown : Critical_System Failure
    }
    state Recovery_Mode {
        entry / Power Up Non-Essential; Restore State
        Recovery_Mode --> Normal_Operation : Full Power Up
    }
    state Emergency_Shutdown {
        entry / Immediate Power Off All Islands
        Emergency_Shutdown --> [*]
    }

    [*] --> Normal_Operation

Derivative 5.2: Ultra-Low Leakage Data Retention Mode with Shadow Registers

• Enabling Description: For power islands that contain critical state information (e.g., register values, configuration settings) that must be preserved during extended power-down periods, this derivative integrates "shadow registers" or non-volatile memory elements within each power island's power control circuitry. When the power manager determines an action to power off an island (or enter deep sleep), it first triggers a sequence to automatically save the state of all volatile registers within that island to the adjacent shadow registers. The power-off action then proceeds, reducing the island's Vdd to near-zero or completely disconnecting it, minimizing static leakage. Upon power-up, the power manager triggers the restoration sequence, reloading the saved state from the shadow registers, allowing for rapid and energy-efficient resumption of operation without full re-initialization. This includes the use of multi-threshold (Vt) transistors (high Vt for shadow registers, low Vt for active logic) and anti-glitch circuits during transitions.

sequenceDiagram
    participant PM as Power Manager
    participant SPM as Slave Power Manager
    participant PI as Power Island (Active Logic)
    participant SR as Shadow Registers (Non-Volatile)
    participant VDDM as Vdd Multiplexer

    PM->>SPM: PowerOffRequest(IslandID)
    SPM->>PI: Signal_Freeze_State
    PI->>SR: Save_State_Data
    SPM->>VDDM: Switch_Vdd_Off
    VDDM->>PI: Vdd_Off
    SPM->>PM: PowerOffAck

    PM->>SPM: PowerOnRequest(IslandID)
    SPM->>VDDM: Switch_Vdd_On
    VDDM->>PI: Vdd_On
    SPM->>SR: Restore_State_Data
    SR->>PI: Reload_State
    PI->>SPM: Ready_For_Operation
    SPM->>PM: PowerOnAck

Derivative 5.3: Self-Healing Power Islands with Reconfigurable Redundancy

• Enabling Description: This derivative implements power islands with built-in reconfigurable redundancy and self-healing capabilities. Each power island comprises multiple sub-islands or redundant functional blocks. The power manager (MPM/IPM/SPM) continuously monitors the operational health and power integrity of each sub-island (e.g., current transients, voltage drops, internal clock errors, temperature deviations). Upon detecting a localized failure or performance degradation within a sub-island (e.g., due to aging or soft error), the power manager determines an action to isolate the failing sub-island, re-route power and signals to a redundant healthy sub-island, and/or dynamically adjust the power and frequency of the remaining healthy sub-islands to compensate for the lost capacity. This enables continuous operation, albeit potentially at a reduced performance or increased power level, in the presence of internal faults, enhancing reliability in critical applications.

graph TD
    A[Power Manager] --> B{Power Island with Redundancy}
    B -- Health Monitoring --> C1[Sub-Island A]
    B -- Health Monitoring --> C2[Sub-Island B (Redundant)]
    B -- Health Monitoring --> C3[Sub-Island C]

    C1 -- Fault Detected --> D(Fault Detection & Isolation)
    D --> E{Reconfiguration Logic}
    E -- Power/Signal Reroute --> C2
    E -- Adjust Power/Freq --> C1, C3
    C2 -- Active --> F[Compensated Operation]

    F -- Status --> B
    B -- Aggregated Status --> A

---

Combination Prior Art Scenarios with Open-Source Standards

Here are three scenarios combining the concepts of US Patent 7051306 with existing open-source standards, demonstrating how these integrations would constitute obvious advancements.

1. US7051306 and the AMBA (Advanced Microcontroller Bus Architecture) Standard

• Enabling Description: An integrated circuit (IC) is divided into multiple power islands, each with independently controlled power consumption. A hierarchical power management system, as described in US7051306 (comprising MPM, IPMs, and SPMs), is implemented. The communication backbone for control signals, status reporting, and power management commands between the power managers and the power islands themselves is built upon the ARM AMBA (Advanced Microcontroller Bus Architecture) 5 AHB/AXI protocol. Specifically, the power manager firmware (e.g., running on a CPU in a power island) communicates with the MPM via memory-mapped registers accessible over an AMBA AXI interconnect. The SPMs within each power island utilize an AMBA AHB-Lite interface for low-latency, dedicated communication with their respective IP blocks and local power control circuitry. This integration standardizes the communication infrastructure, making power management instructions and telemetry interoperable across diverse IP cores within the SoC.
  • Open-Source Standard: AMBA (Advanced Microcontroller Bus Architecture) Specification, freely available from ARM Holdings.

2. US7051306 and the RISC-V Instruction Set Architecture (ISA)

• Enabling Description: An integrated circuit includes several power islands. The power manager, instead of being a generic microcontroller or hardwired logic, is implemented as a dedicated RISC-V processor core (e.g., a simple RV32I core for SPMs or a more complex RV64GC core for the MPM). This RISC-V core executes the power manager firmware (as described in US7051306, e.g., PM firmware 224, GPAL 214, PMCL 218). The RISC-V processor dynamically determines target power levels, orchestrates actions (frequency scaling, voltage modification, power gating) for its respective power islands by writing to control registers of the power control circuitry. The extensibility of the RISC-V ISA allows for custom instructions to be added specifically for fast power state transitions or complex power monitoring operations.
  • Open-Source Standard: RISC-V Instruction Set Architecture, maintained by RISC-V International.

3. US7051306 and the OpenTelemetry Standard for Monitoring

• Enabling Description: An integrated circuit features multiple power islands, each monitored by a Slave Power Manager (SPM). These SPMs collect detailed power consumption levels, temperature, voltage, and frequency data. Instead of proprietary status messages, each SPM integrates an OpenTelemetry SDK (Software Development Kit). The SPMs export their collected telemetry data (metrics like current, voltage, temperature, events for power state changes) in an OpenTelemetry-compatible format (e.g., OTLP - OpenTelemetry Protocol) to a local OpenTelemetry Collector running within the integrated circuit (perhaps as part of an IPM or MPM). This collector then forwards the aggregated, standardized power management telemetry to an external observability system for analysis, debugging, and visualization. This enables seamless integration with existing monitoring tools and simplifies the development of diagnostic and optimization applications for complex SoCs.
  • Open-Source Standard: OpenTelemetry, an Apache 2.0 licensed project from the Cloud Native Computing Foundation (CNCF).

---