Proposals/canton-foundation/canton-dev-fund/#663GitHub ↗

Incoming#663 Pull request

Proposal: KAILEdge - Hardware-Anchored Measurement and Verification for Carbon Credit Assets on Canton

Opened 24-08-2026 by Zenoffi, last activity 11d ago

Incoming58 days so far24-08-2026
Needs Championnot reached
In Reviewnot reached
Ready for Votenot reached
Voting Livenot reached
Approvednot reached

Waiting on: Waiting on the committee; idle 11 days
Funding request: $120,000
Champion named in file: Parth Chaturvedi (Canton Foundation)
Discussion: 6 messages; 3 people

Development Fund Proposal Submission

Proposal file: rfps/financial-markets-standards-verification/kailedge-cbg-measurement-pilot.md

Summary

This proposal funds Phase 1 of a proposed pilot at a Compressed Biogas (CBG) facility at Harohalli, Karnataka, India, operated by Sustainable Impacts. The pilot would build a reference integration that takes device-sealed plant measurements into Daml contracts on Canton, and would publish the open-source integration layer.

RFP category: Primary: RWA Standards (RFP 12), area "Daml and Institutional RWA Workflow Standards". Secondary: Public verifiability (RFP 11). No published RFP covers physical-world measurement. If the committee considers this outside the RFPs, we ask that it be considered as an individual initiative.

Champion: Parth Chaturvedi (Canton Foundation)

The problem

Carbon credit verification is retrospective. Physical production is continuous, but evidence is assembled after the fact from fragmented telemetry, batch sampling and modelling assumptions. This leaves producers waiting for verification and gives buyers evidence that is hard to check independently.

217 CBG plants were functional in India as of 31 July 2026 (GOBARdhan portal figure reported in The Hindu BusinessLine, https://www.thehindubusinessline.com/markets/commodities/india-achieves-105-cbg-blending-in-fy26-ahead-of-1-target/article71314065.ece). The SATAT programme plans 5,000 plants by 2030 (PNGRB CBG Cell, https://pngrb.gov.in/pdf/confluence/SESSION-8-CBG-CELL.pdf).

The solution

KAILEdge is designed to run a deterministic physics model at the edge, compute physical state from calibrated sensors, and seal each result in TPM 2.0 hardware. The certificate is designed to record device identity, model version, calibration lineage, input hash, output state vector and signature. An authorised party with the inputs and the model version should be able to replay the computation and check the hash. The approach is tamper-evident, replayable and independently checkable. It is trust-minimised, not trustless. Sensing is read-only and online. KAILEdge is a monitoring and verification layer: it does not issue or manage carbon credits. Issuance stays with registries, verifiers and issuers.

Why the ecosystem needs this

Canton's RWA direction needs ways to bring evidence about physical assets into Daml contracts. Tokens that represent carbon credits or industrial production commonly depend on data supplied by a reporter, which attests to a value rather than to the computation that produced it. We are not aware of an open reference implementation on Canton that checks a device-sealed, replayable measurement on ledger.

Who benefits: registries, verifiers and issuers (a replayable record to reference), buyers (evidence they can check), CBG producers (continuous rather than batch evidence), and Canton developers (open templates, schema, conformance tests and replay guide).

This pilot would produce:

A Daml contract layer that verifies a hardware-sealed certificate
Measurement-backed asset templates that a registry, verifier or issuer can reference (issuance authority stays with them)
A documented replay pathway that a verifier or buyer can follow
A reusable certificate schema and conformance tests for later environmental assets

Public good: the integration layer is Apache 2.0 and designed for any issuer or registry. Only KAILEdge's calibrated physics parameters are proprietary.

Related work on Canton: Xpansiv announced a phased initiative to enable tokenization of environmental assets on Canton (https://www.xpansiv.com/xpansiv-to-enable-tokenized-environmental-asset-infrastructure-via-canton-network/), and ClimateTrade has announced joining as a network validator (https://www.linkedin.com/posts/climatetrade_activity-7394333673728221184-Korb). Neither has agreed to use KAILEdge.

Market context

The Union Cabinet approved the GOBARdhan scheme for compressed biogas on 6 August 2026, with an outlay of Rs 23,731 crore for FY 2026-27 to FY 2035-36. The Government states the scheme aims to increase domestic CBG production nearly ten-fold (PIB, https://www.pib.gov.in/PressReleasePage.aspx?PRID=2295480). Business Standard reports the target as about 5 million standard cubic metres per day (https://www.business-standard.com/industry/news/ril-plans-1-trn-investment-to-set-up-cbg-projects-in-andhra-pradesh-126100200970_1.html).

On 2 October 2026, Anant Ambani, Executive Director of Reliance Industries, said Reliance plans to invest Rs 1 trillion in compressed biogas plants in Andhra Pradesh (Business Standard, same link; Moneycontrol, https://www.moneycontrol.com/news/india/anant-ambani-announces-rs-1-lakh-crore-reliance-investment-in-andhra-pradesh-backs-tech-led-farming-14043461.html).

These are public statements about the CBG market. They do not indicate any relationship with, or interest in, KAILEdge or this pilot.

Potential credit buyers and marketplaces

KAILEdge does not issue credits, and no party below has an agreement or stated interest in this pilot. These are platforms a registry or issuer could use for credits backed by measured data from the pilot, as published by each party:

Xpansiv: announced (8 Oct 2025) a phased initiative to enable tokenization of environmental assets on Canton Network, starting with a pilot (https://www.xpansiv.com/xpansiv-to-enable-tokenized-environmental-asset-infrastructure-via-canton-network/). Canton link: announced, pilot stage.
ClimateTrade: carbon, plastic and biodiversity credit marketplace (https://www.climatetrade.com/en/about-us/). Announced on 12 Nov 2025 that it joined Canton Network as a network validator (https://www.linkedin.com/posts/climatetrade_activity-7394333673728221184-Korb).
Carbonplace: bank-founded platform for credit management and trading, 14 registries (https://carbonplace.com/). No Canton link found.
Toucan: tokenized carbon credits, says $100 million onchain (https://docs.toucan.earth/). No Canton link found.

Whether any of them accepts Indian CBG credits has not been checked.

Pilot status

We are in discussions with Sustainable Impacts about hosting the pilot. No agreement is signed. Hardware is specified and supplier quotations are being collected. Nothing has been ordered or installed. T1 begins only after a signed host-site agreement.

The grant is requested to fund the hardware, installation and field deployment. Total ask: $120,000 across four tranches. Our separate Canton Foundation application is under evaluation and is not secured.

Milestones and funding

Tranche	Amount	Deliverable
T1 - Hardware deployment and first data	$60,000	Three device sets (207 devices) procured, 69 installed at Harohalli, first certificates generated and checked on Canton testnet
T2 - Physics validation and certificate pipeline	$25,000	Equations calibrated against field data, methane mass uncertainty characterised, TPM-sealed certificates verified by independent replay
T3 - Canton integration	$20,000	Daml contracts deployed on testnet, certificates consumed by smart contracts, measurement-backed asset template exercised end to end on testnet
T4 - Reference case and acceptance	$15,000	One full production cycle complete, verifier evidence package assembled, public reference case published, outside review of schema and replay guide sought and published
Total	$120,000

What $60,000 in T1 covers

207 devices across three complete sets (69 for the live installation plus 138 held as calibrated spares)
Installation and commissioning at the Harohalli CBG facility
Calibration of Class A reference sensors by an NABL-accredited laboratory (to be engaged)
Edge compute units (Intel N305, TPM 2.0, IP67-rated)
Site engineering and field support during deployment

Open and proprietary components

Open source (Apache 2.0): Daml templates and interfaces, certificate schema, conformance tests, replay guide. Proprietary: the calibrated physics parameters. A third party can verify a certificate's signature, integrity and chain without KAILEdge software. Reproducing KAILEdge's calibrated outputs requires KAILEdge. We are not claiming otherwise.

Honest position on Canton adoption

No external Canton application currently consumes KAILEdge certificates. The pilot would create a reference integration and produce a reusable schema, Daml contract templates and a documented replay pathway for later applications. We are not claiming existing Canton adoption.

Acceptance criteria

The Tech & Ops Committee will evaluate completion against:

A reproducible, TPM-sealed certificate is generated from calibrated field sensors at the Harohalli facility.
An independent reviewer can replay the certificate from the recorded inputs and model version and verify the hash.
The certificate is consumed by a Daml smart contract on Canton testnet.
Documentation and knowledge transfer are complete: reference case published, replay guide delivered, open-source integration layer released.
The reference case reports measured results against the stated targets, including misses.

Key risks

Host-site agreement not signed (T1 does not start until it is). Calibration delays. Testnet integration complexity. Field performance of the physics model against the 8% uncertainty target. Methodology and registry acceptance is not assumed. Full table in the proposal file.

Checklist

Proposal file added (path above)
Milestones and funding amounts defined
Acceptance criteria included
Alignment with Canton priorities described

Notes for Reviewers

The earlier comments on this PR (container shipping, parametric insurance, reinsurer engagement, performance figures, the earlier $100,000 request) describe the previous scope and are superseded by this description and the proposal file. On the 8 Sep adoption question: no Canton application consumes KAILEdge today, and the proposal says so.

Contact

Prasad Gopal, Founder, KAILEdge, prasad@kailedge.com

Milestones 4

Hardware deployment and first dataMonth 1-2
Physics validation and certificate pipelineMonth 2-3
Canton integrationMonth 3-4
Reference case and acceptanceMonth 4-5

Discussion6 comments, 0 reviews, 0 inline comments

Showing automated messages too. Hide bot messages

Showing 1 to 9 of 9, oldest first.

github-actions[bot]commented24-08-2026 · 1mo agoGitHub ↗

SIG labels auto-detected and applied: financial-workflows-composability

If this is incorrect, you can update the labels from the right sidebar.

github-actions[bot]commented24-08-2026 · 1mo agoGitHub ↗

Champion identified Parth Chaturvedi, Canton Foundation

The committee will verify this champion during review.

Zenofficommented24-08-2026 · 1mo agoGitHub ↗

I've added a high-level pipeline overview to the supporting repository: https://github.com/Zenoffi/kailedge-canton-integration/blob/main/docs/PIPELINE.md

The pipeline describes the full flow from physical sensors to TPM-sealed forensic certificates. The open-source Daml integration layer is the settlement interface; the proprietary physics engine is the verification layer.

Zenofficommented25-08-2026 · 1mo agoGitHub ↗

Performance Benchmarks

The following benchmarks validate the 41M data points per second throughput and 1.8 µs core physics reflex claimed in the proposal.

KAIL Edge Efficiency Breakdown

[Image 1]

Memory bandwidth: Optimized data movement
SIMD: Full AVX2 utilization
Cache: L1/L2 residency confirmed
Overall: System is bottleneck-free

Performance Contribution Breakdown

[Image 2]

Optimization	Speedup
Cache Optimization	~8×
SoA Memory Layout	~7×
AVX2 Vectorization	~5×
Lookup Tables	~6×
Batch API	~4×

Full benchmark methodology available in the dossier (Technical Annex B).

Zenofficommented26-08-2026 · 1mo agoGitHub ↗

KAILEdge Architecture Diagrams

The following diagrams illustrate the complete KAILEdge workflow — from physical sensors to Canton settlement.

Full Architecture Workflow

flowchart TB
    subgraph Physical["Physical Layer (Asset)"]
        S1[PT1000 Temperature Sensor]
        S2[Humidity Sensor]
        S3[Vibration Sensor]
        S4[GPS]
        S5[Thermal Camera 32×24]
        S6[RGB Camera]
        S7[LiDAR]
        S8[Door Sensor]
    end

    subgraph Edge["KAILEdge Edge Device (LattePanda IOTA)"]
        subgraph Pipeline["Data Pipeline"]
            direction LR
            IN[("Sensor Ingest<br>~1 µs")]
            PHYS[("Deterministic Physics<br>1.8 µs")]
            ML[("ML Classifier<br>~5 µs")]
            TPM[("TPM Seal<br>~100 µs")]
            CERT[("Forensic Certificate<br>~1 µs")]
            
            IN --> PHYS --> ML --> TPM --> CERT
        end
        
        subgraph Physics["Physics Engine (62 CAs)"]
            direction TB
            CA1["CA 1-10: Thermodynamics"]
            CA2["CA 11-20: Mechanical"]
            CA3["CA 21-30: Electrical/Battery"]
            CA4["CA 31-40: Fluid/Thermal"]
            CA5["CA 41-48: Environmental"]
            CA6["CA 49-56: Sensor Degradation"]
            CA7["CA 57-62: Safety & Thermal Runaway"]
        end
        
        subgraph Coupling["21×21 Cross-Coupling Matrix"]
            C1["Temperature → Moisture (0.047)"]
            C2["Moisture → Microbial (0.012)"]
            C3["Microbial → Enzymatic (0.89)"]
            C4["Temperature → Microbial (0.23)"]
            C5["Moisture → Enzymatic (0.005)"]
            CX["...3,844 coefficients"]
        end
        
        subgraph ML_Class["ML Classifier"]
            F1["8 Physics Features"]
            H1["16 Neurons"]
            H2["16 Neurons"]
            OUT["4 Classes<br>NORMAL·CAUTION·CRITICAL·SPOILED"]
            F1 --> H1 --> H2 --> OUT
        end
        
        subgraph TPM_Core["TPM 2.0"]
            PCR["PCR 12 Extension"]
            SHA["SHA3-256 Hash"]
            QUOTE["TPM Quote"]
            PCR --> SHA --> QUOTE
        end
        
        subgraph Certificate["Forensic Certificate"]
            CID["Certificate ID"]
            DEV["Device ID"]
            TS["Timestamp"]
            GPS["GPS Location"]
            SL["Shelf Life %"]
            ST["State"]
            TPM_DIGEST["TPM PCR Digest"]
            SIG["Signature"]
            PREV["Previous Cert Hash"]
        end
    end

    subgraph Connectivity["Connectivity Layer"]
        STARLINK["Starlink Satellite"]
        LTE["LTE-M / 5G"]
        OFFLINE["Offline Queue<br>50,000 Certificates"]
    end

    subgraph Oracle["Oracle Coordination Layer"]
        CHAINLINK["Chainlink CRE"]
        VALIDATE["Signature Validation"]
        ROUTE["Route to Contracts"]
        CCIP["Cross-Chain CCIP"]
    end

    subgraph Settlement["Settlement Layer"]
        CANTON["Canton Network"]
        POLICY["Parametric Policy"]
        ATTEST["Breach Attestation"]
        PAYOUT["Payout Authorization"]
        FINALITY["<2s Finality"]
    end

    subgraph Stakeholders["Stakeholders"]
        REINSURER["Reinsurer<br>Mandates KAILEdge"]
        SHIPPER["Shipper<br>Pays for Kit"]
        INSURER["Insurer<br>Issues Policy"]
        DATA["Data Companies<br>Bloomberg·Reuters·S&P"]
    end

    S1 --> IN
    S2 --> IN
    S3 --> IN
    S4 --> IN
    S5 --> IN
    S6 --> IN
    S7 --> IN
    S8 --> IN
    
    IN --> PHYS
    PHYS --> Coupling
    PHYS --> Physics
    
    PHYS --> ML
    ML --> ML_Class
    
    ML --> TPM
    TPM --> TPM_Core
    TPM --> CERT
    
    CERT --> OFFLINE
    CERT --> STARLINK
    CERT --> LTE
    
    STARLINK --> CHAINLINK
    LTE --> CHAINLINK
    OFFLINE --> CHAINLINK
    
    CHAINLINK --> VALIDATE
    VALIDATE --> ROUTE
    ROUTE --> CCIP
    
    CCIP --> CANTON
    CANTON --> POLICY
    POLICY --> ATTEST
    ATTEST --> PAYOUT
    PAYOUT --> FINALITY
    
    REINSURER -->|"Mandates"| SHIPPER
    SHIPPER -->|"Pays for Kit"| Edge
    INSURER -->|"Issues Policy"| POLICY
    
    CERT -->|"Data Feed"| DATA
    DATA -->|"Licensing"| REINSURER
    DATA -->|"Trading Signals"| SHIPPER

    style Physical fill:#1a3a5c,color:#fff
    style Edge fill:#0d2b3e,color:#fff
    style Connectivity fill:#2d4a6b,color:#fff
    style Oracle fill:#3a5a7a,color:#fff
    style Settlement fill:#4a6a8a,color:#fff
    style Stakeholders fill:#1a4a3c,color:#fff

Sequence Diagram: End-to-End Flow

sequenceDiagram
    participant Sensors as Physical Sensors
    participant Engine as KAILEdge Engine
    participant TPM as TPM 2.0
    participant Cert as Forensic Certificate
    participant Chainlink as Chainlink CRE
    participant Canton as Canton Network
    participant Contract as Daml Contract
    participant Reinsurer as Reinsurer

    Note over Sensors, Reinsurer: 1.8 µs Physics → 108.8 µs Total → 60-Second Settlement

    Sensors->>Engine: Raw Data (41M pts/sec)
    Engine->>Engine: 21-62 CAs Cross-Coupled
    Engine->>Engine: 50+ Non-Linear Equations (1.8 µs)
    Engine->>TPM: State Vector + ML Classification
    TPM->>TPM: PCR 12 Extension (SHA3-256)
    TPM->>Cert: TPM Quote + Signature
    Cert->>Chainlink: Push Certificate (Offline Queue)
    Chainlink->>Chainlink: Validate Signatures
    Chainlink->>Canton: Route to Contracts
    Canton->>Contract: Submit Attestation
    Contract->>Contract: Validate Policy Rules
    Contract->>Contract: Verify Certificate
    Contract->>Contract: Check Sensor Binding
    Contract->>Contract: Check Coverage Window
    Contract->>Contract: Check Breach Rule
    Contract->>Reinsurer: Automatic Payout (60 Seconds)
    Note over Reinsurer: No Adjuster. No Dispute. No 647 Days.

Value Layer: Who Benefits

quadrantChart
    title KAILEdge Value Proposition
    x-axis "Low Impact" --> "High Impact"
    y-axis "High Value" --> "Low Value"
    quadrant-1 "Strategic Advantage"
    quadrant-2 "Operational Efficiency"
    quadrant-3 "Compliance"
    quadrant-4 "Experimental"
    Reinsurers: [0.85, 0.15]
    Shipping Companies: [0.75, 0.25]
    Carbon Registries: [0.70, 0.30]
    Infrastructure Owners: [0.65, 0.35]
    Insurers: [0.60, 0.40]
    Hedge Funds: [0.55, 0.45]
    Regulators: [0.50, 0.50]
    Data Companies: [0.45, 0.55]

Performance Metrics

Stage Time Sensor Ingest ~1 µs Physics Computation ~1.8 µs ML Classification ~5 µs TPM Sealing ~100 µs Certificate Generation ~1 µs Total Pipeline ~108.8 µs Settlement 60 Seconds

What This Shows

Diagram Purpose Architecture Workflow Complete system from sensors to settlement Sequence Diagram End-to-end flow with timing and stakeholders Value Layer Who benefits and where the value is

Jatinp26membercommented08-09-2026 · 1mo agoGitHub ↗

@Zenoffi I'd like to know the need / evidence of adoption for this toolings in canton ecosystem, do you have any existing canton projects who'd be wanting this?

How's this gonna be a public good Infra for builders and developers or in general users on canton network?
The equations are public, the architecture is open-source, but the calibration data is proprietary. This means any Canton developer can use the integration layer but cannot reproduce KAILEdge's actual physics outputs without engaging KAILEdge commercially.

Overall, I am not able to get the requirement and impact of this proposal/tooling for current ecosystem builders/

Zenofficommented08-09-2026 · 1mo agoGitHub ↗

@Jatinp26 Hi Jatin,

Thanks for taking time to review the proposal. Let me address your questions directly.

The Beachhead and the Autonomous Endgame

Canton Network is built for tokenized real-world assets. But a tokenized asset is blind without a verifiable physical state.

Global supply chain is our beachhead. It is the perfect vertical to prove the architecture because the industry loses $1.2 trillion annually to spoilage and 647-day dispute cycles by relying on probabilistic cloud data instead of proof. The problem is well-known, the pain is acute, and the solution is immediate.

But the true endgame is the autonomous era. Over the next decade, millions of humanoid robots, autonomous vehicles, and industrial drones will come online. Every autonomous action creates a liability risk. To be insurable, these systems require a deterministic, court-admissible audit trail of their physical state.

Probabilistic AI cannot be cross-examined in court. TPM-sealed physics can.

KAILEdge is the mandatory liability and verification layer for this $5T+ autonomous economy. By anchoring KAILEdge to Canton, Canton becomes the default settlement layer for autonomous liability and RWA tokenization.

This is the public good.

As autonomous systems scale, developers from across the ecosystem — robotics companies, autonomous vehicle fleets, drone operators, industrial automation firms — will need to integrate KAILEdge into their Canton-based applications.

KAILEdge provides the shared infrastructure layer that every autonomous system needs, making it a foundational public good for builders and developers on the Canton Network.

Public Good Infrastructure

The integration layer—Daml contracts, API bindings, and parametric insurance templates—will be open-sourced under Apache 2.0.

This gives every Canton developer the scaffolding to build physics-verified applications without needing to rebuild the core physics engine. The reference implementation is designed to accelerate institutional adoption on Canton.

The open-source layer ensures that any builder can leverage KAILEdge's verification capabilities without being locked into a proprietary system. The reference implementation serves as a template for the ecosystem, reducing the time and cost for developers to build physics-verified applications.

The Open / Proprietary Split

You are correct regarding the calibration data. It is strictly proprietary. The split is as follows:

· Open-source (Apache 2.0): Daml contracts, API bindings, integration templates — for ecosystem adoption and developer access. · Proprietary: Physics engine — core IP. · Trade Secret: Calibration data (activation energies, coupling coefficients) — the moat.

A developer uses the open-source layer to interface with KAILEdge commercially. The equations are public, but the calibrated parameters are the proprietary moat. This is exactly how oracle networks operate.

Status and the Adoption Path

The core physics engine is already built, benchmarked, and running at a 1.8 µs core reflex. We are not in the R&D phase; we are in the deployment phase.

We are currently in active engagement with global reinsurers, and we are in talks with hardware manufacturers to meet international ruggedization standards (IP67, industrial isolation) for mass deployment.

The deployment model is mandate-led.

When a reinsurer mandates KAILEdge for parametric coverage, every policyholder deploys hardware. The premium reserves and payout pools for those policies are settled on-chain—driving institutional TVL to Canton.

One mandate will give access to over 10,000+ containers in the shipping industry for the first year, with a target of 50,000 units. This integration brings massive, mandated institutional liquidity to the network.

The Scope of the Grant

The $100,000 grant is not funding R&D or theoretical claims. It funds the bridge: the hardware test rigs, the Daml integration engineering, and the IP filings required to lock the priority date.

The acceptance criterion is absolute:

Demonstrate one reproducible, cryptographically attested physical breach flowing through a Daml contract on Canton to one correctly authorized settlement outcome.

The architecture is complete. This grant accelerates its arrival on Canton.

Best regards,

Prasad Gopal Founder & Chief Architect, KAILEdge

github-actions[bot]commented22-09-2026 · 28d agoGitHub ↗

The Canton Foundation Dev Fund is moving to Dev Fund 2.0

Thanks for making and opening a PR for your proposal. The Developer Fund is transitioning, which shifts toward RFP-scoped proposal system and clearer problem & adoption requirement framing.

Full details on Dev Fund 2.0: https://github.com/canton-foundation/canton-dev-fund/blob/main/2026-2028-strategic-roadmap.md#requests-for-proposals

To keep your proposal PR active, please update it by within the next 10 days with:

The relevant RFP category it maps to
Why the ecosystem needs it i.e. who benefits from it and how it drives adoption

Proposals not updated by then will get their PRs closed and will not be considered for the next Dev Fund cycle.

BhaskarPeruricommented10-10-2026 · 11d agoGitHub ↗

Hi guys, I’ll be handling the technical implementation of this project, including Daml smart contracts on Canton, backend development, and frontend integration. My focus will be on building the end-to-end integration between the measurement certificate pipeline and Canton, ensuring reliable on-ledger verification and a smooth application workflow.

Looking forward to contributing to the KAILEdge project and the broader Canton ecosystem!

Proposal file rfps/financial-markets-standards-verification/kailedge-cbg-measurement-pilot.md, 18 KB

Development Fund Proposal

Author: Prasad Gopal, Founder, KAILEdge
Status: Draft
Created: 2026-10
Label: financial-workflows-composability
Champion: Parth Chaturvedi (Canton Foundation)
RFP Category: Primary: RWA Standards (RFP 12), area "Daml and Institutional RWA Workflow Standards". Secondary: Public verifiability (RFP 11). No published RFP covers physical-world measurement. If the committee considers the work outside the RFPs, we ask that it be considered as an individual initiative.

Abstract

This proposal funds Phase 1 of a proposed pilot at a Compressed Biogas (CBG) facility at Harohalli, Karnataka, India, operated by Sustainable Impacts. The pilot would build a reference integration from device-sealed plant measurements to Daml contracts on Canton, and release the open-source integration layer.

The total funding request is $120,000 across four tranches. The pilot is planned to run one full production cycle (about 5 months). Dates depend on a signed host-site agreement.

Specification

1. Objective

Problem. Carbon credit verification is retrospective. Physical production is continuous, but evidence is assembled after the fact from fragmented telemetry, batch sampling and modelling assumptions. Producers wait for verification and buyers get evidence that is hard to check independently.

India's CBG sector: 217 plants were functional as of 31 July 2026 (https://www.thehindubusinessline.com/markets/commodities/india-achieves-105-cbg-blending-in-fy26-ahead-of-1-target/article71314065.ece). The SATAT programme plans 5,000 plants by 2030 (https://pngrb.gov.in/pdf/confluence/SESSION-8-CBG-CELL.pdf).

Intended outcome. Show that a hardware-anchored measurement pipeline can run in field conditions and be consumed on Canton, and leave a reusable open reference implementation. The pipeline has three parts:

A deterministic physics model running at the edge (design: 62 cross-coupled cellular automata across 21 coupled physical domains).
TPM 2.0 hardware-sealed certificates (design: 128-byte records).
Daml smart contracts on Canton that verify certificates and support measurement-backed asset workflows.

KAILEdge is a monitoring and verification layer. It does not issue or manage carbon credits. Issuance stays with registries, verifiers and issuers.

2. Implementation Mechanics

Three-layer architecture at the Harohalli CBG facility. Sensing is read-only: no actuators and no connection to plant control systems. Measurement is online only.

Layer 1 - Instrumentation. PT100 temperature probes (digester, gas line, ambient); pressure transmitters (gas line, compression); pH and ORP electrodes (digester health); methane flow reference measurement; NDIR CH4/CO2 sensors and H2S detectors (trend monitoring); MEMS vibration sensors (equipment health); Modbus energy meters (subsystem power monitoring); feedstock mass measurement.

Measurement classes:

Class A: reference sensors, calibrated by an NABL-accredited laboratory, used to validate the physics model.
Class B: operational telemetry, continuous monitoring, not certified.
Class C: derived quantities computed by the physics engine.

Layer 2 - Deterministic physics. 62 cross-coupled cellular automata; 21 physical domains coupled through a 21x21 matrix; 50+ non-linear equations. Target: combined methane mass uncertainty of 8% or less during the pilot. This is a target, not a measured result. T2 characterises the uncertainty from field data.

Layer 3 - Hardware seal. The state vector is sealed in TPM 2.0 (PCR 12 extension, SHA3-256 hash, device-bound signature, hardware-anchored identity). Output: a 128-byte certificate designed to be tamper-evident and independently replayable. Trust-minimised, not trustless: calibration, installation and methodology still require trust.

Canton side (Daml contracts, planned).

Verify the certificate signature and integrity.
Record the verified measurement against a measurement-backed asset template.
Where a methodology is applied, compute derived quantities (for example CO2-equivalent and conservativeness deduction) as that methodology defines. The methodology is not yet confirmed and depends on the project's registry and verifier.
Expose the verified record to a registry, verifier or issuer, who keep issuance authority.
Support transfer and retirement workflows through template interfaces. The templates are intended to implement the Canton Network Token Standard (CIP-0056, https://docs.canton.network/overview/reference/cip-0056) interfaces where applicable.

Edge compute platform. Intel N305 (8C/8T, 15W TDP), 16 GB RAM, on-board TPM 2.0, fanless IP67-rated enclosure. During connectivity loss, certificates are queued locally.

Hardware scope. Three complete sets of 69 devices, 207 devices to be procured under T1. 69 installed live; 138 held as calibrated spares so measurement continues during calibration windows. Status today: specified, supplier quotations being collected, nothing ordered or installed.

Open and proprietary. Apache 2.0: Daml templates and interfaces, certificate schema, conformance tests, replay guide. Proprietary: the calibrated physics parameters. A third party can verify a certificate's signature, integrity and chain without KAILEdge software. Reproducing calibrated outputs requires KAILEdge. We are not claiming otherwise.

3. Architectural Alignment

Canton RWA direction. The pilot would extend Canton's RWA work into measurement-backed environmental assets. The Daml layer is written for multiple issuers, registries and applications, not a single KAILEdge-specific contract.

Related environmental asset work. Xpansiv announced a phased initiative to enable tokenization of environmental assets on Canton (https://www.xpansiv.com/xpansiv-to-enable-tokenized-environmental-asset-infrastructure-via-canton-network/). ClimateTrade has announced joining as a network validator (https://www.linkedin.com/posts/climatetrade_activity-7394333673728221184-Korb). KAILEdge's measurement layer is intended to complement platforms like these. Neither has agreed to use it.

Token standard. Templates intended to implement CIP-0056 interfaces where applicable.

4. Backward Compatibility

No backward compatibility impact. New deployment at a new facility. No change to Canton protocol components.

Milestones and Deliverables

Dates assume a signed host-site agreement with Sustainable Impacts. Month 1 starts at signature.

Milestone T1: Hardware deployment and first data

Estimated Delivery: Month 1-2
Focus: Get calibrated sensors and edge nodes running at the plant.
Deliverables / Value Metrics: 207 devices procured (three complete sets); 69 installed at the Harohalli CBG facility; edge compute units deployed with TPM 2.0; Class A reference sensors calibrated by an NABL-accredited laboratory; first data transmitted from plant to edge node; first certificates generated and checked on Canton testnet.

Milestone T2: Physics validation and certificate pipeline

Estimated Delivery: Month 2-3
Focus: Validate the model against field data.
Deliverables / Value Metrics: Physics engine deployed on the edge node; equations calibrated against field data; methane mass uncertainty characterised (target 8% or less); first TPM-sealed certificates verified by independent replay; calibration chain documented.

Milestone T3: Canton integration

Estimated Delivery: Month 3-4
Focus: Consume certificates on Canton testnet.
Deliverables / Value Metrics: Daml contracts deployed on testnet; certificate verification, measurement-backed asset, transfer and retirement templates; certificates consumed by smart contracts; workflow exercised end to end on testnet.

Milestone T4: Reference case and acceptance

Estimated Delivery: Month 4-5
Focus: Complete one production cycle and publish the reference case.
Deliverables / Value Metrics: One full production cycle completed; verifier evidence package assembled; reference case published; replay guide and developer documentation delivered; open-source integration layer released; written review of the certificate schema and replay guide sought from at least one registry, verifier or Canton application developer outside KAILEdge, and published in the repository.

Use of Funds (T1, $60,000)

207 devices across three complete sets (69 for the live installation plus 138 held as calibrated spares)
Installation and commissioning at the Harohalli CBG facility
Calibration of Class A reference sensors by an NABL-accredited laboratory (to be engaged)
Edge compute units (Intel N305, TPM 2.0, IP67-rated)
Site engineering and field support during deployment

T2 to T4 cover engineering, calibration analysis, Daml development, documentation and the reference case.

Possible follow-on uses (not funded by this grant, no commitments)

Spot and forward trading of verified carbon credits, tokenized CBG contracts, energy efficiency attestations, NPK-verified fertilizer records, and data tokens built on sealed measurement streams. These depend on registries, issuers and buyers. Nothing in this proposal assumes them.

Acceptance Criteria

The Tech & Ops Committee will evaluate completion based on:

Deliverables completed as specified for each milestone
Demonstrated functionality or operational readiness
Documentation and knowledge transfer provided
Alignment with stated value metrics

Project-specific conditions:

A reproducible, TPM-sealed certificate is generated from calibrated field sensors at the Harohalli CBG facility.
An independent reviewer can replay the certificate from the recorded inputs and model version and verify the hash.
The certificate is consumed by a Daml smart contract on Canton testnet.
Documentation and knowledge transfer are complete: reference case published, replay guide delivered, open-source integration layer released.
The published reference case reports measured results against the stated targets, including where they were missed.

Funding

Total Funding Request: $120,000

Payment Breakdown by Milestone

T1 Hardware deployment and first data: $60,000 upon committee acceptance
T2 Physics validation and certificate pipeline: $25,000 upon committee acceptance
T3 Canton integration: $20,000 upon committee acceptance
T4 Reference case and acceptance: $15,000 upon final release and acceptance

Our separate Canton Foundation application is under evaluation and is not secured.

Volatility Stipulation

The project duration is 5 months. Should the timeline extend beyond 6 months due to Committee-requested scope changes, any remaining milestones must be renegotiated to account for USD/CC price volatility.

Sustainability

After the grant period, KAILEdge would maintain the open-source Daml templates, certificate schema and replay guide, and accept outside contributions through the repository. The maintenance period and response commitment are to be agreed with the champion and committee. Operation of the plant-side hardware stays with KAILEdge and the host site under the host-site agreement. The grant does not fund ongoing operation after T4.

Co-Marketing

Upon release, KAILEdge would collaborate with the Foundation on an announcement, a technical write-up of the reference case, and ecosystem promotion. Any public mention of Sustainable Impacts depends on its agreement.

Motivation

Who benefits:

Registries, verifiers and issuers get a replayable measurement record they can reference.
Buyers get evidence they can check without trusting a single reporter.
CBG producers get a route to supply continuous evidence rather than periodic batch evidence.
Canton developers get open templates, a certificate schema, conformance tests and a replay guide for later environmental assets.

Value to the Canton ecosystem:

A reference integration from device-sealed environmental measurements to Daml contracts, tested on testnet.
A second vertical for the same measurement architecture (designed for cold chain, not yet built there).
Reusable open-source templates, schema, conformance tests and replay guide.
A route for CBG producers to supply verifiable evidence to registries, verifiers and buyers who use Canton.
Any resulting Canton activity would generate network fees. No volume is assumed.

Public good. The integration layer is Apache 2.0 and is designed to work for any issuer or registry. The proprietary part is limited to KAILEdge's calibrated physics parameters. A verifier can check signature, integrity and chain without KAILEdge.

Honest position on Canton adoption. No external Canton application currently consumes KAILEdge certificates. The pilot would create a reference integration and a reusable schema, Daml templates and a documented replay pathway. We are not claiming existing Canton adoption. We are in discussions with Sustainable Impacts about hosting the pilot. No agreement is signed.

Market context

These are public statements about the CBG market. They do not indicate any relationship with, or interest in, KAILEdge or this pilot.

Potential credit buyers and marketplaces

KAILEdge does not issue credits. No party below has an agreement, partnership or stated interest in this pilot. These are marketplaces and platforms that a registry or issuer could use for credits backed by measured data from the pilot. Each statement is as published by that party. Whether any of them accepts credits from Indian CBG plants depends on registry and methodology, and has not been checked.

Platform	What its own source says	Canton link	Source
Xpansiv	Announced on 8 Oct 2025 a phased initiative to enable tokenization of environmental assets and their performance data on Canton Network. It begins with a pilot program on asset identifiers.	Announced, pilot stage	https://www.xpansiv.com/xpansiv-to-enable-tokenized-environmental-asset-infrastructure-via-canton-network/
ClimateTrade	A marketplace where companies and individuals buy carbon, plastic and biodiversity credits and EACs directly from project developers. Announced on LinkedIn on 12 Nov 2025 that it joined Canton Network as a network validator.	Validator, announced by ClimateTrade	https://www.climatetrade.com/en/about-us/ and https://www.linkedin.com/posts/climatetrade_activity-7394333673728221184-Korb
Carbonplace	Founded by nine global banks. A platform for carbon credit management and trading, with access to 14 registries through one account.	None found on its site	https://carbonplace.com/
Toucan	Tokenizes carbon credits (TCO2 tokens and carbon pools). Says it has brought $100 million in carbon credits onchain. Offers a bridge for Puro.earth removal credits.	None found in its docs	https://docs.toucan.earth/

"None found" means not found in the sources checked, not that none exists.

Rationale

Why this approach. A sealed, replayable record separates what was measured from who reported it. Hardware sealing at the source, with replay from recorded inputs, lets a third party check the result without trusting the reporter.

Alternatives. Oracle-supplied values (simple, but attest to a number, not its computation). Periodic third-party measurement (existing practice, retrospective). KAILEdge does not replace either. It sits beneath registries, verifiers and issuers.

Fit with existing tooling. The Daml layer is new and does not duplicate existing token standards. It intends to use CIP-0056 interfaces for transfer and retirement rather than define its own.

Risks and Mitigations

Risk	Mitigation
Host-site agreement not signed	T1 does not start until it is signed
Sensor calibration delays	An NABL-accredited laboratory will be engaged during procurement
Canton testnet integration complexity	Daml contracts developed in parallel with hardware deployment
Physics engine performance in field conditions	The pilot tests the architecture before any scale commitment. Results are reported against the targets, including misses
Regulatory compliance	A hazardous-area zoning assessment will be conducted before installation
Offline operation	Certificates queued locally. Backup power design to be confirmed
Methodology and registry acceptance	Not assumed. Registry, verifier and methodology are confirmed with the project, not by this grant

Contact

Prasad Gopal, Founder, KAILEdge, p***@kailedge.com, www.kailedge.com

← Back to Proposals Open on GitHub ↗

Development Fund Proposal Submission

Summary

The problem

The solution

Why the ecosystem needs this

Market context

Potential credit buyers and marketplaces

Pilot status

Milestones and funding

What $60,000 in T1 covers

Open and proprietary components

Honest position on Canton adoption

Acceptance criteria

Key risks

Checklist

Notes for Reviewers

Contact

Milestones 4

Discussion6 comments, 0 reviews, 0 inline comments

Performance Benchmarks

KAIL Edge Efficiency Breakdown

Performance Contribution Breakdown

Proposal file rfps/financial-markets-standards-verification/kailedge-cbg-measurement-pilot.md, 18 KB

Development Fund Proposal

Abstract

Specification

1. Objective

2. Implementation Mechanics

3. Architectural Alignment

4. Backward Compatibility

Milestones and Deliverables

Milestone T1: Hardware deployment and first data

Milestone T2: Physics validation and certificate pipeline

Milestone T3: Canton integration

Milestone T4: Reference case and acceptance

Use of Funds (T1, $60,000)

Possible follow-on uses (not funded by this grant, no commitments)

Acceptance Criteria

Funding

Payment Breakdown by Milestone

Volatility Stipulation

Sustainability

Co-Marketing

Motivation

Market context

Potential credit buyers and marketplaces

Rationale

Risks and Mitigations

Contact

Journey on the board 1 move

Timeline 14 events