# Halowerk Two-Channel Global Fourth-Cumulant ICA Projection

> Halowerk Two-Channel Global Fourth-Cumulant ICA Projection is a paid API for AI agents from detect.halowerk.com, paid per call via x402, $0.003/call, status unknown (last checked 2026-11-04).

Performs independent component analysis on two-channel signal data using fourth-cumulant (kurtosis-based) methods to recover non-Gaussian source projections and stationary-contrast artifacts.

## Facts

- Endpoint: POST https://detect.halowerk.com/v1/F-TD2000-GEO-INDUSTRY-B013/two-channel-global-fourth-cumulant-independent-component-projection?utm_source=zero.xyz
- Price: $0.003/call
- Payment: x402
- Status: unknown
- Last checked: 2026-11-04
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/halowerk-two-channel-global-fourth-cumulant-ica-projection-08a47e42
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_Y338whzYwFLY8s1u22SOb

Status and success rate cover calls made through Zero and Zero's own probes. Third-party monitors may report differently.

## How to call it through Zero

Zero handles the 402 payment challenge and records the run. With the Zero CLI installed (`npm i -g @zeroxyz/cli`):

```sh
zero fetch --capability halowerk-two-channel-global-fourth-cumulant-ica-projection-08a47e42 -d '<json body>'
```

Example prompt: Run a two-channel fourth-cumulant ICA on my signal observations — I have about 32 paired samples (each value between -10 and 10) and I need the independent component projection and stationary-contrast artifact recovered.

## When to prefer this

Choose this endpoint when you need fast, single-call blind source separation on exactly two-channel mixed signal data using a kurtosis/fourth-cumulant statistical approach rather than second-order PCA methods. Best suited for non-Gaussian source signals with up to 64 observation samples per call.

## Known failure modes

- Fewer than 8 observation pairs provided (minItems violation)
- More than 64 observation pairs provided (maxItems violation)
- Observation values outside the [-10, 10] range
- Malformed ICA object missing required 'data' or 'claim' fields
- Numerically degenerate input causing ICA to fail (e.g., constant or perfectly correlated channels)
- Missing 'observations' array inside 'data' object

## How this service works

Recover a complete two-channel nonGaussian component projection and stationary-contrast artifact.

## Output

Returns the recovered two-channel non-Gaussian independent component projection along with the stationary-contrast artifact, computed via global fourth-cumulant (kurtosis-based) ICA methods applied to the supplied observation pairs.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "artifact": {
   "type": "object",
   "required": [
    "ica"
   ],
   "properties": {
    "ica": {
     "type": "object",
     "required": [
      "data",
      "claim"
     ],
     "properties": {
      "data": {
       "type": "object",
       "required": [
        "observations"
       ],
       "properties": {
        "observations": {
         "type": "array",
         "items": {
          "type": "array",
          "items": {
           "type": "number",
           "maximum": 10,
           "minimum": -10
          },
          "maxItems": 2,
          "minItems": 2
         },
         "maxItems": 64,
         "minItems": 8
        }
       },
       "additionalProperties": false
      },
      "claim": {}
     },
     "additionalProperties": false
    }
   },
   "additionalProperties": false
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/halowerk-two-channel-global-fourth-cumulant-ica-projection-08a47e42/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from detect.halowerk.com](https://www.zero.xyz/host/detect.halowerk.com/llms.txt)