Table of Contents

1 · What is RDS?

RDS (Radio Data System) is a digital communication standard that transmits data as an inaudible signal modulated onto FM broadcasts (87.5–108 MHz). It was standardised under ETSI EN 50067 and has been in use across Europe since the 1980s.

RDS carries among others the following information:

RDS transmits at 1,187.5 bit/s as a BPSK signal on a 57 kHz subcarrier. A complete PS transmission takes approximately 0.35 seconds under error-free conditions.

2 · Why an AI Plugin?

During long-distance FM reception (DX) – especially via tropospheric ducting or Sporadic-E propagation – signals are often weak and affected by multipath interference or co-channel interference. This results in:

• Packet loss: Entire RDS groups are lost in transit
• Bit errors: Individual bits are decoded incorrectly
• Incomplete PS names: Only 2–3 of the 8 characters are received correctly
• Unknown PI codes: Block A contains errors, PI unreadable

3 · Architecture – The Two Plugin Components

4 · Simple Explanation – How the Plugin Works

Imagine listening to an announcement in a busy train station with a lot of background noise. The first time you might only catch "...NR...", the second time "...NRJ...". After hearing it 10 times, you can say with high confidence: the announcement said "NRJ".

That is exactly what the plugin does with RDS data: it listens to many erroneous packets, collects them all, and votes statistically on which character at which position is most likely the correct one.

4.1 · Learning by Voting

The station name (PS) consists of 8 character positions. Each position is voted on separately:

After enough votes, a reliable station name emerges. Through its database the plugin already knows thousands of stations – for recognised ones, the full name is available immediately after just 2 confirmed PI code packets.

4.2 · Frequency Change and Pre-Cache

When tuning to a new frequency, the following sequence runs:

4.3 · Dynamic vs. Static Stations

Some stations scroll their PS name or change it regularly (e.g. to display song information). The plugin detects this automatically:

4.4 · RDS Follow Mode

The plugin operates in two modes:

4.5 · The Panel – Display Elements

5 · Technical Deep Dive

5.1 · RDS Group Structure and Error Correction

An RDS group consists of 4 blocks of 26 bits each (16 bits payload + 10 bits CRC checksum). The error correction code is a shortened cyclic code capable of detecting up to 5 bit errors and correcting up to 2 bit errors per block.

The plugin receives these error levels as the array errB[0..3] and converts them into confidence weights:

Group types and their relevance to the plugin

5.2 · Voting Engine – Compact Aggregation Format

Since version 4.0, a compact aggregated format is used instead of a list of individual timestamps:

Format per character position (JSON)

// db[pi].ps[position][character]
{
  "w":         147.3,         // current weighted score (lazy-decayed)
  "count":     18,            // raw number of votes received
  "firstSeen": 1773554046944, // Unix timestamp of first vote (ms)
  "lastSeen":  1773554677188  // Unix timestamp of last vote (ms)
}

Lazy Exponential Decay

The weight w decays exponentially with a half-life of 7 days. The calculation is performed lazily – only on the next access, not as a background process:

function applyDecay(v, now) {
  const ageMs  = now - v.lastSeen;
  const halfMs = 7 * 86400000;              // 7 days in milliseconds
  v.w        = v.w * Math.pow(0.5, ageMs / halfMs);
  v.lastSeen = now;
}

Values below 0.1 weight with fewer than 3 votes are deleted on the next write access (noise suppression).

Consistency Boost

When a character already clearly dominates (weight > 2× all competitors combined), the new vote receives a multiplier of 1.5×. This accelerates convergence for unambiguous stations:

if (posVotes[char].w > competitorW * 2) {
  finalWeight = weight * 1.5;   // CONSISTENCY_BOOST
}

5.3 · Confidence Calculation

The confidence per PS slot is composed from three components:

conf = share      * 0.5   // winner's share of total weight
     + dominance  * 0.3   // gap between winner and runner-up
     + voteFactor * 0.2   // experience factor (saturates at 30+ votes)

// Where:
share      = bestWeight / totalWeight
dominance  = (best - second) / totalWeight   // when >1 candidate
voteFactor = Math.min(1, totalCount / 30)    // saturates at 30 votes

// Upper bound: 0.97 (never 100% confidence from DB alone)

Confidence Scale – Colour Mapping in the Panel

RGB value = 20 + conf × 220 (normal) · 15 + conf × 30 (bigram)

5.4 · PI Verification Pipeline

From version 4.4 onwards every PI code passes through a strict verification pipeline before being delivered to any output:

5.5 · Database Format and Storage Management

File structure (rdsm_memory.json)

{
  "_meta": {
    "rdsFollowMode": false,     // restored on restart
    "savedAt": 1773556231184
  },
  "D3C3": {
    "freq":        "104.00",       // rounded frequency (±0.1 MHz)
    "ps":          { "0": { "N": {w,count,firstSeen,lastSeen} }, ... },
    "psResolved":  "NRJ     ",    // current best voting result
    "psConf":      [0.97,0.95,0.93,0,0,0,0,0],
    "psIsDynamic": false,
    "psLastRaw":   ["N","R","J"," "," "," "," "," "],
    "psLastRawTs": 1773554677188,
    "pty":         10,
    "tp":          true,
    "ta":          false,
    "ms":          1,
    "stereo":      true,
    "ecc":         "E1",           // Extended Country Code (hex)
    "seen":        1773554677188,  // last seen (Unix ms)
    "seenCount":   847             // total groups received
  }
}

Retention periods and limits

5.6 · WebSocket Protocol

Communication between the server plugin and the browser runs over WebSocket on the path /data_plugins. All messages are JSON.

Example rdsm_ai payload

{
  "type":  "rdsm_ai",
  "pi":    "D3C3",
  "ps": [
    { "char": "N", "conf": 0.97, "src": "ai-cached-db"  },
    { "char": "R", "conf": 0.95, "src": "ai-voted-high" },
    { "char": "J", "conf": 0.93, "src": "ai-voted-high" },
    { "char": "x", "conf": 0.05, "src": "ai-bigram"     },
    // ... 4 more slots
  ],
  "rt": { "text": "Dua Lipa - Levitating", "score": 0.72, "src": "ai-rt-live" },
  "stats": { "seenCount": 847, "psVoteTotal": 312.4, "freq": "104.00",
             "psIsDynamic": false, "pty": 10 },
  "ts":   1773556231184
}

PS source types

5.7 · dataHandler Patch

The plugin hooks into the native WebSDR server's handleData() function using a JavaScript function replacement (monkey-patch):

const orig = dh.handleData;
dh.handleData = function(wss, receivedData, rdsWss) {

  // 1. Our processing runs first
  interceptLines(receivedData);

  // 2. Exclusive mode: bypass native decoder entirely
  if (aiExclusiveMode) return;

  // 3. Follow mode: strip RDS lines from the stream
  const dataForNative = nativeRDSDisabled
    ? stripRDSLines(receivedData)
    : receivedData;

  // 4. Call native decoder with (possibly filtered) data
  const result = orig.call(this, wss, dataForNative, rdsWss);

  // 5. Follow mode: overwrite native data with AI data
  if (rdsFollowMode && piConfirmed) applyFollowToDataHandler();

  return result;
};
dh._rdsm_patched = true;  // prevents double-patching

5.8 · Bigram Predictor

When no database entry exists for a PS position, a statistical estimate is made based on the preceding character (bigram model). The table contains typical frequencies of two-character combinations found in station names:

// Excerpt from the bigram table:
BIGRAMS = {
  ' ': { 'R':25, 'F':20, 'M':15, ... },  // after space, R is common
  'F': { 'M':40, 'R':12, 'L': 8, ... },  // after F, M is common (FM...)
  'R': { 'D':20, 'A':15, 'I':12, ... },  // after R, D is common (Radio)
}

// Confidence of a bigram guess:
conf = (bestScore / totalScore) * 0.25  // max. 25% – clearly identifiable as a guess

6 · REST API

The plugin registers two HTTP endpoints for external queries:

Example response for /api/rdsm/pi/D3C3

{
  "freq":         "104.00",
  "psResolved":   "NRJ     ",
  "psConf":       [0.97, 0.95, 0.93, 0, 0, 0, 0, 0],
  "psVoteTotal":  312,
  "psIsDynamic":  false,
  "psLastRaw":    "NRJ     ",
  "psLastRawAge": "42s",
  "pty":          10,
  "tp":           true,
  "ta":           false,
  "ecc":          "E1",
  "seen":         1773554677188,
  "seenCount":    847
}

7 · Configuration Parameters

RDS AI Decoder · Documentation v1.0 · Author: Highpoint · March 2026
github.com/Highpoint2000/RDS-AI-Decoder