GPS Geofencing for Venue Blocking: Accuracy and Privacy

Whistl's GPS geofencing detects when you approach gambling venues and automatically activates protective measures. This technical guide explains geofence accuracy, battery optimisation, privacy safeguards, and how location-based blocking prevents impulses in the critical moments before they happen.

How GPS Geofencing Works

Geofencing creates virtual boundaries around physical locations. When you cross these boundaries, Whistl triggers protective actions:

1. Venue database: 5,000+ gambling venues mapped across Australia
2. Geofence creation: 500m radius around each venue
3. Location monitoring: Background GPS tracking
4. Boundary detection: Entry/exit event triggers
5. Protection activation: Automatic blocking + intervention

Venue Database

Whistl maintains a comprehensive database of gambling venues:

Venue Categories

Data Sources

• Government registers: State liquor and gaming commissions
• OpenStreetMap: Community-mapped venue locations
• Google Places API: Verified business locations
• User submissions: Crowdsourced venue additions

Geofence Configuration

Each venue has configurable geofence parameters:

Default Settings

{
  "venue_id": "crown_melbourne",
  "coordinates": {"lat": -37.8225, "lng": 144.9586},
  "geofence_radius": 500,  // meters
  "trigger_on_entry": true,
  "trigger_on_exit": true,
  "dwell_time_threshold": 300,  // 5 minutes
  "accuracy_threshold": 50,  // meters
  "priority": "high"
}

Radius Selection

500m radius balances accuracy with battery life:

• 100m: Too late—user already inside venue
• 500m: Optimal—early warning, time to intervene
• 1000m: Too early—false positives from passing by

Research shows 500m provides 3-5 minutes advance warning at walking speed.

Location Monitoring Implementation

Whistl uses native location services for efficient background monitoring:

iOS Core Location

import CoreLocation

class VenueGeofenceManager: NSObject, CLLocationManagerDelegate {
    private let locationManager = CLLocationManager()
    private var monitoredGeofences: Set = []
    
    func startMonitoring() {
        locationManager.delegate = self
        locationManager.desiredAccuracy = kCLLocationAccuracyHundredMeters
        locationManager.distanceFilter = 100  // Update every 100m
        
        // Request background permission
        locationManager.requestAlwaysAuthorization()
        locationManager.allowsBackgroundLocationUpdates = true
        locationManager.pausesLocationUpdatesAutomatically = false
        
        locationManager.startUpdatingLocation()
        startMonitoringVenues()
    }
    
    private func startMonitoringVenues() {
        let venues = loadNearbyVenues()
        
        for venue in venues {
            let region = CLCircularRegion(
                center: venue.coordinates,
                radius: 500,
                identifier: venue.id
            )
            region.notifyOnEntry = true
            region.notifyOnExit = true
            
            locationManager.startMonitoring(for: region)
            monitoredGeofences.insert(venue.id)
        }
    }
    
    func locationManager(_ manager: CLLocationManager, didEnterRegion region: CLRegion) {
        guard let circularRegion = region as? CLCircularRegion else { return }
        handleVenueEntry(venueId: circularRegion.identifier)
    }
    
    func locationManager(_ manager: CLLocationManager, didExitRegion region: CLRegion) {
        guard let circularRegion = region as? CLCircularRegion else { return }
        handleVenueExit(venueId: circularRegion.identifier)
    }
}

Android Geofencing API

import android.location.Location
import com.google.android.gms.location.*

class VenueGeofenceManager(private val context: Context) {
    private val geofencingClient = LocationServices.getGeofencingClient(context)
    private val geofenceList = mutableListOf()
    
    private val geofencePendingIntent: PendingIntent by lazy {
        val intent = Intent(context, GeofenceBroadcastReceiver::class.java)
        PendingIntent.getBroadcast(
            context, 0, intent,
            PendingIntent.FLAG_UPDATE_CURRENT or PendingIntent.FLAG_MUTABLE
        )
    }
    
    fun startMonitoring(venues: List) {
        for (venue in venues) {
            val geofence = Geofence.Builder()
                .setRequestId(venue.id)
                .setCircularRegion(venue.lat, venue.lng, 500f)
                .setExpirationDuration(Geofence.NEVER_EXPIRE)
                .setTransitionTypes(Geofence.GEOFENCE_TRANSITION_ENTER or Geofence.GEOFENCE_TRANSITION_EXIT)
                .setLoiteringDelay(300000)  // 5 minutes
                .build()
            geofenceList.add(geofence)
        }
        
        val geofencingRequest = GeofencingRequest.Builder()
            .setInitialTrigger(GeofencingRequest.INITIAL_TRIGGER_ENTER)
            .addGeofences(geofenceList)
            .build()
        
        geofencingClient.addGeofences(geofencingRequest, geofencePendingIntent)
            .addOnSuccessListener { /* Monitoring started */ }
            .addOnFailureListener { e -> /* Handle error */ }
    }
}

Accuracy and Reliability

GPS accuracy varies based on conditions:

Accuracy Levels

Accuracy Enhancement Techniques

• WiFi positioning: Supplement GPS with nearby WiFi networks
• Cell tower triangulation: Use cellular signal for coarse location
• Sensor fusion: Combine accelerometer, gyroscope for dead reckoning
• Map matching: Snap locations to known roads/paths

False Positive Prevention

To avoid triggering when simply passing by:

• Dwell time threshold: Must remain in geofence for 5+ minutes
• Heading analysis: Check if moving toward or away from venue
• Speed filtering: Ignore if moving at vehicle speed (>30 km/h)
• Historical patterns: Skip if this is regular commute route

Battery Optimisation

Continuous GPS monitoring drains battery. Whistl implements multiple optimisations:

Adaptive Location Updates

func calculateUpdateInterval() -> TimeInterval {
    let nearbyVenues = countVenuesWithin(distance: 2000)
    
    switch nearbyVenues {
    case 0:
        return 300  // 5 minutes (no venues nearby)
    case 1...3:
        return 60   // 1 minute (few venues)
    case 4...10:
        return 30   // 30 seconds (moderate density)
    default:
        return 10   // 10 seconds (high density area)
    }
}

Significant Location Change Monitoring

iOS provides low-power significant change monitoring:

• Power consumption: ~1% battery per day
• Update trigger: ~500m movement
• Use case: Background monitoring when app not active

Region Monitoring vs. Continuous GPS

Privacy Safeguards

Location data is highly sensitive. Whistl implements strict privacy protections:

On-Device Processing

• Local geofence checks: All venue detection happens on device
• No location transmission: Coordinates never sent to servers
• Encrypted storage: Location history stored in secure enclave
• Automatic deletion: History purged after 30 days

User Controls

• Granular permissions: "Always Allow" required for background monitoring
• Temporary disable: Pause location tracking for set duration
• Venue whitelist: Exclude specific venues (e.g., work location)
• Export/delete data: Full control over location history

Permission Transparency

<key>NSLocationAlwaysAndWhenInUseUsageDescription</key>
<string>Whistl monitors your location to detect when you're near gambling venues and activate protective blocking. Your location data never leaves your device.</string>

<key>NSLocationWhenInUseUsageDescription</key>
<string>Whistl uses your location to identify nearby gambling venues and provide context-aware support.</string>

Protection Activation

When a geofence is triggered, Whistl activates multiple protective measures:

Immediate Actions

1. SpendingShield elevation: Risk level jumps to ORANGE/RED
2. VPN activation: DNS blocking enabled immediately
3. Push notification: "You're near [Venue]. Your risk is elevated."
4. Partner alert: Optional notification to accountability partner
5. AI intervention: Neural impulse predictor recalculates with venue proximity

Intervention Message

"We noticed you're near Crown Casino. Your risk is elevated right now. Remember your Bali goal—you're $840 away. Want to call your sponsor instead?"

Edge Cases and Handling

Real-world usage presents challenges:

Multi-Venue Complexes

Some locations have multiple venues (e.g., Crown Entertainment Complex):

• Solution: Treat as single large geofence
• Radius: Extended to 800m for complexes
• Notification: Generic "entertainment complex" message

Temporary Venues

Pop-up betting shops during major events:

• Solution: Dynamic venue database updates
• Source: Event calendars + user reports
• Expiration: Auto-remove after event ends

GPS Spoofing

Users may attempt to fake location:

• Detection: Check for mock location flags
• Response: Disable location features, notify user
• Alternative: Fall back to DNS-only protection

Performance Metrics

Geofencing performance from 10,000+ users:

User Testimonials

"I walked past the TAB on my way home and my phone buzzed. Whistl knew before I even thought about going in. That's wild." — Jake, 31

"The geofencing saved me at Crown. I was about to walk in, got the notification, and turned around. Best feature by far." — Marcus, 28

"I was worried about battery drain but it's maybe 5% per day. Totally worth it for the protection." — Sarah, 34

Conclusion

GPS geofencing provides powerful location-based protection while respecting privacy and battery life. By detecting venue proximity early, Whistl intervenes at the critical moment—before the impulse becomes action.

All location processing happens on-device, ensuring your movements remain private while still enabling life-saving intervention.

Related: DNS Filtering Implementation | 27 Risk Signals | Biometric Spending Protection