Skip to content

GPS_GENERATOR.md

Latest commit

 

History

History
56 lines (43 loc) · 3.01 KB

GPS_GENERATOR.md

File metadata and controls

56 lines (43 loc) · 3.01 KB
Raw

High Performance GPS Simulator

Overview

The GPS Simulator is a standalone .NET Console Application designed to mimic the behavior of a fleet of IoT devices. It focuses on high performance, realistic movement patterns, and robust error handling.

Key Architectural Patterns

1. Orchestration vs. Logic (IPositionSimulator)

The simulator logic is decoupled from the data submission flow.

  • GpsDataGenerator: Acts as the orchestrator. It fetches vehicle states, coordinates the simulation tasks, and aggregates results.
  • AthensPositionSimulator: Contains the pure logic for path generation, boundary enforcement, and speed calculation.

2. Parallel Processing (Concurrency)

To ensure scalability, the simulator processes vehicles in parallel using the Task-based Asynchronous Pattern (TAP).

var tasks = vehicles.Select(v => ProcessSingleVehicleAsync(v));
var results = await Task.WhenAll(tasks);

This allows the system to handle hundreds of vehicles simultaneously without sequential blocking, maximizing throughput.

3. Resilience & Error Isolation

The system is designed to be robust against partial failures:

  • Isolated Tasks: Each vehicle is processed in its own asynchronous task. If one vehicle fails (e.g., due to invalid data or a specific API error), it does not stop the simulation for the remaining vehicles.
  • Detailed Diagnostics: Every failure is captured with a full StackTrace and contextual data (Vehicle ID) using Serilog, ensuring that issues can be debugged without interrupting the service.

4. Dependency Injection (Autofac)

Every component is injected via interfaces, making the system highly modular and testable.

  • IVehicleApiClient: Handles HTTP communication.
  • IPositionSimulator: Handles path calculation.
  • IBoundingBoxProvider: Defines geographic limits.
  • ILogger: Provides structured telemetry.

Movement Logic

The simulator uses a "Random Walk within Bounds" algorithm:

  1. Starting Point: If a vehicle has a last known position, it starts from there. Otherwise, it uses a default starting point (Athens city center).
  2. Path Generation: Generates N positions per iteration.
  3. Speed Calculation: Realistic distance-based speed calculation (km/h) based on the time interval between readings.
  4. Boundary Enforcement: If a vehicle exceeds the Athens Bounding Box, it is automatically redirected towards the center to prevent "lost" vehicles.

Structured Logging (Serilog)

Instead of plain text, the simulator records data in a structured format:

_logger.Information("Iteration #{Iteration} completed. Vehicles: {Processed}", iteration, count);

This enables real-time monitoring and advanced filtering based on properties like VehicleId or ProcessedCount.

Configuration

Settings are managed via App.config:

  • ApiBaseUrl: Target Web API address.
  • IntervalSeconds: Delay between simulation bursts.
  • PositionsPerVehicle: Number of GPS points generated per burst.
  • RadiusMeters: Maximum movement distance per point.