High-Fidelity Sensor Telemetry Engine (C++) with Mandatory Phase Decorrelation for HIL Android Testing

We are seeking an expert Systems and Digital Signal Processing (DSP) Engineer to design and develop a high-fidelity, native C++ sensor simulation library. The module will be integrated into our internal automated smartphone testing farm (Hardware-in-the-Loop QA infrastructure) to generate statistically realistic inertial telemetry for devices that remain physically static on server racks. The objective is to procedurally synthesize continuous accelerometer and gyroscope streams that mimic natural, real-world handheld dynamics for automated UI and hardware state validation. ### Core Technical Requirements & Scope: * **Mandatory Phase Decorrelation:** Periodicity and cross-layer leaks are the most critical points of failure in synthetic signals. To guarantee that the combined multi-axis output absolutely never enters a phase-lock state over extended simulation runtimes, rigorous and mandatory phase decorrelation mechanisms must be implemented across all simulation layers and axes (e.g., using isolated SplitMix64 PRNG streams with independent seeds and spatial salts, or irrational phase offsets). Cross-coherence must remain at the statistical estimation floor, with zero linear correlation between layers. * **Target Architecture:** Native C++17 module optimized for 64-bit ARM (arm64-v8a) targeting Android API level 30 (Android 11) and above. * **Real-Time Performance:** Pure O(1) execution complexity during the runtime hot-path loop with absolutely zero heap allocations (`new`/`malloc`) and a stack-only design to guarantee sub-microsecond processing overhead. * **Procedural Signal Layers:** Synthesis of layered stochastic processes including 16-octave pink noise (Voss-McCartney), continuous Ornstein-Uhlenbeck drift, dynamic Ballistocardiogram (BCG) impulse modeling for physiological micro-movements, and active 8–12Hz tremor filter emulation. * **Dynamic Sample Rate Adaptability:** The execution function must receive a continuous nanosecond timestamp delta ($\Delta t$) and seamlessly handle dynamic runtime sensor refresh transitions between 60Hz, 90Hz, and 120Hz. Internal state history (IIR filter memory) must be preserved or warped instantly across frequency shifts to eliminate signal amplitude jumps or phase spikes. * **Validation Suite:** Delivery must include a mathematical validation test harness to verify the telemetry output against baseline statistical expectations (including Power Spectral Density (PSD) log-linear slopes, autocorrelation floors, and cross-coherence estimation analysis). ### Project Deliverables: * Complete, clean, and well-commented native C++ source files (`.cpp`/`.h` or single-header). * Architectural blueprint and clean API surface definition (`void step(IMUSample& out, uint64_t timestamp_ns);`). * Comprehensive unit tests and statistical benchmarking harness. --- ## Required Skills * C++ Programming * Android NDK * Digital Signal Processing (DSP) * Algorithm * Embedded Software

Project ID: 40465946