🔷 Quaternionic Signal Processing (QSP)¶
Status: Active development · Production-ready architecture
Apply quaternion mathematics to real-world signal processing.
QSP is a quaternion-native signal processing framework built on qsp-core. It provides FFT, filtering, modulation, and orientation estimation through a unified API that treats signals as quaternion-valued sequences.
Stack Architecture¶
Package Roles¶
-
qsp-core— Foundation
Defines the quaternion-valued signal primitives, number types, and shared utilities that every other
qsp-*package depends on.- Single source of truth for quaternion arithmetic
- No external signal-processing dependencies
- Designed for numerical correctness and testability
-
qsp-fft— Quaternionic FFT
Implements the quaternion Fourier transform and its inverse, enabling frequency-domain analysis of quaternion-valued signals.
- Quaternion FFT (QQFT) and inverse
- Spectral analysis for multi-channel signals
- Depends on
qsp-core
-
qsp-filter— Signal Filtering
Provides FIR and IIR filter designs that operate natively on quaternion signals, preserving inter-channel phase relationships.
- Quaternion FIR and IIR filters
- Hilbert-transform-based analytic signals
- Depends on
qsp-core
-
qsp-modulation— Modulation & Demodulation
Encodes and decodes information using quaternion phase-shift keying and amplitude modulation schemes.
- Quaternion phase-shift keying (QPSK extensions)
- Amplitude and phase modulation
- Depends on
qsp-core
-
qsp-orient— Orientation Estimation
Estimates rigid-body orientation from inertial or multi-antenna sensor streams using quaternion-domain filtering.
- Attitude estimation from IMU data
- Quaternion Kalman and complementary filters
- Depends on
qsp-core
-
qsp-stack— Umbrella Package
A convenience meta-package that installs the complete QSP ecosystem in a single step.
🌐 Website: https://rqmtechnologies.com