SIGNAL PROCESSING
Digital signal processing (DSP) systems are widely used in medical devices, video processing, telecommunications, and industrial products. Syprosoft has deep theoretical and empirical skills with real-time DSP algorithms and implementation on actual commercial DSP hardware. Trade-offs and optimizations needed to meet execution throughput and resource usage are routinely handled. Algorithms such as FIR filters, IIR filters and FFT have been implemented. Advanced techniques of adaptive filtering, signal detection, echo cancellation, and speech signal processing have been implemented. Descriptions of some sample projects follow.
Vibration AnalysisCollected and analyzed time series data on the unwanted vibrations of a robot used during surgery to improve performance.
Cardiac Signal AnalysisDeveloped a robust algorithm to accurately detect the dicrotic notch from a pressure waveform to demarcate the end of systole and the beginning of diastole while the patient is undergoing external counter pulsation therapy for angina.
Human GaitUsing a single sensor worn on the chest determined human gait parameters and walking posture for the purpose of pre-operative diagnostics and post-operative rehabilitation. Developed an algorithm to predict fall risk.
Resonance Frequency Designed an optical system to collect metrology on a precision machine. Performed DFT analysis to extract resonance frequencies and identify means to shift them to benign regions.
Speech Signal ProcessingDesigned a Linear Predictive Coding filter to use past values of a speech signal to predict future values. Used the LPC coefficients to reduce the bandwidth needed for the signal.
Tone DetectionDeveloped an efficient algorithm for the extraction of multiple tones from a coded signal in the presence of high levels of noise. Achieved very high levels of accuracy.