Advanced signal processing
Unlike traditional reliance on dedicated DSP processors, Innosonix opted for FPGA (field- programmable gate arrays) to achieve the highest flexibility since the actual DSP engine is created at every bootup in hardware and can be easily exchanged with every firmware update. This guarantees low latency and massive parallel signal processing for those high channel counts.
Acoustic designers and Integrators welcome the flexibility offered by their unique FIR (finite impulse response) solution. Besides all the classical processing elements like EQ, the ability to load FIR-filter with up to 2048 taps per channel allows acoustic designers to fine-tune each room for optimal performance. The FIR filters are convoluted at a sample basis, to achieve low latency if desired and not on a block basis. The total FIR calculation power results in almost 4 Giga MACs/sec and about 202 Gigabit/ sec memory bandwidth which can be easily handled by the FPGA DSP core.
In addition to the user-controllable DSP chain, an additional speaker-dsp-chain allows you to load prepared speaker corrections from manufacturers or even build your own library to speed up the deployment and tuning process of hundreds of speakers.
Innosonix also offers a service to create custom speaker presets for any provided speaker type. The goal is to provide integrators with the DSP settings they need to optimize their system performance for speakers that pair with their amplifiers.
A shorter FIR-filter of 512 taps can also be placed inside the speaker-dsp to phase linearize any tweeter and still leave the full room acoustic FIR-filter untouched.
Monitoring and maintenance
Ease of monitoring and maintenance is vital in a challenging marine environment. Innosonix offers a comprehensive set of self-diagnoses in each amplifier to guarantee remote management and predicted maintenance. Each amplifier output sampled back with 24-bit audio grade ADCs to measure the precise output voltage and current of each connected speaker.
The measured voltage is compared to the desired output signal to guarantee its correctness. In combination with inserted pilot tones, the line integrity can continuously be observed. An exciting new feature that will be released in Q3/2022 - impedance monitoring of the connected speaker by calculating the impedance magnitude and phase curves based on any available audio output signal.
Thus, even a broken or altered midrange speaker can be detected by playing music which isn’t normally possible by simple pilot tones.
The values and ethos of the team at Innosonix are shared by Qualtron and our partners. Both companies seek to realize our clients’ dreams with meticulous attention to detail at every stage.
For more details
T: +49 4331 43 79 057
E: firstname.lastname@example.org W: www.qualtron.com