IP Library › Granted Patent US 11,902,745
Granted Patent B2
US 11,902,745 · App. 17/767,331 · Granted Feb 13, 2024

System of processing devices to perform an algorithm

Inventors: Nun Mendez Rodriguez (Tenerife, ES); Amaury Hazan (Barcelona, ES); Nicolas Wack (Barcelona, ES); Kamil Budzynski (Barcelona, ES); Jacques Kinsbergen (Antwerp, BE)
Assignee: JACOTI BV
H04R25/502H04R29/00
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 11,902,745
App. No.
17/767,331
Granted
Feb 13, 2024
Kind
B2
Abstract

A system comprising a plurality of processing devices is arranged for performing an algorithm having a plurality of computational tasks. The plurality of processing devices are interconnected with each other. A control logic module is adapted to select an implementation of the algorithm among a plurality of implementations. At least a part of the control logic module resides in one of the processing devices and the implementation is selected according to at least one criterion of (latency requirements, computational power, resource availability, data locality, privacy requirements, precision requirements, hardware capabilities, distance, location, acoustical performance requirements, energy efficiency, orientation) and in the selected implementation the plurality of computational tasks of the algorithm is distributed over at least two of the processing devices.

Claims (19)

1. A system comprising:

a plurality of processing devices arranged for performing an algorithm comprising a plurality of computational tasks, said plurality of processing devices being interconnected with each other,

a control logic module adapted to select an implementation of said algorithm among a plurality of implementations,

wherein at least a part of said control logic module resides in one of said processing devices and

wherein said implementation is selected according to hardware capabilities of the system, allowing a choice between an implementation of said algorithm using signal values in a floating-point representation and an implementation of said algorithm using signal values in a fixed-point representation, and wherein in the selected implementation said plurality of computational tasks of said algorithm is distributed over at least two of said processing devices.

2. The system as in claim 1 , wherein one of said processing devices is an ear-level audio processing device and another of said processing devices is a further portable computing device or a cloud computing service.

3. The system as in claim 2 , wherein said ear level audio processing device is arranged for noise monitoring and/or sound generation.

4. The system as in claim 2 , wherein said ear-level audio processing device is a hearing aid.

5. The system as in claim 1 , wherein said control logic module is distributed over said plurality of processing devices.

6. The system as in claim 1 , wherein at least one of said processing devices comprises a graphical user interface.

7. The system as in claim 1 , wherein output produced by performing a computational task of said plurality of computational tasks in a first processing device can be received as input data in at least two other processing devices of said plurality of processing devices.

8. The system as in claim 1 , wherein said control logic module can assign a computational task of said algorithm simultaneously to at least two different processing devices of said plurality.

9. The system as in claim 1 , wherein said control logic module is arranged to steer a data stream in or to processing devices of said system by using a command from a set of commands.

10. The system as in claim 1 , wherein selecting said implementation is performed on the fly.

11. The system as in claim 1 , wherein in said implementation with said fixed point representation numbers representing a gain to be applied to signal values are normalized by subtracting a maximum gain value from said gain and limiting said signal values to a given negative threshold.

12. The system as in claim 11 , wherein said given negative threshold equals the negative of said maximum gain value.

13. The system as in claim 1 , wherein further at least one criterion of (privacy requirements, precision requirements, distance), is taken into account when selecting the implementation.

14. The system as in claim 13 , wherein selecting said implementation is performed based on a weighted sum of two or more of said criteria.

15. The system as in claim 14 , wherein weights of said weighted sum are determined by interrogating one or more of said processing devices.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Apr 7, 2022
From: MENDEZ RODRIGUEZ, NUN; HAZAN, AMAURY; WACK, NICOLAS; BUDZYNSKI, KAMIL; KINSBERGEN, JACQUES
To: JACOTI BV
Reel/Frame 059535/0167 →
Continuity (2)
Provisional Application 62912873 · Oct 9, 2019
Related Publication 20220377469A1 · Nov 24, 2022
Cited By (1)
US 12,669,051