Patents
Literature
Patsnap Eureka AI that helps you search prior art, draft patents, and assess FTO risks, powered by patent and scientific literature data.

3results about "Computations using stochastic pulse trains" patented technology

System for generating probability streams

PendingEP4760484A1Random number generatorsComputations using stochastic pulse trains
The invention relates to a system (100) for generating multiple probability streams with varying probabilities for stream processing applications. The system comprises at least one probabilistic bit (110) configured to produce a fixed probability output and a digital logic circuit (120) designed to process the outputs from the probabilistic bit. The digital logic circuit performs logic operations to transform the fixed probability outputs into multiple streams with different probabilities. This approach reduces the complexity associated with analog control of probabilistic bits by shifting the probability adjustment to the digital domain. The system enables efficient and scalable probability stream generation suitable for applications such as stochastic computing, bit stream computing, and neural networks. Its versatility and reduced hardware requirements make it an ideal solution for error-tolerant stream processing tasks, providing a balance between accuracy, speed, and implementation simplicity.
Owner:INTERUNIVERSITAIR MICRO ELECTRONICS CENT (IMEC VZW)

System for generating probability streams

PCT designated stageWO2026125602A1Random number generatorsComputations using stochastic pulse trains
The invention relates to a system (100) for generating multiple probability streams with varying probabilities for stream processing applications. The system comprises at least one probabilistic bit (110) configured to produce a fixed probability output and a digital logic circuit (120) designed to process the outputs from the probabilistic bit. The digital logic circuit performs logic operations to transform the fixed probability outputs into multiple streams with different probabilities. This approach reduces the complexity associated with analog control of probabilistic bits by shifting the probability adjustment to the digital domain. The system enables efficient and scalable probability stream generation suitable for applications such as stochastic computing, bit stream computing, and neural networks. Its versatility and reduced hardware requirements make it an ideal solution for error-tolerant stream processing tasks, providing a balance between accuracy, speed, and implementation simplicity.
Owner:INTERUNIVERSITAIR MICRO ELECTRONICS CENT (IMEC VZW)

Generation of furniture assembly instructions

A computer-implemented method (100) for generating furniture assembly instructions is provided. The method (100) comprises receiving (110, 120) first and at least one additional sets (10, 20) of furniture assembly instructions comprising instructions for assembling corresponding first and at least one additional furniture modules (12, 22a- n). The method (100) further comprises retrieving (130) assembly metadata (11, 21a-n) being indicative of respective relationships between furniture components (14, 24a-n) of the first and at least one additional furniture modules (12, 22a-n) to be included in a combined furniture module (32). The method (100) further comprises generating (140) a combined set (30) of furniture assembly instructions for assembling the combined furniture module (32) by combining the first set (10) and the at least one additional set (20a-n), the combined set (30) providing instructional media of recommended assembly steps (50) to be followed during an assembly procedure of the combined furniture module (32).
Owner:INTER IKEA SYST