US 12,271,252 B2
Increasing the robustness of electronic systems against SEU and other radiation effects
Luis Angel Maestro Ruiz De Temino, San Francisco, CA (US); Jerzy Kolek, Wlosan (PL); Slawomir Cichon, Paszkowka (PL); Pratibha Gupta, Menlo Park, CA (US); Gustav Derkits, New Providence, NJ (US); and Kiran Patel, Piscataway, NJ (US)
Assigned to Nokia Solutions and Networks Oy, Espoo (FI)
Filed by Nokia Solutions and Networks Oy, Espoo (FI)
Filed on Nov. 3, 2022, as Appl. No. 17/979,940.
Prior Publication US 2024/0152416 A1, May 9, 2024
Int. Cl. G06F 11/00 (2006.01); G06N 7/01 (2023.01)
CPC G06F 11/004 (2013.01) [G06F 11/008 (2013.01); G06N 7/01 (2023.01); G06F 2201/805 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method, comprising:
fetching, by an electronic device, first information from memory devices concerning errors associated with radiation effects in the memory devices, wherein the errors are assumed to be caused by radiation, and storing the fetched first information in storage;
fetching, by the electronic device, second information about system performance associated with the electronic device, and storing the fetched second information in the storage;
monitoring, by the electronic device, current parameters of the memory devices, and storing monitored current parameters in the storage;
calculating, by the electronic device in real-time using an algorithm, parameter values for configuration of the memory devices based on the fetched first information, the fetched second information, and the monitored current parameters to determine calculated parameter values, the calculating performed to adjust the parameter values to improve a metric of the system performance against the errors associated with the radiation effects, wherein the algorithm uses a list of system performance metric values determined using a machine learning model that has been trained with data collected during testing of the electronic device under a radiation flux, wherein the calculating comprises learning new parameters of the machine learning model using retrieved stored first information, retrieved stored second information, and retrieved stored monitored current parameters to revise the system performance metric values in the list;
implementing by the electronic device the calculated parameter values for the configuration of the memory devices;
iterating the fetching the first information, fetching the second information, monitoring, and calculating; and
modifying, in real-time using the algorithm that uses the revised system performance metric values from the machine learning model with the new parameters, configurations of the memory devices to address the errors that are assumed to be caused by radiation.