Intelligent Sensor Interrupts for Silicon Testing Efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current silicon life cycle management and infield testing of integrated circuit devices face inefficiencies due to the need to continuously monitor all sensors, leading to high volumes of non-anomaly data, increased probing time, limited telemetry endpoints, and potential permanent failures from abnormal parameter spikes during testing.
Innovation Solution
Intelligent sensors with digital and analog portions that detect physical characteristics and send interrupts only when anomalies occur, allowing selective probing and reducing unnecessary monitoring, thereby improving data quality and reducing probing time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If all sensors are continuously monitored to ensure complete coverage of silicon life cycle parameters, then measurement completeness is improved, but probing time and system complexity increase significantly
Solution Approach 1:
Sensors monitor themselves and automatically generate interrupts when anomalies are detected, eliminating the need for external continuous probing. The sensor's own resources are used to detect its own state and trigger notifications, reducing the monitoring burden on the test system.
Solution Approach 2:
Sensors provide feedback about their operational state through interrupts triggered by anomaly detection. This feedback mechanism allows the system to respond to actual sensor conditions rather than requiring continuous active monitoring, reducing probing time while maintaining monitoring effectiveness.
2Loss of information
If all sensors are continuously probed to collect complete data, then data completeness is improved, but the volume of non-anomaly data increases and monitoring efficiency decreases
Solution Approach 1:
Sensors self-monitor and self-report only when anomalies occur, eliminating unnecessary data transmission for normal operations. This reduces the volume of non-anomaly data while maintaining complete information about actual problems.
Solution Approach 2:
Instead of continuous probing, the system uses periodic interrupt-driven communication where sensors actively notify the test system only when anomalies are detected. This maintains data completeness for critical events while dramatically improving monitoring efficiency by eliminating continuous data transfer for normal operations.
3Speed
If high-frequency probing is performed to detect rapid parameter changes, then detection speed is improved, but the risk of permanent failure from abnormal parameter spikes increases
Solution Approach 1:
Sensors monitor their own parameters and trigger interrupts only when anomalies are detected, eliminating the need for external high-frequency probing. This maintains rapid anomaly detection capability while reducing the stress on the device that would result from continuous external probing.
Solution Approach 2:
The system replaces continuous high-frequency probing with periodic interrupt-driven monitoring. Sensors actively check their own state and notify only when anomalies occur, maintaining detection speed for critical events while eliminating the harmful effects of continuous external probing on device reliability.
4Measurement precision
If multiple telemetry endpoints are added to improve monitoring coverage, then measurement completeness is improved, but device complexity and the number of components increase
Solution Approach 1:
Sensors are designed to monitor multiple parameters and trigger interrupts for various anomaly conditions using a unified mechanism. This multi-functionality provides comprehensive monitoring coverage without requiring separate dedicated components for each monitoring function, reducing overall system complexity.
Data Source
AI summary
Methods and apparatus relating to intelligent sensors for high quality silicon life cycle management as well as efficient infield structural and/or functional testing are described. In an embodiment, one or more registers store configuration data. A sensor having sensor event detection logic circuitry detects an event based at least in part on one or more sensor signals and the stored configuration data. The sensor event detection logic circuitry generates a signal to cause interrupt generator logic circuitry of the sensor to generate an interrupt. Other embodiments are also disclosed and claimed.