An adaptive current-threshold determination for IDDQ testing based on Bayesian process parameter estimation

Application of IDDQ testing to LSIs fabricated using advanced process technology is becoming increasingly difficult due to large variability of scaled devices. In this paper, we propose a novel technique that adaptively determines per-chip current-threshold for IDDQ testing to enhance test accuracy. In the proposed technique, process condition of a chip and fault-sensitization vector are first estimated based on measured IDDQ currents through Bayesian inference. Then, using the estimated process condition, a statistical distribution of the leakage current for each test pattern is calculated and suitable current-threshold is determined by the distribution. Simulation experiments demonstrate that the proposed technique can successfully detect a very small leakage fault, down to 16% of the nominal IDDQ current with the test escape ratio of 3.1 %.

[1]  Lawrence T. Pileggi,et al.  Projection-based statistical analysis of full-chip leakage power with non-log-normal distributions , 2006, 2006 43rd ACM/IEEE Design Automation Conference.

[2]  David Blaauw,et al.  Statistical analysis of subthreshold leakage current for VLSI circuits , 2004, IEEE Transactions on Very Large Scale Integration (VLSI) Systems.

[3]  David Blaauw,et al.  Parametric yield estimation considering leakage variability , 2004, Proceedings. 41st Design Automation Conference, 2004..

[4]  Theo J. Powell,et al.  Delta Iddq for testing reliability , 2000, Proceedings 18th IEEE VLSI Test Symposium.

[5]  Anura P. Jayasumana,et al.  Clustering based evaluation of I/sub DDQ/ measurements: applications in testing and classification of ICs , 2000, Proceedings 18th IEEE VLSI Test Symposium.

[6]  Robert C. Aitken,et al.  Current ratios: a self-scaling technique for production IDDQ testing , 2000, Proceedings International Test Conference 2000 (IEEE Cat. No.00CH37159).

[7]  C. D. Gelatt,et al.  Optimization by Simulated Annealing , 1983, Science.

[8]  David Blaauw,et al.  Modeling and analysis of parametric yield under power and performance constraints , 2005, IEEE Design & Test of Computers.

[9]  W. Robert Daasch,et al.  Defect screening using independent component analysis on I/sub DDQ/ , 2005, 23rd IEEE VLSI Test Symposium (VTS'05).

[10]  Marcel J. M. Pelgrom,et al.  Matching properties of MOS transistors , 1989 .

[11]  Takashi Sato,et al.  A Bayesian-based process parameter estimation using IDDQ current signature , 2012, 2012 IEEE 30th VLSI Test Symposium (VTS).

[12]  Claude Thibeault.  Improving Delta-I/sub DDQ/-based test methods , 2000, Proceedings International Test Conference 2000 (IEEE Cat. No.00CH37159).

[13]  James McNames,et al.  Neighbor selection for variance reduction in I/sub DDQ/ and other parametric data , 2001, Proceedings International Test Conference 2001 (Cat. No.01CH37260).