Using integer equations for high level formal verification property checking

Alizadeh, Bijan; Kakoee, Mohammad Reza

doi:10.1109/isqed.2003.1194711

“…Also path coverage on LTL properties can be obtained based on this word-level model efficiently unlike of other models [11,12]. As mentioned before, our approach does not need to solve integer equations or do satisfiability checking despite of other approaches [3,4,5,6,9].…”

Section: Discussionmentioning

confidence: 99%

Word level functional coverage computation

Alizadeh¹

2006

Proceedings of the 2006 Conference on Asia South Pacific Design Automation - ASP-DAC '06

2

0

View full text Add to dashboard Cite

This paper proposes a word-level coverage metric to determine the completeness of a set of properties verified by a word-level method. An algorithm is presented to compute a functionality based coverage metric for a sequence property as specification. Control, intermediate and output signals are represented by a multiplexer based structure of linear integer equations, and RT level properties are directly applied to this representation. A set of integer equations are symbolically simulated based on the specified property in a predictable time. We used a canonical form of linear Taylor Expansion Diagram.

show abstract

“…CHIRE is a format that for every hardware description language construct has a distinct C++ object. CHIRE format is then transformed to a Data Flow Graph (DFG) [11] format and subsequently to BDD [12] representation.…”

Section: Lemma 1 Given An Hdl Model Of a Design + With A Set Of Assigmentioning

confidence: 99%

Optimized Assignment Coverage Computation in Formal Verification of Digital Systems

Nabi

¹

,

Shojaei

²

,

Mohammadi

³

2007

16th Asian Test Symposium (ATS 2007)

0

View full text Add to dashboard Cite

Model checking thoroughly verifies the design correctness with respect to a specification. When the verification process succeeds, we can only postulate the correctness of the design relative to the given specification. How far can we affirm the verified design implements all the behavior of the desired system? With this regard we need to estimate the completeness of the properties by using some coverage metrics. In this paper, we have proposed a new metric called assignment coverage and an optimized method to overcome the intensive computations required for the multiple transformations among the abstract layers in the verification tool. The proposed coverage computation method provides adequate information to complete the set of properties. Finally, we have applied the proposed metric to some verification benchmark to reveal the effectiveness of this metric in finding undetected coverage holes.16th IEEE Asian Test Symposium

show abstract

“…Also path coverage on LTL properties can be obtained based on this word-level model efficiently unlike of other models [11,12]. As mentioned before, our approach does not need to solve integer equations or do satisfiability checking despite of other approaches [3,4,5,6,9].…”

Section: Discussionmentioning

confidence: 99%

“…For this work, we used VHDL to describe a design and a sequence format to describe its properties. We extract Data Flow Graph (DFG) for the design [9], convert it to LTED, and prove the design property symbolically [10].…”

Section: Introductionmentioning

confidence: 99%

Word level functional coverage computation

Alizadeh¹

Asia and South Pacific Conference on Design Automation, 2006.

Self Cite

View full text Add to dashboard Cite

This paper proposes a word-level coverage metric to determine the completeness of a set of properties verified by a word-level method. An algorithm is presented to compute a functionality based coverage metric for a sequence property as specification. Control, intermediate and output signals are represented by a multiplexer based structure of linear integer equations, and RT level properties are directly applied to this representation. A set of integer equations are symbolically simulated based on the specified property in a predictable time. We used a canonical form of linear Taylor Expansion Diagram.

show abstract

Using integer equations for high level formal verification property checking

Abstract: Abstract

Cited by 10 publications

References 29 publications

Word level functional coverage computation

Word level functional coverage computation

Optimized Assignment Coverage Computation in Formal Verification of Digital Systems

Word level functional coverage computation

Contact Info

Product

Resources

About