Functional verification of an MPEG-4 decoder design using a random constrained movie generator

Abstract

The advent of the new VLSI technology and SoC design methodologies has brought about an explosive growth to the complexity of modern electronic circuits. As a result, functional verification has become a major bottleneck in any digital design flow. Functional verification has two major problems: choose good stimuli and measure coverage. This paper present a movie generator called RandMovie used together with a coverage library in order to implement a generator capable of producing a synthetic random constrained movie. With such a movie generator one is capable of generating good random constrained movies, increasing coverage and simulating all specified functionalities. A case study for an MPEG-4 decoder design has been used to demonstrate the effectiveness of this approach.

Genetic Algorithm applied to the Functional Verification in Digital Systems

“…The experimental results were obtained using the new approach to data generation in VeriSC methodology in three experiments: DPCM project and two modules of an MPEG-4 decoder [12]. The MPEG design used was a Simple Profile Level 0 with two external memories.…”

“…Fig. 7 describes the MPEG-4 decoder and the modules used [12]. These MPEG-4 modules were developed by Brazil IP group [12].…”

“…7 describes the MPEG-4 decoder and the modules used [12]. These MPEG-4 modules were developed by Brazil IP group [12]. The data synchronization between modules was performed using FIFO (First-In, First-Out) data structure, which ensure the correct order of transmitted data.…”

“…The output structure is generated after applying the DCT transform on the input structure reconstructing the output matrix coefficients. In this module, four buckets were described to perform the functional verification [12]. The buckets, variables and conditions are described in Table II.…”

“…The variable level indicates the value of the pixel and the variable run defines the number of coefficients with values equal to 0 before the value of variable level. The last coefficient of the matrix is defined by the variable last [12].…”