Jin-Hyun Cho

Constrained-random bitstream generation for H.264/AVC decoder conformance test

We propose a novel method of video test bitstream for the H.264/AVC decoder conformance test without using input image sequences. Although the bitstreams can be generated from encoding the natural or artificial image sequences by a reference video encoder, their function coverage is not high enough because strong spatial and temporal correlations exist in the pixels and video parameters. The proposed method employs the constrained-random techniques to produce highly randomized video parameters in its bitstream. According to the experimental results, it substantially improves the function coverage and reduces the test time compared to those of normative conformance test bitstreams. Moreover, we can also generate the arbitrary bitstreams for stress test with the extreme values in the allowable ranges of the decoder parameters.

RTL generation of channel architecture templates for a template-based SoC design flow

In this paper, we propose the design methodology for communication channel templates from formal specification to RTL description. In this flow, design and verification start from one source, LTL property. We constructed LTL-to-TRS, which is translator from LTL property sets to Bluespec term-rewriting system (TRS) description. And, we use a Bluespec compiler as a synthesizer from TRS to RTL. Also, to match the implementation with the formal specification, we use a VIS solver as a model checker. And then, channel instances generated by proposed design method are transformed into channel template-generators for communication channel library. These channel templates can be used in DSE process in SoC design flow.