Efficient communication support in predictable heterogeneous MPSoC designs for streaming applications
Citations
12 citations
Cites background or methods from "Efficient communication support in ..."
...Previously, a Communication Assist (CA) for homogeneous general purpose processors [5] and for accelerators [6] had also been introduced, but it was without a complete framework....
[...]
...The ARM PE CA was implemented through AXI DMA IP [3] while the accelerator CA implementation is from our previous work [6]....
[...]
...Additionally, actors Proj, Eros and Bin also have hardware accelerator implementations [6]....
[...]
...Actors wr and rd are for modeling write and read latencies of a word and actor d1 is for the de-serialization of words that belong to the same token [6]....
[...]
2 citations
2 citations
Cites background from "Efficient communication support in ..."
...The graphs can be used to analyze, at design-time, the temporal behavior and resource requirements (buffer sizes) of applications [24]....
[...]
2 citations
1 citations
References
37,017 citations
"Efficient communication support in ..." refers methods in this paper
...The front end of the pipeline applies OTSU optimal threshold [20] to segment the OLED structures from the background....
[...]
1,985 citations
"Efficient communication support in ..." refers background or methods in this paper
...Synchronous Data Flow (SDF) is a model of computation commonly used to model streaming applications [1]....
[...]
...The Synchronous Data Flow (SDF) model-of-computation is a very powerful model for analyzing streaming applications [1]....
[...]
[...]
305 citations
"Efficient communication support in ..." refers background or methods in this paper
...The analysis tool that MAMPS uses is called SDF3 [8]....
[...]
..., through SDF analysis tools such as SDF3 [8]....
[...]
...It calculates buffer assignments, and predicts the throughput of this mapping [8]....
[...]
...By integrating this SDF model onto our SDF analysis tool, SDF3 [8], worst-case system properties, such as throughput, latency, and buffer sizes can be conservatively analyzed at design time....
[...]
300 citations
284 citations