Misalignment-aware delay modeling of narrow on-chip interconnects considering variability

Najafi, Atabak; Bamberg, Lennart; Najafi, Ardalan; García-Ortiz, Alberto

doi:10.1109/mocast.2018.8376593

Cited by 4 publications

(3 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The signals alignment is a relatively well-known effect, especially in global segmented interconnects [25]- [27]. Traditional energy and delay models intrinsically assume that the signals switching on each wire of a segment occur at the same time.…”

Section: B Signals Alignment In Narrow Busesmentioning

confidence: 99%

“…Traditional energy and delay models intrinsically assume that the signals switching on each wire of a segment occur at the same time. However, as discussed in [27], misalignment of signals in the presence of noise and dynamic variations can result in delay misprediction of up to 40%. In the same way, intrinsic misalignment can affect the number of errors and the integervalue deviation.…”

Section: B Signals Alignment In Narrow Busesmentioning

confidence: 99%

“…The misalignment effect should be considered when a coding technique is developed. According to [27], the standard delay models overestimate the improvement of the forbidden transition free CAC coding technique by up to 12%. Therefore, a realistic delay and error probability estimation is only feasible when the misalignment effect is considered.…”

Section: B Signals Alignment In Narrow Busesmentioning

confidence: 99%

See 2 more Smart Citations

Integer-Value Encoding for Approximate On-Chip Communication

et al. 2019

Self Cite

View full text Add to dashboard Cite

Approximate computing has been broadly investigated in the computing domain to maintain the benefits of technology scaling. Recently, it has been extended to the communication domain. The idea is to trade transmission quality for energy or/and performance efficiency. Exploiting the integervalue representation of the transmitted signals in parallel buses, we propose two memoryless encoding approaches for approximate communications in order to reduce the integer value deviation. First, restricted to area-constrained applications, we propose a Combined Integer-Value (CIV) coding technique based on the swap and inversion of the input signals. Second, a Crosstalk-Avoidance-based Integer-Value (CAIV) coding technique for applications with a more relaxed area constraint is presented. The optimal mapping of the data words and codewords combined with the selective inversion of the input signals minimize the error magnitude in this coding technique. A comprehensive experimental result using a 65 nm commercial technology evaluates the proposed coding approaches. For example, our proposed CIV coding scheme improves the quality of the received images and the sampled radio communication signals with different modulation up to factor of 2 8 and 2 7 , respectively, as it compares with conventional transmission of signals with no coding. CAIV coder can improve the data transmission accuracy by about 2 18 as it compares with the conventional transmission. Furthermore, to assess the applicability of the proposed encoders, we carried out two case studies employing the scale-invariant feature transform algorithm and SOBEL edge detector. Approximate communication, on-chip interconnects, integer-value encoding, error efficiency, computer vision. INDEX TERMS

show abstract

Section: B Signals Alignment In Narrow Busesmentioning

confidence: 99%

Section: B Signals Alignment In Narrow Busesmentioning

confidence: 99%

Section: B Signals Alignment In Narrow Busesmentioning

confidence: 99%

See 1 more Smart Citation