作者：
Son, KyungrakChoi, WanHUFS
Dept Informat & Commun Engn Yongin 17035 South Korea SNU
Dept Elect & Comp Engn Seoul 08826 South Korea SNU
Inst New Media & Commun Seoul 08826 South Korea
Despite its vital role in intelligent IoT networks, distributed computing was not studied well in wireless networks. Notably, the distortions introduced by wireless channels and noise can significantly undermine the a...
详细信息
Despite its vital role in intelligent IoT networks, distributed computing was not studied well in wireless networks. Notably, the distortions introduced by wireless channels and noise can significantly undermine the accuracy of distributed computations. Unfortunately, existing distributed computing schemes inadequately address this critical issue. In this paper, we aim at resolving the numerical stability issue associated within wireless networks and building a new computationally accurate design framework which concretely integrates distributed computing and communication. To this end, we first explore the previous group algebra based coded matrix computation scheme, suitable for numerically stable computation in noisy network, in orthogonal multiple access channel and perform a thorough analysis of computational errors. Furthermore, to leverage the inherent communication structure constructed in distributed computing and thus more efficiently utilize limited wireless resources, we propose to combine the Compute-and-Forward scheme with coded matrix computation. Furthermore, we devise a novel transmission scheme for distributed computing called the Broadcast-and-Compute scheme, in which a master node broadcasts the same information, not transmitting differently encoded submatrices to worker nodes. We analyze the computationerrors of Compute-and-Forward and Broadcast-and-Compute based coded matrix computation. The superiority of the proposed schemes is validated by various simulations.
It has been recently demonstrated that digital signal processing systems may possibly leverage unconventional voltage overscaling (VOS) to reduce energy consumption while maintaining satisfactory signal processing per...
详细信息
It has been recently demonstrated that digital signal processing systems may possibly leverage unconventional voltage overscaling (VOS) to reduce energy consumption while maintaining satisfactory signal processing performance. Due to the computation-intensive nature of most signal processing algorithms, the energy saving potential largely depends on the behavior of computer arithmetic units in response to overscaled supply voltage. This paper shows that different hardware implementations of the same computer arithmetic function may respond to VOS very differently and result in different energy saving potentials. Therefore, the selection of appropriate computer arithmetic architecture is an important issue in voltage-overscaled signal processing system design. This paper presents an analytical method to estimate the statistics of computer arithmetic computationerrors due to supply voltage overscaling. Compared with computation-intensive circuit simulations, this analytical approach can be several orders of magnitude faster and can achieve a reasonable accuracy. This approach can be used to choose the appropriate computer arithmetic architecture in voltage-overscaled signal processing systems. Finally, we carry out case studies on a coordinate rotation digital computer processor and a finite-impulse-response filter to further demonstrate the importance of choosing proper computer arithmetic implementations.
暂无评论