System level DSP synthesis using voltage overscaling, unequal error protection & adaptive quality tuning

نویسندگان

  • Georgios Karakonstantis
  • Debabrata Mohapatra
  • Kaushik Roy
چکیده

In this paper, we propose a system level design approach considering voltage over-scaling (VaS) that achieves error resiliency using unequal error protection of different computation elements, while incurring minor quality degradation. Depending on user specifications and severity of process variations/channel noise, the degree of vas in each block of the system is adaptively tuned to ensure minimum system power while providing "just-theright" amount of quality and robustness. This is achieved, by taking into consideration system level interactions and ensuring that under any change of operating conditions only the "lesscrucial" computations, that contribute less to block/system output quality, are affected. The design methodology applied to a DCT/IDCT system shows large power benefits (up to 69%) at reasonable image quality while tolerating errors induced by varying operating conditions (VaS, process variations, channel noise). Interestingly, the proposed IDCT scheme conceals channel noise at scaled voltages. Index Terms Low power, error resilient, supply voltage scaling

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Vlsi Implementation of a Low-energy Soft Digital Filter

In this paper, we present a VLSI implementation of an energyefficient digital filtering algorithm developed using the soft DSP framework. Soft DSP refers to overscaling the supply voltage without sacrificing speed and employing algorithmic error-control to restore the resulting performance degradation. It is shown that delay imbalance at the circuit level inherent in existing arithmetic structu...

متن کامل

Low-power Signal Processing via Error-cancellation

We present an algorithmic noise-tolerance (ANT) technique for designing low-power DSP systems. The proposed technique achieves substantial energy savings via voltage overscaling, whereby the supply voltage is scaled beyond the minimum supply voltage Vdd−crit at which the architecture operates correctly for a given throughput specification. The resulting input-dependent soft errors are corrected...

متن کامل

Tuning of Extended Kalman Filter using Self-adaptive Differential Evolution Algorithm for Sensorless Permanent Magnet Synchronous Motor Drive

In this paper, a novel method based on a combination of Extended Kalman Filter (EKF) with Self-adaptive Differential Evolution (SaDE) algorithm to estimate rotor position, speed and machine states for a Permanent Magnet Synchronous Motor (PMSM) is proposed. In the proposed method, as a first step SaDE algorithm is used to tune the noise covariance matrices of state noise and measurement noise i...

متن کامل

Robust Scalable Video Streaming over Internet with Network-Adaptive Congestion Control and Unequal Loss Protection

Video communication with Quality of Service (QoS) is an important and challenging task. To have QoS provision or guarantee at application level in the current best-effort Internet, congestion control and error control are two effective approaches. In this paper, we propose a new network-adaptive congestion control scheme for scalable video streaming in conjunction with Unequal Loss Protection (...

متن کامل

A Low-power Digital Filter IC via Soft DSP f

In this paper we present an integrated circuit implementation of a soft DSP based low-power digital filter in 0.35pm, 3.3V CMOS process. Soft DSP is a low-power technique that employs voltage overscaling (VOS) and algorithmic noise-tolerance (ANT) to push the limits of energy-efficiency beyond that achievable by voltage scaling alone. VOS refers. to scaling the supply voltage beyond the limit i...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2009