Hardware support for deep convolutional neural networks (CNNs) is critical to advanced computer vision in mobile and embedded devices. Current designs, however, accelerate generic CNNs; they do not exploit the unique characteristics of real-time vision. We propose to use the temporal redundancy in natural video to avoid unnecessary computation on most frames. A new algorithm, activation motion ...

In the RESUME project we explore the use of reconfigurable hardware for the design of portable multimedia systems by developing a hardware-friendly scalable wavelet-based video codec. Our scalable video codec provides the ability to decoded the video stream with reduced frame rate, resolution or image quality directly from the original encoded video stream. This is important for portable device...

Multiple Input Multiple Output Orthogonal Frequency Division Multiplexing (MIMO-OFDM) technology is an attractive transmission technique for wireless communication systems with multiple antennas at receiver and transmitter. The reliability of Data rate and diversity can be increased along with the stability for multi-path signals. The efficient implementation of a Low-Power 64-point Pipeline FF...

This paper demonstrates that an adaptive computing system (ACS) is a good platform for implementing robotic control algorithms. We show that an ACS can be used to provide both good performance and high dependability. An example of an FPGAimplemented dependable control algorithm is presented. The flexibility of ACS is exploited by choosing the best precision for our application. This makes it po...

This paper describes how MEADEP, a system level dependability prediction tool, and CASRE, a software reliability growth prediction tool can be used together to predict system reliability (probability of failure in a given time interval), availability (proportion of time service is available), and performability (reward-weighted availability) for a. The system includes COTS hardware, COTS softwa...

Modular redundancy, the traditional approach to fault tolerance, is prohibitively expensive because of the overhead in replicating the hardware. In this paper we discuss alternative methods for obtaining fault-tolerant architectures for discrete-time (DT) linear time-invariant (LTI) filters. We view such filters as dynamic systems and provide fault tolerance to them by embedding their state int...

In this paper, the authors propose an algorithm to find all the minimal signed digit (MSD) representations of a constant and present an algorithm to synthesize digital filters based on the MSD representation. The hardware complexity of a digital signal processing system is dependent on the number system used for the implementation. Although the canonical signed digit (CSD) representation is wid...

Cross-layer fault-tolerance solutions are the key to effectively and efficiently increase the reliability in future safety-critical real-time systems. Replicated software execution with hardware support for error detection is a cross-layer approach that exploits future many-core platforms to increase reliability without resorting to redundancy in hardware. The performance of such systems, howev...

This paper presents an Orthogonal Frequency Division Multiplexing (OFDM) transceiver that makes use of a low power Fast Fourier Transform (FFT) along with a Least Mean Square (LMS) filter. The folded FFT is developed via folding transformation and register minimization techniques with real values as inputs which leads to reduction in hardware complexity by exploiting the redundancy present in c...

The particular attraction of hardware voting is that a high degree of fault-tolerance should be achievable in a largely software transparent manner and with a minimal performance penalty. The proposals are particularly applicable to systems using multiple transputers, allowing both the natural parallelism of the application and the redundancy demanded by replication to be exploited in a consist...