This paper reports on the application of the Walsh-Hadamard transform (WHT) for data compression in brain-machine/brain-computer interfaces. Using the proposed technique, the amount of the neural data transmitted off the implant is compressed by a factor of at least 63 at the expense of as low as 4.66% RMS error between the signal reconstructed on the external host and the original neural signa...

The development of rf systems capable of producing high peak power (hundreds of megawatts) at relatively short pulse lengths (0.1-5 microseconds) is currently being driven mainly by the requirements of future high energy linear colliders, although there may be applications to industrial, medical and research linacs as well. The production of high peak power rf typically involves four basic elem...

Wireless video-surveillance networks are gaining increasing popularity due to the number of applications they make possible to carry out. In this paper, we address the problem of designing a wireless video-surveillance network so as to optimize its performance. In particular, we investigate the possible trade-offs between energy consumption and image quality. We provide simulation results showi...

Data compression and learning are, in some sense, two sides of the same coin. If we paraphrase Occam’s razor by saying that a small theory is better than a larger theory with the same explanatory power, we can characterize data compression as a preoccupation with small, and learning as a preoccupation with better. Nevill-Manning et al. (1994) presented an algorithm, since dubbed SEQUITUR, that ...

The networked visual sharing application in multi-party collaboration environment needs compression of video streams due to network bandwidth limitation. For interactive real-time sharing, real-time compression of high-quality video as well as audio echo cancellation are required, which commonly depend on the availability of high-cost hard-to-setup specialized compression and echo-cancellation ...

The combination of a source coder and a digital modu-lator is in this article viewed as an analog-to-analog converter with signal compression ability. A reference channel with deened available bandwidth, total received power, and noise statistics, can transmit one analog source signal. Compression is here deened as the number of such source signals that the digital system can transmit over the ...

This paper presents the application of spectral analysis methods that are based on the FFT algorithm to the study of the surface EMG power spectrum compression. Some preferred characteristic frequencies used to monitor the spectral compression are the mean and median frequencies. In this work, we show that the spectral density estimation by the Blackman-Tukey method associated with the weightin...

In this paper, a D-band power amplifier (PA) based on 0.5-μm InP DHBT is presented. Wilkinson combiners with broadband stepped-impedance matching are used, and the eight-way PA is designed for wideband power performance. As input power is fixed at 3 dBm, the PA exhibits a saturated output power of 16.8 dBm and 7.9% PAE at 150 GHz with >45-GHz 3-dB power bandwidth from 125 GHz to 170 GHz. Meanwh...

We consider the enciphering of a data stream while being compressed by a LZ algorithm. This has to be compared to the classical encryption after compression methods used in security protocols. Actually, most cryptanalysis techniques exploit patterns found in the plaintext to crack the cipher; compression techniques reduce these attacks. Our scheme is based on a LZ compression in which a Vernam ...

| The output of recently developed imaging devices is more often in a digital format and both storage space and transmission times of these images proot from data compression. However, for legal and functional reasons, loss of information should be avoided, introducing the need for lossless compression techniques. Recent evolutions in computing power have given rise to the birth of several new ...