Frequency domain channel correlation can be exploited to reduce feedback in limited feedback beamforming multiple-input multiple-output orthogonal frequency division multiplexing wireless systems. Prior methods rely o...
详细信息
Frequency domain channel correlation can be exploited to reduce feedback in limited feedback beamforming multiple-input multiple-output orthogonal frequency division multiplexing wireless systems. Prior methods rely on downsampling, interpolation, or clustering the channel state information in the frequency domain. The resulting compressed samples are quantized using one-shot quantization on the Grassmann manifold. The resolution, unfortunately, is limited. We propose a new frequency domain compression technique to obtain high resolution channel state information. The key idea is to use predictive coding on the Grassmann manifold, exploiting the correlation between adjacent subcarriers.
This paper proposes a novel 3D piecewise planar reconstruction algorithm, which utilizes the statistical error between a particular frame and its prediction to refine a coarse 3D piecewise planar representation. The a...
详细信息
This paper proposes a novel 3D piecewise planar reconstruction algorithm, which utilizes the statistical error between a particular frame and its prediction to refine a coarse 3D piecewise planar representation. The algorithm aims utilization of 3D scene geometry to remove the visual redundancy between frame pairs in any predictive coding scheme. This approach associates the rate increase with the quality of representation for determining an efficient description for a given budget. The preliminary experiments on synthetic and real data indicate the validity of the rate-distortion based approach.
Motivated by theoretical analysis of the curve fitting problem based on equivalent kernel, in this paper we propose a local adaptive learning and fusion model for side information interpolation in distributed video co...
详细信息
Motivated by theoretical analysis of the curve fitting problem based on equivalent kernel, in this paper we propose a local adaptive learning and fusion model for side information interpolation in distributed video coding. In the proposed model, each pixel in the interpolated frame is approximated as the linear combination of samples within a local spatio-temporal window using kernel parameters as weight. The size of training window can be adaptive to the motion characteristic of video, from samples in which the kernel parameters can be locally learned. In order to further improve the quality of interpolated frames, we introduce a belief-projection based fusion strategy with adaptive weights for multiple interpolated results which are with the same time index. Experimental results demonstrate that the proposed learning and fusion model is effective in performance for side information interpolation in distributed video coding.
A study was performed to evaluate various techniques for improving the robustness of an Adaptive predictive Coder operating in the presence of channel errors. The communications environment consisted of a packetized, ...
详细信息
A study was performed to evaluate various techniques for improving the robustness of an Adaptive predictive Coder operating in the presence of channel errors. The communications environment consisted of a packetized, symbol oriented system in which channel errors were manifested as both 5-bit symbol errors in the received bit stream and the occasional loss of an entire packet of data. To reduce the effects of symbol errors forward error correction was applied to a selected subset of the transmitted parameter bits. At the receiver, smoothing strategies were employed to replace those parameters in which errors could be detected but not corrected and to cope with situations in which entire packets were lost. The introduction of these techniques to the basic APC vocoder added 900-1100 bps to the bit rate and resulted in significantly improved robustness in channel errors.
A new hybrid predictive coding scheme for lossy data compression is introduced and studied here. The proposed technique attempts to improve the compression performance of a conventional adaptive predictive coder throu...
详细信息
A new hybrid predictive coding scheme for lossy data compression is introduced and studied here. The proposed technique attempts to improve the compression performance of a conventional adaptive predictive coder through three levels of prediction. The first level estimates and removes a time-varying mean signal. The second level uses DFT and IDFT to estimate and remove the predictable high frequency subband component of the signal, whereas the third level performs the adaptive predictive coding of the smoothed signal. The resultant residual is then quantized using an adaptive quantizer with 3 and 5 levels. Experimental results based on a group of test signals seem to indicate that the proposed hybrid scheme can improve the lossy compression achievable using a conventional adaptive predictive coder.
The purpose of the research reported here was to find out some acoustical and perceptual characteristics of the French nasal vowels. Acoustic analysis was made using a Pole-Zero Analysis-by-Synthesis procedure : below...
详细信息
The purpose of the research reported here was to find out some acoustical and perceptual characteristics of the French nasal vowels. Acoustic analysis was made using a Pole-Zero Analysis-by-Synthesis procedure : below 4 kHz (in the case of male adult informants) the main acoustic correlates of the nasal vowels are the formant shifts and the introduction of two pole-zero pairs, one below F1, the other between F2 and F3. By means of a pole-zero synthesizer, synthetic stimuli were generated : two of them have the spectral characteristics of the real vowels [ \varepsilon ] and [ \tilde{\varepsilon} ], while the others are stimuli whose spectral characteristics are between those of [ \varepsilon ] and [ \tilde{\varepsilon} ]. Phonemic and phonetic perceptual tests were conducted. Their results show that formant shifts and pole-zero separation contribute almost equally to the decision on the phonemic identity of / \varepsilon / versus / \tilde{\varepsilon} /, whereas pole-zero separation has a stronger effect than formant shifts in the phonetic judgment on nasality.
This article argues for an externalist self-model in self-knowledge, as an answer to the question of how to defend self-knowledge in the face of evident errors in self-cognition such as self-illusions. The externalist...
详细信息
This article argues for an externalist self-model in self-knowledge, as an answer to the question of how to defend self-knowledge in the face of evident errors in self-cognition such as self-illusions. The externalist model is contrasted with an internalist model, although both are based on a predictive coding mechanism. The main thesis is that for preserving the status of self-knowledge as knowledge an external element is needed - namely, some feedback from the environment - which corrects not only cognitive prediction errors regarding the world but also errors in self-cognition by forcing the proper action of the subject (agent) and dynamic adaptation of the model of the embodied self. The mechanism of self-corrections is best described by externalist models.
The mental states of other people are components of the external world that modulate the activity of our sensory epithelia. Recent probabilistic frameworks that cast perception as unconscious inference on the external...
详细信息
The mental states of other people are components of the external world that modulate the activity of our sensory epithelia. Recent probabilistic frameworks that cast perception as unconscious inference on the external causes of sensory input can thus be expanded to enfold the brain's representation of others' mental states. This paper examines this subject in the context of the debate concerning the extent to which we have perceptual awareness of other minds. In particular, we suggest that the notion of perceptual presence helps to refine this debate: are others' mental states experienced as veridical qualities of the perceptual world around us? This experiential aspect of social cognition may be central to conditions such as autism spectrum disorder, where representations of others' mental states seem to be selectively compromised. Importantly, recent work ties perceptual presence to the counterfactual predictions of hierarchical generative models that are suggested to perform unconscious inference in the brain. This enables a characterisation of mental state representations in terms of their associated counterfactual predictions, allowing a distinction between spontaneous and explicit forms of mentalising within the framework of predictive processing. This leads to a hypothesis that social cognition in autism spectrum disorder is characterised by a diminished set of counterfactual predictions and the reduced perceptual presence of others' mental states. (C) 2015 Elsevier Inc. All rights reserved.
Aron Gurwitsch's theory of the structure and dynamics of consciousness has much to offer contemporary theorizing about consciousness and its basis in the embodied brain. On Gurwitsch's account, as we develop i...
详细信息
Aron Gurwitsch's theory of the structure and dynamics of consciousness has much to offer contemporary theorizing about consciousness and its basis in the embodied brain. On Gurwitsch's account, as we develop it, the field of consciousness has a variable sized focus or "theme" of attention surrounded by a structured periphery of inattentional contents. As the field evolves, its contents change their status, sometimes smoothly, sometimes abruptly. Inner thoughts, a sense of one's body, and the physical environment are dominant field contents. These ideas can be linked with (and help unify) contemporary theories about the neural correlates of consciousness, inattention, the small world structure of the brain, meta-stable dynamics, embodied cognition, and predictive coding in the brain. Published by Elsevier Inc.
Statistical learning (SL) is an innate mechanism by which the brain automatically encodes the n-th order transition probability (TP) of a sequence and grasps the uncertainty of the TP distribution. Through SL, the bra...
详细信息
Statistical learning (SL) is an innate mechanism by which the brain automatically encodes the n-th order transition probability (TP) of a sequence and grasps the uncertainty of the TP distribution. Through SL, the brain predicts a subsequent event (en+1) based on the preceding events (en) that have a length of “n”. It is now known that uncertainty modulates prediction in top-down processing by the human predictive brain. However, the manner in which the human brain modulates the order of SL strategies based on the degree of uncertainty remains an open question. The present study examined how uncertainty modulates the neural effects of SL and whether differences in uncertainty alter the order of SL strategies. It used auditory sequences in which the uncertainty of sequential information is manipulated based on the conditional entropy. Three sequences with different TP ratios of 90:10, 80:20, and 67:33 were prepared as low-, intermediate, and high-uncertainty sequences, respectively (conditional entropy: 0.47, 0.72, and 0.92 bit, respectively). Neural responses were recorded when the participants listened to the three sequences. The results showed that stimuli with lower TPs elicited a stronger neural response than those with higher TPs, as demonstrated by a number of previous studies. Furthermore, we found that participants adopted higher-order SL strategies in the high uncertainty sequence. These results may indicate that the human brain has an ability to flexibly alter the order based on the uncertainty. This uncertainty may be an important factor that determines the order of SL strategies. Particularly, considering that a higher-order SL strategy mathematically allows the reduction of uncertainty in information, we assumed that the brain may take higher-order SL strategies when encountering high uncertain information in order to reduce the uncertainty. The present study may shed new light on understanding individual differences in SL performa
暂无评论