In this paper, a fully Bayesian algorithm for endmember extraction and abundance estimation for hyperspectral imagery is introduced. Following the linear mixing model, each pixel spectrum of the hyperspectral image is...
详细信息
ISBN:
(纸本)9781424451807
In this paper, a fully Bayesian algorithm for endmember extraction and abundance estimation for hyperspectral imagery is introduced. Following the linear mixing model, each pixel spectrum of the hyperspectral image is decomposed as a linear combination of pure endmember spectra. The estimation of the unknown endmember spectra and the corresponding abundances is conducted in a unified manner by generating the posterior distribution of the unknown parameters under a hierarchical Bayesian model. The proposed model accounts for non-negativity and full-additivity constraints, and exploits the fact that the endmember spectra lie on a lower dimensional space. A Gibbs algorithm is proposed to generate samples distributed according to the posterior of interest. Simulation results illustrate the accuracy of the proposed joint Bayesian estimator.
A hybrid acoustic vector sensorarray (HAVSA) is consisted mainly of pressure hydrophones and a few vector sensors within. It can acquire good direction-finding results as well as the ability to discriminate the left-...
详细信息
In this paper, we study the source localization problem in wireless sensor networks where the location of the source is estimated according to the quantized measurements received from sensors in the field. We propose ...
详细信息
ISBN:
(纸本)9781424451791
In this paper, we study the source localization problem in wireless sensor networks where the location of the source is estimated according to the quantized measurements received from sensors in the field. We propose an energy efficient iterative source localization scheme, where the algorithm begins with a coarse location estimate obtained from a set of anchor sensors. Based on the available data at each iteration, we approximate the posterior probability density function (pdf) of the source location using a Monte Carlo method and we use this information to activate a number of non-anchor sensors that minimize the Conditional Posterior Cramer Rao Lower Bound (C-PCRLB). Then we also use the Monte Carlo approximation of the posterior pdf of the source location to compress the quantized data of each activated sensor using distributed data compression techniques. Simulation results show that the proposed iterative method achieves the mean squared error that gets close to the unconditional Posterior Cramer Rao Lower Bound (PCRLB) for a Bayesian estimate based on quantized data from all the sensors within a few iterations. By selecting only the most informative sensors, the iterative approach also reduces the communication requirements significantly and resulting in energy savings.
The real-time ultrasonic imaging system can be achieved using a digital beamforming (DBF) method The critical part of the DBF is the real-time sampled-delay focusing (SDF) which requires a large number of memories (FI...
详细信息
ISBN:
(纸本)076951944X
The real-time ultrasonic imaging system can be achieved using a digital beamforming (DBF) method The critical part of the DBF is the real-time sampled-delay focusing (SDF) which requires a large number of memories (FIFO) to store the scanned information. The sampled-delay focusing technique is used to eliminate the use of the analog delay lines. This paper concerns the design and implementation of a pipelined sampled-delay architecture for ultrasonic digital beamforming. The design uses a minimum size look-up memory to store the initial scan information as opposed to previous approaches. The circuit is implemented in CMOS 0.18 mum technology and the resulting active layout area is 0. 14 mm(2) while its total power consumption is below 40 mW.
Space-time multiuser detection (STMD) is an emerging technique for wideband CDMA mobile communication systems, where the problems of the multiuser access interference (MUAI) and the inter-symbol interference (ISI) can...
详细信息
ISBN:
(纸本)0780367200
Space-time multiuser detection (STMD) is an emerging technique for wideband CDMA mobile communication systems, where the problems of the multiuser access interference (MUAI) and the inter-symbol interference (ISI) can be solved simultaneously. However, the complexity becomes one of the key problems in its practical implementation because of the exponential increment of computation burden due to increased number of users. In this paper, we propose the sectorized STMD scheme that confines the detection within a given sector. In the proposed scheme, the number of users in the sector becomes smaller, and, subsequently, the system complexity becomes lower. Due to the sectorized beamforming, the signals to noise ratio of the users' multipath rays are enhanced, which lead to the performance improvements of the signature waveforms' estimation and the detection.
This paper describes a pulse based signalprocessing technique for VLSI implementation of fully programmable analogue arrays. A pulse width modulation signal representation and a basic Palmo(1) analogue cell are intro...
详细信息
ISBN:
(纸本)0780342402
This paper describes a pulse based signalprocessing technique for VLSI implementation of fully programmable analogue arrays. A pulse width modulation signal representation and a basic Palmo(1) analogue cell are introduced. The equivalence of the Palmo cell to a switched capacitor building block is demonstrated and the sources of harmonic distortion are discussed. VLSI results showing Ist, 2nd, 3rd order IIR filters and a 24th order FIR filter are presented. Improvements for a fully programmable analogue array currently under testing are discussed.
Single Snapshot R-D Unitary Tensor-ESPRIT Using an Augmentation of the Tensor Order (SS-U-TE-ATO) is a subspace-based parameter estimation technique for R-dimensional (R-D) undamped harmonics using a single snapshot o...
详细信息
ISBN:
(纸本)9781424451791
Single Snapshot R-D Unitary Tensor-ESPRIT Using an Augmentation of the Tensor Order (SS-U-TE-ATO) is a subspace-based parameter estimation technique for R-dimensional (R-D) undamped harmonics using a single snapshot of data. In SS-U-TE-ATO the measurement data is packed into a measurement tensor to which spatial smoothing is applied. We propose the construction of R higher-order tensors from the spatially smoothed tensor by exploiting the inherent structure of the spatially smoothed tensor. In the next step of SS-U-TE-ATO, R enhanced real-valued signal subspace estimates, one for each dimension, are obtained from the R complex-valued higher-order tensors. We show that SS-U-TE-ATO performs significantly better than Unitary Tensor-ESPRIT (U-TE) directly applied to the spatially smoothed tensor. Moreover, for each dimension, SS-U-TE-ATO is almost insensitive to changes in the number of sensors per subarray provided that the number of subarrays is greater than the number of sources. Thereby we avoid the problem of selecting the optimum subarray size for a given source configuration.
作者：
Lie, S.Leyman, R.Chew, Y.H.
Digital Wireless Department TeleTech Park Singapore Science Park II 20 Science Park Road #02-34/37 Singapore117674 Singapore
This paper proposes a closed-form estimation technique for a wideband constant amplitude polynomial-phase signal embedded in white Gaussian noise. It is shown that it is possible to jointly estimate the highest order ...
详细信息
We propose a novel sensor interface for detecting notes in the musical audio signals, particularly with reference to polyphonic music of percussive pitched musical instruments. Our sensor interface is able to transfor...
详细信息
ISBN:
(纸本)9781424447084
We propose a novel sensor interface for detecting notes in the musical audio signals, particularly with reference to polyphonic music of percussive pitched musical instruments. Our sensor interface is able to transform acoustic pressure, caused by a sound wave, into a musical score, that is a symbolic representation of musical notes. We focuses on note events and their main characteristics: the onset (note attack instant) and the pitch (note name). signalprocessing techniques based on the Constant-Q Transform (CQT) are used to create a time-frequency representation of the signal. In particular, we propose a supervised classification methods based on Support Vector Machine (SVM) to detect pitch. Instead, our onset detection algorithm exploits a Short Time Fourier Transform (STET) representation of the audio signal. Finally, to validate our method, we present a collection of experiments using a wide number of musical pieces of heterogeneous styles.
The development of microelectronic lab-on-a-chip devices (LOACs) can now be pursued thanks to advances in silicon technology. As these kinds of devices may integrate different functional units, much care has to be put...
详细信息
ISBN:
(纸本)076951944X
The development of microelectronic lab-on-a-chip devices (LOACs) can now be pursued thanks to advances in silicon technology. As these kinds of devices may integrate different functional units, much care has to be put in the design of control units which have to provide real-time control capabilities in order to deal with complex systems composed of sensors, actuators, signal conditioning and processing circuits. Moreover reconfigurability, and expandability, are key design features to get a flexible and reusable architecture. The FPGA implementation of a control system for an existing LOAC is presented as a case study with emphasis on the advantages of a programmable device approach. The presented system has been implemented on an Altera EPF10K200S device and the achieved operating frequency is 20MHz. The device was successfully tested and experimental results are hereby shown.
暂无评论