An Error Free Passive UHF RFID System using a New Form of Wireless Signal Distribution

A wide area and error free ultra high frequency (UHF) radio frequency identification (RFID) interrogation system based on the use of multiple antennas used in cooperation to provide high quality ubiquitous coverage, is presented. The system uses an intelligent distributed antenna system (DAS) whereby two or more spatially separated transmit and receive antenna pairs are used to allow greatly improved multiple tag identification performance over wide areas. The system is shown to increase the read accuracy of 115 passive UHF RFID tags to 100% from <60% over a 10m x 8m open plan office area. The returned signal strength of the tag backscatter signals is also increased by an average of 10dB and 17dB over an area of 10m x 8m and 10m x 4m respectively. Furthermore, it is shown that the DAS RFID system has improved immunity to tag orientation. Finally, the new system is also shown to increase the tag read speed/rate of a population of tags compared with a conventional RFID system.

Mixed Wireless Communications and RFID Sensor Networks over Fibre

A second harmonic suppression scheme allowing RoF links to support communications and passive UHF RFID is reviewed. Using RoF distributed antenna system techniques, the coverage and location accuracy of passive UHF RFID are significantly improved.

Passive UHF RFID Interrogation System Using Wireless RFID Repeater Nodes

This paper presents a new wireless radio frequency identification (RFID) repeater system, facilitating remote interrogation without the need for arrays of wired antennas, despite using entirely passive, low-cost ultra high frequency (UHF) RFID tags. The proposed system comprises a master RFID reader with both transmit and receive functions, and multiple RFID repeaters to receive, amplify and retransmit tag-to-reader and reader-to-tag communications. This expands the area over which the master RFID reader may provide coverage for a given maximum transmit power at each antenna. We first demonstrate a single hop wireless repeater system to allow similar read performance to a standard commercial passive UHF RFID reader. Finally, a proof of principle system demonstrates that a single wireless repeater node can allow an extension in range.

Evaluation of Setup Procedures on Mobile Devices Based on Users’ Initial Experience

Users’ initial perceptions of their competence are key motivational factors for further use. However, initial tasks on a mobile operating system (OS) require setup procedures, which are currently largely inconsistent, do not provide users with clear, visible and immediate feedback on their actions, and require significant adjustment time for first-time users. This paper reports on a study with ten users, carried out to better understand how both prior experience and initial interaction with two touchscreen mobile interfaces (Apple iOS and Google Android) affected setup task performance and motivation. The results show that the reactions to setup on mobile interfaces appear to be partially dependent on which device was experienced first. Initial experience with lower-complexity devices improves performance on higher-complexity devices, but not vice versa. Based on these results, the paper proposes six guidelines for designers to design more intuitive and motivating user interfaces (UI) for setup procedures. The preliminary results indicate that these guidelines can contribute to the design of more inclusive mobile platforms and further work to validate these findings is proposed.

Structure Discovery in Nonparametric Regression through Compositional Kernel Search

Despite its importance, choosing the structural form of the kernel in nonparametric regression remains a black art. We define a space of kernel structures which are built compositionally by adding and multiplying a small number of base kernels. We present a method for searching over this space of structures which mirrors the scientific discovery process. The learned structures can often decompose functions into interpretable components and enable long-range extrapolation on time-series datasets. Our structure search method outperforms many widely used kernels and kernel combination methods on a variety of prediction tasks.

Kernel conditional quantile estimation via reduction revisited

Quantile regression refers to the process of estimating the quantiles of a conditional distribution and has many important applications within econometrics and data mining, among other domains. In this paper, we show how to estimate these conditional quantile functions within a Bayes risk minimization framework using a Gaussian process prior. The resulting non-parametric probabilistic model is easy to implement and allows non-crossing quantile functions to be enforced. Moreover, it can directly be used in combination with tools and extensions of standard Gaussian Processes such as principled hyperparameter estimation, sparsification, and quantile regression with input-dependent noise rates. No existing approach enjoys all of these desirable properties. Experiments on benchmark datasets show that our method is competitive with state-of-the-art approaches. © 2009 IEEE.

Holographic beam steering a directly modulated twoelectrode high brightness tapered laser diode for optical wireless communications

An error-free free space communication link with 3 angular coverage and 1.25GHz modulation bandwidth is demonstrated by beam steering an ultra high modulation efficiency bright tapered laser diode using a Liquid Crystal Spatial Light Modulator. © OSA 2012.

Thin film coil design considerations for wireless power transfer in flat panel display

Wireless power transfer is experimentally demonstrated by transmission between an AC power transmitter and receiver, both realised using thin film technology. The transmitter and receiver thin film coils are chosen to be identical in order to promote resonant coupling. Planar spiral coils are used because of the ease of fabrication and to reduce the metal layer thickness. The energy transfer efficiency as a function of transfer distance is analysed along with a comparison between the theoretical and the experimental results. © 2012 Materials Research Society.

Hybrid microstrip and carbon nanotubes based patch antenna for wireless applications

In this we have looked at the concept of introducing carbon nanotubes on the surfaces of the microstrip patch antennas. We examined the performance improvements in a patch antenna through finite difference time domain simulations to increase the efficiency of the antenna. The results suggest that carbon nanotubes lead to a higher gain due to their electrical properties. A high gain antenna with low power requirements resulted in achieving a higher overall bandwidth. The designed antenna's gain, bandwidth and directivity are analyzed before and after introducing carbon nanotubes. © 2013 IEEE.

Determinantal Clustering Processes - A Nonparametric Bayesian Approach to Kernel Based Semi-Supervised Clustering

Semi-supervised clustering is the task of clustering data points into clusters where only a fraction of the points are labelled. The true number of clusters in the data is often unknown and most models require this parameter as an input. Dirichlet process mixture models are appealing as they can infer the number of clusters from the data. However, these models do not deal with high dimensional data well and can encounter difficulties in inference. We present a novel nonparameteric Bayesian kernel based method to cluster data points without the need to prespecify the number of clusters or to model complicated densities from which data points are assumed to be generated from. The key insight is to use determinants of submatrices of a kernel matrix as a measure of how close together a set of points are. We explore some theoretical properties of the model and derive a natural Gibbs based algorithm with MCMC hyperparameter learning. The model is implemented on a variety of synthetic and real world data sets.

Using drift nets to capture early life stages and monitor spawning of the Yangtze River Chinese sturgeon (Acipenser sinensis)

P>A sampling system for capturing sturgeon eggs using a D-shaped bottom anchored drift net was used to capture early life stages (ELS) of Chinese sturgeon, Acipenser sinensis, and monitor annual spawning success at Yichang on the Yangtze River, 1996-2004, before and just after the Three Gorges Dam began operation. Captured were 96 875 ELS (early life stages: eggs, yolk-sac larvae = eleuthero embryos, and larvae); most were eggs and only 2477 were yolk-sac larvae. Most ELS were captured in the main river channel and inside the bend at the Yichang spawning reach. Yolk-sac larvae were captured for a maximum of 3 days after hatching began, indicating quick dispersal downstream. The back-calculated day of egg fertilization over the eight years indicated a maximum spawning window of 23 days (20 October-10 November). Spawning in all years was restricted temporally, occurred mostly at night and during one or two spawning periods, each lasting several days. The brief temporal spawning window may reduce egg predation by opportunistic predators by flooding the river bottom with millions of eggs. During 1996-2002, the percentage of fertilized eggs in an annual 20-egg sample was between 63.5 to 94.1%; however, in 2003 the percentage fertilized was only 23.8%. This sudden decline may be related to the altered environmental conditions at Yichang caused by operation of the Three Gorges Dam. Further studies are needed to monitor spawning and changes in egg fertilization in this threatened population.

The Random Forest Kernel and other kernels for big data from random partitions

We present Random Partition Kernels, a new class of kernels derived by demonstrating a natural connection between random partitions of objects and kernels between those objects. We show how the construction can be used to create kernels from methods that would not normally be viewed as random partitions, such as Random Forest. To demonstrate the potential of this method, we propose two new kernels, the Random Forest Kernel and the Fast Cluster Kernel, and show that these kernels consistently outperform standard kernels on problems involving real-world datasets. Finally, we show how the form of these kernels lend themselves to a natural approximation that is appropriate for certain big data problems, allowing $O(N)$ inference in methods such as Gaussian Processes, Support Vector Machines and Kernel PCA.