John Devlin

Adaptive linear prediction for lossless coding of greyscale images

We present a new prediction method for lossless image coding. In this prediction scheme, the prediction for each pixel is formed by adaptively combining the predicted values from a set of least squares based linear predictors. The combination scheme follows the main idea of the Bayesian model averaging to reduce prediction error due to model uncertainty. A lossless image coding algorithm based on this prediction method is also presented. Experimental results show that the compression performance of this algorithm is better than that of the TMW 0.51. It is also close to that of a newly improved version of TMW.

Toward Refocused Optical Mouse Sensors for Outdoor Optical Flow Odometry

This paper investigates the use of refocused optical mouse sensors for odometry in the field of outdoor robotic navigation. Optical mouse sensors like the ADNS-2610 are small, inexpensive, non-contact devices, which integrate a complementary metal-oxide semiconductor camera and DSP hardware to provide 2-D optical displacement measurements. Current research indicates that vertical height variance contributes as a dominant cause of systematic error to horizontal displacement measurements, which raises significant problems for irregular environments encountered in outdoor robotic navigation. In this paper, we propose two approaches to mitigate this systematic error induced by height variance. The efficacy and robustness of the proposed approaches are tested by experimentation on an asphalt concrete road surface and by simulation.

The Tasman international geospace environment radar (TIGER) - current development and future plans

The Tasman international geospace environment radar is a dual HF radar system with overlapping footprints designed to map ionospheric motions by detecting ionospheric scatter. The first radar was set up on Bruny Island, Tasmania at the end of 1999 and development of the second radar to be placed near Invercargill, NZ, has begun. TIGER is part of the super dual auroral radar network (SuperDARN) which currently consists of 15 radars deployed in the northern and southern hemispheres. TIGER is located more equatorward than other SuperDARN radars, enabling it to observe new phenomena, such as auroral westward flow channels (AWFCs). This paper describes TIGERs capabilities and presents examples of observations, including an AWFC. Plans to develop digital transmitters and receivers are discussed as is a proposal to extend the network to even lower latitudes by deploying two additional radars.

FPGA-Based Implementation of Multiple Modes in Near Field Inductive Communication Using Frequency Splitting and MIMO Configuration

Conventional near field inductive wireless power transfer theory shows that systems suffer from splitting frequency behaviors when strong coupling condition exists between the transmitter and the receiver. However, this characteristic has not been explored for communication. Our analysis demonstrates that the splitting behaviour of frequency creates multiple frequencies that support inductive communication in MIMO configuration. As a result, we implement a binary chirp modulation on an FPGA and validate two channel communication using splitting. This paper introduces the use of chirp signals to spread data and excite inductive MIMO systems. The simulation and experiment show that the splitting frequency depends on a quality factor and the flux coupling condition between the data source and receiver. In other words, the degree of mutual coupling defines the splitting mode. This paper proves that multi-channel communication using splitting can be used for data transmission. The results show that data rates of 50 Mbps or 69 Kbps can be achieved for each channel between the transmitters and receivers when the transmitter and receiver operate at the original resonant frequency of 13.56 MHz or 28 KHz, respectively and the distance between them varies from about 1 cm to 10 cm.

Associated geomagnetic and ionospheric variations

Abstract An investigation has been conducted into the association between geomagnetic micropulsation activity and ionospheric disturbances. It has been found that both correlated oscillations and associated irregularities may be observed at various times.

Least squares approach for lossless image coding

The least squares (LS) approach-a powerful tool for estimation is incorporated into a predictive lossless image compression algorithm and is proven very effective in reducing the prediction error. Based on the observation that sometimes LS has also limitations, a scheme which adaptively combines the LS-based prediction with a simpler prediction is proposed. Experiments show that the proposed algorithm can further improve the performance of the LS approach. The compression performance of the proposed technique is the same as that of TMW, the best published technique for lossless image coding. However, in reaching this, this technique requires much less computation than TMW.

Determination of ionospheric parameters in real time using SuperDARN HF Radars

A technique for determining key ionospheric parameters from high-frequency (HF) over-the-horizon radar ground scatter data is investigated using two Southern Hemisphere SuperDARN radars and also a Northern Hemisphere SuperDARN radar with reliable elevation angle-of-arrival capability. Ground scatter data are analyzed over a range of frequencies from 8 to 18 MHz to determine the maximum usable frequency and the vertical critical frequency over a wide geographical area within the radar field of view. The technique is shown to be well suited to middle to high latitudes where backscatter echoes from the ground dominate over those from ionospheric scattering targets. However, the technique is shown to break down during the winter months and away from solar maximum. It is shown that the use of reliable elevation angles can greatly enhance such methods allowing discrimination between ground scatter propagating via the E and F regions. It is also shown that contamination from very low velocity ionospheric scatter and ground scatter originating from the back lobe of the radar can be effectively filtered out, with the use of reliable elevation angles. This greatly improves the reliability of the ionospheric data products and allows for a high degree of automation of the process.

Multipoint visualization of ULF oscillations using the Super Dual Auroral Radar Network

We present a technique for the automatic detection of Pc5 (150 s to 600 s period) ultralow frequency (ULF) pulsations in ground and ionospheric backscatter from the Super Dual Auroral Radar Network (SuperDARN). This new technique enables rapid identification and visualization of ULF oscillations over the very wide geographical coverage of SuperDARN. The technique detects ULF oscillations using the Lomb-Scargle periodogram method, which provides a natural test for periodic behavior against the null hypothesis of a pure noise distribution. This does not require any interpolation across data gaps and is thus appropriate for SuperDARN data. We demonstrate the application of the technique to SuperDARN data for March 2014 and find that Pc5 pulsations are frequently observed by multiple radars simultaneously. A preliminary investigation using data from all Northern Hemisphere SuperDARN radars indicates that Pc5 pulsation activity occurs most often on the nightside of the magnetosphere, with a mean frequency of about 2 mHz.

Implementation of the MFCC front-end for low-cost speech recognition systems

Speech recognition front-end implemented using a high-end floating-point processor is expensive both in terms of computer resources and cost. This paper presents a new small footprint MFCC front-end design that is suitable for low-cost speech recognition systems. By exploiting the overlapping nature of the input frames and by adopting a simple pipeline structure, the implemented design only utilizes approximately 10% total resources of a low-cost and modest-size FPGA device, thus leaving significant space for speech recognition post-processing.

Small footprint implementation of dual-microphone delay-and-sum beamforming for in-car speech enhancement

For effective speech processing in an automotive environment, speech enhancement is necessary due to significant levels of background noise. In this paper, we present a cost effective small footprint implementation of one particular speech enhancement technique: dual microphone delay-and-sum beamforming. In order to save resources, the implementation utilizes the overlapping frame property used in speech processing systems. The implementation also exhibits a simple interconnection structure leading to even greater resource saving. Experiment results show that the proposed design can produce enhanced output very close to that generated by a theoretical (floating-point) model while only requiring a modest hardware resource usage.