Arthur J. Redfern

Receiver window design for multicarrier communication systems

Multicarrier modulation formats have been chosen for some wireline and wireless communication systems. These systems use the Fourier transform to convert a convolutive channel into multiple memoryless subchannels, which allows for simple equalization at the receiver. However, because the received signal is windowed in time, noise is spread between subchannels. As a result, narrowband interference sources can degrade the performance over a large number of subchannels. Receiver windowing has been proposed as a computationally efficient technique for reducing noise spreading in multicarrier communication systems. We propose a receiver window design algorithm which minimizes the noise power of the demodulated multicarrier signal. Simulations demonstrate the effectiveness of the windows on a variety of different channels and noise sources.

Compressive Sensing for Sparse Touch Detection on Capacitive Touch Screens

Capacitive touch screens are ubiquitous in todays electronic devices. Improved touch screen responsiveness and resolution can be achieved at the expense of the touch screen controller analog hardware complexity and power consumption. This paper proposes an alternative compressive sensing based approach to exploit the sparsity of simultaneous touches with respect to the number of sensor nodes to achieve similar levels of responsiveness. It is possible to reduce the analog data acquisition complexity at the cost of extra digital computations with less total power consumption. Using compressive sensing, in order to resolve the positions of the sparse touches, the number of measurements required is related to the number of touches rather than the number of nodes. Detailed measurement circuits and methodologies are presented along with the corresponding reconstruction algorithm.

A dual-path TEQ structure for DMT-ADSL systems

Discrete Multitone (DMT) modulation is the industry standard for Asymmetrical Digital Subscriber Loop (ADSL) modems. DMT modulation allows for simplified equalization when the effective duration of the channel memory is shorter than the predefined length of the cyclic prefix (CP). A single time-domain equalizer (TEQ) is typically used to truncate the channel memory such that the data rate loss due to the use of the CP is reduced. Because the single TEQ is optimized over all sub channels, it is difficult to find an ideal solution in which the number of bits supported on each sub channel is optimized. This paper proposes a dual-path TEQ, which adds a second TEQ that is biased to selected sub channels to increase transmission throughput. The implementation complexity is lower than other band-partition equalization methods such as per tone equalization.

Multicarrier modulation with data dependent frequency domain redundancy

Discrete multitone (DMT) modulation uses a time domain cyclic prefix to allow for simplified equalization at the receiver. It is shown that data dependent frequency domain redundancy can be added at the transmitter which allows standard DMT demodulation at the receiver without requiring a cyclic prefix. Benefits include the ability to add redundancy to subchannels with small capacities such that the resulting data rate loss is minimized with respect to methods which use a cyclic prefix, and that the prefix length can be dynamically adjusted for different channel conditions while maintaining a fixed frame and transform size.

Blind volterra system linearization with applications to post compensation of ADC nonlinearities

Blind Volterra system linearization is a challenging problem with a variety of applications where nonlinearities are present and training signals are not. This paper focuses on digital post compensation of nonlinear analog-to-digital converters (ADCs) and proposes a blind linearization approach based on the Volterra series. Assuming that the input signal does not occupy the full Nyquist bandwidth, i.e., there exists an energy-free band, the Volterra series coefficients are estimated by minimizing the linearized signal power in the energy-free band. Simulation results are presented which demonstrate the effectiveness of the proposed method.

Sparse touch sensing for capacitive touch screens

Improved capacitive touch screen responsiveness can be achieved at the expense of the touch screen controller analog hardware complexity and power consumption. This paper proposes a compressive sensing based approach to exploit the sparsity of simultaneous touches (e.g., 10 or less per person) with respect to the number of sensor nodes (e.g., 100s) to achieve similar levels of responsiveness with lower levels of analog complexity and power consumption. This is done by showing that the number of measurements required for touch detection is related to the number of touches rather than the number of nodes.

Using delta-sigma modulators in visible light OFDM systems

Visible light communications (VLC) are motivated by the radio-frequency (RF) spectrum crunch and fast-growing solid-state lighting technology. VLC relies on white light emitting diodes (LEDs) to provide communication and illumination simultaneously. Simple two-level on-off keying (OOK) and pulse-position modulation (PPM) are supported in IEEE standard due to their compatibility with existing constant current LED drivers, but their low spectral efficiency have limited the achievable data rates of VLC. Orthogonal frequency division multiplexing (OFDM) has been applied to VLC due to its high spectral efficiency and ability to combat inter-symbol-interference (ISI). However, VLC-OFDM inherits the disadvantage of high peak-to-average power ratio (PAPR) from RF-OFDM. Besides, the continuous magnitude of OFDM signals requires complicated mixed-signal digital-to-analog converter (DAC) and modification of LED drivers. We propose the use of delta-sigma modulators in visible light OFDM systems to convert continuous magnitude OFDM symbols into LED driver signals. The proposed system has the communication theory advantages of OFDM along with the practical analog and optical advantages of simple two level driver signals. Simulation results are provided to illustrate the proposed system.

Digital background calibration for pipelined ADCs

Pipelined ADCs with more than 12 bits of resolution typically require linearity enhancement. This paper shows that the gain errors of both the sub DACs and interstage amplifiers in a pipelined ADC can be digitally corrected. A background calibration method is developed employing an energy free methodology for both single and multi stage errors and simulation results are presented to demonstrate the effectiveness of the proposed method.

A DSL customer-premise equipment modem SoC with extended reach/rate for broadband bridging and routing

A highly-integrated DSL modem for residential gateway applications is described. The chip integrates a C62x based DSL PHY, AFE, line driver and receiver, power management, and broadband controller subsystem. A 0.13/spl mu/m 5M CMOS process is used to implement the 2.3W chip. Supplies are 1.5V for digital and 3.3V for analog subsystems.

Quantization noise shaping for information maximizing ADCs

ADCs sit at the interface of the analog and digital worlds and fundamentally determine what information is available in the digital domain for processing. This paper shows that a configurable ADC can be designed for signals with non constant information as a function of frequency such that within a fixed power budget the ADC maximizes the information in the converted signal by frequency shaping the quantization noise. Quantization noise shaping can be realized via loop filter design for a single channel delta sigma ADC and extended to common time and frequency interleaved multi channel structures. Results are presented for example wireline and wireless style channels.