Y Yan Wu

Performance comparison for illumination and visible light communication system using buck converters

This paper presents a new analysis of the illumination and communication performance for visible light communication (VLC) with variable pulse position modulation (VPPM). In particular, it addresses the extra power losses due to the data modulation if a traditional buck converter LED driver is adapted for VLC. We compare VLC systems using two previously proposed driver schemes in the aspects of efficiency, illumination power, data rate and BER.

Experimental Evaluation of an Adaptive Nonlinear Interference Suppressor for Multimode Transceivers

In multimode transceivers, the transmitter for one communication standard may induce a strong interference in the receiver for another standard. Using linear filtering techniques to suppress this interference requires a receiver with a very large dynamic range, leading to an excessive power consumption. A much more power efficient approach suppresses the interference using an adaptive nonlinear interference suppressor (NIS). In previous work an ideal model was used to derive an adaptation method and study the receiver performance afforded by the NIS. In this paper, we present experimental results of a receiver that uses an implementation of the NIS, fabricated in 140 nm complementary metal-oxide-semiconductor technology. Main imperfections that limit the NIS performance are identified, simple models are developed that explain the experimental results, and for the key imperfections, low-complexity digital compensation and calibration methods are proposed. These digital methods permit the use of lower-performance analogue circuits, thus further reducing the transceiver cost and power consumption. The experimental results show that the NIS can achieve a substantial interference suppression at attractive complexity and power dissipation.

System study of a 60 GHz wireless-powered monolithic sensor system

State-of-the-art batteryless wireless sensors have separate modules for sensing and energy scavenging. Such separation increases the size and cost of sensors and limits their robustness. To overcome these limitations, we propose a 60 GHz wireless sensor system, which we call the PREMISS system, based on monolithic sensors with on chip sensing, tranceiving, integrated antenna and energy scavenging. In the PREMISS system, a high-power central controller transmits RF energy and information to many low-power low-cost sensors via pencil beams and receive and detect the information from these sensors. In this paper, we present a system study on the PREMISS system highlighting design challenges and practical implementation considerations. From the link budget calculation, we show that in the PREMISS system, a signal to noise ratio of 17 dB at the central controller receiver may be achieved from a sensor 5 meters away. In addition, we also identified that one key challenges in the PREMISS system is the design of good efficiency rectifiers for input power and voltage levels far below those in state-of-the-art 2.4 GHz systems.

Effect of buck driver ripple on BER performance in visible light communication using LED

This paper analyses the communication performance for visible light communication (VLC) with Manchester encoded amplitude modulation. In particular, it considers the ripple generated by the LED driver as an important noise contribution for VLC. The ripple depends on the oscillation frequency of the converter which typically is chosen to satisfy other performance criteria such as power conversion efficiency. This paper models the ripple as an additional noise if a buck converter LED driver is adapted for VLC. We argue that ripple can as a worst case be modeled as a random binary offset in the bit detection that randomly affects the distance between the data signal and the decision threshold. We derive expressions for the BER, propose approximations and we compare these with a simulation.

Epileptic seizure detection on patients with mental retardation based on EEG features : a pilot study

Mental retardation (MR) is one of the most common secondary disabilities in people with Epilepsy. However, to our knowledge there are no reliable seizure detection methods specified for MR-patients. In this paper we performed a pilot study on a group of six patients with mental retardation to assess what EEG features potentially work well on this group. A group of EEG features on the time, frequency and spatio-temporal domain were extracted, the modified wrapper approach was then employed as an improved feature subset selection method. Results show high variance on obtained features subset across this group, meanwhile there exist some common features which characterize the high-frequency components of epileptic EEG signals.

A mm-wave analog adaptive array with genetic algorithm for interference mitigation

In the future dense 60 GHz wireless environments, co-channel interference (CCI) is a potential problem that degrades the link quality. In this paper, a genetic algorithm (GA) assisted analog adaptive array is proposed for 60 GHz indoor applications to mitigate the CCI. The GA optimizes the weights on each receiver path to attenuate CCI from different directions and improve signal-to-interference-plus-noise ratio (SINR) with robustness. We make two further improvements for better spatial re-use. First, we take into account the trade-off between array gain at the desired direction and interference suppression by proposing adaptivity on the weights control in GA. By adaptively increasing phase perturbation, near-optimum SINR can be achieved for different interference situations. Secondly, in some cases, the link requires less receiver paths than available, e.g. for line-of-sight links. So instead of using fixed antennas, we further propose antenna selection in the GA. A further improvement on spatial re-use can be achieved compared to arrays with fixed uniform spacing.

Seizure detection using dynamic warping for patients with intellectual disability

Electroencephalography (EEG) is paramount for both retrospective analysis and real-time monitoring of epileptic seizures. Studies have shown that EEG-based seizure detection is very difficult for a specific epileptic population with intellectual disability due to the cerebral development disorders. In this work, a seizure detection method based on dynamic warping (DW) is proposed for patients with intellectual disability. It uses an EEG template of an individual subjects dominant seizure type, to extract the morphological features from EEG signals. A linear discriminant analysis (LDA) classifier is used to perform the seizure detection. Results show that the DW-based feature in the frequency domain is superior than that in the time domain, and the features extracted using wavelet transform method.

Modeling of RF energy scavenging for batteryless wireless sensors with low input power

RF energy scavenging enables batteryless operation of wireless sensors. In particular, a system with a central controller that transfers wireless energy to and exchanges information with RF energy scavenging sensors is very suitable for a wide range of applications. State-of-the-art analysis of RF energy scavenging is mostly based on RF-DC rectifier models operating with relatively high input power to achieve high rectification efficiency. However, to enable larger distance between the central controller and sensors and/or to increase the operating frequencies, which can lead to small and low-cost smart dust like sensors, a good model describing the RF-DC rectification with low input power is needed to aid system design and optimization. In this paper, we develop such a model. Using the model, we derive closed-form solutions for the equilibrium voltage and the input resistance of the rectifier. We further propose a quasi-static model to describe the dynamic charging of the capacitor in the rectifier. A comparison with circuit simulations using Cadence Virtuoso Spectre circuit simulator shows good match between our model and the circuit simulation.

Performance of barcode scanner using peak detection with interference from LED lamps

This paper presents a new model for barcode scanners that suffer from the interference of LED lamps, to quantify the system performance in terms of Timing Signal-to-Interference Ratio (TSIR), particularly as a function of modulation depth and frequency of the flicker in the LED lighting. Physical characteristics for laser scanner channels are presented, including laser beam path and LED interference path. The barcode and LED interference are described in detail based on the scanning process. For the commonly used peak detection algorithm in the scanner, we derive expressions for the TSIR and we present numerical results for a typical system.

Analysis of an Adaptive Nonlinear Interference Suppressor for Wireless Multimode Transceivers

In multimode transceivers, the transmitter for one communication standard may induce strong interference in the receiver for another standard, often exceeding the desired signal by many tens of decibels. To linearly suppress this interference, the receiver requires a very large linear dynamic range, resulting in excessive power consumption. In a recent paper, a nonlinear block, which requires an adaptation signal proportional to the envelope of the received interference, has been used to strongly suppress the interference. In that work, the required adaptation signal for the nonlinear block has been determined analytically. In this paper, we quantify the required accuracy for the adaptation signal to properly suppress the interference while keeping the degradation to the receiver symbol error rate (SER) negligible. To provide the required accuracy, we propose a closed-loop method that calculates the adaptation signal based on a model, which describes the received interference in terms of the locally available baseband interference. We propose a method to adapt this model during the operation of the transceiver such that the power of the residual interference at the output of the nonlinear block is minimized. Our analysis shows that the proposed method can strongly suppress the interference while a SER close to that of an exactly linear receiver is achieved. Simulation results for a practical scenario validate this analysis.