Showing papers by "Moises V. Ribeiro published in 2010"

Linear and periodically time-varying PLC channels estimation in the presence of impulsive noise

Abstract: This work aims to analyze and discuss the performance of channel estimation techniques based on OFDM pilot signals, applied to data transmission over PLC (power line communication) channels. PLC channel is modeled as linear and periodically time-varying (LPTV), which presents an additive impulsive Gaussian-like noise (AIGN). Non-adaptive channel estimators, such as Least Square and Transform Domain associated with channel equalizers based on minimum mean squared error criterion and adaptive ones, such as LMS and RLS are investigated. Simulation results indicate interesting research directions to improve estimation techniques based on pilot signals when LPTV PLC channels are corrupted by AIGN, such as i) adaptive channel estimation techniques suffer considerable performance degradation due to periodically-stationary behavior of PLC channel, and ii) the periodically-stationarity of PLC channel can causes more performance degradation than the impulsive noise.

A discussion about the suitability of UWB modulation for outdoor power line communication

Abstract: The emergent technology known as ultra wideband (UWB) is addressed in this contribution. UWB pulses are used to transmit data through a low-voltage outdoor time-varying power line communications (PLC) channel corrupted by the presence of impulsive noises. The attained results reveal a deep relationship between the pulse design and the PLC channel attenuation profile. In addition, numerical results indicate that the detection of the transmitted symbol with a matched filter based on the pulse response of the PLC channel is a good strategy to improve the system performance, specially when pulse amplitude modulation (PAM) is the modulation scheme adopted.

Coded CP-SC communication scheme for outdoor power line communications

Abstract: In this paper, we consider a cyclic prefixed single-carrier communication scheme with frequency domain equalization (CP-SC) and low-density parity check code (LDPC) over power line communication (PLC) channel. The aim is to evaluate the error performance when the PLC channel is corrupted by impulsive noise. The simulation results show that the best performance is obtained when the LDPC code is considered in association with the frequency domain equalizer based on the MMSE criterion. Additionaly, comparative results between the CP-SC and SC-DFE are presented. The results indicate that CP-SC with FDE-MMSE offer better performance than SC-DFE. Moreover, we present a simulation result that indicates that if the channel decoder has correct information about the impulsive noise variance, then error performance improvements can be obtained.

HOS-based impulsive noise detection technique for power line communication systems

Abstract: This paper discusses a detection technique for impulsive noise in power line cables. Essentially, a reduced set of features (signature vector), which is selectively extracted from the power line signal feeds a detection technique. The features are higher-order statistics and the Fisher's discriminant ratio is the selection technique. The designed detectors are a multilayer perceptron neural network and a Bayes implemented according the maximum likelihood criterium. Simulation results indicate that improved performance can be attained if multilayer perceptron neural network is considered as a nonlinear detector. Also, the results reveal that higher-order statistics is a very interesting technique to extract a reduced and representative signature vector of impulsive noise.

Analysis of a hybrid OFDM synchronization algorithm for power line communication

Abstract: This paper outlines a hybrid OFDM synchronization technique for powerline communication. Initially, the timing and frequency offset are estimated by searching for the peak of an evaluated metric provided by a reference block transmitted at the beginning of each OFDM frame, and subsequently, autocorrelation properties induced at cyclic prefix on the time domain are exploited to maintain estimates of the timing and frequency offset without the use of a new reference block for synchronization purpose. This strategy is interesting because once a first synchronization of the first OFDM block (the start point of the frame) is accomplished, any loss in spectral efficiency will occur subsequently. Additionally, we propose an interesting strategy with low computational complexity to improve the timing and frequency estimates after synchronization is carried out at the beginning of the frame. Basically, this strategy consists of averaging the peaks spacing between successive peaks of the evaluated autocorrelation metric. Simulation results reveal good performance in PLC scenarios, specially in the cases of occurrences of impulsive noise.

Time-varying channel characterization based on SVM under the impulsive noise presence

Abstract: The purpose of this work is to present an off-line technique for characterization of time-varying power line communications channels in the presence of additive impulsive noise when they are used for broadband data transmission. We propose and analyze a channel estimation technique based on Support Vector Machine (SVM). Simulation results reveal that the proposed method is more robust to impulsive noise and has better convergence rate for time-varying channels when compared to classic techniques like least squares (LS).

A Low-Cost Implementation of High-Order Square M-QAM Detection/Demodulation in a FPGA Device

Abstract: This paper outlines a low-cost technique for detection/demodulation of high-order squareM -ary quadrature amplitude modulation(M -QAM), M = 4, 16, 64, 256, 1024, and 4096, in afield programmable gate array(FPGA) device. Hence, it is very interesting for advancing high-speed power line communication technology. The proposed technique is based on a so-called heuristic decision region approach combined with a nonlinear function that allows to simplify the hardware implementation of detection and demodulation for a high-order square M -QAM symbols. The performance evaluation of the proposed technique is carried out when the channel isadditive white Gaussian noise(AWGN) and additive impulsive Gaussian noise(AIGN). The attained results, in terms of bit error rate vs. energy per bit to noise power spectral density ratio ( Eb/N0), indicate that the use of the proposed technique offers reduc ed performance losses. Also, its implementation in a FPGA devi ce demands the lowest hardware resource and latency. I. I NTRODUCTION Currently, the development of spectral efficient power line communication(PLC) systems for broadband application in lowand medium-voltage electric power grids demands the use of high-order digital modulation techniques, like M aryquadrature amplitude modulation (M -QAM), capable of transmitting a high amount of bits per symbol. Consequently , the process of detection in the demodulator, which represen ts a computational cost, must be carefully considered to desig n such systems. Although themaximum likelihood(ML) criterion is an optimum detection technique for additive white Gaussian noise (AWGN) channels presenting messages of equal probability, this technique requires large computational requirements to calculate the Euclidean distance between the received symbol and all other points of the QAM constellation [1]. One of the main challenges in modern digital communication systems is to reduce the computational complexity. In this regard, this paper outlines a low-cost detection/demodulation technique for high-order square M -QAM constellations. The proposed technique is based on the so-called heuristic decision region(HDR) approach that allows a considerable reduction of hardware resource in a FPGA device [2]. Additionally, it makes use of non-linear function that simpli fy the implementation of HDR approach with finite-precision (i.e., unsigned fixed-point). The performance of the proposed technique is analyzed when the signal is corrupted by the presence of background AWGN andadditive impulsive Gaussian noise (AIGN) [3]. The numerical results indicate that reduced performance losses are noted for M -QAM, M = 4, 16, 64, 256, 1024, and 4096. The FPGA implementation of the proposed technique is accomplished withardware description language (HDL). The analysis of hardware resource demand reveals that the proposed technique is the one with the lowest hardware resource demand and latency. This paper is organized as follows. In Section II, the problem formulation is arisen. HDR approach is shown in Section III. Section IV provides a description of the proposed detection strategies for high-order M -QAM used in the FPGA implementation. The simulation results and the computational burden analysis for FPGA-based implementation of the proposed strategies are shown in Section V, and finally, Section VI presents the conclusions of this paper. II. PROBLEM FORMULATION A modulation and demodulation scheme for digital communication can be depicted in Fig. 1. In this figure, functions f(.) and g(.) implement the modulation and detection/demodulation techniques, respectively. Also, the ch annel output, which is corrupted by additive noise, can be expressed by y = xi + v, (1) wherey is the channel output vector; xi, i = 0, 1, ..., M − 1, is the i-th constellation point associated with the binary messagemi, which lenght islog2(M), andv is the additive noise vector. For this contribution, we assume that xi can be a point in a squareM -QAM constellation, such that M = 2, n = 0, 2, 4, 6, 8, 10, 12. Additionally, a Gray code similar to the one employed in IEEE 802.16 standard [4] is considered. Fig. 2 presents the Gray-mapped constellation for 16-QAM, in which the least significant bit is b0 and the The 7th International Telecommunications Symposium (ITS 2010)