Cheolwoo You

Nonlinear blind equalization schemes using complex-valued multilayer feedforward neural networks

Among the useful blind equalization algorithms, stochastic-gradient iterative equalization schemes are based on minimizing a nonconvex and nonlinear cost function. However, as they use a linear FIR filter with a convex decision region, their residual estimation error is high. In this paper, four nonlinear blind equalization schemes that employ a complex-valued multilayer perceptron instead of the linear filter are proposed and their learning algorithms are derived. After the important properties that a suitable complex-valued activation function must possess are discussed, a new complex-valued activation function is developed for the proposed schemes to deal with QAM signals of any constellation sizes. It has been further proven that by the nonlinear transformation of the proposed function, the correlation coefficient between the real and imaginary parts of input data decreases when they are jointly Gaussian random variables. Last, the effectiveness of the proposed schemes is verified in terms of initial convergence speed and MSE in the steady state. In particular, even without carrier phase tracking procedure, the proposed schemes correct an arbitrary phase rotation caused by channel distortion.

Performance of multi-relay collaborative hybrid-ARQ protocols over fading channels

According to collaborative hybrid automatic-repeat-request (ARQ) protocols, relay stations that have been able to decode the original message from previous transmissions, collaborate with the source in future retransmissions by jointly sending a space-time codeword. In this letter, analysis of the average number of retransmissions and throughput of collaborative hybrid-ARQ Type I (i.e., without memory) and chase combining (i.e., with memory) protocols is provided for any number of relays.

Multicarrier CDMA systems using time-domain and frequency-domain spreading codes

For wideband code-division multiple-access systems, the paper introduces a multicarrier modulation scheme that performs the spreading simultaneously in the time and frequency domains. This scheme attains higher flexibility and spectrum efficiency because system parameters can be selected at will. The performance is compared with that of a single carrier RAKE system by calculating the probability of error over a frequency-selective Rayleigh fading channel. The proposed scheme outperforms the single carrier RAKE system if the system parameters are selected properly for given conditions, such as bandwidth and delay spread.

A decision feedback recurrent neural equalizer as an infinite impulse response filter

An adaptive decision feedback recurrent neural equalizer (DFRNE), which models a kind of an IIR structure, is proposed. Its performance is compared with the traditional linear and nonlinear equalizers with FIR structures for various communication channels. The small size and high performance of the DFRNE makes it suitable for high-speed channel equalization.

Performance Of Neural Equalizers On Partial Erasure Model

The increase in the capacity of the digital magnetic recording systems inevitably causes severe intersymbol interference (ISI) and nonlinear distortions in the magnetic channel. In this paper, to cope with severe ISI and nonlinear distortions a neural decision feedback equalizer (NDFE) is applied to the digital magnetic recording channel-partial erasure model. In the performance comparison between the NDFE and the conventional decision feedback equalizer (DFE) via simulations, it has been found that as nonlinear distortions increase the NDFE has more SNR (Signal to Noise Ratio) advantage over the conventional DFE.

Adaptive equalization using the complex backpropagation algorithm

For decreasing intersymbol interference (ISI) due to band-limited channels in digital communication, the uses of equalization techniques are necessary. Among adaptive equalization techniques, because of their ease of implementation and nonlinear capabilities, the neural networks have been used as an alternative for effectively dealing with the channel distortion, especially the nonlinear distortion. The complex backpropagation (BP) neural networks are proposed as nonlinear adaptive equalizers that can deal with both QAM and PSK signals of any constellation size (e.g. 32-QAM, 64-QAM and MPSK), and the complex BP algorithm for the new node activation functions having multi-output values and multi-saturation regions is presented. We also show that the proposed complex BPN provides, compared with the linear equalizer using the least mean squares (LMS) algorithm, an interesting improvement concerning bit error rate (BER) when channel distortions are nonlinear.

The neural decision feedback equalizer for the nonlinear digital magnetic recording systems

The increase of the capacity of the digital magnetic recording systems inevitably causes severe intersymbol interference (ISI) and nonlinear distortions in the magnetic channel. Thus to achieve large capacity magnetic recording systems it is necessary to compensate these problems. In order to cope with the severe ISI and nonlinear distortions a neural decision feedback equalizer (NDFE) is applied to the digital magnetic recording channel. In the performance comparison between the NDFE and the conventional decision feedback equalizer (DFE) via simulations, it is found that as nonlinear distortions increase the NDFE has a greater SNR (signal to noise ratio) advantage over the DFE.

A practical transmit beamforming strategy for closed-loop MIMO communication

A new beamforming strategy is proposed for multiuser systems with N transmit antennas at the transmitter and M ⩽ N single antenna receivers. The proposed scheme remarkably improves on the classical spatial division multiple access, and achieves the same data rates as spatial multiplexing for all users but with significantly superior performance/diversity gain. When compared with the Bell labs layered space–time system, the symbol rate is the same and the performance is much superior because of the presence of diversity gain. In addition, unlike the Bell labs layered space–time system, the receivers do not need to know each others vector channels. Finally, the proposed algorithm is based on dirty-paper coding, but does not require much complexity and is implementable. Copyright

A quadratic sigmoid neural equalizer for nonlinear digital magnetic recording channels

A new neural equalizer is proposed in order to compensate for intersymbol interference and to mitigate nonlinear distortions in digital magnetic recording systems. The proposed equalizer uses the quadratic sigmoid function as the activation function. The performance of the proposed equalizer is compared to those of a decision-feedback equalizer (DFE) and a neural decision feedback equalizer (NDFE) in terms of bit-error rate in nonlinear digital magnetic recording channels. Simulation results demonstrate that the proposed equalizer outperforms both DFE and NDFE.

Intercell Interference Coordination Using Threshold-Based Region Decisions

IMT-Advanced mobile communication systems make it possible for any devices to access high-speed networks anytime and anywhere. To meet the needs of IMT-Advanced systems, cellular systems must solve the problem of intercell interference caused by frequency reuse. Intercell interference problems become severe when orthogonal frequency division multiplexing (OFDM) transmission, which is a key technology for 4G communication systems, is used in a cellular system. In this paper, a zone-based intercell interference coordination (ICIC) scheme with high flexibility and low cost is proposed, and its performance is evaluated through multicell system-level simulations carried out according to the simplified 3GPP (3rd Generation Partnership Project) Long Term Evolution (LTE) system parameters. In the proposed algorithm, each cell is divided into several regions based on threshold values. Each region reuses frequencies in different ways, and the regions have different maximum transmit (TX) powers according to the interference environment. Even though the proposed scheme can be implemented with low complexity by using only the existing user equipment (UE) measurement, simulation results have confirmed that it provides significant improvements in geometry distribution.