Juraj Gazda

Receiver technique for iterative estimation and cancellation of nonlinear distortion in MIMO SFBC-OFDM systems

Multiple-input multiple-output (MIMO) techniques and orthogonal frequency-division multiplexing (OFDM) can be combined to obtain a promising candidate for next generation wireless communications. However, like a single- input single output (SISO) systems one main disadvantage of MIMO-OFDM is high sensitivity to nonlinear distortion that results in significant loss in power efficiency and performance. In order to improve overall system performance this paper introduces a new iterative method for the detection of nonlinearly distorted symbols received by the receiver of the space-frequency block coded (SFBC) MIMO-OFDM transmission system. The proposed approach is based on the Bussgang theorem application and consists in iterative estimation and accurate cancellation of the nonlinear distortion due to the high power amplifier of the transmitter. The numerical performance evaluation shows that the proposed scheme can achieve significant performance improvement compared to conventional receivers.

Receiver based compensation of nonlinear distortion in MIMO-OFDM

Multiple antenna orthogonal frequency division multiplex (MIMO-OFDM) is transmission technique being employed in the future software defined/cognitive radio solutions. The main disadvantage of using MIMO-OFDM is its high sensitivity to the nonlinear amplification due to the specific characteristic of the high power amplifier (HPA) in the transmitter front-end. This phenomena results in the large overall performance degradation in terms of the bit error rate (BER). In this contribution, we present an iterative receiver scheme to reduce the BER of MIMO-OFDM systems undergoing nonlinear distortion. The technique consist in the iterative estimation and compensation of the nonlinear distortion introduced by transmitters HPA. As it will be shown by means of the numerical performance evaluation, the proposed technique reduces significantly BER of MIMO-OFDM systems.

An experimental comparison of feature selection methods on two-class biomedical datasets

Feature selection is a significant part of many machine learning applications dealing with small-sample and high-dimensional data. Choosing the most important features is an essential step for knowledge discovery in many areas of biomedical informatics. The increased popularity of feature selection methods and their frequent utilisation raise challenging new questions about the interpretability and stability of feature selection techniques. In this study, we compared the behaviour of ten state-of-the-art filter methods for feature selection in terms of their stability, similarity, and influence on prediction performance. All of the experiments were conducted on eight two-class datasets from biomedical areas. While entropy-based feature selection appears to be the most stable, the feature selection techniques yielding the highest prediction performance are minimum redundance maximum relevance method and feature selection based on Bhattacharyya distance. In general, univariate feature selection techniques perform similarly to or even better than more complex multivariate feature selection techniques with high-dimensional datasets. However, with more complex and smaller datasets multivariate methods slightly outperform univariate techniques.

Tax optimization in an agent-based model of real-time spectrum secondary market

The wireless communication industry is an essential sector boosting economic progress worldwide. The structure of the legacy wireless communication market, characterised by static licensing schemes, is moving towards real-time secondary spectrum markets. While the technological body of spectrum trading has been discussed in detail, from an economic perspectives there are still a lot of gaps in understanding how these transactions affect the economy of future communication standards. A challenging aspect of the real-time spectrum market deployment is the implementation of the appropriate tax system that impacts the market structure. With regards to this, we aim to build an agent-based model of the real-time secondary spectrum market in which various taxes including value-added tax, corporate tax, consumption tax and fixed tax, are employed. The relations between selected tax type rates and the hypothetical revenue of the national regulator is established using Laffer curves. The results of the analysis confirm the existence of a tax distortion, i.e. a system deviation from the efficient system functioning affected by the tax introduction. To measure the complexity of the tax strategies and the emergent tax distortion, an original approach based on Euclidean metrics defined over a vector space of the system performance indicators was proposed. This approach was later applied in parallel with the traditional Harberger’s triangle methodology. We found that the constrained optimisation with the tax distortion restrictions provide satisfactory results regarding the stability of the tax distortion measure. Therefore, we propose the application of the most effective corporate tax optimisation complemented by selected additional tax types.

Comparative evaluation of OFDMA and SC-FDMA based transmission systems

New evolving standards of the cellular systems, like e.g. Long Term Evolution (LTE), WiMax and Advanced LTE, consider Orthogonal Frequency Division Multiplex Access (OFDMA) as a mature and suitable solution to cope with the Inter Symbol Interference (ISI) due to the multi-path propagation. However, a major drawback of OFDMA which largely limited its application in the real environment is large envelope fluctuation which results in strong nonlinear distortion due to the nonlinear characteristic of power amplifier (PA). One possible solution to tackle this problem has been introduced by the the 3rd Generation Partnership Project (3GPP) consortium. This solution is based on spreading the base-band modulated signal before the application of OFDMA by using Discrete Fourier Transformation (DFT) that leads eventually to lower envelope fluctuation in comparison with that of OFDMA. The presented method is widely recognized as Single Frequency Division Multiplex (SC-FDMA). This paper aims to provide general comparison of both conventional OFDMA and SC-FDMA and draws important conclusions from this comparison.

M-APSK modulation for SC-FDMA communication systems

This paper investigates the application of Amplitude Phase Shift Keying (APSK) modulation in the nonlin-early distorted Long Term Evolution (LTE) uplink. APSK is recognized as spectrally efficient baseband modulation scheme, which is already adopted in several standards for digital video broadcasting, including e.g. Digital Video Broadcasting Satellite-Second Generation (DVB-S2). However, its application in the LTE standard has not been studied yet. Therefore, the motivation of this paper is to elaborate on the APSK performance in the LTE uplink and to provide the comparison with already standardized QAM modulation scheme. As the numerical results presented in this paper indicate, the applicaton of APSK modulation scheme is beneficial specially in highly nonlinearly distorted scenarios.

An Agent-Based Economy Model of Real-Time Secondary Market for the Cognitive Radio Networks

The evolution of the cognitive radio network theory often presents economics-based methods and free market principles in the administration of network functioning. We present here, an application of the agent-based model in the case of an economy consisting of five primary users and secondary users (SUs) uniformly distributed in the linear region. We show that the price-based demand evolution and price competition play a significant role only in an “economizable” network loading band, while in both the low-loading and the over-loading state, the market rules fail. On the contrary, in the economizable loading state, the price competition, demand sensitivity and specific SUs’ locations work well, underpinning the known economic rules.

MC-CDMA performance analysis for different spreading codes at HPA Saleh model

Multi-carrier code division multiple access (MC- CDMA) is a powerfull modulation technique that is being considered in many emerging broadband communication systems. However the high peak to average power ratio (PAPR) of the MC-CDMA signal, a special drawback of multicarrier transmission, has limited its wider aplication due to inefficient use of power amplifiers. PAPR strongly depends on particular spreading sequences considered for transmission. Therefore, the performance analysis of uplink MC-CDMA for different spreading sequences and for different receivers in the presence of Saleh model of high power amplifier (HPA) is presented. The results of our analysies show that the aplication of Golay codes in conduction with micro statistic multi-user receiver (MSF-MUD) provide significant better result as the other investigated spreading sequences and receivers.

Agent-based modeling of the cooperative spectrum management with insurance in cognitive radio networks

We propose and numerically analyze an agent-based simulation model of the spectrum frequency trading mechanism, where the heterogeneous agents take on the role of primary users. The interactions with the demand of the secondary users are considered. The model is constructed on the basis of Bak-Sneppen model of coevolution where the extremal dynamics is used to activate the low profitable users. Here, the strategies of the primary users are coevolving. They are characterized by the spectrum prices and cooperation intensity levels. The primary users interact indirectly by means of the demand stimulation of the secondary users and an insurance pool, which is provided by the spectrum exchange management system. The existence of the insurance pool is motivated by the needs of avoidance of the financial losses. The simulation results indicate the reliability of the insurance mechanism. In addition, several notable phenomena have emerged from the interactions of agents. The price increase resulting from the spontaneously formed oligopolistic practices of agents is considered as the most emergent feature of the model.

Iterative algorithm for nonlinear noise cancellation and channel re-estimation in nonlinearly distorted OFDM system

This paper presets a new algorithm for nonlinear distortion cancellation and channel re-estimation in an iterative manner. Whereas in practical OFDM systems, in which nonlinear amplifiers are used, data as well as pilot symbols are nonlinearly distorted. As perfect channel state information is often assumed in many works and papers, the effect of nonlinear distortion on pilot symbols is often not accounted for, and only data symbols are considered. However, as pilot symbols are used to obtain channel state information, distortion inflicted to pilot symbols by nonlinear amplification is of great importance, and for this reason should be accounted for. To address this issue, simple algorithm will be introduced for joint data recovery and channel re-estimation in an iterative manner. This algorithm iteratively estimates nonlinear distortion and then uses it to compensate this nonlinear distortion present at data and pilot symbols, which are then used for subsequent and more precise channel re-estimation. As it is shown in this paper, proposed algorithm can be used to reduce nonlinear distortions even in a situation, when highly nonlinear amplifiers are applied.