Jussi Poikonen

Error Models for the Transport Stream Packet Channel in the DVB-H Link Layer

Digital video broadcasting for hand held terminals (DVB-H) is a broadcast system designed for high-speed data transmission in highly dispersive mobile channel conditions. In this paper, methods of reproducing the statistical properties of measured DVB-H packet error traces are presented. Statistical and finite-state modeling approaches are found to be suitable for simulating the error performance of a DVB-H system operating in typical urban channel conditions. Evaluation of these models focuses on the accuracy of the models in replicating the high-order statistical properties of measured DVB-H transport stream error traces. Also, the effect of these error statistics on the DVB-H link layer frame error rate is considered.

Validation of a DVB-H dynamic system simulator using field measurements

In this paper we discuss the need for dynamic system-level simulations for DVB-H (Digital Video Broadcast - Handheld), specifically to evaluate the overall system performance perceived by mobile users dynamically over time. Such simulations can be used as a complement of traditional radio coverage planning tools for analyzing quality of service and radio resource management aspects of the DVB-H network. We describe a general simulation structure and the main models required, and validate, using vehicular urban field measurements, a DVB-H physical layer performance model that enables these analyses.

Coarse time- and frequency synchronization applied in a mobile WiMAX system

In this paper, we present coarse synchronization algoritms for OFDM systems, specifically applied with mobile WiMAX transmissions. We consider computationally efficient synchronization algorithms, which are applicable also without full knowledge of the properties of the transmitted signal at the receiver. Although the proposed algorithms are considered for a mobile WiMAX system, their application is not limited to the specific WiMAX signal structure. Indeed, they can be used on any OFDM system with certain type of training sequence. Approaches are considered both for time and for frequency-domain synchronization. In the time-domain case, the proposed algorithm improves the accuracy of a previously studied coarse synchronization approach without significant increase in computational complexity. Furthermore, the considered frequency-domain synchronization method provides computationally efficient and accurate results without explicit prior information on the transmitted signal contents, also in the presence of severe phase error caused by timing inaccuracy.

A chaotic memristor circuit

We propose and analyze a circuit comprising of three memristors, three switches and a driver. The circuit computes the logistic map and thus acts as an example of a simple memristive circuit with chaotic behavior. For the analysis of the circuit we also describe how analog computation can be performed with certain type of voltage controlled memristors. Simulation results validating the theory are presented.

A finite-state simulation model for OFDM over frequency-selective fast fading channels

In this paper, a computationally efficient, algorithmically simple simulation model for transmission of orthogonal frequency division multiplexing (OFDM) signals over time-variant, frequency-selective wireless propagation channels is proposed. The considered approach is based on applying a set of finite-state models with parameters derived explicitly from propagation channel and transmission system characteristics. The validity of the proposed model is studied using theoretical capacity analysis, and by comparing results with simulations performed using a reference propagation model. The obtained results indicate good performance with all considered criteria.

Increasing the convergence rate of distributed beamforming

In this paper we discuss techniques for Distributed Beamforming (DBF). We specify a generic communication scenario in which a DBF scheme should function to be of practical significance. We identify need for improvement in recently proposed DBF schemes in order to operate reliably in realistic scenarios, and propose a new algorithm which incorporates such improvements. We show by simulations that our new algorithm significantly outperforms the previous approaches in terms of speef of convergence and reliability of operation.

Distributed beamforming in OFDM-based systems over frequency-selective channels

Earlier publications on distributed beamforming have discussed only single carrier systems. Because orthogonal frequency division multiplexing has several advantages over single-carrier modulation in wideband systems, a logical direction of research is to investigate how such algorithm performs with OFDM. In this paper, we present a modified and further optimized version of a recently proposed distributed beamforming algorithm and incorporate it to a system using OFDM. The functionality of the algorithm is verified in simulations using realistic, frequency-selective channels. Based on extensive simulation studies, we recommend specific parameter values for the beamforming algorithm with the aims of maximizing the speed of convergence and asymptotic received power, and minimizing the amount of feedback required.

Analysis of channel estimation error for OFDM reception over severely time-dispersive channels

The accuracy of equalization in the demodulation of orthogonal frequency division multiplexing signals is directly dependent on the density of pilot carriers in the signal and the delay dispersion of the propagation channel. In this paper, the error introduced by an insufficient pilot density with respect to the channel delay dispersion is theoretically analyzed. An expression is derived for the average channel estimation error given an OFDM pilot density and channel power delay profile. Based on this expression, the effect of channel estimation error on the perceived signal-to-noise ratio in the receiver demodulation is approximated. Results of the theoretical analysis and approximation are compared to simulations of OFDM transmitted over time-dispersive channels.

A CNN approach to computing arbitrary Boolean functions

In this paper, a novel approach to computing arbitrary Boolean functions using a binary-state cellular neural/nonlinear/nanoscale network (CNN) architecture with local static memory is presented. We define explicitly how to map a given Boolean function and its input values to the cells of a specific type of binary CNN, and the global rules used to perform parallel calculations. Each of the computation steps can be performed asynchronously. Additionally, the total CNN area is readily split into subsections, each of which perform individual computations of different Boolean functions. The main benefits of our approach are simple implementation of arbitrary Boolean functions, built-in parallelism both in local and global scale of the computation and the possibility for asynchronous operation.

Effects of the number of discrete delay components on the accuracy of wireless channel simulation

In this paper we investigate the effect of the number and delays of discrete channel impulse response components on the accuracy of wireless system simulations. The subject is relevant in considering large systems such as wireless networks, where realistic results are required, but it is also necessary to apply models that are simple enough to enable practically realizable simulation times. We present basic criteria for estimating the accuracy of discrete channel profiles, and use these to examine both analytical channel models, and models derived directly from channel measurements. We use average simulation performance, frequency response correlation, and theoretical capacity as performance measures for evaluating the effect of varying the number of discrete model components used.