In mobile communications, channel side information at transmitters can increasecapacity. For moving relays nodes, local nodes placed on buses and trams in urbanareas, the channel state information is outdated for control delays of severalmilliseconds, as in the LTE system. Prediction of the channel based on statistic is notadequate for vehicular velocities. In this thesis, prediction made with an additionalantenna, a ``predictor antenna", placed in front of the main antenna is evaluated. Thepredictor utilises that the channel of the predictor antenna is highly correlated to thechannel experienced by the main antenna somewhat later, when the main antenna hasmoved to the position previous occupied by the predictor antenna. A normalisedcorrelation of up to 0.98 could be measured between the channels of the antennasfor an antenna separation of several wavelengths, but it was found that the closeenvironment and the antenna pattern have a big impact on the correlation. Thepredictions made with the antenna are also combined with predictions based onstatistics of past measurements from the main antenna to see if a better result can beachieved. For a prediction range of 0.5 carrier wavelengths, a prediction as good as anormalised mean square error (NMSE) of -13.9 dB could be seen. This is sufficient togive a gain in the performance when using link adaptation and opportunistic multi-userscheduling, based on channel state information at transmitter. The evaluations isbased on measurements on a 20 MHz downlink channel at 2.68 GHz.

One of the currently explored strategies for interference avoidance and improving Signal to Noise Ratio (SNR) for mobile communication systems is Coordinated MultiPoint (CoMP) transmission. The general idea of the strategy is to let two or more base stations serve the same user. Due to delay factors, the channels from each serving base station needs to be predicted to obtain an adaptive CoMP system. In this thesis, a user interface is created to act as an experimental platform for a set of measured downlink channel data. The user interface supports editing of the channel data, model estimation, Kalman filtering and prediction and evaluation of the channel statistics. The user interface and the measured channel downlink data is then used to examine how well we can predict the weakest channel in a CoMP setup with three base stations. The predictions are carried out using an m-step Kalman predictor which uses an AR4 model, estimated from previous channel data. For the investigation, the user moves at pedestrian speed and the signals from the three different base stations use orthogonal Common Reference Signals (CRS). A comparison of different CRS patterns is also included in the investigation. It is concluded that 5 ms predictions of the weakest channel achieves a normalized mean squared error (NMSE) of -8 dB or lower provided that the weakest signal has an SNR of at least 5 dB and is no more than 15 dB lower than the combined received signal. Additionally, it is found that predictions are more accurate for CRS patterns spread over time than over subcarriers.

<p>In an optimal power control scheme for a Code Division Multiple Access (CDMA) system all mobile stations signals should arrive to the base station at equal power. If not, stronger singals may cause too much interference and block out weaker ones. Commonly used power control schemes utilizes the Signal to Interference Ratio (SIR) to design a Power Control Command (PCC) to adjust the transmit power of the mobile station. A significant problem within the conventional methods is the slow SIR recovery due to deep channel fades. Conventional methods base the PCC on the previous channel state when in fact, the channel state may have significantly changed when transmission occurs. These channel changes may cause the SIR to drop or rise drastically and lead to uncontrollable Multi Access Interference (MAI) resulting in power escalation and making the system unstable. In order to overcome power escalation and improve the recovery from deep fades a novel power control method has been developed. Based on Linear Quadratic Control and Kalman filtering for channel prediction this method designs the PCC based on the coming channel state instead of the current. This optimizes the PCC for the channel state where transmission occurs. Simulations show that this control scheme outperforms previous methods by making the impacts of the deep fades less severe on the SIR and also improves the overall SIR behaviour.</p>

In an optimal power control scheme for a Code Division Multiple Access (CDMA) system all mobile stations signals should arrive to the base station at equal power. If not, stronger singals may cause too much interference and block out weaker ones. Commonly used power control schemes utilizes the Signal to Interference Ratio (SIR) to design a Power Control Command (PCC) to adjust the transmit power of the mobile station. A significant problem within the conventional methods is the slow SIR recovery due to deep channel fades. Conventional methods base the PCC on the previous channel state when in fact, the channel state may have significantly changed when transmission occurs. These channel changes may cause the SIR to drop or rise drastically and lead to uncontrollable Multi Access Interference (MAI) resulting in power escalation and making the system unstable. In order to overcome power escalation and improve the recovery from deep fades a novel power control method has been developed. Based on Linear Quadratic Control and Kalman filtering for channel prediction this method designs the PCC based on the coming channel state instead of the current. This optimizes the PCC for the channel state where transmission occurs. Simulations show that this control scheme outperforms previous methods by making the impacts of the deep fades less severe on the SIR and also improves the overall SIR behaviour.

In this thesis, the neural network, a powerful tool which has demonstrated its ability in many fields, is studied for the indoor localization system and channel prediction system. This thesis first proposes a received signal strength indicator (RSSI) fingerprinting-based indoor positioning system for the widely deployed WiFi environment, using deep neural networks (DNN). To reduce the computing time as well as improve the estimation accuracy, a two-step scheme is designed, employing a classification network for clustering and several regression networks for final location prediction. A new fingerprinting, which utilizes the similarity in RSSI readings of the nearby reference points (RPs) is also proposed. Real-time tests demonstrate that the proposed algorithm achieves an average distance error of 43.5 inches. Then this thesis extends the ability of the neural network to the physical layer communications by introducing a recurrent neural network (RNN) based approach for real-time channel prediction which uses the recent history channel state information (CSI) estimation for online training before prediction, to adapt to the continuously changing channel to gain a more accurate CSI prediction compared to the other conventional methods. Furthermore, the proposed method needs no additional knowledge, neither the internal properties of the channel itself nor the external features that affect the channel propagation. The proposed approach outperforms the other methods in a changing environment in the simulation test, validating it a promising method for channel prediction in wireless communications. / Graduate

The field of wireless communication is under massive development and the demands on the cellular system, especially, are constantly increasing as the utilizing devices are increasing in number and diversity. A key component of wireless communication is the knowledge of the channel, i.e, how the signal is affected when sent over the wireless medium. Channel prediction is one concept which can improve current techniques or enable new ones in order to increase the performance of the cellular system. Firstly, this report will investigate the concept of a predictor antenna on new, extensive measurements which represent many different environments and scenarios. A predictor antenna is a separate antenna that is placed in front of the main antenna on the roof of a vehicle. The predictor antenna could enable good channel prediction for high velocity vehicles. The measurements show to be too noisy to be used directly in the predictor antenna concept but show potential if the measurements can be noise-filtered without distorting the signal. The use of low-pass filter and Kalman filter to do this, did not give the desired results but the technique to do this should be further investigated. Secondly, a interpolation technique will be presented which utilizes predictions with different prediction horizon by estimating intermediate channel components using interpolation. This could save channel feedback resources as well as give a better robustness to bad channel predictions by letting fresh, local, channel predictions be used as quality reference of the interpolated channel estimates. For a linear interpolation between 8-step and 18-step Kalman predictions with Normalized Mean Square Error (NMSE) of -15.02 dB and -10.88 dB, the interpolated estimates had an average NMSE of -13.14 dB, while lowering the required feedback data by about 80 %. The use of a warning algorithm reduced the NMSE by a further 0.2 dB. It mainly eliminated the largest prediction error which otherwise could lead to retransmission, which is not desired.

ITC/USA 2008 Conference Proceedings / The Forty-Fourth Annual International Telemetering Conference and Technical Exhibition / October 27-30, 2008 / Town and Country Resort & Convention Center, San Diego, California / Adaptive modulation is a communication technique capable of maximizing throughput while guaranteeing a fixed symbol error rate (SER). However, this technique requires instantaneous channel state information at the transmitter. This can be obtained by predicting channel states at the receiver and feeding them back to the transmitter. Existing algorithms used to predict single-input single-output (SISO) channels with recurrent neural networks (RNN) are extended to multiple-input multiple-output (MIMO) channels for use with adaptive modulation and their performance is demonstrated in several examples.

En els sistemes de comunicacions on el transmissor disposa de certa informació sobre l'estat del canal (CSI), es possible dissenyar esquemes lineals de precodificació que assignin la potència de manera òptima induint guanys considerables, sigui en termes de capacitat, sigui en termes de la fiabilitat de l'enllaç de comunicacions. A la pràctica, aquest coneixement del canal mai és perfecte i, per tant, el senyal transmès es veurà degradat degut al desajust entre la informació que el transmissor disposi del canal i el seu estat real.En aquest context, aquesta tesi estudia dos problemes diferents però alhora estretament relacionats: el disseny d'un esquema pràctic de seguiment del canal en transmissió per canals variants en temps, i el disseny d'esquemes lineals de precodificació que siguin robustos a la incertesa del canal. La primera part de la tesi proposa el disseny d'un esquema de seguiment de canal que, mitjançant un enllaç de retorn de baixa capacitat, proporcioni al transmissor una informació acurada sobre el seu estat. Històricament, aquest tipus d'esquemes han rebut fortes crítiques degut a la gran quantitat d'informació que és necessari transmetre des del receptor cap el transmissor. Aquesta tesi, doncs, posa especial èmfasi en el disseny d'aquest canal de retorn. La solució que es proposa, basada en el filtre de Kalman, utilitza un esquema que recorda al transmissor DPCM. Les variacions del canal són tractades mitjançant dos predictors lineals idèntics situats en el transmissor i en el receptor, i un canal de retorn que assisteix el transmissor amb l'error de predicció. L'interès d'aquest esquema diferencial és que permet seguir les variacions del canal amb només dos o quatre bits per coeficient complex, fins i tot en canals ràpidament variants.La resta de la tesi cobreix el segon objectiu, l'estudi de diferents esquemes d'assignació de potències quan el coneixement del canal en transmissió no és perfecte. El problema es planteja per a un sistema MIMO OFDM com a formulació més general, incloent els casos d'una sola antena, de l'esquema beamforming i del canal multiplicatiu com a casos particulars.Primerament s'ha plantejat l'optimització dels criteris de mínim error quadràtic mig (MMSE) i mínima BER sense codificar. La innovació en el treball presentat a la tesi, respecte a altres treballs que segueixen els mateixos criteris de disseny, ha estat la formulació Bayesiana del problema per al disseny dels algoritmes robustos.La tesi continua amb el plantejament d'estratègies robustes d'assignació de potència destinades a minimitzar la BER codificada. Per aquesta tasca s'han utilitzat criteris de teoria de la informació. Possiblement una de les principals contribucions d'aquesta tesi ha estat el plantejament del cut-off rate com a paràmetre de disseny. Aquest criteri s'introdueix com alternativa a la capacitat de canal o a la informació mutual per al disseny del transmissor quan s'inclou codificació de canal. La ultima part de la tesi proposa un interleaver adaptatiu de baixa complexitat que, utilitzant el coneixement del canal disponible en el transmissor, assigna estratègicament els bits no només per combatre les ràfegues d'errors, sinó també per lluitar contra els esvaïments que puguin presentar les diferents portadores del canal per a una realització concreta. El disseny d'aquest interleaver, anomenat "interleaver RCPC" està basat en els codis Rate-Compatible Punctured Convolutional Codes. Com s'il·lustra a partir del resultats numèrics, l'ús d'aquest interleaver millora les prestacions dels algoritmes quan es comparen amb les que s'obtindrien si s'utilitzes un interleaver de bloc o un interleaver pseudo-aleatori. / When the transmitter of a communication system disposes of some Channel State Information (CSI), it is possible to design linear precoders that optimally allocate the power inducing high gains either in terms of capacity or in terms of reliable communications. In practical scenarios, this channel knowledge is not perfect and thus the transmitted signal suffers from the mismatch between the CSI at the transmitter and the real channel.In that context, this thesis deals with two different, but related, topics: the design of a feasible transmitter channel tracker for time varying channels, and the design of optimal linear precoders robust to imperfect channel estimates.The first part of the thesis proposes the design of a channel tracker that provides an accurate CSI at the transmitter by means of a low capacity feedback link. Historically, those schemes have been criticized because of the large amount of information to be transmitted from the receiver to the transmitter. This thesis focuses, thus, the attention in an accurate design of the return link. The proposed solution is based on the Kalman filter and follows a scheme that reminds the well known DPCM transmitter. The channel variability is processed by two identical linear predictors located at the transmitter and at the receiver, and a feedback link that assists the transmitter with the prediction error. The interest of this differential scheme is that allows to track the channel variations with only two or four bits per complex channel coefficient even in fast time-varying channels.The rest of the thesis covers the second topic, studying different robust power allocation algorithms when the CSI is not perfectly known at the transmitter. For the sake of generality, the problem is formulated for the general MIMO OFDM case, encompassing the single antenna transmission, the beamforming schemes and the frequency-flat fading channels as particular cases. First, the minimum MSE and the minimum uncoded BER parameters are chosen to be optimized, evaluating the performance of the algorithms in terms of uncoded BER. The basic novelty with respect to previous works that considers the same strategies of design is the proposal of a Bayesian approach for the design of the robust algorithms.Next the study is extended by proposing robust power allocation strategies focused on the minimization of the coded BER. For this purpose, information-theoretic criteria are used. Probably, one of the main contributions in the thesis is the proposal of the cut-off rate as a parameter of design whose maximization is directly related to the coded BER. This criterion is introduced as an alternative to the channel capacity and the mutual information for the design of optimal transceivers in the presence of any channel coding stage. The last part of the thesis proposes a low complexity adaptive interleaver that, making use of the CSI available at the transmitter, reallocates the bits not only to combat the bursty channel errors but also to combat the specific distribution of the faded subcarriers as a function of the channel response. The design of this interleaver, named as "RCPC interleaver", is based on the Rate-Compatible Punctured Convolutional Codes. As shown by numerical results, the use of this interleaver improves the performance of the algorithms when they are compared with the classical block interleavers and pseudo-random interleavers.

<p>Prediction of the rapidly fading envelope of a mobile radio channel enables a number of capacity improving techniques like fast resource allocation and fast link adaptation. This thesis deals with linear prediction of the complex impulse response of a channel and unbiased quadratic prediction of the power. The design and performance of these predictors depend heavily on the correlation properties of the channel. Models for a channelwhere the multipath is caused by clusters of scatterers are studied. The correlation for the contribution from a cluster can be approximated as a damped complex sinusoid. A suitable model for the dynamics of the channel is an ARMA-process. This motivates the use of linear predictors.</p><p>A limiting factor in the prediction are the estimation errors on the observed channels. This estimation error, caused by measurement noise and time variation, is analyzed for a block based least squares algorithm which operates on a Jakes channel model. Efficient noise reduction on the estimated channel impulse responses can be obtained with Wienersmoothers that are based on simple models for the dynamics of the channel combined with estimates of the variance of the estimation error.</p><p>Power prediction that is based on the squared magnitude of linear prediction of the taps will be biased. Hence, a bias compensated power predictor is proposed and the optimal prediction coefficients are derived for the Rayleigh fading channel. The corresponding probability density functions for the predicted power are also derived. A performance evaluation of the prediction algorithm is carried out on measured broadband mobile radio channels. The performance is highly dependent on the variance of the estimation error and the dynamics of the individual taps.</p>

Prediction of the rapidly fading envelope of a mobile radio channel enables a number of capacity improving techniques like fast resource allocation and fast link adaptation. This thesis deals with linear prediction of the complex impulse response of a channel and unbiased quadratic prediction of the power. The design and performance of these predictors depend heavily on the correlation properties of the channel. Models for a channelwhere the multipath is caused by clusters of scatterers are studied. The correlation for the contribution from a cluster can be approximated as a damped complex sinusoid. A suitable model for the dynamics of the channel is an ARMA-process. This motivates the use of linear predictors. A limiting factor in the prediction are the estimation errors on the observed channels. This estimation error, caused by measurement noise and time variation, is analyzed for a block based least squares algorithm which operates on a Jakes channel model. Efficient noise reduction on the estimated channel impulse responses can be obtained with Wienersmoothers that are based on simple models for the dynamics of the channel combined with estimates of the variance of the estimation error. Power prediction that is based on the squared magnitude of linear prediction of the taps will be biased. Hence, a bias compensated power predictor is proposed and the optimal prediction coefficients are derived for the Rayleigh fading channel. The corresponding probability density functions for the predicted power are also derived. A performance evaluation of the prediction algorithm is carried out on measured broadband mobile radio channels. The performance is highly dependent on the variance of the estimation error and the dynamics of the individual taps.