Browsing by Department "Department of Electrical and Communications Engineering"

Filter results by typing the first few letters
Now showing 1 - 20 of 387
  • Adaptive methods for blind equalization and signal separation in MIMO systems
    (2002-08-23) Enescu, Mihai
     Doctoral dissertation (article-based)
    This thesis addresses the problems of blind source separation (BSS) and blind and semi-blind communications channel equalization. In blind source separation, signals from multiple sources arrive simultaneously at a sensor array, so that each sensor output contains a mixture of source signals. Sets of sensor outputs are processed to recover the source signals from the mixed observations. The term blind refers to the fact that specific source signal values and accurate parameter values of a mixing model are not known a priori. Application domains for the material in this thesis include communications, biomedical, and sensor array signal processing. The goal of this thesis is development of blind and semi-blind algorithms which require little or no prior information about source signal or mixing system parameter values in order to process the data. We start with the problem of extracting unknown input signals from measured outputs of instantaneous multiple-input multiple-output (I-MIMO) systems with constant parameter values. Suggested solutions are then extended to time-varying I-MIMO systems and also to constant finite impulse response multiple-input multiple-output (FIR-MIMO) systems. Another goal is to find a practical solution for the more challenging case of time-varying FIR-MIMO systems. The source separation techniques proposed in this thesis are based on state-space models and on recursive estimation. Blind separation algorithms based on Kalman filters are proposed. The source signals are treated using low-order autoregressive models. Projections along signal subspace eigenvectors are used to reduce the dimensionality of observations and also for spatial decorrelation of sources. Any changes that occur in the signal subspace can be tracked online. When considering slowly time-varying FIR-MIMO systems, fractional sampling can be used to derive a set of slowly time-varying I-MIMO systems. Thus, the proposed recursive BSS algorithms for I-MIMO systems can be used for blind equalization of slowly time-varying FIR communications channels. The problem of equalization of time-varying FIR MIMO systems is also addressed in this thesis. The proposed solutions involve semi-blind algorithms which work in two stages. First, a channel estimate is derived, and then the observation sequence is equalized. The algorithms estimate the otherwise-unknown noise statistics, and as a result achieve performance close to that of an optimal Kalman-based algorithm. A non-connected decision feedback equalization algorithm is derived for FIR-MIMO systems, using a minimum mean square error criterion. Simulation results show that the algorithm is able to track time and frequency selective channels and also to mitigate intersymbol and interuser interference.
  • Adaptive methods for score function modeling in blind source separation
    (2002-08-26) Karvanen, Juha
     Doctoral dissertation (article-based)
    In signal processing and related fields, multichannel measurements are often encountered. Depending on the application, for instance, multiple antennas, multiple microphones or multiple biomedical sensors are used for the data acquisition. Such systems can be described using Multiple-Input Multiple-Output (MIMO) system models. In many cases, several source signals are present at the same time and there is only limited knowledge of their properties and how they contribute to each sensor output. If the source signals and the physical system are unknown and only the sensor outputs are observed, the processing methods developed for recovering the original signals are called blind. In Blind Source Separation (BSS) the goal is to recover the source signals from the observed mixed signals (mixtures). Blindness means that neither the sources nor the mixing system is known. Separation can be based on the theoretically limiting but practically feasible assumption that the sources are statistically independent. This assumption connects BSS and Independent Component Analysis (ICA). The usage of mutual information as a measure of independence leads to iterative estimation of the score functions of the mixtures. The purpose of this thesis is to develop BSS methods that can adapt to different source distributions. Adaptation makes it possible to separate sources without knowing the source distributions or even the characteristics of source distributions. Special attention is paid to methods that allow also asymmetric source distributions. Asymmetric distributions occur in important applications such as communications and biomedical signal processing. Adaptive techniques are proposed for the modeling of score functions or estimating functions. Three approaches based on the Pearson system, the Extended Generalized Lambda Distribution (EGLD) and adaptively combined fixed estimating functions are proposed. The Pearson system and the EGLD are parametric families of distributions and they are used to model the distributions of the mixtures. The strength of these parametric families is that they contain a wide class of distributions, including asymmetric distributions with positive and negative kurtosis, while the estimation of the parameters is still a relatively simple procedure. The methods may be implemented using existing ICA algorithms. The reliable performance of the proposed methods is demonstrated in extensive simulations. In addition to symmetric source distributions, asymmetric distributions, such as Rayleigh and lognormal distribution, are utilized in simulations. The score adaptive methods outperform commonly used methods due to their ability to adapt to asymmetric distributions.
  • Adaptive power control in CDMA cellular communication systems
    (2005-11-18) Rintamäki, Matti
     Doctoral dissertation (monograph)
    Power control is an essential radio resource management method in CDMA cellular communication systems, where co-channel interference is the primary capacity-limiting factor. Power control aims to control the transmission power levels in such a way that acceptable quality of service for the users is guaranteed with lowest possible transmission powers. All users benefit from the minimized interference and the preserved signal qualities. In this thesis new closed loop power control algorithms for CDMA cellular communication systems are proposed. To cope with the random changes of the radio channel and interference, adaptive algorithms are considered that utilize ideas from self-tuning control systems. The inherent loop delay associated with closed loop power control can be included in the design process, and thus alleviated with the proposed methods. Another problem in closed-loop power control is that extensive control signaling consumes radio resources, and thus the control feedback bandwidth must be limited. A new approach to enhance the performance of closed-loop power control in limited-feedback-case is presented, and power control algorithms based on the new approach are proposed. The performances of the proposed algorithms are evaluated through both analysis and computer simulations, and compared with well-known algorithms from the literature. The results indicate that significant performance improvements are achievable with the proposed algorithms.
  • Advanced CMP processes for special substrates and for device manufacturing in MEMS applications
    (2006-10-20) Kulawski, Martin
     Doctoral dissertation (article-based)
    The present work reports on studies and process developments to utilize the chemical mechanical planarization (CMP) technology in the field of micro electrical mechanical systems (MEMS). Approaches have been undertaken to enable the manufacturing of thick film SOI (silicon-on-insulator) substrates with a high degree of flatness as well as utilizing CMP for the formation of several novel MEMS devices. Thick film SOI wafers are of high interest in MEMS manufacturing as they offer obvious benefits as a starting material or foundation for more complex structures. Precise control of the SOI layer thickness as well as the removal uniformity is of critical importance to fully utilize the benefits of this technology. By combining fixed abrasive (FA) pads for polishing and novel grinding techniques it is shown that major improvements can be achieved over the standard manufacturing sequence. Analysis of the material removal rate (MRR) dependency on several process parameters is made. Together with the FA pad vendor a suitable consumable set for SOI is generated, which shows long term stability in the generated process. A comparison with standard methods is undertaken to prove the surface and crystalline quality of the resulting substrate material is equivalent. Analysis is done to understand the microscopic mechanism of removal. The CMP process is applied to several MEMS structures to smooth deposited oxide films and to enable direct wafer bonding (DWB) at low temperatures. This allows the design of bonded multiple stack layers including heat sensitive materials such as metals. FA CMP is applied to large pattern MEMS for total planarization but also for smoothing of the surface of single protruding structures while minimizing edge rounding and preserving the original intended pattern shape. With dedicated CMP steps thick film polysilicon smoothing is demonstrated enabling DWB. The chemo-mechanical particularities of the FA pad are investigated in detail.
  • Advanced fiber components for optical networks
    (2004-05-27) Ylä-Jarkko, Kalle
     Doctoral dissertation (article-based)
    Due to the tremendous growth in data traffic and the rapid development in optical transmission technologies, the limits of the transmission capacity available with the conventional erbium-doped amplifiers (EDFA), optical filters and modulation techniques have nearly been reached. The objective of this thesis is to introduce new fiber-optic components to optical networks to cope with the future growth in traffic and also to bring down the size and cost of the transmission equipment. Improvements in performance and in scalability of the optical networks are studied through simulations and experimental network set-ups. High-power single-mode laser sources operating at 980 nm are important in pumping EDFAs and Raman amplifiers. In this thesis, two new practical, fiber-coupled configurations of stable high-power cladding-pumped Yb-doped fiber sources operating at 977 nm are presented: a fiber laser and an ASE (amplified spontaneous emission) or superfluorescent source. Sources are based on high numerical aperture Yb-doped jacketed air-clad fiber and high brightness pump diodes. L-band EDFAs are used to expand amplification bandwidth beyond the C-band wavelengths. Traditional L-band EDFAs are costly devices, which are core-pumped with expensive high-power single-mode diodes. Cladding-pumping technology brings down the cost of the pump diodes in L-band EDFAs, since high-power but low-cost multimode pump diodes can then be used. Additionally, the flexibility in designing erbium-doped fiber is improved. In this thesis, a new design for L-band EDFA based on GTWave cladding-pumping technology is introduced. Simultaneous noise reduction and transient suppression in the amplifier is achieved by using a gain-clamping seed-signal. To increase the spectral efficiency of the optical transmission systems optical filters having square spectral response and linear phase, leading to zero dispersion both in-band and out-of-band, are required. The application of inverse scattering technique in conjunction with advanced fiber Bragg grating writing technique significantly reduces in-band dispersion and greatly improves grating characteristics. In this thesis, the in-band and out-of-band dispersion penalty of a cascade of linear-phase fiber Bragg grating (FBG) filters is experimentally measured and compared to the results with conventional apodized FBG filters. Fiber Bragg grating based distributed feedback fiber lasers (DFB FL) are attractive alternatives to semiconductor lasers. Output power and efficiency of DFB FLs can be significantly increased by using a master-oscillator-and-power-amplifier (MOPA) configuration, consequently degrading optical signal to noise ratio (OSNR) and RIN of the master source. These trade-offs are studied in several MOPA configurations using core-pumped and cladding-pumped EDFAs as power amplifiers and compared to the results with a high-power stand-alone DFB-FLs, i.e. DFB FLs pumped with a high-power pump source. Finally, the performance and scalability of a bidirectional and a high-density metropolitan WDM ring networks is analyzed. Results show that the scalability limitation imposed by the amplified RIN arising from the Rayleigh backscattering in bidirectional WDM ring networks can be avoided by using low gain shared-pump EDFAs and directly modulated transmitters. In high-density metropolitan WDM networks based on non-zero dispersion shifted fibers the main limiting nonlinearity is four-wave mixing. In metropolitan areas distributed Raman amplification (DRA) is the most effective means reduce the effect of four-wave mixing.
  • Advanced receiver structures for mobile MIMO multicarrier communication systems
    (2006-04-06) Roman, Timo
     Doctoral dissertation (article-based)
    Beyond third generation (3G) and fourth generation (4G) wireless communication systems are targeting far higher data rates, spectral efficiency and mobility requirements than existing 3G networks. By using multiple antennas at the transmitter and the receiver, multiple-input multiple-output (MIMO) technology allows improving both the spectral efficiency (bits/s/Hz), the coverage, and link reliability of the system. Multicarrier modulation such as orthogonal frequency division multiplexing (OFDM) is a powerful technique to handle impairments specific to the wireless radio channel. The combination of multicarrier modulation together with MIMO signaling provides a feasible physical layer technology for future beyond 3G and fourth generation communication systems. The theoretical benefits of MIMO and multicarrier modulation may not be fully achieved because the wireless transmission channels are time and frequency selective. Also, high data rates call for a large bandwidth and high carrier frequencies. As a result, an important Doppler spread is likely to be experienced, leading to variations of the channel over very short period of time. At the same time, transceiver front-end imperfections, mobility and rich scattering environments cause frequency synchronization errors. Unlike their single-carrier counterparts, multi-carrier transmissions are extremely sensitive to carrier frequency offsets (CFO). Therefore, reliable channel estimation and frequency synchronization are necessary to obtain the benefits of MIMO OFDM in mobile systems. These two topics are the main research problems in this thesis. An algorithm for the joint estimation and tracking of channel and CFO parameters in MIMO OFDM is developed in this thesis. A specific state-space model is introduced for MIMO OFDM systems impaired by multiple carrier frequency offsets under time-frequency selective fading. In MIMO systems, multiple frequency offsets are justified by mobility, rich scattering environment and large angle spread, as well as potentially separate radio frequency - intermediate frequency chains. An extended Kalman filter stage tracks channel and CFO parameters. Tracking takes place in time domain, which ensures reduced computational complexity, robustness to estimation errors as well as low estimation variance in comparison to frequency domain processing. The thesis also addresses the problem of blind carrier frequency synchronization in OFDM. Blind techniques exploit statistical or structural properties of the OFDM modulation. Two novel approaches are proposed for blind fine CFO estimation. The first one aims at restoring the orthogonality of the OFDM transmission by exploiting the properties of the received signal covariance matrix. The second approach is a subspace algorithm exploiting the correlation of the channel frequency response among the subcarriers. Both methods achieve reliable estimation of the CFO regardless of multipath fading. The subspace algorithm needs extremely small sample support, which is a key feature in the face of time-selective channels. Finally, the Cramér-Rao (CRB) bound is established for the problem in order to assess the large sample performance of the proposed algorithms.
  • Advanced receivers for high data rate mobile communications
    (2006-12-08) Melvasalo, Maarit
     Doctoral dissertation (article-based)
    Improving the spectral efficiency is a key issue in the future wireless communication systems since the spectrum is a scarce resource. Both the number of users as well the demanded data rates are increasing all the time. Furthermore, in mobile communications the wireless link is required to be reliable even when the mobile is in a fast moving vehicle. Using Multiple-Input Multiple-Output (MIMO) antennas is a well known technique to provide higher spectral efficiency as well as better link reliability. Additionally, higher order modulation methods can be used to provide higher data rates. In order to benefit from these enhancements in practise, sophisticated signal processing methods as well as accurate estimates of time-varying wireless channel parameters are needed. This thesis addresses the problem of designing multi-antenna receivers in high data rate systems. The case of multiple transmit antennas is also considered. System specific features of High Speed Downlink Packet Access (HSDPA) which is part of 3rd generation (3G) Wideband Code Division Multiple Access (WCDMA) evolution are exploited in channel estimation methods and in MIMO receiver design. Additionally, complexity reduction methods for Minimum Mean Square Error (MMSE) equalization are addressed. Blind channel estimation methods are spectrally efficient, since no extra resources are needed for pilot signals. However, in mobile communications accurate estimates are needed also in fast fading channels. Consequently, semi-blind channel estimation methods where the receiver combines blind and pilot based channel estimation are an appealing alternative. In this thesis blind and semi-blind channel estimation methods based on knowledge of multiple spreading codes are derived. A novel semi-blind combining scheme for code multiplexed pilot signal and blind estimation is proposed. Another important factor in receiver design criteria is the structure of interference in the received signals. Interference mitigation techniques in MIMO systems have been shown to be potential methods for providing improved performance. A chip level inter-antenna interference cancellation method has been developed in this thesis for HSDPA. Furthermore, this multi-stage ordered interference canceler is combined with the semi-blind channel estimation scheme to enhance the system performance further.
  • Advanced spherical antenna measurements
    (2005-12-15) Laitinen, Tommi
     Doctoral dissertation (article-based)
    Concrete guidelines for effectively performing spherical antenna measurements and for designing multi-probe systems will be provided. The work will mainly be restricted to antennas whose maximum cross-section dimension is in the order of 1-2 λ or less. Specific design guidelines for a very fast radiation pattern measurement system for mobile phone models will be provided. Information on practical aspects related to such a system will be provided by building a demonstrator system and testing it. Firstly, the errors in the total radiated power and the maximum electric field are illustrated by simulations of near-zone spherical antenna measurements of electrically relatively small AUTs (antennas under test) for various applied truncation numbers and for different measurement distances [P1]. Secondly, a novel iterative matrix method is presented that is shown to provide, for a fixed relatively small number field samples, a lower uncertainty in the determination of the radiated field of an AUT model than the traditional matrix method [P2]. Thirdly, it is shown that, for a fixed relatively small number field samples, the radiation pattern can generally be determined with a lower uncertainty from the complex data than the amplitude-only data [P3]. It is shown in [P4] that a high-order probe correction becomes increasingly significant with an increasing ratio between the radius of the minimum sphere of the AUT and the measurement distance. It is shown in [P5] that by enclosing the head phantom with a mobile phone inside the minimum sphere, and the calculation of the truncation number for the spherical wave expansion of the radiated field based on the radius of this minimum sphere in wavelengths, leads to an overestimation of the truncation number. It is illustrated by simulations for a mobile phone that by multiplying the truncation number for the mobile phone without a head phantom by a factor of approximately 1.2 leads to a reasonable truncation number for the mobile phone with the head phantom. It is demonstrated in [P6], by building and testing a spherical fully 3-D measurement system for mobile phone models (RAMS), that the radiation pattern of a typically-sized mobile phone model at approximately 1.8 GHz can be determined without its rotation with a relatively small uncertainty from the complex-valued signals gathered from only 32 dual-port probes on a spherical surface. Information on the reflectivity level inside RAMS will be provided. It is shown in [P7] that the complex radiation pattern of a mobile phone model can be determined without taking advantage of the field-disturbing radio-frequency feed cable to the mobile phone model during the measurement. It is shown in [P8] that, instead of a single spherical wave expansion, the use of multiple spherical wave expansions (MSWE) for the field characterization can lead to a smaller number required spherical modes for reaching a desired level of uncertainty in the determination of the radiation pattern. It will further be shown, that using the MSWE technique can also lead to the smaller number of required measurement locations.
  • An agent-based method for self-study interactive web-based education
    (2006) Rahkila, Martti
     Licentiate thesis
    This thesis deals with computer-based education of acoustics and digital signal processing. The focus throughout the thesis is on interactive, self-study web-based applications even though many issues are of more general nature as well. The emphasis is especially on describing interactivity while using educational applications and the use of log information for evaluation of learning. The goal for the thesis has been to develop a web-based solution for audio signal processing education with emphasis on advanced, intelligent interactivity. The basis for this interactivity is the double agent architecture for web applications. The architecture allows the control of the interaction process by means of logs and using them as a basis for content adaptation. Furthermore, the novelty of this method is its applicability to evaluation of learning. The log information, provided by the architecture, can be used for on-line evaluation of users' requests and thus provides means for content adaptation. Furthermore, the log information can also be post-processed and used for off-line evaluation of the learning process by both teachers as well as students themselves. The latter has also pedagogical importance supporting the development of self-reflection and metacognitive skills.
  • Airborne sound insulation of wall structures : measurement and prediction methods
    (2000-12-01) Hongisto, Valtteri
     Doctoral dissertation (article-based)
    Protection against noise is one of the six essential requirements of the European Construction Product directive. In buildings, airborne sound insulation is used to define the acoustical quality between rooms. In order to develop wall structures with optimal sound insulation, an understanding of the physical origins of sound transmission is necessary. The purpose of this thesis was, firstly, to study and compare the validity of existing physical models to predict the sound insulation of wall structures, and, secondly, to study the benefits of the sound intensity measurement method for determining the sound insulation. To develop the kind of knowledge that is applicable to the improvement of real wall and door structures was the motive behind this study. Five main results are summarized in the following. 1. It was possible to measure wall structures with a considerably, up to 22 dB, higher sound reduction index with the intensity method than with the pressure method. Thus, the intensity method enables the determination of sound insulation in the presence of strong flanking where the pressure method gives only an underestimate. 2. The sound transmission through doors was modelled by two separate paths: a structural path through the door leaf and a leaking path through the door slits. The structural path was predicted using Sharp's model. The agreement with measurements was reasonably good except at high frequencies where overestimations were obtained. The leaking path was predicted using the model of Gomperts and Kihlman. The agreement with measurements was good for free apertures. 3. Thirteen existing prediction models of double panels were compared. The variations in predicted sound reduction indices were high, 20 ... 40 dB. Further work is needed to rank different models according to their reliability for practical structures. In addition, there is an obvious need to develop a hybrid model where all the important parameters are considered. 4. A new flanking mechanism could be observed in situ for a floating floor covering over a concrete slab. Identical floor structures in adjacent dwellings led to strong flanking transmission at the double panel resonance frequency of the floors. Strong flanking could be avoided by modifying the double structure in one dwelling. 5. In general, the most typical design fault of sound insulating double structures was strong mechanical connections, either in the form of rigid interpanel connections (studs) or in the form of bonded cavity absorbent (sandwich structures). In the case of door structures, efforts are usually wasted on the development of the structure, while the leak transmission may be the main transmission path. The results of this study are useful when the intensity method is used in the presence of strong flanking sound, the sound insulation of wall and door structures are predicted or improved and when prediction models are developed.
  • Algorithms and performance evaluation methods for wireless networks
    (2006-09-29) Penttinen, Aleksi
     Doctoral dissertation (article-based)
    The performance of wireless networks depends fundamentally on the characteristics of the radio resource. In this thesis we study methods that can be used to improve performance of wireless networks. We also study methods that can be used to analyze the performance of such networks. In the first part of the thesis, we propose algorithms for multicast routing and max-min fair link scheduling in wireless multihop networks. The multicast routing problem is to find a minimum-cost sequence of transmissions which delivers a message from a given source node to a set of destination nodes. We propose three efficient multicast routing algorithms for certain common instances of the problem. The first algorithm assumes fixed transmission costs and constructs an efficient multicast tree in a centralized fashion. The second algorithm can be used to minimize only the number of transmissions in the multicast tree, but it has a distributed implementation. The last algorithm is applicable in scenarios where the network nodes can control their transmission range and the objective is to minimize the power consumption of the multicast tree. In the max-min fair link scheduling problem one attempts to assign transmission rights to flows in a wireless multihop network so that the long-term flow rates become max-min fair. We present a low-complexity, low-overhead distributed algorithm for the problem. The second part comprises of the flow-level performance analysis of elastic data traffic in wireless networks. The network is modeled in a dynamic setting, in which flows (e.g., file transfers) arrive stochastically and depart upon completion. We discuss how a recently introduced resource allocation concept, balanced fairness, can be applied to wireless networks and devise an efficient computational scheme for solving the resulting joint problem of scheduling and resource allocation. Additionally, we propose an alternative method to approximate the flow throughput under balanced fairness in arbitrary networks. Finally, we adapt balanced fairness to a model where flows are indexed by a continuous variable. The model can capture, e.g., location-dependent features of flows.
  • Älykäs potilasvuode tehohoitopotilaan seurannassa
    (2007) Reijula, Jori
     Master's thesis
    Sairaaloiden tehohoitoyksiköissä työskentelevät sairaanhoitajat ja lääkärit joutuvat päivittäin tekemisiin potilasvuoteiden kanssa, joissa on runsaasti potilaiden valvontaan tarvittavia laitteita ja johtoja. Sängyssä makaavan potilaan irroittaminen johdoista ja johtojen uudelleenkiinnittäminen hidastaa hoitohenkilöstön työskentelyä lisäten sairaalassa tarvittavan henkilökunnan määrää ja sairaanhoitokustannuksia. Kaapelimäärän vähentäminen paitsi helpottaisi sairaalahenkilökunnan työskentelyä, se myös nopeuttaisi ja tehostaisi sairaalaan saapuvien, kriittisessä tilassa olevien potilaiden hoitoa ja vähentäisi myös hoitotapahtumiin liittyviä virheitä ja todennäköisesti jopa potilaskuolemia. Langaton potilasmonitorointi tarjoaa ratkaisun kaapelimäärän vähentämiseen. Tässä diplomityössä esitetään tämän hetken käyttökelpoisimmat tekniikat sairaalaympäristön langattomaan tiedonsiirtoon. Varteenotettavimman ratkaisun tarjoavat lyhyen kantaman langattomat tiedonsiirtoverkot, jotka hyödyntävät teknologioita kuten mm. Bluetooth, UWB ja ZigBee. Lisäksi esitellään lyhyesti keskipitkän kantaman WLAN-verkot ja pitkän kantaman WAP-pohjainen tiedonsiirto. Tämän lisäksi tässä työssä on valmisteltu "älykkään potilasvuoteen" prototyyppi. Se on langaton potilasvuode, joka monitoroi potilasseurannan tärkeimpiä parametreja, joita ovat mm. EKG, verenpaine, pulssin etenemisaika, lämpötila, hengityksen monitorointi, liikkeen monitorointi, hapettuminen, pulssioksimetria sekä hiilidioksidin osapaineen mittaus. Työssä käydään kukin edellä mainituista läpi lyhyen teoriakatsauksen muodossa. Työssä pohditaan myös langattoman tiedonsiirron vaikutusta nykysairaalateknologiaan; miten se näkyy tehohoitopotilaiden seurannassa tällä hetkellä, ja miten langattomuus tulee kehittymään tulevaisuuden sairaaloissa.
  • Ampumamelun leviäminen, mittaaminen ja arviointi
    (2006) Markula, Timo
     Master's thesis
    Ampumamelua tässä työssä käsiteltiin lähinnä ampumaratojen ja alueiden ympäristön kannalta, mahdollisena ympäristöhaittana. Työssä tarkasteltiin sekä kevyiden että raskaiden aseiden melua. Ampumamelu on hyvin impulsiivista ja voimakasta ja se eroaa monin tavoin merkittävästi muista ympäristömelulajeista. Erityisiä menetelmiä tarvitaan tarkkaan tehtävässä ampumamelun mittauksessa, leviämisen mallinnuksessa ja häiritsevyyden arvioinnssa. Ihmisen kuulon äänekkyyden aikaintegrointi ja vakioäänekkyyskäyrästöt tukevat F- ja A-painotusten käyttöä. Mittaus- ja mallinnusteknisestä näkökulmasta energiapohjaisten äänitasojen käyttö on suositeltavampaa kuin enimmäisäänitasojen käyttö. Tässä työssä tehdyt kahden kilpailevan emissiomittausmenetelmän simuloinnit ja mittaukset osoittivat, että kevyiden aseiden melupäästöt pitäisi mitata kovalla maalla mikrofoni lähellä maanpintaa Nordtestin menetelmän mukaisesti. Mikrofonin korkeuden nostaminen aiheuttaa ei-toivotun interferenssin. Raskailla aseilla mikrofoni kuitenkin pitää käytännössä nostaa ylös, jolloin maaheijastuksen vaikutus pitää laskea jälkikäteen pois tuloksesta. Tämän työn mittaustulosten analyysit osoittavat, että ympäristön aiheuttama sironta levittää impulssia ajassa etäisyyden kasvaessa, mikä vaikuttaa enimmäisäänitasoihin merkittävästi. Tavalliset laskentamallit ennustavat leviämistä melko tarkasti myötätuuliolosuhteissa energiatasoja laskettaessa. Pohjoismaisessa ja ISO:n laskentamalleissa ampumamelun etenemisen laskenta on samanlainen estevaimennusta lukuun ottamatta. Teorian ja mittausten perusteella luodin aiheuttama yliäänipamaus voi olla jopa voimakkaampi kuin suupamaus ainakin kevyillä aseilla.
  • Analog baseband circuits for WCDMA direct-conversion receivers
    (2003-06-27) Jussila, Jarkko
     Doctoral dissertation (monograph)
    This thesis describes the design and implementation of analog baseband circuits for low-power single-chip WCDMA direct-conversion receivers. The reference radio system throughout the thesis is UTRA/FDD. The analog baseband circuit consists of two similar channels, which contain analog channel-select filters, programmable-gain amplifiers, and circuits that remove DC offsets. The direct-conversion architecture is described and the UTRA/FDD system characteristics are summarized. The UTRA/FDD specifications define the performance requirement for the whole receiver. Therefore, the specifications for the analog baseband circuit are obtained from the receiver requirements through calculations performed by hand. When the power dissipation of an UTRA/FDD direct-conversion receiver is minimized, the design parameters of an all-pole analog channel-select filter and the following Nyquist rate analog-to-digital converter must be considered simultaneously. In this thesis, it is shown that minimum power consumption is achieved with a fifth-order lowpass filter and a 15.36-MS/s Nyquist rate converter that has a 7- or 8-bit resolution. A fifth-order Chebyshev prototype with a passband ripple of 0.01 dB and a −3-dB frequency of 1.92-MHz is adopted in this thesis. The error-vector-magnitude can be significantly reduced by using a first-order 1.4-MHz allpass filter. The selected filter prototype fulfills all selectivity requirements in the analog domain. In this thesis, all the filter implementations use the opamp-RC technique to achieve insensitivity to parasitic capacitances and a high dynamic range. The adopted technique is analyzed in detail. The effect of the finite opamp unity-gain bandwidth on the filter frequency response can be compensated for by using passive methods. Compensation schemes that also track the process and temperature variations have been developed. The opamp-RC technique enables the implementation of low-voltage filters. The design and simulation results of a 1.5-V 2-MHz lowpass filter are discussed. The developed biasing scheme does not use any additional current to achieve the low-voltage operation, unlike the filter topology published previously elsewhere. Methods for removing DC offsets in UTRA/FDD direct-conversion receivers are presented. The minimum areas for cascaded AC couplings and DC-feedback loops are calculated. The distortion of the frequency response of a lowpass filter caused by a DC-feedback loop connected over the filter is calculated and a method for compensating for the distortion is developed. The time constant of an AC coupling can be increased using time-constant multipliers. This enables the implementation of AC couplings with a small silicon area. Novel time-constant multipliers suitable for systems that have a continuous reception, such as UTRA/FDD, are presented. The proposed time-constant multipliers only require one additional amplifier. In an UTRA/FDD direct-conversion receiver, the reception is continuous. In a low-power receiver, the programmable baseband gain must be changed during reception. This may produce large, slowly decaying transients that degrade the receiver performance. The thesis shows that AC-coupling networks and DC-feedback loops can be used to implement programmable-gain amplifiers, which do not produce significant transients when the gain is altered. The principles of operation, the design, and the practical implementation issues of these amplifiers are discussed. New PGA topologies suitable for continuously receiving systems have been developed. The behavior of these circuits in the presence of strong out-of-channel signals is analyzed. The interface between the downconversion mixers and the analog baseband circuit is discussed. The effect of the interface on the receiver noise figure and the trimming of mixer IIP2 are analyzed. The design and implementation of analog baseband circuits and channel-select filters for UTRA/FDD direct-conversion receivers are discussed in five application cases. The first case presents the analog baseband circuit for a chip-set receiver. A channel-select filter that has an improved dynamic range with a smaller supply current is presented next. The third and fifth application cases describe embedded analog baseband circuits for single-chip receivers. In the fifth case, the dual-mode analog baseband circuit of a quad-mode receiver designed for GSM900, DCS1800, PCS1900, and UTRA/FDD cellular systems is described. A new, highly linear low-power transconductor is presented in the fourth application case. The fourth application case also describes a channel-select filter. The filter achieves +99-dBV out-of-channel IIP2, +45-dBV out-of-channel IIP3 and 23-μVRMS input-referred noise with 2.6-mA current from a 2.7-V supply. In the fifth application case, a corresponding performance is achieved in UTRA/FDD mode. The out-of-channel IIP2 values of approximately +100 dBV achieved in this work are the best reported so far. This is also the case with the figure of merits for the analog channel-select filter and analog baseband circuit described in the fourth and fifth application cases, respectively. For equal power dissipation, bandwidth, and filter order, these circuits achieve approximately 10 dB and 15 dB higher spurious-free dynamic ranges, respectively, when compared to implementations that are published elsewhere and have the second best figure of merits.
  • Analog parallel processor solutions for video encoding
    (2005-12-16) Koskinen, Lauri
     Doctoral dissertation (monograph)
    This thesis deals with Cellular Nonlinear Network (CNN) analog parallel processor networks and their implementations in current video coding standards. The target applications are low-power video encoders within 3rd generation mobile terminals. The video codecs of such mobile terminals are defined by either the MPEG-4/H.263 or H.264 video standard. All of these standards are based on the block-based hybrid approach. As block-based motion estimation (ME) is responsible for most of the power consumption of such hybrid video encoders, this thesis deals mostly with low-power ME implementations. Low-power solutions are introduced at both the algorithmic and hardware levels. On the algorithmic level, the introduced implementations are derived from a segmentation algorithm, which has previously been partly realized. The first introduced algorithm reduces the computational complexity of ME within an object-based MPEG-4 encoder. The use of this algorithm enables a 60% drop in the power consumption of Full Search ME. The second algorithm calculates a near-optimal block-size partition for H.264 motion estimation. With this algorithm, the use of computationally complex Lagrange optimization in H.264 ME is not required. The third algorithm reduces the shape bit-rate of an object-based MPEG-4 encoder. On the hardware level a CNN-type ME architecture is introduced. The architecture includes connections and circuitry to fully realize block-based ME. The analog ME implemented with this architecture is capable of lower power than comparable digital realizations. A 9×9 test chip has also been realized. Additionally implemented is a digital predictive ME realization that takes advantage of the introduced partition algorithm. Although the IC layout of the ME algorithm was drawn, the design was verified as an FPGA.
  • Analysis and compensation of nonlinear power amplifier effects in multi-antenna OFDM systems
    (2007-11-19) Gregorio, Fernando Hugo
     Doctoral dissertation (article-based)
    The high peak to average power ratio (PAPR) levels associated with multicarrier systems require the use of linear power amplifiers (PA) with large dynamic range. Unfortunately, linearity and power efficiency are two conflicting specifications for practical PAs. However, systems with high power efficiency are required, for example, in case of battery operated mobile terminals. Thus, the PA needs to operate in a high power efficiency region which compromises the linearity constraints. The evaluation of the effect of nonlinear amplification on the system performance is an important issue that must be considered in a realistic system design. This thesis first addresses the performance analysis of multiuser OFDM systems (space division multiple access (SDMA)-OFDM) employing nonlinear PAs. Bit error probability (BEP) and capacity expressions are derived for an OFDM system with diversity, assuming a memoryless PA. Moreover, BEP expressions are derived for an orthogonal space-time coded (OSTBC)-OFDM system that include the effect of a broadband nonlinear PA and imperfect memory compensation. Thereafter, techniques to reduce the effects of power amplifier distortion are proposed. We consider techniques that can be implemented either in the receiver, the transmitter, or as a combination of both, all with the same goal of relieving the linearity constaints and improving the power efficiency. A low PAPR OFDM implementation is addressed where subcarriers are allocated to different users in such a way that the generated intermodulation distortion is kept at minimum. Then, we propose the combination of multiuser detection and nonlinear distortion mitigation techniques. Initially, knowledge of the PA model (modelled as memoryless) is required for the implementation. The extension to broadband PAs and to PA model estimation in the receiver are also made. Efficient channel estimation strategies are developed that work in the presence of nonlinear distortion. Finally, a split predistorter structure to remove nonlinear distortion caused by a broadband PA is proposed. The idea is to reduce the transmitter complexity by distributing the predistorter tasks such that the transmitter compensates for the static nonlinearity, and the receiver equalizes the PA memory.
  • Analysis and optimization of photonic crystal components for optical telecommunications
    (2005-07-08) Huttunen, Anu
     Doctoral dissertation (article-based)
    Photonic crystals are periodic dielectric structures where the period is of the same order of magnitude than the wavelength of light. As a result of interference, there exist band gaps for light, i.e., light of certain range of frequencies is not allowed to exist inside the photonic crystal, which can be used to control and confine light. In this thesis, photonic crystals were studied with computer simulations and concepts of new components for optical telecommunications were proposed. An all-optical switch, based on the properties of a Kerr-nonlinear one-dimensional photonic crystal, was investigated. The band gap was shown to change as a function of light intensity inside the nonlinear photonic crystal. Thus, it performs an all-optical switching function, where one signal can be dynamically reflected depending on another signal. Two parallel waveguides in a two-dimensional photonic crystal were considered. In general, adjacent waveguides are coupled and a light pulse traveling in one waveguide will oscillate between the waveguides. We found complete decoupling between the waveguides for certain geometries. This can be utilized in integrated optics when cross-talk between contiguous channels is not desired. The band gap of a thin slab of two-dimensional photonic crystal was shown to depend strongly on the material on top/below the slab, and thus a new type of waveguide was proposed. Instead of the conventional defect waveguide, a waveguide can be made by patterning the photonic crystal slab by a suitable material. The waveguiding was shown to be due to band gap difference in some cases and due to average refractive index difference in others. Microstructure and dual-core fibers were optimized to achieve large negative dispersion and large mode area simultaneously. Negative dispersion fibers are needed for dispersion compensation and pulse compression. They usually have small mode areas and are highly nonlinear, which is a problem for high-intensity applications. This can be avoided with the fiber geometries proposed in this thesis. The effect of the wavelength dependence of gain, nonlinearity and dispersion to the propagation of short pulses in high-gain efficiency photonic crystal fiber amplifiers was studied. It resulted in asymmetric spectrum and chirp, and reduction of the pulse broadening. Wavelength dependence of the nonlinearity was demonstrated to have the most effect, compared to that of dispersion or gain.
  • Analysis of induction motors based on the numerical solution of the magnetic field and circuit equations
    (1987-12-18) Arkkio, Antero
     Doctoral dissertation (monograph)
    A method for the analysis of induction motors is presented. The analysis is based on the combined solution of the magnetic field equations and the circuit equations of the windings. The equations are discretized by the finite element method. The magnetic field is assumed to be two-dimensional. The three-dimensional features i.e. the skew of the rotor slots and the end-region fields are taken into account within the two-dimensional formulation. The general time-dependence of the field and the motion of the rotor are modelled correctly in a step-by-step solution. The amount of computation is reduced significantly if the time-dependence is assumed to be sinusoidal and phasor quantities are used in the solution. The method is applied to the calculation of a cage rotor motor and of a solid rotor motor. The sinusoidal approximation gives good results in the computation of steady-state locked-rotor quantities, but it does not model the motion of the rotor properly. The step-by-step method is used for computing machine quantities in steady and transient states. For instance the operation of the solid rotor motor supplied by a static frequency converter is simulated. The results obtained by the method agree well with the measured ones.
  • Analysis of the existing visual performance based mesopic models and a proposal for a model for the basis of mesopic photometry
    (2007-12-14) Viikari, Meri
     Doctoral dissertation (article-based)
    The work started with outlining the current status of photometry. The origins of the photopic spectral luminous efficiency function V(λ) were extensively investigated. The work continued to examine the ability of the existing photopic spectral luminous efficiency functions V(λ), VM(λ), or V10(λ) to describe peripheral photopic vision. The conducted reaction time measurements indicated that none of the existing functions described the peripheral vision correctly. A new photopic spectral luminous efficiency function for peripheral vision Vper(λ) is proposed for the photopic end with which the mesopic photometric model should merge at the upper luminance limit of the mesopic region. The work continued to review the existing mesopic photometric systems. The previously proposed mesopic models are mainly based on brightness matching, which does not preserve additivity. The two recently proposed mesopic models, MOVE-model by the MOVE consortium (Eloholma 2005) and the X-model by Rea et al (2004), are based on visual task performance. Performance based mesopic models are claimed to preserve additivity within the given light level. The two performance based models are currently being analyzed by the CIE TC 1-58 in order to result in an internationally agreed system for performance based mesopic photometry. The work continued to generate contrast threshold and reaction time data in order to compare the existing performance based mesopic models. The differences between the mesopic luminances predicted by the two models are evident. Also the upper transition points of the models between the mesopic and photopic regions are very different. Finally, the work concluded by proposing a new performance based mesopic model. The new model is based on the same experimental data as the MOVE-model. Thus the newly proposed model is assigned as a modified MOVE-model. The modified MOVE-model was compared along with the other two performance based models using three independent experimental visual performance data sets. The comparison indicated that the modified MOVE-model described the experimental data best.
  • Analysis, synthesis, and perception of spatial sound : binaural localization modeling and multichannel loudspeaker reproduction
    (2006-08-11) Merimaa, Juha
     Doctoral dissertation (monograph)
    In everyday audio environments, sound from several sources arrives at a listening position both directly from the sources and as reflections from the acoustical environment. This thesis deals, within some limitations, with analysis of the resulting spatial sound field, reproduction of perceptually relevant features of the sound as measured in a chosen listening position, as well as with modeling of the related auditory localization. For the localization, the auditory system needs to independently determine the direction of each source, while ignoring the reflections and superposition effects of any possible concurrently arriving sound. A modeling mechanism with these desired properties is proposed. Interaural time difference (ITD) and interaural level difference (ILD) cues are only considered at time instants when only the direct sound of a single source has non-negligible energy within a critical band and, thus, when the evoked ITD and ILD represent the direction of that source. It is shown how to identify such time instants as a function of the interaural coherence (IC). The source directions suggested by the selected ITD and ILD cues are also shown to imply the results of a number of published psychophysical studies. Although the room reflections are usually suppressed in auditory localization, they contribute to the perception of the acoustical environment. The reviewed physical analysis techniques and psychoacoustical knowledge on spatial hearing are applied in development of the Spatial Impulse Response Rendering (SIRR) method. SIRR aims at recreating ITD, ILD, IC, and monaural localization cues by using a perceptually motivated analysis-synthesis method. The method is described in the context of multichannel loudspeaker reproduction of room responses with convolving reverberators. The analyzed quantities consist of the time- and frequency-dependent direction of arrival and diffuseness of sound. Based on the analysis data and a recorded omnidirectional signal, multichannel responses suitable for reproduction with any chosen surround loudspeaker setup are synthesized. In formal listening tests, it is shown that SIRR creates a more natural spatial impression than can be achieved with conventional techniques.