Recent findings suggest to split the impulse response of the radio channel into discrete paths and the "diffuse multipath" (DMP). This diffuse part can be described by an exponentially decaying power delay profile. This paper shows how to improve current radio channel models using the DMP concept. From MIMO channel measurements, we find that the DMP parameters are strongly correlated with the parameters of the discrete paths. This holds for various (indoor) environments. We provide a simple way to model the statistics of the DMP parameters. Finally, we include the DMP concept in a novel MIMO model, the Random-Cluster model. We find that including DMP significantly improves the model fit in terms of mutual information and channel diversity.

Essential parameters of physical, propagation-based MIMO channel models are the fading statistics and the directional spread of multipath clusters. In this paper we determine these parameters in the azimuth-of-arrival/azimuth-of-departure (AoA/AoD) domain based on comprehensive indoor MIMO measurements at 5.2 GHz in a cluttered office environment using the SAGE algorithm for parameter estimation. Due to cluster identification in AoA/AoD-domain we found a greater number of clusters than those reported in previous publications. Regarding the fading statistics of clusters, so far not studied, strong (obstructed-)line-of-sight clusters show Rician fading, corresponding to few dominant propagation paths, whereas most clusters exhibit Rayleigh fading, corresponding to many paths with approximately equal powers and uncorrelated phases. Root-mean-square cluster azimuth spreads (CASs) were estimated with a novel method by appropriately restricting the support of the cluster azimuth distribution. We found that the estimated CASs are different when seen from transmitter or receiver, i.e. their ranges are from 2deg to 9deg and from 2deg to 7deg at the transmitter side and the receiver side, respectively

This paper presents a complete solution to the problem of how to parametrise cluster-based stochastic MIMO channel models from measurement data, with minimum user intervention. The method comprises the following steps: (i) identify clusters in measurement data, (ii) identify the optimum number of clusters, (iii) track clusters over consecutive time snapshots, (iv) estimate cluster parameters. These parameters are given as estimated probability density functions of the cluster power, cluster positions, delay and angular spreads of clusters and the number of paths within a cluster. Applied to high-resolution indoor MIMO measurement data at 5.2 GHz and at 2.55 GHz, the method yields coherent results of the obtained cluster parameters.

We present a solution to the problem of identifying clusters from MIMO measurement data in a data window, with a minimum of user interaction. Conventionally, visual inspection has been used for the cluster identification. However this approach is impractical for a large amount of measurement data. Moreover, visual methods lack an accurate definition of a "cluster" itself. We introduce a framework that is able to cluster multi-path components (MPCs), decide on the number of clusters, and discard outliers. For clustering we use the K-means algorithm, which iteratively moves a number of cluster centroids through the data space to minimize the total difference between MPCs and their closest centroid. We significantly improve this algorithm by following changes: (i) as the distance metric we use the multi- path component distance (MCD), (ii) the distances are weighted by the powers of the MPCs. The implications of these changes result in a definition of a "cluster" itself that appeals to intuition. We assess the performance of the new algorithm by clustering real-world measurement data from an indoor big hall environment.

Fixed-beam smart antennas are a simple but effective method to boost the downlink capacity of UMTS FDD. In this paper we compare the two possible strategies. Using a four element uniform linear antenna array, we find the optimum number of fixed beams per 120/spl deg/ sector dependent on the direction of departure (DoD) spread at the base station. For the first method we find the optimum number of beams to be four for low DoD spreads and two or three for large DoD spreads. For the second method the optimum number of beams per sector is seven for small DoD spreads and goes down to four or five beams per sector for large DoD spreads depending on base station spacing. By extensive system level simulations, we show, for 1 km inter base station distance, a capacity gain of more than 160% over a conventional 3-sectored reference system by both fixed beam methods.

Abstract-This paper presents a novel stochastic channel model for multiple-input multiple-output (MIMO) wireless radio channels. In contrast to state-of-the-art stochastic MIMO channel models, the spatial correlation properties of the channel are not divided into separate contributions from transmitter and receiver. Instead, the joint correlation properties are modeled by describing the average coupling between the eigenmodes of the two link ends. The necessary and sufficient condition for the proposed model to hold is that the eigenbasis at the receiver is independent of the transmit weights, and vice versa. The authors discuss the mathematical elements of the model, which can be easily extracted from measurements, from a radio propagation point of view and explain the underlying assumption of the model in physical terms. The validation of the proposed model by means of measured data obtained from two completely different measurement campaigns reveals its ability to better predict capacity and spatial channel structure than other popular stochastic channel models.

An important parameter of MIMO channel models is the cluster root-mean-square (rms) directional spread. In this paper we determine this parameter in the azimuth-of-arrival/azimuth-of-departure (AoA/AoD) domain based on comprehensive indoor MIMO measurements at 5.2 GHz in a cluttered office environment. This is done in a four-step procedure: (i) the SAGE algorithm is used to extract propagation paths, (ii) clusters of estimated paths in the double-azimuth domain are defined, (iii) the estimated propagation paths are allocated to the clusters, (iv) the cluster directional spreads are estimated based solely on propagation paths within the clusters. We found that the spreads are different when seen from transmitter or receiver due to different propagation conditions resulting in AoD rms cluster spreads lying in the range from 2 to 9 degrees and AoA rms cluster spreads in the range from 2 to 7 degrees.

A diffuse spectrum estimation method is presented that is suitable for direction finding for channels with diffuse electromagnetic scattering. The diffuse channel may be the result of physical phenomena (e.g., rough-surface scattering) or limited angular and temporal discrimination of the channel probing equipment. The method decomposes the ideal noncoherent spectrum into a sum of power-weighted basis functions. The basis function coefficients are obtained through a linear programming solution. The method is applied to recent radio channel data collected on the Vienna University of Technology campus. Parameters for a path-based model are extracted using the technique. Site-specific spectra and global capacity comparisons indicate the utility of this new method.

We present a simple beam switching scheme to increase the downlink capacity of the UMTS FDD-mode. At the base station, this method only requires one additional transceiver chain, i.e. a separate one for each beam, a beamforming network and a uniform linear array. The mobiles need no changes at all. We analyze the performance of our method by means of system level simulations for different angular spreads. We found an increase in downlink capacity of more than 75% over a conventional 3-sectored system for small direction of departure spreads. Compared to a 6-sectored system our scheme gives a capacity gain of more than 25% for small direction of departure spreads.

In mobile communication systems using CDMA, common channels needed for proper network operation share the same resource as the dedicated channels for data transmission. However, the 3GPP standard does not specify the amount of transmit power for the common channels. So it is up to the network operator to assign appropriate transmit powers to the various common channels. We study the influence of the power assigned to the common channels on the achievable capacity in UMTS FDD systems that use beam switching. For a system with five beams per 120/spl deg/ sector, we show that a change of the common-channel power per beam by only 1 watt can boost capacity by about 70%.

We measure the capacity of multiple-input multiple-output radio systems in microcellular environments. We use a new data evaluation method that allows to evaluate the cumulative distribution function of the capacity from a single measurement. This method is based on an extraction of the parameters of the multipath components and, thereafter, a synthetic variation of their phases. In the analyzed environments, we find capacities to be about 30% smaller than would be anticipated from an idealized model. In addition, the frequency selectivity of the channel makes the CDF of the capacity steeper and, thus, increases the outage capacity, compared with the frequency-flat case, but the influence on the mean capacity is small.

We present a statistical analysis of wideband three-dimensional channel measurements at base station locations in an urban environment. Plots of the received energy over azimuth, elevation, and delay planes suggest that the incident waves group to clusters in most measured transmitter positions. A super-resolution algorithm (Unitary ESPRIT) allows one to resolve individual multipath components in such clusters and hence enables a detailed statistical analysis of the propagation properties. The origins of clusters-sometimes even individual multipath components-such as street apertures, large buildings, roof edges, or building corners can be localized on the city map. Street guided propagation dominates most of the scenarios (78%-97% of the total received power), while quasi-line-of-sight over-the-rooftop components are weak(3%-13% of the total received power). For this measurement campaign, in 90% of the cases, 75% of the total received power is concentrated in the two strongest clusters, but only 55% in the strongest one. Our analysis yields an exponential decay of power with 8.9 dB//spl mu/s, and a standard deviation of the log-normally distributed deviations from the exponential of 9.0 dB. The power of cross-polarized components is 8 dB below copolarized ones on average (vertical transmission).

We present a robust smart antenna processor that is based on directions-of-arrival (DOA) estimation at the uplink. For optimum BER performance in a cellular mobile radio environment we resolve all relevant paths for the intended user, not only a single one. To increase the system's reliability, we apply separate trackers for uplink and downlink and use a beamforming algorithm that places broad nulls to suppress interferers as they appear realistically in the mobile radio channel. A performance evaluation of the eight-element array based on a semi-stochastic directional channel model shows that the DOA estimation accuracy is not of great concern, but estimation robustness strongly influences the BER. By adaptation within every GSM frame, the processor yields a mean noise and interference ratio (SNIR) improvement of up to 35 dB in an interference limited mobile radio macro-cell environment. The tracking algorithm gains 10 dB in mean input SNIR at a raw BER of 1%. Our beamformer with broad nulls has up to an order of magnitude better BER performance than a conventional beamformer that places sharp nulls. Angular diversity in the uplink and spatial interleaving in the downlink are implemented to improve the link quality.

We propose a semi-blind algorithm for smart antennas that, in addition to some structural signal properties, utilises user identifiers that are available in existing cellular systems. Our algorithm estimates, first, the row span of the adequately composed data matrix and then projects the obtained basis vectors iteratively to the finite alphabet (FA) constellation. We call the projection part of our algorithm DILSF (decoupled iterative least-squares with subspace fitting). Assuming two co-channel users in a GSM-like TDMA system and a Rayleigh fading channel with finite angular spread, we obtained a BER<10/sup -3/ with only 3 antennas, spaced 10/spl lambda/ apart. We also demonstrate that co-channel signals are resolvable even if they are not separable in angle. We gain a factor of up to 70 in computational complexity by employing subspace tracking algorithms and demonstrate the robustness of the method against array imperfections.

This paper studies various diversity arrangements for handset antennas at about 1900 MHz by means of measurements and simulations. The antennas used in the diversity arrangements are radiation coupled dual-L antennas (RCDLA), which are strong candidates for internal antennas for mobile terminals. The measurements show that good pattern and polarisation diversity can be obtained with the studied diversity arrangements. It is also shown that the position of the antenna on the case strongly affects the magnitude of the detuning effect that the hand of the user causes. The problems caused by the hand of the user can be mitigated by using diversity antennas at the mobile.

We analyze the error floor of unequalized personal communications systems using minimum-shift keying (MSK) with training-sequence-based adaptive sampling in a two-delay fading channel. We include Gaussian filtering of the input, finite accuracy in determining the optimum sampling time, and filtering at the receiver. We prove that for pure MSK and low time dispersion, training-sequence-based adaptive sampling can completely avoid errors caused by intersymbol interference. The actual errors are caused by "secondary" effects (filtering and finite-resolution sampling) in conjunction with the channel time dispersion. Errors occur if the normalized phasor of the channel impulse response falls into certain error regions; the computation of these regions gives physical insights into the error mechanism and allows a highly efficient computation of the average bit-error rate (BER). The average BER varies as K/spl middot/(S/T)/sup 2/, where S is the RMS delay spread and T is the bit length. The proportionality constant K depends on the Gaussian filtering in the transmitter, the receiver filtering, and the amount of oversampling. The BER can be orders of magnitude lower than for the (quasi-) fixed sampling case, in which timing is derived solely based on the channel delay power profile. For two-branch diversity reception, we show that the BER is proportional to (S/T)/sup 4/.

We give an overview of the work of COST 231 on the "error floor", i.e. errors introduced by the time dispersion of the mobile radio channel. The error floor depends mainly on the RMS delay spread S and on the sampling scheme; the error probability is in general K/spl middot/(S/T)/sup 2/, where T is the bit length. For fixed sampling, the proportionality constant K is about 0.5, and depends only weakly on the filters. For adaptive sampling, K is one or more orders of magnitude smaller, depending strongly on the filters in the system. Errors occur in a "bursty" way in the fading dips.

The proper combination of architecture and integration concept will decide the success of a transceiver realization. The authors discuss three realization concepts of monolithic integrated transceivers: (i) a modular multichip solution, (ii) a dedicated few chip solution and (iii) a single chip solution of full custom library modules. The authors compare them with respect to size, power consumption, and production cost. Applied to heterodyne and direct conversion transceiver architectures, they find that direct conversion is more complex but superior for integration and cost. They present first-time experimental results of a prototype 1.5 GHz fully integrated direct conversion transceiver chip that uses a VCO at twice the required LO frequency for reduction of DC offset. A novel, mainly digital high-dynamic-range limiting baseband detector complements a direct conversion solution.

Cellular network operators are greatly interested in radio propagation in tunnels and other wave guiding structures. Road tunnels are commonly served by leaky feeders to provide VHF radio coverage. For UHF and up, we present an attractive alternative to this solution, using discrete antennas in- or outside the tunnel, depending on the tunnel's length. The key to coverage predictions is a "critical distance", after which higher-order modes excited near the antenna have died out. This distance depends on tunnel cross-section dimensions and wavelength. A propagation model using this critical distance can predict power level versus length for distances larger than the critical one (e.g. 300 m for 900 MHz in a Euro-standard dual-lane tunnel). Two reference measurements are needed as input for the prediction. Validation measurements taken in a dual-lane road tunnel at 900 and 1800 MHz showed the prediction to be useful at various traffic rates. Traffic does not significantly hamper propagation, if only the vehicles leave enough space in the cross section, nor do bends with curvature radii commonly used in motorway tunnels.

We performed 950 and 1800 MHz CW-measurements of radio wave propagation into various urban buildings in small-cell and microcell propagation scenarios in Vienna. We introduced a new empirical building penetration model that accounted for the power outdoors in a more detailed way than simpler previous ones. In both urban cell types, radio coverage of buildings can successfully be estimated from (i) measured or modelled signal strength on street level and from (ii) wall attenuation parameters. In comparison to small cells, street canyon microcells exhibited, due to predominantly grazing incidence, much higher penetration loss values. >

The authors investigated experimentally, analytically, and by simulation the irreducible errors such as error floor for high signal-to-noise ratio that are introduced by time dispersion in a slowly fading Rayleigh channel. A multipath indoor propagation environment with controlled delay spread was created to validate this approach. The system is close to the Digital European Cordless Telecommunications (DECT) standard with 1 Mb/s GMSK-modulated signals being transmitted in FDM/TDMA/TDD format and received by a differential direct-conversion receiver. The measured and simulated error probability floor is Pe approximately=5*10/sup -3/ as a consequence of delay spread that is 10% of the bit length. This value is approximated by a closed-form analytical description of phase jitter due to large group delay bursts. Simulation results obtained previously by others for comparable cases are partially corroborated and partially disproved.>

By analytical derivation, supported by simulation, the authors develop a closed form theory of the error floor probability of differentially detected MSK in a slowly fading Rayleigh channel with small delay spread. As the major error source they identify the phase jitter generated by group delay bursts coinciding with the frequency deviation of the modulated signal, rather than the timing jitter following the instantaneous mean delay. Such errors turn up in the fades even when the channel is short-term invariant and the receiver is stationary. For experimental verification they apply controlled delay spreads (over the air) in a DECT-like test bed. The results confirm that the error probability is, to first order, a function of the delay spread to bit-length ratio only and varies with the square of this ratio.>

We have transferred the principle of the Costas-type nonlinear phase-locked loop to the\lambda = 10.6 \mum wavelength of CO2lasers. The ability of the optical Costas loop to regenerate the carrier of a binary phase-shift-keyed input signal and to perform coherent demodulation was demonstrated at a data rate of 20 Mbits/s.