Frank Frederiksen

A stochastic MIMO radio channel model with experimental validation

Theoretical and experimental studies of multiple-input/multiple-output (MIMO) radio channels are presented. A simple stochastic MIMO model channel has been developed. This model uses the correlation matrices at the mobile station (MS) and base station (BS) so that results of the numerous single-input/multiple-output studies that have been published in the literature can be used as input parameters. The model is simplified to the narrowband channels. The validation of the model is based upon data collected in both picocell and microcell environments. The stochastic model has also been used to investigate the capacity of MIMO radio channels, considering two different power allocation strategies, water filling and uniform and two different antenna topologies, 4/spl times/4 and 2/spl times/4. Space diversity used at both ends of the MIMO radio link is shown to be an efficient technique in picocell environments, achieving capacities within 14 b/s/Hz and 16 b/s/Hz in 80% of the cases for a 4/spl times/4 antenna configuration implementing water filling at a SNR of 20 dB.

LTE Capacity Compared to the Shannon Bound

In this paper we propose a modification to Shannon capacity bound in order to facilitate accurate benchmarking of UTRAN long term evolution (LTE). The method is generally applicable to wireless communication systems, while we have used LTE air-interface technology as a case study. We introduce an adjusted Shannon capacity formula, where we take into account the system bandwidth efficiency and the SNR efficiency of LTE. Separating these issues, allows for simplified parameter extraction. We show that the bandwidth efficiency can be calculated based on system parameters, while the SNR efficiency is extracted from detailed link level studies including advanced features of MIMO and frequency domain packet scheduling (FDPS). We then use the adjusted Shannon capacity formula combined with G-factor distributions for macro and micro cell scenarios to predict LTE cell spectral efficiency (SE). Such LTE SE predictions are compared to LTE cell SE results generated by system level simulations. The results show an excellent match of less that 5-10% deviation.

Carrier aggregation for LTE-advanced: functionality and performance aspects

Carrier aggregation is one of the key features for LTE-Advanced. By means of CA, users gain access to a total bandwidth of up to 100 MHz in order to meet the IMT-Advanced requirements. The system bandwidth may be contiguous, or composed of several non-contiguous bandwidth chunks that are aggregated. This article presents a summary of the supported CA scenarios as well as an overview of the CA functionality for LTE-Advanced with special emphasis on the basic concept, control mechanisms, and performance aspects. The discussion includes definitions of the new terms primary cell (PCell) and secondary cell (SCell), mechanisms for activation and deactivation of CCs, and the new cross-CC scheduling functionality for improved control channel optimizations. We also demonstrate how CA can be used as an enabler for simple yet effective frequency domain interference management schemes. In particular, interference management is anticipated to provide significant gains in heterogeneous networks, envisioning intrinsically uncoordinated deployments of home base stations.

Prediction of future fading based on past measurements

The possibility of predicting the complex radio channel ahead in time when the terminal is moving in a straight line is investigated. It is assumed that a number of samples with constant spatial sampling period are available. The Doppler spectrum and the amplitude of the complex scatterers are determined using an ESPRIT-type algorithm, and the signals are then extrapolated into the future assuming that the scatterers remain constant. Synthetic and real data both indicate that for a large number of scatterers, a continuous Doppler spectrum, the signal may be predicted about a wavelength ahead. This assumes that the sampling is sufficiently dense.

Performance aspects of WCDMA systems with high speed downlink packet access (HSDPA)

In this paper, we consider the link and network layer performance aspects of a WCDMA/UTRA system with high speed downlink packet access. The study considers packet scheduling and the tradeoff among user fairness and cell throughput. We show performance numbers for different network setups and study the applicability of proportional scheduling methods. Even with conservative system and traffic settings, the best effort methods produce high user data rates and cell throughput.

Extension of the ITU channel models for wideband (OFDM) systems

This paper outlines a procedure to extend, or upsample, the ITU power delay profiles (PDP) to improve the frequency correlation properties while keeping the same mean delay and almost the same rms delay spread. Realistic frequency correlation properties are of particular importance for the evaluation of wideband system concepts with frequency dependent characteristics, e.g. frequency domain link adaptation and packet scheduling, both of which are likely to be part of future wideband systems such as based on OFDM. With the suggested procedure the frequency correlation can be kept approximately at or below 0.6 for frequency separations up to 25 MHz for the exemplified ITU Vehicular A and Pedestrian B channel profiles. The compatibility in terms of link level performance with WCDMA rake processing and the representation of the profiles on different sampling grids is briefly discussed in the paper. ITU Channel models; frequency correlation function; wideband systems; power delay profile; beyond 3G system concepts

An overview of downlink radio resource management for UTRAN long-term evolution

Radio resource management algorithms ranging from bearer admission control to semi-persistent and dynamic packet scheduling, fast link adaptation, and transmission control of multi-antenna transmission modes are addressed in this article for UTRAN long-term evolution. First, a high-level system overview of LTE is given, with special emphasis on the important components related to RRM. The quality of service parameter framework is outlined, as one of the main objectives for the families of RRM algorithms is to maximize system capacity while serving all users according to their minimum QoS constraints. It is demonstrated how the collocation of the RRM algorithms at the base station with easy access to air interface measurements offers opportunities for efficient cross-functional optimization between layers 1, 2, and 3. Examples of performance results for different traffic mixes and antenna transmission schemes are also presented, and the article is concluded with recommendations on how to operate the various RRM options under different load and traffic conditions.

Frequency Domain Scheduling for OFDMA with Limited and Noisy Channel Feedback

In this study we analyze the downlink OFDMA system level performance for three different channel quality indicator (CQI) reporting schemes. The effect of terminal measurement and estimation errors, quantization from formatting and compression, and uplink reporting delays and detection errors are included. We find that a simple threshold-based CQI scheme provides an attractive trade-off between downlink system level performance and uplink CQI signaling overhead, as compared to using a best-M scheme. When applied to the UTRAN LTE system in a 10 MHz bandwidth, we find that a frequency domain packet scheduling gain of 40% is achievable with a CQI word size of only 30-bits. Finally, the effect of applying a so-called outer loop link adaptation algorithm is reported.

Experimental investigation of multipath richness for multi-element transmit and receive antenna arrays

The multi-element antenna arrays concept with M elements at the mobile station (MS) in combination with N elements at the base station (BS) is experimentally investigated. Forschini (1996) has shown very promising results to improve the spectral efficiency in a rich scattering environment. The performance of the M/spl times/N concept is evaluated in terms of the number of independent parallel channels, diversity gain and total capacity in an outdoor to indoor microcellular environment. It is shown that the eigenanalysis provides a tool to describe the effective number of parallel channels in a multi-element array configuration. Practical results on spectral efficiency are presented for different antenna setups applied to different propagation scenarios. Also it is shown that polarization diversity is an attractive solution to achieve decorrelated antenna elements and subsequently provide a more robust system in terms of spectral efficiency within the microcell. Results show that a total capacity of 27.9 b/s/Hz can be achieved for an uncorrelated propagation environment and 17 b/s/Hz for a correlated one with a mean signal to noise ratio (SNR) of 30 dB in the case of a 4/spl times/4 antenna set-up.

Enhanced inter-cell interference coordination in co-channel multi-layer LTE-advanced networks

Different technical solutions and innovations are enabling the move from macro-only scenarios towards heterogeneous networks with a mixture of different base station types. In this article we focus on multi-layer LTE-Advanced networks, and especially address aspects related to interference management. The network controlled time-domain enhanced inter-cell interference coordination (eICIC) concept is outlined by explaining the benefits and characteristics of this solution. The benefits of using advanced terminal device receiver architectures with interference suppression capabilities are motivated. Extensive system level performance results are presented with bursty traffic to demonstrate the eICIC concepts ability to dynamically adapt according to the traffic conditions.