Darren McNamara

Modeling of wide-band MIMO radio channels based on NLoS indoor measurements

In this paper, we first verify a previously proposed Kronecker-structure-based narrow-band model for nonline-of-sight (NLoS) indoor multiple-input-multiple-output (MIMO) radio channels based on 5.2-GHz indoor MIMO channel measurements. It is observed that, for the narrow-band case, the measured channel coefficients are complex Gaussian distributed and, consequently, we focus on a statistical description using the first- and second-order moments of MIMO radio channels. It is shown that the MIMO channel covariance matrix can be well approximated by the Kronecker product of the covariance matrices, seen from the transmitter and receiver, respectively. A narrow-band model for NLoS indoor MIMO channels is thus verified by these results. As for the wide-band case, it is observed that the average power-delay profile of each element of the channel impulse response matrix fits the exponential decay curve and that the Kronecker structure of the second-order moments can be extended to each channel tap. A wide-band MIMO channel model is then proposed, combining a simple COST 259 single-input-single-output channel model and the Kronecker structure. Monte Carlo simulations are used to generate indoor MIMO channel realizations according to the models discussed. The results are compared with the measured data using the channel capacity and good agreement is found.

Second order statistics of NLOS indoor MIMO channels based on 5.2 GHz measurements

Herein, results from measurements conducted by the University of Bristol are presented. The channel characteristics of multiple input multiple output (MIMO) indoor systems at 5.2 GHz are studied. Our investigation shows that the envelope of the channel for non-line-of-sight (NLOS) indoor situations are approximately Rayleigh distributed and consequently we focus on a statistical description of the first and second order moments of the narrowband MIMO channel. Furthermore, it is shown that for NLOS indoor scenarios, the MIMO channel covariance matrix can be well approximated by a Kronecker product of the covariance matrices describing the correlation at the transmitter and receiver side respectively. A statistical narrowband model for the NLOS indoor MIMO channel based on this covariance structure is presented.

Spatial correlation in indoor MIMO channels

The paper presents the analysis of spatial correlation in MIMO channels, calculated from data measured in office environments at 5.2 GHz. Results are compared with those from channels generated using a stochastic MIMO channel model and the effect of different comparison metrics is shown. The suitability of the stochastic model under different propagation conditions is also investigated. The performance of the models has been shown to be good under non-line-of-sight (NLOS) propagation, although one model was shown to fail under conditions where the correlation amongst the elements of one array was not independent of the antenna element at the other array. It is shown how situations in which array correlation varies with array element can occur in line-of-sight (LOS) conditions, and how under these extreme circumstances, both stochastic models are unsuitable.

A wideband statistical model for NLOS indoor MIMO channels

Herein, results of 5.2 GHz wideband indoor multiple input multiple output (MIMO) channel measurements under the EU IST SATURN project are reported. Our investigation shows that for non-line-of-sight (NLOS) cases, the average power delay profiles fit the exponentially decaying curve quite well, therefore a simple wideband model for single-input single-output (SISO) proposed in COST259 has been used in our model. Furthermore, the investigations show that the MIMO channel covariance matrix of each normalized tap of the impulse response could be well approximated by the Kronecker product of the covariance matrices seen from the transmitter and receiver respectively. Based on the above results, a wideband statistical model is presented. Monte-Carlo simulations show reasonably good agreement between the measured data and our model. Finally, we use this model to show some capacity characteristics of Hiper-LAN/2 channels in NLOS indoor scenarios.

Performance of multiple-receive multiple-transmit beamforming in WLAN-type systems under power or EIRP constraints with delayed channel estimates

Downlink beamforming in a WLAN-type system employing access points and mobiles equipped with multiple antennas and associated receivers and transmitters are considered. The beamforming aims at maximizing the performance under constraints on transmit power or equivalent isotropic radiated power (EIRP). Solutions for the two constraints are derived and investigated using simulated and measured channels. Our simulation and experimental results shows that performance gains of 8-10 dB when using four directional transmitter antennas and two receive antennas (as compared with a base-line one-transmit two-receive), are possible under both constraints. For simulated channels, a delay between channel estimation and use of the same channel of up to 10% of the (inverse of the) Doppler frequency only degrades performance some tenths of a dB. In our measurements, very small degradations are seen with delays of up to 130 ms. The measurements were made under relatively stationary conditions with only occasionally moving people. Two different strategies for updating the beamforming vectors: sounding and ping-pong, are also considered in the paper.

Initial investigation of multiple-input multiple-output (MIMO) channels in indoor environments

Theoretical investigation of multiple-input multiple-output (MIMO) channels has shown the potentially huge increases in spectral efficiency that this architecture can offer. The performance of such MIMO systems is dependent on the propagation environment exhibiting sufficient multipath scattering. This paper presents some initial analysis of MIMO channel measurements taken in indoor environments at 5.2 GHz. Outage capacities are calculated and compared between different locations and for different antenna array configurations. Surprisingly, it is shown that the highest channel capacities were actually calculated from data measured in an empty room.

Initial characterisation or multiple-input multiple-output (MIMO) channels for space-time communication

Previous theoretical studies have demonstrated the substantial capacity or diversity gains possible when multi-element arrays are employed at both a transmitter and receiver. This is only possible when the radio channel exhibits sufficient scattering to induce independent fading at each receive element. An appreciation of both the temporal and spatial variation of such multiple-input, multiple-output (MIMO) channels is therefore required in order to investigate the performance of this architecture in real environments. In this paper we present initial results from indoor MIMO channel measurements taken within the coherence time of the channel. It is shown how capacity analysis of the MIMO channel response matrix alone can be misleading and how the combination of the variation of this with signal to noise ratio in real environments is particularly important.

Downlink beamforming with delayed channel estimates under total power, element power and equivalent isotropic radiated power (EIRP) constraints

A WLAN-type scenario where a base-station (access point) equipped with multiple antennas is transmitting and a mobile with only a single antenna is receiving, is considered. Three approaches are investigated: grid-of-beams (GOB), maximal-ratio (MR) and equal-gain (EG), each with one of the following constraints 1) total transmit power over all antenna elements, 2) maximum power on any antenna element 3) equivalent isotropic radiated power (EIRP). The median diversity gain for the MR approach is estimated to be 14.8 dB, 10.6 dB, and 11.4 dB using constraint 1-3, respectively at 5.2 GHz in a modern university building at 5-50 meter range. For GOB corresponding numbers are 9.4 dB, 9.4 dB, 2.3 dB and for EG 13.9 dB, 13.9 dB, 10.3 dB, respectively. These results are virtually independent of delay between channel estimation and use of the same estimate, when the delay is less than 130 ms. This result is obtained although there were people moving in the environment. The diversity gain under the EIRP constraint is encouraging and shows that coverage improvements are possible even under EIRP limitations.

MIMO detection in analog VLSI

In this paper we propose an analog VLSI approach to maximum a posteriori (MAP) detection in multiple-input multiple-output (MIMO) systems. This detector can be seen as an extension of the well known analog decoding concept for error correcting codes, as it is constructed using similar building blocks. Therefore, it can naturally interact with analog decoders in order to perform turbo detection in MIMO systems. First transistor-level simulations for a small analog MIMO detector in a 0.25mum BiCMOS process agree well with floating-point digital simulations

On the performance of recursive space–frequency codes and iterative decoding in wideband OFDM‐MIMO systems: simulated and measured results

In this paper, we study the performance of a bandwidth efficient space–frequency turbo encoding scheme over wideband channels. Results are presented for simulated wideband MIMO channels consisting of two transmit antennas and up to two receive antennas. In addition, wideband channel measurements undertaken with practical multi-element antenna structures at both the access point (AP) and mobile terminal (MT) are presented. Analysis is in terms of channel capacity, 10% channel outage capacity and space–frequency iterative decoding for an lEEE802.11a physical layer complaint modem. It is shown when operating with a spectral efficiency of 1.2 bits/s/Hz, the iterative decoded space–time codes comes within approximately 4.7 dB of 10% outage capacity over Rayleigh fading wideband channels with two transmit and two receive antennas. Over measured channels the iterative decoding scheme performs within 7.7 dB 10% of outage capacity. Losses due to channel state information estimation are also investigated. Copyright