Report itu-r m. 2038 Technology trends
Download 490.99 Kb.
|
- Navigate this page:
- 3 Spectral efficiency of MIMO systems in isolated links
- 4 Spectral efficiency gains in an interference limited cellular system
- 5 Variation in spectrum efficiency gains using MIMO techniques for different scattering environments
- 6 Variation of MIMO link capacity within a cellular network
2 MIMO antenna schemesIn a typical propagation environment in the bands used for mobile systems there are of course many paths for the RF energy to pass along between the transmitter and receiver antenna elements. In a MIMO link there will therefore be many separate paths between the different pairings of antenna elements and Fig. 15 shows a diagram of some of these paths which may exist between the transmit and receive antenna element arrays. 
For adaptive beam forming techniques the signals to and from the antenna elements are adjusted in phase and amplitude to form a beam to select the best single path between the transmit array and the receive array, while minimizing the antenna gain in the direction of the unwanted interfering signals. With multiple antenna elements at both ends of the link it is also possible to transmit different data streams on some, or all, of the transmit elements and separate the different transmissions by several receivers which each maximize one signal while minimizing the others. This process can give significant capacity advantages over the simple beam forming approach for complex scattering environments and the bounds to the capacity achievable with this type of arrangement in typical mobile propagation environment are derived in [Foschini and Gans, 1998]. The signal handling in these MIMO systems will generally be as shown in Fig. 16. The data streams may be coded independently, which makes for simpler decoding at the receiver [Foschini, 1996], or the full potential of the method can be achieved by including all of the data streams in a two dimensional coding scheme [Foschini, et al., 1999]. The ultimate capacity is achieved when a feedback from the receiver is used to adjust the power split, coding and modulation at the transmitter.
T
 hese MIMO techniques are currently the subject of a great deal of research activity and the following sections are included as general guides to the gains that can be achieved, and the type of propagation environments where these techniques have the greatest potential to increase the capacity of a network. They are not intended as definitive results and more detailed analyses are available in the references given.
3 Spectral efficiency of MIMO systems in isolated linksThe capacity (bit/s/Hz) of an [M,N] MIMO link is given by: where:
We have assumed perfect channel state information at the receiver (in other words, the entries of the matrix H are known exactly. In a rich scattering environment, the entries in H are independent and identically distributed complex Gaussian random variables. Under this condition and the additional condition that r is much larger than one, as the number of antennas gets large and M = N, the capacity approaches: On the other hand, using beam forming the array becomes more directive as M increases, resulting in a linear increase in S/N. Hence the capacity is given by: where hn is the complex amplitude at the n-th receiver. Hence in a sufficiently complex scattering environment the capacity is directly proportional to the number of antennas at each end of the link for the [N,N] MIMO system operating, but only proportional to the log of the number of antennas for phased array beam forming techniques. (See footnote for a grossly simplified explanation of the basis of this effect.)4
Similar related results for an intereference limited mobile network environment can be found in [Catreux et al., 2001]. 
5 Variation in spectrum efficiency gains using MIMO techniques for different scattering environmentsIn order to achieve the potential gains of MIMO systems it is necessary to have a complex scattering environment and so it is essential to quantify the gains for different scattering environments. A simple demonstration of the gains possible is shown in Fig. 18 where the results are shown of some simulations of the spectral efficiency achievable for a single link of a mobile system operating in a range of scattering environments. The simulated environments are modelled as Rician channels with a range of K factors to cover the cases from fairly clear line of sight (K > 10) down to rich scattering environments (K < 0.1). These results were obtained under idealized conditions of perfect channel state information at both the transmitter and receiver at 10 dB S/N. 
6 Variation of MIMO link capacity within a cellular networkIn order to be able to plan the networks which employ MIMO techniques it is necessary to have models for the distribution of capacity for the links to the individual terminals to enable engineering estimates of system capacity, total throughput, and capacity outage probability. This topic is the focus for a substantial amount of research work, e.g. in [Smith and Shati, 2001 and 2002b] which shows that the distribution of MIMO capacity for classical Rayleigh channels tends towards a Gaussian distribution as the number of transmit and receive antennas increases. Simulations in these papers show that the Gaussian approximation to the capacity is reasonably accurate even for systems with three or more antennas at each end of the link and exact expressions for the mean and variance of the Gaussian capacity for any given numbers of transmit and receive antennas are derived in [Smith and Shati, 2001]. It is shown [Shati and Smith, 2002] that the distribution of MIMO capacity for Rician fading conditions can also be expressed as a Gaussian random variable with a similar dependence on the number of transmit and receive antennas. These topics are further developed in [Smith and Shati, 2002a] which examines the MIMO capacity under the influence of various power allocation algorithms at the transmitter and it is shown that the Gaussian approximation still holds good for the different power allocation strategies studied. These included equal power for each antenna, the classical “water filling” technique with perfect knowledge of the channel conditions and a new strategy giving better capacity than the first two for less than ideal estimates of the channel conditions. All of these power allocation methods are shown to give link capacity that can still be approximated by a single Gaussian random variable for both Rayleigh and Rician fading channels. Directory: dms pub -> itu-r -> opb -> rep rep -> Report itu-r m. 2243 (11/2011) rep -> Report itu-r bt. 2053-2 (11/2009) L rep -> Report itu-r sa. 2312-0 (09/2014) rep -> Report itu-r rep -> Report itu-r sm. 2179 (09/2010) Short-range radiocommunication devices measurements rep -> Report itu-r rep -> Report itu-r rs. 2194 (10/2010) rep -> Report itu-r rs. 2165 (09/2009) rep -> Rep. Itu-r bs. 2104 Download 490.99 Kb. Share with your friends:
|
