The Effect Of Antenna Layout On MIMO Ergodic Capacity

With the use of multiple antennas, the space becomes a new resource to be exploited towards increasing the capacity of the communication systems. However, in practical applications, the space allocated to the transmitter and the receiver is in general limited. In a limited region, different antenna layout may have different antenna correlation and hence the different performance.In this paper the main innovation points are as follows.1. A spatial antenna channel model based on Jakes fading model with the power azimuth spectrum (PAS) is introduced and the expression of spatial correlation is given.2. The channel gain of SIMO system, one side correlated MIMO system and both side correlated MIMO system are deduced and simulated.3. The MIMO ergodic capacity with various typical antenna layout cases of fixed antenna region diameter and area have been simulated and compared respectively. The results show that the circular antenna layout case has the best performance under the condition of both diameter constraints and area constraints. Hence, when the antenna spacing region is limited, dispersing antennas at the edge of the antenna spacing region approaching circular layout will get the best performance especially when angle spread is serious and the antenna region diameter is smaller than a wavelength. If antenna layout is not centre-symmetric, the antenna region diameter should be vertical with the LOS signal to get a better performance except when the rays are uniformly distributed among all directions.4. The paper studies the optimal antenna layout which has the best MIMO capacity through simulated annealing algorithm when maximum antenna distance is limited as dmax. Because of the complexity and uncertainty of the Monte-Carlo's ergodic capacity, we use ergodic capacity bounds instead of ergodic capacity as the SA cost function. The difference between uniform layout capacity and optimal layout capacity is compared. Results show that when dmax /λ≤0.5, the optimal layout is uniform layout in most cases; but optimal layout capacity may undergo deep loss in some cases. The optimal layout capacity will reach the maximum as dmax/λ=1 of 8×8-2D, dmax/λ=0.8 of 4×4-2D. In antenna space larger than these sizes, antennas should also be placed as the optimal layout at the right dmax5. The studies extend to 3D space. The paper studies the optimal antenna layout which has the best MIMO capacity through simulated annealing algorithm when maximum antenna distance is limited as dmax in 3D space. The difference between uniform layout capacity and optimal layout capacity is compared. Results show that the optimal layout capacity will reach the maximum as 8×8-3D and dmax/λ=0.5 of 4×4-3D. In antenna space larger than these sizes, antennas should also be placed as the optimal layout at the right dmax...