Tag Archives: Correlation

Fundamentals of Direction of Arrival Estimation

Direction of Arrival (DOA) estimation is a fundamental problem in communications and signal processing with application in cellular communications, radar, sonar etc. It has become increasingly important in recent times as 5G communications uses DOA to spatially separate the users resulting in higher capacity and throughput. Direction of Arrival estimation can be thought of as the converse of beamforming. As you might recall from the discussion in previous posts, in beamforming you use the steering vector to receive a signal from a particular direction, rejecting the signals from other directions. In DOA estimation you scan the entire angular domain to find the required signal or signals and estimate their angles of arrival and possibly the ranges as well.

Continue reading Fundamentals of Direction of Arrival Estimation

Massive MIMO and Antenna Correlation

Some Background

In a previous post we calculated the Bit Error Rate (BER) of a Massive MIMO system using two different channel models namely deterministic and probabilistic. The deterministic channel model is derived from the geometry of the array (ULA in this case) and the distribution of users in the cell. Whereas probabilistic channel model assumes that the channel is flat fading and can be modeled, between each transmit receive pair, as a complex, circularly symmetric, Gaussian random variable with mean of zero and variance of 0.5 per dimension.

Continue reading Massive MIMO and Antenna Correlation

Building an LTE Channel Simulator

As discussed previously building an LTE fading simulator is a three step procedure.

1. Generate a temporally correlated Rayleigh fading sequence. This step would be repeated for each channel tap and transmit receive antenna combination e.g. for a 2×2 MIMO system and EPA channel model with 7 taps the number of fading sequences to be generated is 4×7=28. The temporal correlation of these fading sequences is controlled by the Doppler frequency. A higher Doppler frequency results in faster channel variations and vice versa.

2. Introduce spatial correlation between the parallel paths e.g. for a 2×2 MIMO system a 4×4 antenna corelation matrix would be used to introduce spatial correlation between the 4 parallel paths h11, h12, h21 and h22. This can be thought of as a weighted average. A channel coefficient between Tx-1 and Rx-1 would be calculated as h11=w1*h11+w2*h12+w3*h21+w4*h22. In this case the weight ‘w1’ would have a value of 1 whereas the other weights would have a value less than 1. If w2=w3=w4=0 there is no correlation between h11 and other channel coefficients.

2-Transmit 2-Receive Channel Model
2-Transmit 2-Receive Channel Model

3. Once the sequences with the desired temporal and spatial correlation have been generated their mean power would have to be adjusted according to the power delay profile of the selected channel model (EPA, EVA or ETU). The number of channel coefficients increases exponentially with the number of transmit and receiver antennas e.g. for a 4×4 MIMO system each filter tap would have to be calculated after performing a weighted average of 16 different channel taps. And this step would have to be repeated for each filter tap resulting in a total of 16×7=112 fading sequences.

We have already discussed step 1 in detail. We would now elaborate on step 2 i.e. generation of spatially correlated fading sequences.