top of page
Writer's pictureHamza Bouchebbah

Deep dive into Beamforming and Massive MIMO

Updated: 6 hours ago


Beamforming is an effective technique that enables precise directional coverage while significantly minimizing interference and noise from other directions.


Beamforming & MIMO are a crucial aspect of modern communication systems in 5G, where significant challenges like increased path loss and throughput demands.

Massive MIMO and beamforming techniques are extensively used to enhance coverage, capacity, and spectral efficiency.


Antenna array patterns


  • Antenna elements spacing


The physical size of an antenna array depends on the spacing between the individual antenna elements and the number of AE which are important to maximize gain and directivity.

A linear array will consist of equally spaced antenna elements on horizontal or/and vertical straight so considering the distance between AE about half of the wavelength it is considered to be inversely proportional to the frequency :


FR1 --> for example on 3 Ghz --> λ= c/f​ = 10 cm. --> λ/2= 5 cm .

FR2 --> for example for 30 Ghz -->λ= c/f​ = 10 mm --> λ/2= 5 mm .


Increasing the distance between antenna elements will result in higher gain and performance; however, increasing the spacing between antenna elements 

above > λ/2 results in additional lobes called grating lobes.


Generally, having a λ/2 spacing between antenna elements is a good way to balance coverage and interference, however, using a larger spacing like 2λ could be a viable option in certain situations. This would help reduce the half-power beam width and increase the gain of the main lobe with the trade-off of spreading power more widely away from the main lobe (interference). This approach may be suitable, for example, in MU-MIMO scenarios with fewer antenna elements narrower beams are essential for efficient beam steering and UE discrimination


Beamforming relies on constructive interference between the transmitted signals by each antenna element, and an array composed of a higher number of antenna elements provides additional gain and relatively narrow main lobes the beamforming gain could be modeled in the case of the lossless array as


Array Gain (dB) = Element gain + 10 log10 (N) where N: represents the number of array elements.


we have simulated the following linear array of dipole antenna using the Matlab antenna toolbox ( Figure 1 ) which shows the change of pattern, gain, and directivity as the number of antenna elements increases for 4, 8,16, and 32.


Figure 1: Plot using different antenna element configuration


beamforming linear array of dipole antenna simulated using  the Matlab antenna toolbox

 



  • Beam Scanning


Assuming a phased array of a fixed number of antenna elements (AEs) positioned fixedly;

by manipulating the RF signal phases and amplitudes supplied to these elements, it becomes feasible to dynamically modify and mold the array factor (which reflects the collective impact of signals from all antenna elements). This modification leads to a particular radiation pattern that concentrates energy in preferred directions while reducing it in others.

This is the foundation of beamforming, where the signal directionality is dynamically adjusted without physically moving the antennas. Instead, signal steering is achieved by precisely controlling the phases and amplitudes at each antenna element.


However, for beam scanning 2 characteristics to be considered :


Scan Loss: As the beam of a phased array antenna is steered away from the boresight direction Θ = 0 the antenna's effective aperture gain decreases, this is known as scan loss which follows the cosine law with k a numeric value, typically in the 1.3 range, which accounts for the non-ideal isotropic behavior of the embedded element gain :


So for around 30-degree scanning, for example, a drop in directivity of about 0.9 dB is to be expected note to be considered for SSB sweeping.



Massive MIMO AE mapping


Antenna elements can be used for beamforming or transmitting multiple streams, or a combination of both as described in Figure 3 with a reduced number of antenna elements available for beamforming,


Figure 3: Massive MIMO 32 AE mapping

massive MIMO beamforming design


On high frequencies especially in the FR2 range suffer from significant path loss and penetration loss so large-scale antenna arrays may be an excellent alternative for enhancing the coverage as simulated in Figure 1 however when the antenna element number is large relying on simple digital beamforming may not be an appropriate option due to cost power consumption and design complexity so hybrid analog-digital beam-forming, as shown bellow could be an excellent option to reduce cost and complexity of 5G equipment so RF Chain will not be one to one mapping this is achieved by mapping each RF Chain only to the antenna elements belonging to the same subarray and this is achieved by dividing antenna elements into multiple subarrays.


Analog beamforming

Analog beamforming

Digital beamforming

Digital beamforming

Hybrid beamforming

Hybrid beamforming

The hybrid beamforming system could be modeled before rx beamforming as described in Figure 4 :

Y=H*P*W*S+n


  • S: Modulated symbole or data stream

  • W: Represents digital precoding matrix applied in the baseband,

  • P: Represents analog beamforming matrix applied via phase shifters

  • H: Represents the wireless channel coefficient matrix.

  • n: noise added to the received signal

Figure 4: Hybrid beamforming

Hybrid beamforming model

To tackle the beamforming challenges associated with rising costs and energy usage, the earlier approach combined digital beamforming at the baseband level with analog beamforming in the RF domain as each RF chain is digitally controlled and mapped to a single subarray .




Hope to see you all again and if you have any questions please reach out to me at the following Email address or you can drop a comment below






10 views0 comments

Comments


bottom of page