1. Introduction
Our dependence on wireless communications has reached unprecedented levels. From streaming our favourite entertainment content on our smartphones to relying on the Internet of Things (IoT) for smart homes and businesses. The requirements for faster, more reliable wireless connectivity continue to grow, and as our demands soar, so do the challenges faced by traditional wireless networks.
This quest for faster, more reliable, and energy-efficient networks has led to the emergence of many groundbreaking technologies. Massive multiple-input multiple-output (mMIMO) is one of those technologies, aiming at advancing wireless communication systems by equipping each base station (BS) with hundreds of antenna elements – a significant upgrade from the traditional 4 to 8 antennas. As a result, when using a larger number of antennas, e.g. 128 or 256 per BS, one can attain superior link reliability (seamless connection), and spectral efficiency (enhanced capacity), due to the following reasons:
- Link reliability: By exploiting the spatial diversity gain, the same information streams can be transmitted on several antennas to the same User Equipment (UE).
- Spectral efficiency: By adjusting the spatial multiplexing gain, a hostile severe fading channel (e.g., Rayleigh fading channel) is turned into several independent channels that can transmit multiple information streams simultaneously.
- Energy efficiency: By transmitting much narrower beams towards the intended receiver rather than radiating power across the entire cell, the required transmission energy is dramatically reduced.
Hence, mMIMO is considered as one of the key enabling technologies for future wireless networks and its effectiveness can be further enhanced via intelligent beamforming techniques that exploit multipath information and enable steering signals in the direction of specific user equipment (UE).
In this blog, we explore beamforming strategies crucial for mMIMO, and delve into real-world case studies from London and San Francisco to highlight the role that rigorous and reliable network planning tools network planning software can have in deploying an effective mMIMO-enabled wireless network.
2. Beamforming Strategies for mMIMO
Beamforming is a spatial signal processing technique that utilizes antenna arrays for directional signal transmission and reception. It is achieved by properly controlling the phase and relative amplitude of the same transmitted signal at each antenna through a dedicated device known as a beamformer. In the receiving beamformer, each antenna signal is amplified with appropriate scale-factors or phase-shifts to regenerate the composite signal.
There are two main types of beamforming:
- Digital Beamforming: Each antenna element has its own radio frequency (RF) chain, which means that the signal is processed in the digital domain. The phase and amplitude of the signals are adjusted digitally before being combined and transmitted or received. Digital beamforming allows to precisely control the emitted and received waveforms but can be complex and expensive in terms of hardware requirements.
- Analog Beamforming: Analog beamforming, on the other hand, controls the phase and amplitude of the signals through analog components such as phase shifters. It is often simpler and more cost-effective than digital beamforming but may offer less flexibility and control.
Traditional digital and analog beamforming methods are impractical for mMIMO systems and thus, a hybrid beamforming technique has been identified as a key enabler.
Hybrid beamforming
Hybrid beamforming seamlessly combines the strengths of analog beamformers in the RF domain with the precision of digital beamformers in the baseband. By interlaying a limited number of RF chains between the analog and digital beamformers, the hybrid approach orchestrates a harmonious convergence of cost-effectiveness and peak performance. Notably, the judicious use of phase shifters with constant amplitude in the analogue beamformer plays a key role in simplifying the intricate landscape of mMIMO systems, all while preserving their formidable capabilities.
Figure 1: Beamforming strategy of mMIMO systems
Based on the connection configuration between the analog beamformer and RF chains, there are two widely used hybrid beamforming structures:
- Fully-connected structure: each RF chain is connected to all the antennas via phase shifters.
- Partially-connected structure: each RF chain is connected to a unique subset of antennas via phase shifters. Less phase shifters are required but its performance is inferior.
The capital expenditure (CAPEX) and operational expenditure (OPEX) of a hybrid-beamforming mMIMO system are primarily influenced by the number of RF chains and antenna elements. On the one hand, each dedicated RF chain includes a low-noise amplifier, an analog-digital converter (ADC) and a power amplifier, thus its number directly affects the CAPEX [5]. Additionally, RF chains can be power-hungry, consuming up to 70% of the total transceiver power [6]. In fact, some studies suggest that a BS equipped with 256 RF chains consumes roughly ten times the power of a conventional Long-Term Evolution (LTE) BS [7] equipped with a lower number of antennas. This directly affects OPEX, particularly in terms of energy consumption. On the other hand, the design complexity of analog beamformers is highly related to the number of antenna elements, as high-speed phase shifters are required, and its quantity should be at least equal to the number of antennas [8]. Therefore, the number of antenna elements affects the CAPEX of the mMIMO system.
The performance of a hybrid-beamforming mMIMO system is tied to the number of RF chains and data streams. When the number of RF chains is no smaller than the number of data streams, hybrid beamforming can approach performance levels comparable to fully-digital beamforming baselines, especially in the flat-fading channels.
In light of these considerations, it becomes evident that the beam and antenna selection, encompassing both the total number of antennas and RF chains, have a profound impact on the CAPEX/OPEX, the sustainability and the overall performance of the mMIMO network.
To navigate these complexities and optimize mMIMO network deployment, software tools like Ranplan Professional have become indispensable. They empower network planners to model the entire radio access network topology, simulating with high fidelity cutting edge technologies, such as hybrid beamforming, helping to save on CAPEX and OPEX while enhancing design productivity. These tools are invaluable in achieving the delicate balance between cost-effectiveness and optimal network performance that today's complex wireless ecosystem landscape demands.
3. Five Case Study Examples of mMIMO beamforming strategies
To evaluate the mMIMO beamforming strategies, we constructed 5 realistic scenarios from London and San Francisco and their corresponding digital twins in Ranplan Professional. These scenarios are designed following the guidance from the Ericsson Massive MIMO handbook [9], which are listed as follows:
Figure 2: Site A~E in Ericsson Massive MIMO handbook
Site A: High-traffic (RED), Dense Urban.
Priority: High.
Inter Site Distance (ISD): ~200m.
Distribution of building heights: High.
Site B: Medium-traffic (BLUE) Urban.
Priority: Medium.
ISD: ~350m.
Distribution of building heights: Low.
Candidate site for sharing equipment with other service providers.
Site C: High-traffic (RED) Suburban.
Priority: High.
ISD: ~800m.
Distribution of building heights: Low.
Site D: Low-traffic (GREEN) Dense Urban.
Priority: Medium.
ISD: ~300m.
Distribution of building heights: Medium.
Site E: Low-traffic (GREEN), Suburban.
Priority: Low.
ISD: ~600m.
Distribution of building heights: Low.
For Site A which has a high traffic load and growth, superior mMIMO products are required. Since the distribution of building heights assumes large values, high steerability in the vertical dimension is needed. Thus, a mMIMO system with 64 antennas operating on 3.5 GHz with 100 MHz bandwidth is recommended.
Site B has a similar condition to site A in terms of traffic and traffic growth, but the building height distribution exhibits smaller values. Thus, the benefits of high steerability are not as pronounced and a 32-antenna mMIMO system will yield similar capacity performance, along with lower CAPEX and OPEX.
For Site C, the location represents a suburban area with a rather large ISD, high traffic load and high expectations for traffic growth. The distribution of building heights is low, thus a 32-antenna mMIMO system is recommended.
Site D is an urban area with low ISD. The actual traffic and the expected traffic growth is low. Therefore, a conventional solution is preferred, utilizing an 8-antenna MIMO system, without requiring the deployment of a vast number of antennas.
For Site E which is a suburban area with low traffic load and low traffic growth, a lower cost solution is preferred, compared to Site D (lower cost RRU, less capacity demand from baseband licenses, less energy consumption, less complex and cheaper antenna).
Based on the recommended mMIMO configuration, the downlink user throughput for each scenario is analyzed in Ranplan Professional with the following simulation parameters:
Transmitter Power: 43 dBm
Carrier Frequency: 3.5 GHz
Bandwidth: 100 MHz
Antenna Quantity for Site A~B: 64
Antenna Quantity for Site C~E: 32
Beamforming Strategy: Hybrid beamforming
RF Chain Number: 10
Beamforming Algorithm: Zero-forcing
Minimal Receiver Signal: -85.7 dBm
Signal to Interference Ratio: -20 dB
Figure 3: User downlink throughput of Site A in a suburban area of San Francisco
Figure 3 illustrates the simulated heatmap of the attainable downlink user throughput throughout Site A, i.e., a suburban area of San Francisco. The area exhibits a user density of 2000 users/km2, experiencing heavy traffic load and significant traffic growth. Given these challenging traffic demand conditions, leveraging advanced Massive MIMO products and software features to fully exploit the mid-band potential is recommended, especially considering the area's high average building height, necessitating strong vertical steerability.
Figure 4: User throughput of Site B in a suburban area of London
Fig. 4 illustrates the downlink user throughput of Site B for the individual users within a suburban area of London, where the user density reaches 1000 user/km2. The traffic and traffic growth conditions are similar to those of site A, albeit the lower average building height.
Figure 5 User throughput of Site C in an urban area of San Francisco
Figure 5 depicts the achievable downlink end user throughput within Site C, a suburban area of London, with a user density of 2000 user/km2. The location features a relatively large ISD and experiences a high traffic load, along with high traffic growth expectations. Additionally, the average building height in this location is relatively low.
Figure 6: User throughput of Site D in a suburban area of San Francisco
Figure 6 shows the downlink end user throughput in Site D, which represents an urban area of San Francisco with a user density of 500 user/km2. Compared to Site C, Site D exhibits a relatively low ISD. It is shown that both the current traffic and the anticipated traffic growth in this area are low.
Figure 7: User throughput of Site E in an urban area of London
Fig. 7 shows the downlink user throughput of individual users in Site E, an urban area of London, characterized by a user density of 500 user/km2. This site experiences low traffic volume, large ISD, and minimal expected traffic growth.
4. Conclusion
The deployment of mMIMO offers a transformative opportunity for enhancing downlink user throughput across diverse suburban and urban scenarios. Through meticulous consideration of crucial factors such as CAPEX/OPEX, ISD, the number of antennas, and RF chain configurations, the effectiveness of mMIMO deployments can be significantly optimized to meet the unique demands and building characteristics of each scenario.
By strategically tailoring mMIMO configurations to specific traffic requirements and building characteristics, substantial improvements can be achieved in investment returns. For instance, scenarios with high traffic loads and growth projections, coupled with tall building structures, necessitate advanced mMIMO products with larger antenna arrays and sophisticated beamforming strategies to fully exploit the mid-band potential. On the other hand, scenarios characterized by lower traffic volumes and building heights may benefit from more cost-effective mMIMO solutions with fewer antenna elements and simplified beamforming techniques.
Furthermore, the importance of leveraging advanced network planning software, such as Ranplan Professional, cannot be overstated. These tools empower network operators and planners to model complex network topologies and optimize mMIMO deployment strategies, thereby striking a delicate balance between cost-effectiveness and network performance. By emulating high fidelity digital twins of wireless networks equipped with various mMIMO configurations and analysing downlink user throughput under different simulation parameters, network planners can make informed decisions to maximize the efficiency and effectiveness of mMIMO-enabled network deployments.
In essence, the successful deployment of mMIMO hinges upon a comprehensive and in-depth understanding of the unique characteristics and requirements of each deployment scenario, coupled with the strategic utilization of advanced technologies and software tools. By adopting a tailored approach to mMIMO deployment, wireless network operators can unlock the full potential of mMIMO technology, delivering enhanced connectivity and user quality of experience and meeting the ever-growing demands of today's wireless ecosystem.
References
[1] Ranplan, “5G New Radio Network Planning Whitepaper”, December 2018.
[2] J. Navarro-Ortiz, et al., "A Survey on 5G Usage Scenarios and Traffic Models," in IEEE Communications Surveys & Tutorials, vol. 22, no. 2, pp. 905-929, Second quarter 2020
[3] T. E. Bogale and L. B. Le, "Beamforming for multiuser massive MIMO systems: Digital versus hybrid analog-digital," 2014 IEEE Global Communications Conference, 2014, pp. 4066-4071 .
[4] A. F. Molisch et al., "Hybrid Beamforming for Massive MIMO: A Survey," in IEEE Communications Magazine, vol. 55, no. 9, pp. 134-141, Sept. 2017.
[5] Z. Jiang, S. Chen, S. Zhou and Z. Niu, "Joint User Scheduling and Beam Selection Optimization for Beam-Based Massive MIMO Downlinks," in IEEE Transactions on Wireless Communications, vol. 17, no. 4, pp. 2190-2204, April 2018.
[6] O. El Ayach et al., “Spatially Sparse Precoding in Millimeter Wave MIMO Systems,” IEEE Trans. Wireless Commun., vol. 13, no. 3, Mar. 2014, pp. 1499–1513.
[7] S. Han, C. -l. I, Z. Xu and C. Rowell, "Large-scale antenna systems with hybrid analog and digital beamforming for millimeter wave 5G," in IEEE Communications Magazine, vol. 53, no. 1, pp. 186-194, January 2015.
[8] S. Payami, M. Ghoraishi and M. Dianati, "Hybrid Beamforming for Large Antenna Arrays With Phase Shifter Selection," in IEEE Transactions on Wireless Communications, vol. 15, no. 11, pp. 7258-7271, Nov. 2016.
[9] Ericsson, “Massive MIMO handbook”, https://www.ericsson.com/en/ran/massive-mimo, 2023.
