5G, as the next-generation mobile communication technology, has a network architecture, capabilities, and requirements that differ substantially from previous generations. Many technologies are integrated into the standard, reflecting the breadth and depth of modern mobile wireless communications.
Defined 5G NR Use Cases
- Enhanced mobile broadband (eMBB): Targets higher peak and average data rates, larger capacity, and broader coverage compared with legacy mobile broadband. eMBB is specified to enable downlink data rates up to 20 Gbps and uplink data rates up to 10 Gbps.
- Massive machine-type communications (mMTC): Supports billions of connected devices and sensors for IoT use cases. This covers devices with infrequent transmissions and long battery life with low data rates and narrow bandwidth, as well as devices with large bandwidth and high data rates.
- Ultra-reliable low-latency communications (URLLC): Focuses on real-time communications that must avoid failures, such as autonomous driving, industrial IoT, smart grid, infrastructure protection, and intelligent transport systems.
Waveform Improvements
OFDM provides strong multipath mitigation, high spectral efficiency, and relatively simple implementation, and it has been widely used in LTE and LTE-A systems. However, the baseband waveform can be vulnerable to interference. To achieve very high data rates, 5G requires bandwidths on the order of gigahertz, while continuous contiguous spectrum is harder to obtain at lower frequency ranges. The figure below shows the waveform evolution from 2G to 5G.
5G waveforms build on OFDM by adding filtering. Filtered approaches such as FBMC apply filters to each subcarrier, reducing out-of-band emissions. Simulation comparisons between 4G and 5G waveforms show that FBMC can significantly improve adjacent-channel metrics.
Because 5G will be applied to large-scale IoT, with billions of interconnected devices, multiplexing efficiency must improve to handle massive deployments. To prevent adjacent bands from interfering with each other, in-band and out-of-band signal radiation must be minimized.
Massive MIMO
MIMO was already applied in 4G, mainly at the base station side. Terminals could receive multiple streams but typically transmitted on a single path. The figure below illustrates typical 4G MIMO deployments.
In 5G, ultra-high download rates are largely enabled by MIMO, using simultaneous transmission of multiple spatial streams to multiply throughput.
Massive MIMO is considered a key and feasible technology for 5G networks, but implementation has constraints. For example, if orthogonal pilot sequences are used within a cell but the same pilot sets are reused across cells, pilot contamination can occur. This contamination prevents uplink and downlink signal-to-interference-plus-noise ratio from improving proportionally with increased base station antenna count. Deploying large antenna arrays at base stations also increases cost, and in practical scenarios these systems must adapt flexibly to complex antenna propagation environments, which is a technical challenge.
Full-Duplex Technology
Early wireless systems used FDD or TDD. Full-duplex, which allows simultaneous transmission and reception on the same time and frequency resources, has been considered in 5G discussions.
Full-duplex implies DL and UL transmissions occur simultaneously on the same time/frequency resource. Based on the isolation between UL and DL, deployment scenarios can be grouped as follows:
- Case 1: Both the base station and the UE have self-interference cancellers.
- Case 2: The base station has a self-interference canceller, but the UE does not.
In Case 1, self-interference can arise, for example, from a UE's UL transmission leaking into its DL reception or from DL transmissions interfering with the base station's UL reception. The network may display a single UE in a simplified view, but depending on scheduling and MIMO capabilities, multiple UEs may be scheduled in a MU-MIMO-like full-duplex configuration.
In Case 2, UEs do not need cancellers, which reduces UE complexity. UE 1 and UE 2 lack full-duplex (self-interference) cancellers, while the base station has a canceller. Base station scheduling can avoid or reduce interference from UE 2 to UE 1. At the base station, self-interference can be suppressed using its own canceller.
Comparing Case 2 with Case 1, Case 2 imposes lower complexity requirements on the UE. However, Case 1 enables more full-duplex operations regardless of transmitter/receiver capabilities at the UE, giving the base station greater scheduling flexibility. If two UEs are scheduled in the same subframe, Case 1 can achieve up to roughly twice the throughput of Case 2 under ideal conditions.
NR Operating Bands
| NR Operating Band | Uplink (UL) operating band / BS receive / UE transmit | Downlink (DL) operating band / BS transmit / UE receive | Duplex Mode | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| FUL_low – FUL_high | FDL_low – FDL_high | ||||||||||
| n1 | 1920 MHz – 1980 MHz | 2110 MHz – 2170 MHz | FDD | ||||||||
| n2 | 1850 MHz – 1910 MHz | 1930 MHz – 1990 MHz | FDD | ||||||||
| n3 | 1710 MHz – 1785 MHz | 1805 MHz – 1880 MHz | FDD | ||||||||
| n5 | 824 MHz – 849 MHz | 869 MHz – 894 MHz | FDD | ||||||||
| n7 | 2500 MHz – 2570 MHz | 2620 MHz – 2690 MHz | FDD | ||||||||
| n8 | 880 MHz – 915 MHz | 925 MHz – 960 MHz | FDD | ||||||||
| n20 | 832 MHz – 862 MHz | 791 MHz – 821 MHz | FDD | ||||||||
| n28 | 703 MHz – 748 MHz | 758 MHz – 803 MHz | FDD | ||||||||
| n38 | 2570 MHz – 2620 MHz | 2570 MHz – 2620 MHz | TDD | ||||||||
| n41 | 2496 MHz – 2690 MHz | 2496 MHz – 2690 MHz | TDD | ||||||||
| n50 | 1432 MHz – 1517 MHz | 1432 MHz – 1517 MHz | TDD | ||||||||
| n51 | 1427 MHz – 1432 MHz | 1427 MHz – 1432 MHz | TDD | ||||||||
| n66 | 1710 MHz – 1780 MHz | 2110 MHz – 2200 MHz | FDD | ||||||||
| n70 | 1695 MHz – 1710 MHz | 1995 MHz – 2020 MHz | FDD | ||||||||
| n71 | 663 MHz – 698 MHz | 617 MHz – 652 MHz | FDD | ||||||||
| n74 | 1427 MHz – 1470 MHz | 1475 MHz – 1518 MHz | FDD | ||||||||
| n75 | N/A | 1432 MHz – 1517 MHz | SDL | ||||||||
| n76 | N/A | 1427 MHz – 1432 MHz | SDL | ||||||||
| n78 | 3300 MHz – 3800 MHz | 3300 MHz – 3800 MHz | TDD | ||||||||
| n77 | 3300 MHz – 4200 MHz | 3300 MHz – 4200 MHz | TDD | ||||||||
| n79 | 4400 MHz – 5000 MHz | 4400 MHz – 5000 MHz | TDD | ||||||||
| n80 | 1710 MHz – 1785 MHz | N/A | SUL | ||||||||
| n81 | 880 MHz – 915 MHz | N/A | SUL | ||||||||
| n82 | 832 MHz – 862 MHz | N/A | SUL | ||||||||
| n83 | 703 MHz – 748 MHz | N/A | SUL | ||||||||
| n84 | 1920 MHz – 1980 MHz | N/A | SUL | ||||||||
Network Slicing
Network slicing partitions a physical operator network into multiple virtual networks, each adapted to different service requirements. Slices can be differentiated by latency, bandwidth, security, and reliability to meet specific scenarios. By creating multiple logical networks on a single physical infrastructure, network slicing avoids the need to build a dedicated physical network for each service, reducing deployment costs.
Within the same 5G network, operators can create slices for services such as intelligent transport, drones, telemedicine, smart home, and industrial control, and expose those slices to different service providers. Each slice can provide different guarantees in terms of bandwidth and reliability, as well as distinct billing and management schemes. Unlike 4G, where all services share the same network and service characteristics, 5G network slicing can provide differentiated networks, management, services, and billing so that service providers can better utilize 5G capabilities.
