Compatibility Analysis Between RedCap Non-Public Networks and 5G NR in TDD FR1 and FR2 Bands

Abstract

RedCap technology is set to become a critical component in the growth of the Internet of Things (IoT), enabling sensors and wearable devices for medical, industrial, and commercial applications. However, because RedCap primarily operates in non-public networks and does not synchronize its time-division duplexing (TDD) mode with 5G NR networks, interference risks arise. This is particularly concerning as traditional 5G NR networks prioritize downlink communication, whereas RedCap is designed for uplink. This study investigates the potential interference between RedCap non-public networks and 5G NR in TDD FR1 and FR2 frequency bands using Monte Carlo simulation techniques. The results illustrate how RedCap deployments may impact 5G NR performance in urban and suburban environments. Key insights are provided to inform strategies for minimizing interference and ensuring coexistence between these technologies.

1. Introduction

The need to deploy massive machine-type communication (mMTC) and massive Internet of Things (mIoT) services in 5G NR networks has driven the development of RedCap technology. Standardized in Release 17 of the 3GPP Partnership Project, RedCap aims to enhance connectivity for IoT devices and enable seamless machine-to-machine communication [1,2,3,4]. RedCap is strategically positioned as a technology that refines the capabilities of the 5G NR (New Radio) air interface established in Releases 15 and 16 of 3GPP. Its primary objective is to streamline the complexity of 5G NR subscriber devices by leveraging a simplified NR Light air interface. This interface aligns with the standard requirements of Internet networks, bridging the gap between NB-IoT (4G) and URLLC (ultra-reliable low-latency communication).

This article delves into the distinctive features of RedCap technology, highlighting disparities between RedCap-equipped devices and conventional 5G NR solutions. Additionally, it explores the potential prospects for the technology’s evolution and application in Internet networks.

2. Standardization and Market

The 3GPP Partnership Project’s Release 17 phase introduced the standardization of RedCap, a technology designed to deliver simpler and more cost-effective Internet of Things (IoT) devices for battery-powered applications. In the June 2019 meetings, working group members presented multiple proposals to address 5G NR requirements and use cases for Internet of Things services, culminating in the introduction of NR-Light technology. This technology, an offshoot of 5G NR, involves reduced-frequency channel bandwidth, fewer MIMO layers, and diminished downlink modulation levels to simplify User Equipment (UE) complexity [5,6,7]. Evolving from its initial moniker, NR-Light, the technology is now officially recognized as RedCap and has been incorporated into the 3GPP Release 17 technical specification series as of June 2022. Figure 1 illustrates the ongoing and upcoming phases of the 3GPP Partnership Project’s work on RedCap technology [8,9].

The standardization efforts within Release 18 encompass two phases, with the final phase slated for completion in the second quarter of 2024. RedCap technology, standardized in 3GPP Release 18, promises to further reduce subscriber terminal complexity, enabling support for a broader range of data rates. This expansion includes the introduction of positioning, sideline protocols, and the utilization of unlicensed spectrum for diverse use cases [10].

RedCap technology’s market trends showcase commercial testing by over 10 mobile operators across seven countries. Projections indicate that RedCap’s subscriber device connections in 5G NR IoT networks will surpass 100 million within the next three years, driving the development of novel 5G NR applications.

China Mobile, China Telecom, and China Unicom, the three largest mobile operators, have implemented extensive RedCap equipment deployments in over 10 cities, spanning industries such as industrial manufacturing, power generation, and V2X intelligent transport networks.

Anticipated to cost less than USD 10, RedCap subscriber modules in the USA are priced 80% lower than their 5G NR counterparts designed for Enhanced Mobile Broadband (eMBB) services. The MBBF-2023 forum showcased over 10 types of RedCap subscriber modules, including RedCap, DTU, and CPE modules. Qualcomm’s release of the Snapdragon X35 5G NR Modem-RF system in 2023 marked an industry milestone as the first 5G NR-Light (RedCap) modem. Supporting up to two antennas and delivering a peak data rate of 220 Mbps with a 20 MHz channel bandwidth in the FR1 sub-band, this modem represents a significant advancement. Devices featuring RedCap technology are poised to offer tenfold more Internet of Things network capacity than 4G technologies, with RedCap devices consuming 20% less energy than comparable 4G counterparts (NB-IoT and LTE-M).

3. Key Limitations of RedCap’s Technological Capabilities

RedCap technology, also known as NR-Light, represents a breakthrough in creating low-cost, simplified IoT devices. It extends the benefits of 5G NR to battery-powered applications while maintaining efficiency and reducing complexity. Pioneering a departure from the preceding reliance on extant 4G technologies such as NB-IoT or LTE-M for IoT services on 5G NR networks, RedCap technology introduces technical advancements primarily manifested at the Radio Resource Control (RRC) [TS 38.331] and Packet Data Convergence Protocol (PDCP) layers of the 5G NR network, with discernible implications for Internet of Things (IoT) and massive machine-type communication (mMTC) services.

An exploration of the principal innovations within RedCap technology reveals nuanced modifications affecting the throughput, receiving channels, Multiple Input Multiple Output (MIMO) levels in the downlink, downlink modulation order, and duplex operation modes of the subscriber device.

In terms of maximum throughput, the NR basic subscriber device, operating within frequency channels up to 100 MHz wide in the FR1 sub-band (410–7125 MHz) and up to 200 MHz in the FR2 sub-band (24.250–52.600 GHz) for transmission and reception, undergoes a reduction in RedCap technology to 20 MHz and 100 MHz, respectively. This reduction, while constraining frequency channel width, accommodates the reuse of physical channels and signals specified for initial data collection, thereby mitigating adverse impacts on network and device deployments.

RedCap technology minimizes the number of receiving channels required, reducing reliance on antennas and lowering equipment costs.

The maximum number of MIMO levels in the downlink for a RedCap subscriber unit aligns with the number of receiving channels it supports (2 × 2 MIMO DL and 1 UL SISO), deviating from the more expansive capabilities of a basic 5G NR subscriber device. This reduction in MIMO levels is underscored by a comparative analysis against the conventional requirements for a basic 5G NR subscriber device, which may support up to eight levels of Single-User (SU) MIMO.

Considering downlink modulation orders, the RedCap device introduces flexibility by rendering the support for 256QAM in the downlink optional, in contrast to the mandatory requirement for the NR base unit. The uplink for both FR1 and FR2 mandates 64QAM support for the RedCap device, mirroring the baseline device.

In the realm of duplex operation modes, the focal simplification pertains to RedCap’s operation in Frequency Division Duplex (FDD) bands. Unlike the 5G NR base subscriber device, which must support full duplex (FD) operation, including simultaneous transmission and reception at distinct duplex frequencies, the RedCap device exhibits streamlined operation within the FDD bands. This simplification arises from the elimination of the need for simultaneous transmission and reception, resulting in a more straightforward operational framework for RedCap technology in FDD bands. The nuanced intricacies of full-duplex devices, involving the utilization of duplex filters to mitigate interference, are notably alleviated in RedCap’s operational context. The forthcoming investigations in Release 18 aim to refine the specifics of RedCap devices in frequency bands mandating at least four receive channels, addressing ongoing considerations related to the number of required receiving channels in this context.

The abovementioned limitations are similar to those that existed in LTE IoT standards such as NB-IoT or LTE-M; at the same, time there is another issue that arose particularly for RedCap since 5G NR, unlike LTE, is a more Time Division Duplex (TDD)-focused technology. In TDD, it is important to distribute time slots between DL and UL and to synchronize it to avoid inter-link interference between different networks [10] due to the advent of extended MTC in non-public networks (NPNs) within industrial settings. Here, data rate requirements shift from the downlink to the uplink (or at least become symmetrical). While conventional smartphones with human-driven applications emphasize data in the downlink direction, emerging UE types like industrial sensors or closed-circuit television (CCTV) cameras prioritize data transmission in the uplink direction. Consequently, operators may need to reconsider their prevailing TDD UL/DL configuration, historically inclined towards downlink. The dynamic nature of the 5G NR TDD UL/DL configuration permits operators to configure a non-public network (e.g., within a factory) with symmetric UL/DL resources, while outside the factory, downlink resources dominate. This can potentially lead to harmful interference from the private networks that used RedCap to the public networks in the FR1 and FR2 frequency ranges. 5G NR public networks rely highly on the frequency bands n77 and n78 (mostly 3400–3800 MHz) and on band n258 (24.25–27.5 GHz). In this study, we try to estimate the degree to which RedCap private networks’ deployment in the urban environment may affect public TDD 5G networks in the FR1 and FR2 frequency bands.

4. State of the Art

At present, there are no studies that show how the deployment of RedCap may affect the 5G NR networks. There are some studies that show how 5G NR, including IoT, may affect these. Several studies show how 5G NR, including IoT or a low-power wide-area network (LPWAN), may affect other radiocommunication services; for example, [11] studies the impact of 5G NR IoT on the microwave stations operating in the fixed services. In [12], there is an analysis of how IoT of LPWAN may affect radiolocation services2. In [13], the compatibility between 5G NR that uses active antenna systems and LTE, including IoT, is studied, and [14] analyzes the compatibility between 5G networks in a cross-border scenario when TDD is not synchronized, thus partly exploring a similar problem. Thus, while similar studies were conducted, no study has shown how RedCap will affect the current 5G NR public networks.

5. Simulation Parameters

To simulate the interference case between private RedCap systems and public 5G NR networks in the FR1 and FR2 bands, typical characteristics of RedCap and 5G NR were used. The RedCap characteristics were derived from 3GPP specifications, whereas the 5G NR characteristics were derived from the ITU-R documents that were available during the study cycles of the Working Party 5D, and common methodologies for interreference analysis were used [15,16].

The implementation of various IoT standards, such as LPWAN, NB-IoT, LTE-M, and others, demonstrated that the density of these devices can escalate to tens of devices per square kilometer. This proliferation could result in the presence of hundreds of thousands or even millions of potential interfering IoT devices. It is essential to consider the activity factors of these devices when scaling these numbers [17].

In our study, we specifically examine the scenario where 10,000 interferers operate simultaneously; 95% of them are located indoors given that we are studying private networks of RedCap for industrial applications [18]. These interferers are randomly dispersed within the confines of the 5G New Radio (NR) public network area [19,20]. This controlled approach allows us to analyze and assess the potential impact of a considerable number of concurrent interfering devices, providing valuable insights into the challenges associated with IoT device density within the 5G NR environment.

Table 1. RedCap devices’ characteristics in the FR1 and FR2 bands.

Parameter	FR1	FR2
Bandwidth	5/10/15/20 MHz	100 MHz
Deployments	in NR channel (SA only) no min frequency band requirement
Maximum Data Rate	DL: 2 Mbps up to 150 Mbps UL: 2 Mbps up to 50 Mbps
Modulation	Up to 64QAM	Up to 64QAM
Antennas	1 (or 2 RX MIMO)	1 (or 2 RX MIMO)
Device Transmit Power	Class = 3, +23 dBm	Class = 7 TRP: +23 dBm EIRP: +43 dBm
Duplex Mode	FDD Half or Full, TDD

provides the characteristics of the RedCap devices.

5G NR networks for the FR1 and FR2 bands are simulated for the urban deployment.

Table 2. Base station characteristics of the victim 5G NR networks in the FR1 and FR2 bands.

Parameter	FR1	FR2
Antenna height	30 m	6 m
Cell radius	400 m	50 m
Element gain (dBi)	6.4	5
Horizontal/vertical 3 dB beam width of single element (degree)	90° for H 65° for V	65° for both H/V
Horizontal/vertical front-to-back ratio (dB)	30 for both H/V	30 for both H/V
Antenna polarization	Linear ± 45°	Linear ± 45°
Antenna array configuration (Column × Row)	4 × 8 elements	8 × 8 elements
Number of element rows in sub-array, Msub	3	3
Array Ohmic loss (dB)	2	3
Conducted power (before Ohmic loss) per antenna element/sub-array (dBm)	28	10
Base station horizontal coverage range (degrees)	±60	±60
Base station vertical coverage range (degrees)	90–100	90–100
Mechanical downtilt (degrees)	6	10

and

Table 3. User equipment characteristics of the victim 5G NR networks in the FR1 and FR2 bands.

Parameter	FR1	FR2
Indoor user terminal usage	70 m	50 m
Indoor user terminal penetration loss	Rec. ITU-R P.2109
User equipment density for terminals that are transmitting simultaneously	3 UEs per sector
UE height	1.5 m
Typical antenna gain for user terminals	−4 dBi	Rec. ITU-R M.2101 4 × 4 elements with 5 dBi one element
Body loss	3 dB	5 dB
Power control model	Rec. ITU-R M.2101
Maximum user terminal output power, PCMAX	23 dBm	22 dBm
Power (dBm) target value per RB, P0_PUSCH	−92.2	−95
Path loss compensation factor	0.8	1

provide the characteristics of the 5G NR networks in the FR1 and FR2 bands.

Figure 2 presents the antenna pattern of the base station that operates in the FR1 mid-band [21,22,23].

Figure 3 presents the antenna pattern of the base station and user equipment that operate in the FR2 millimeter wave band [16,18].

6. Simulation Methodology

The Monte Carlo method based on the generation of random samples according to ITU-specified probability distributions [19] was used in the study. In interference estimation, we used various random parameters, such as the location of interference sources, their power, direction of antenna patterns, propagation range of signals, etc.

After performing a large number of iterations (10,000 iterations in this study), the resulting mean values, probability distributions and other characteristics help to estimate the probabilities, statistical characteristics and behavior of the system under random impacts.

The Round-Robin model was applied for the victim 5G NR network simulation. This method is used to allocate resources or tasks among several devices or processes in a cyclic order. When devices or subscribers request access to resources, the Round-Robin algorithm allocates access in turn, ensuring that resources are utilized evenly among the participants. This can be useful when resources need to be divided equally among multiple users or devices.

The OFDMA network modeling algorithm applied assumes a full-system load of 100% with full buffer traffic and 1/1 frequency reuse (i.e., a single-frequency network), and takes into account intra-system interference in the reference cell due to UEs located in neighboring cells using the same resource blocks, as well as interference from UEs located in the reference cell using different resource blocks. The methodology assumes that UEs are located randomly throughout the network area according to a homogeneous geographical distribution. The channel model of the public 5G NR networks was in accordance with 3GPP TR 36.821: Solutions for NR to Support Non-Terrestrial Networks (NTNs). Figure 4 shows the simulation of RedCap interference (red dots) to the downlink and uplink of the 5G NR networks, where blue dots are victim receivers and green dots are victim transmitters.

To calculate the throughput loss of uplink and downlink of 5G NR, it is primarily required to calculate the signal-to-noise (SNR) ratio of 5G NR links and interference (I) from RedCap transmitters. Then, the RedCap interference level should be added to the noise level of the 5G NR victim receiver to calculate the signal-to-interference-plus-noise (SINR) ratio (see 3GPP TR 38.901 Study on channel model for frequencies from 0.5 to 100 GHz. Release 17). The obtained SINR levels should be compared with the curves in Figure 4 and throughput loss can be calculated using the following equation:

$$\mathrm{Throughput} \left[\mathrm{Mbit}/\mathrm{s}\right]={\displaystyle \frac{N_{RB\_per\_UE}}{N_{total\_RBs}}} ·S_{BitRate} \left(SINR\right) ·B$$

(1)

where:

BitRate: maximum bitrate in the channel, in bps;

N_{RB_per_UE}: number of research blocks per user;

N_{total_RBs}: total number of resource blocks;

B: channel bandwidth, in MHz;

S_BitRate: bitrate depending on SINR in bps/Hz.

As per 3GPP TR 38.803, International Mobile Telecommunications (IMT)-2020 networks can tolerate a maximum throughput loss of 5%. The following equations approximate the throughput over a channel with a given SINR (dB) when using link adaptation:

$$BitRate\left(SINR\right),\mathrm{bps}/\mathrm{Hz}=\{\begin{array}{c}0 for SINR<SIN{R}_{MIN}\\ \alpha ·S\left(SINR\right) for SIN{R}_{MIN}\le SINR<SIN{R}_{MAX} \\ \alpha ·S\left(SIN{R}_{MAX}\right) for SINR \ge SIN{R}_{MAX} \end{array}$$

(2)

where:

S(SINR): Shannon bound, S(SINR) = log2(1 + 10^SINR/10) (bps/Hz);

α: attenuation factor, representing implementation losses;

SINR_MIN: minimum SINR of the code set, dB;

SINR_MAX: maximum SINR of the code set, dB.

The parameters α, SINR_MIN and SINR_MAX can be chosen to represent different modem implementations and link conditions. The parameters proposed in Table 2 represent a baseline case, which assumes:

-

1:1 antenna configurations;

-

AWGN channel model;

-

Link adaptation (see

Table 4. Parameters describing baseline link level performance for 5G NR.

Parameter	DL	UL	Notes
α factor	0.6	0.4	Represents modem implementation losses based on the 3GPP specifications
SINRMIN, dB	−10	−10	Based on QPSK, 1/8 rate (DL) and 1/5 rate (UL)
SINRMIN, dB	30	22	Based on 256-QAM, 0.93 rate (DL) and 64-QAM, 0.93 rate (UL)

for details of the highest and lowest rate codes);

-

No HARQ.

Based on the equations above, the calculated bitrate mapping curves for uplink and downlink are shown in Figure 5.

7. Results of the Studies

This study’s findings are presented in terms of signal-to-interference-plus-noise ratio levels for both the downlink and uplink channels of the affected 5G NR network. Average throughput calculations are provided both before and after interference, offering insights into the performance impact on the victim 5G NR network. These results enable the estimation of an average throughput loss for 5G NR networks when RedCap is integrated into private networks for urban deployment. The percentage of throughput loss in this study was derived by comparing the throughput values of the 5G NR network before and after interference from RedCap devices. These values were obtained using simulated curves of throughput versus signal-to-interference-plus-noise ratio (SINR) generated for both the uplink and downlink scenarios.

The percentage loss was computed as the difference between the baseline (pre-interference) throughput and the degraded (post-interference) throughput, expressed as a percentage of the baseline.

This approach enabled the quantification of throughput loss for different scenarios, including the uplink and downlink directions and FR1 and FR2 frequency bands, providing a detailed assessment of interference impact.

In Figure 6, the SINR levels and throughput loss of the downlink channels in the FR1 band are illustrated.

In scenarios where RedCap is deployed in the FR1 band, the average throughput loss of the downlink channel for public 5G NR networks is calculated to be 4.5%. While this is in proximity to the threshold level, it remains within an acceptable range of throughput loss.

Figure 7 depicts the SINR levels and throughput loss for the uplink channels of the 5G NR victim network in the FR1 band.

In scenarios where RedCap is deployed in the FR1 band, the average throughput loss of the uplink channel for public 5G NR networks is calculated to be 21%; this throughput loss level significantly exceeds the threshold level.

In Figure 8, the SINR levels and throughput loss of the downlink channels in the FR2 band are illustrated.

In scenarios where RedCap is deployed in the FR2 band, the average throughput loss of the downlink channel for public 5G NR networks is calculated to be 1.5%, which is under the acceptable levels.

Figure 9 depicts the SINR levels and throughput loss for the uplink channels of the 5G NR victim network in the FR2 band.

In scenarios where RedCap is deployed in the FR2 band, the average throughput loss of the uplink channel for public 5G NR networks is calculated to be 0.7%, which is under the acceptable levels.

Based on the studies conducted, it is evident that deploying RedCap in the FR1 band could result in a 4.5% reduction in throughput for both the downlink and uplink channels in public networks. While the downlink throughput reduction remains within acceptable threshold levels, the uplink channel experiences a more substantial decrease, surpassing the tolerable thresholds.

In the case of FR2, RedCap deployment leads to minor throughput reductions of 1.5% and 0.7% for the downlink and uplink channels, respectively. This is attributed to the higher attenuation levels and more precise beamforming techniques inherent in the FR2 band, effectively mitigating interference.

Table 5. Average throughput losses of the 5G public networks in the FR1 and FR2 bands.

Link Direction	FR1	FR2
Downlink	4.5%	1.5%
Uplink	21%	0.7%

outlines the specific throughput loss values for each channel in both the FR1 and FR2 bands of the 5G NR public networks.

8. Conclusions

The widespread adoption of 5G NR has created opportunities for diverse applications, ranging from enhanced mobile broadband (eMBB) to ultra-reliable low-latency communication (uRLLC) and massive machine-type communication (mMTC). RedCap technology, as an advanced IoT solution, bridges the gap between low-complexity 4G-based solutions and the more demanding 5G NR infrastructure. By delivering cost-effective, simplified, and energy-efficient devices, RedCap is well suited for applications such as wearables, industrial automation, and asset tracking.

However, the integration of RedCap into non-public networks raises interference concerns, particularly in urban areas using the TDD FR1 and FR2 frequency bands. Our analysis shows that while downlink throughput losses remain within acceptable thresholds, uplink losses in FR1 exceed the tolerable limits, potentially affecting 5G NR’s quality of service (QoS). Conversely, FR2 bands demonstrate minimal throughput loss due to advanced beamforming techniques and higher attenuation levels.

We propose to consider more complete aspects of this technology in future work. To mitigate interference, we recommend prioritizing RedCap deployment in frequency division duplex (FDD) bands within FR1 and leveraging FR2’s inherent strengths. Strategic deployment planning will allow RedCap and 5G NR networks to coexist effectively, supporting diverse use cases without compromising network performance.