PaperlayerView exact revision on arXiv

Source-linked AI summary

5G Cellular User Equipment: From Theory to Practical Hardware Design

Yiming Huo, Xiaodai Dong, Wei Xu

arXiv:1704.02540v3cs.IT

TL;DR

5G UE hardware must support high throughput while coping with dense integration, antenna constraints, human blockage, path loss, and self-heating. This paper analyzes contemporary UE constraints and proposes a reconfigurable DPA-MIMO architecture, evaluating it through link-budget and throughput analysis. The numerical results report higher long-distance downlink throughput and treatment of human-body blockage compared with prior state-of-the-art works.

  • Problem

    5G UE hardware faces constraints from high throughput demands, dense multifunction integration, antenna design, human blockage, high path loss, and self-heating.

  • Method

    The paper investigates contemporary UE hardware design and proposes a distributed phased array MIMO architecture evaluated using link-budget and throughput analysis.

  • Results

    The numerical analysis reports that DPA-MIMO provides the highest downlink throughput at long distances and addresses the human-body blockage issue.

  • Takeaways & Limitations

    DPA-MIMO is presented as a hardware solution to technical constraints that existing MIMO structures and methods cannot solve.

Abstract

from arXiv · show

Research and development on the next generation wireless systems, namely 5G, has experienced explosive growth in recent years. In the physical layer (PHY), the massive multiple-input-multiple-output (MIMO) technique and the use of high GHz frequency bands are two promising trends for adoption. Millimeter-wave (mmWave) bands such as 28 GHz, 38 GHz, 64 GHz, and 71 GHz, which were previously considered not suitable for commercial cellular networks, will play an important role in 5G. Currently, most 5G research deals with the algorithms and implementations of modulation and coding schemes, new spatial signal processing technologies, new spectrum opportunities, channel modeling, 5G proof of concept (PoC) systems, and other system-level enabling technologies. In this paper, we first investigate the contemporary wireless user equipment (UE) hardware design, and unveil the critical 5G UE hardware design constraints on circuits and systems. On top of the said investigation and design trade-off analysis, a new, highly reconfigurable system architecture for 5G cellular user equipment, namely distributed phased arrays based MIMO (DPA-MIMO) is proposed. Finally, the link budget calculation and data throughput numerical results are presented for the evaluation of the proposed architecture.

I. INTRODUCTION

5G UE design must meet ambitious throughput and spectral-efficiency targets while operating within severe battery, circuit, antenna, and integration constraints. The paper motivates high-GHz bands and proposes DPA-MIMO as a hardware architecture for addressing these challenges.

  • I. INTRODUCTION: 5G peak downlink throughput is expected to reach 10 Gbps, implying at least 100 bits/s/Hz with 100 MHz bandwidth.This requirement can motivate higher-order modulation, more spatial-multiplexing layers, or both.
  • I. INTRODUCTION: High modulation order and wide RF bandwidth require power-hungry, complicated RF and baseband circuits.The passage also identifies antenna dimension, spacing, and radiation efficiency as constraints on high-order MIMO.
  • I. INTRODUCTION: Cellular, Wi-Fi, Bluetooth, NFC, and GNSS systems compete for hardware resources in compact handsets spanning 700 MHz to almost 6 GHz.Integrating these technologies becomes more difficult when high-order MIMO is added within limited device dimensions.
  • I. INTRODUCTION: High-GHz bands can ease 5G UE design challenges because larger continuous RF bandwidth enables higher data rates.The FCC identified licensed bands at 28, 37, and 39 GHz, alongside an unlicensed 64–71 GHz band.
  • I. INTRODUCTION: The paper investigates contemporary UE design constraints and proposes a novel distributed phased array MIMO architecture for 5G hardware.It evaluates the proposed system using link-budget calculations and comparisons with state-of-the-art 5G works.
  • I. INTRODUCTION: Wireless capacity increased around 10,000 times from 1995 to 2014, while battery specific power increased only 4–5 fold.The paper presents this mismatch as a bottleneck for mobile handset devices.

B. Circuit and System Design

Contemporary 5G UE hardware must balance shrinking boards, dense multifunction integration, antenna performance, and electromagnetic compatibility. Antenna dimensions depend on frequency and substrate material, while higher dielectric constants reduce size but degrade radiation performance.

  • B. Circuit and System Design: Deep-nanometer scaling has slowed Moore’s-law improvements, moderating gains in SoC energy efficiency.The paper therefore frames SoC energy efficiency, area, cost, and yield as continuing design priorities.
  • B. Circuit and System Design: Smaller PCB trace widths and spacings allow more chipsets on the main logic board, reducing insertion loss and easing impedance matching.The denser design also worsens signal-integrity and electromagnetic-interference concerns, including spurs and TX-to-RX leakage.
  • C. Antenna and Product Design: Handset antennas must support 700 MHz–6 GHz multi-band and multi-standard operation while enabling diversity and spatial multiplexing.The 5G era adds higher-GHz standards and makes large-scale low-GHz MIMO difficult because adequate antenna spacing is required for isolation.
  • C. Antenna and Product Design: Antenna dimension is mainly determined by frequency and substrate material, and higher dielectric constant reduces size but degrades antenna performance.The paper explains that higher dielectric constant confines more radiation energy inside the substrate instead of radiating it.
  • C. Antenna and Product Design: Metallic casing can deteriorate mmWave antenna performance, while assembled antennas must satisfy both high efficiency and low specific absorption rate.Co-design with the casing and housing is challenging because these structures substantially affect antenna behavior.

D. System Design Trade-offs

5G UE design must balance wireless performance against interference, form factor, and portability constraints across the handset. The paper formulates a cellular UE figure of merit to quantify these trade-offs.

  • Display-induced RF sensitivity degradation requires metallic microwave shielding, but the shield increases handset thickness and degrades form factor.
  • Camera, speaker, finger scanner, battery, and main logic board placements can alter the electromagnetic field and produce undesired effects.
  • The cellular UE figure of merit combines energy-spectral efficiency across non-carrier-aggregation and carrier-aggregation cases with handset volume and weight.
  • The figure of merit has units of bit/Hz/Joule/mm3/gram and excludes material cost for well-defined comparison.

III. 5G CELLULAR UE BASED ON A NOVEL SYSTEM ARCHITECTURE

High-GHz 5G operation faces substantially greater and more complicated propagation loss than sub-6 GHz operation. The paper compares path loss across propagation scenarios at 2.6 GHz, 28 GHz, and 39 GHz.

  • Path loss is compared across propagation scenarios for 2.6 GHz, 28 GHz, and 39 GHz using a 5G channel model with atmospheric absorption and rain attenuation models.

A. Channel Model Analysis

At 28 and 39 GHz, non-line-of-sight propagation produces substantially higher path loss than line-of-sight conditions, making beamforming and careful antenna design necessary. The proposed DPA-MIMO architecture distributes phased-array modules to combine beamforming, spatial multiplexing, blockage mitigation, and heat dissipation in a compact UE.

  • Channel Model Analysis: 28 and 39 GHz path loss is at least 20 dB larger than LTE band 41 across the evaluated scenarios.NLOS loss is much larger than LOS loss, with UMi Street Canyon NLOS being especially severe.
  • Channel Model Analysis: Concrete-wall penetration loss can reach 117 dB at 28 GHz, so beamforming is mandatory at both the BS and UE.The penetration loss increases significantly with frequency for concrete walls.
  • Channel Model Analysis: Beamforming arrays favor element spacing no greater than λ0/2, whereas MIMO spatial multiplexing requires adjacent spacing larger than 1.5λ0 in a uniform square array.These incompatible spacing requirements prevent one M×N array from being perfectly optimized for both functions.
  • Novel Distributed Phased Array Based MIMO Architecture: Eight distributed 8-element phased-array modules are placed in the handset back housing, forming the proposed DPA-MIMO architecture.Each module embeds one RF transceiver chain and realizes an active phased array.
  • Novel Distributed Phased Array Based MIMO Architecture: Eight beamforming modules can process up to eight spatial streams, adding spatial multiplexing gain to increase link throughput by multiple times.The modules can also operate independently or cooperatively across handset positions affected by human-body blockage.
  • Novel Distributed Phased Array Based MIMO Architecture: Distributed phased-array modules mitigate handset self-heating because phased-array power amplifiers have power-added efficiency below 20%.Distributing the modules helps dissipate heat without requiring a cooling device that is difficult to implement in a compact handset.

C. Beamforming Module Hardware Design

The beamforming module design distributes mmWave antenna, chipset, and interconnect components within handset constraints while addressing spacing, thermal, and form-factor requirements.

  • Module arrangement: At least 1.5λ0, or 16 mm, spacing between BF modules is required for sufficient isolation, while additional modules increase achievable MIMO order within limited handset area.The arrangement trades wireless performance against constrained hardware space and resources.
  • Thermal constraints: Poor thermal design can overheat the densely packed power amplifiers, causing BF-module failure and potentially affecting the entire UE system.The paper also identifies possible user-safety consequences in extreme cases.
  • Form factor: The BF-module thickness can be kept below 1.5 mm by combining thin antenna, die, PCB, bump, and coaxial-connector layers.The thickness model is HBF = HANT + Hbump + Hdie + HPCB + Hconnector.
  • Materials and antenna design: BF modules use low-loss materials and configurable antenna structures to construct phased arrays for 28, 37, and 39 GHz 5G bands.Candidate technologies include LTCC, MLO, LCP, and Rogers RO4003C; antenna forms can be tailored to requirements.

D. RF Circuit Design of 5G Cellular UE

The proposed RF design uses distributed BF modules with split-IF conversion to fit handset constraints, reduce mmWave interconnect loss, and support reconfigurable deployments.

  • RF architecture: The split-IF architecture down-converts high-GHz downlink signals and up-converts uplink IF signals within each BF module.The same architecture can support 28, 37, and 39 GHz bands by changing the LO frequency and related hardware.
  • Design motivation: Separating BF modules from the MLB addresses handset form-factor limits and allows more cost-effective use of expensive mmWave-enabled PCB materials.The proposed partition avoids embedding all BF modules on the larger MLB and separates them from conventional FR-4-based design.
  • RF architecture: Immediate mmWave-to-IF conversion on the BF module minimizes front-end insertion loss and improves signal integrity through coaxial-cable connections.This avoids routing high-frequency signals over MLB traces.
  • Reconfigurability: The number and placement of BF modules can be adjusted for different specifications and use cases without reconfiguring the entire wireless system.This flexibility supports different handset deployments while preserving a cost-effective design approach.
  • Frequency plan: For a 28 GHz example, the BF module supports up to 800 MHz through four aggregated 200 MHz sub-bands and uses a 4.4 GHz IF.The LO must cover at least 23.2 to 23.9 GHz for the stated conversion plan.
  • Module implementation: The architecture supports a maximum of NBF streams communicating simultaneously, with each BF module integrating power, switching, amplification, and phase-control functions.PMUs and LDOs distribute supplies, while SPDTs enable time-domain duplexing.

E. Advancement of Data Converter Techniques

The paper examines data-converter requirements for 5G UE and identifies recent ADC/DAC designs that combine wide bandwidth, resolution, energy efficiency, and manageable area.

  • Converter requirements: Supporting 256-QAM requires ADC resolution of 12 bits or above in the proposed 5G cellular UE.The converters must also accommodate wide RF bandwidth and maintain high receiver-end spur-free dynamic range.
  • State-of-the-art designs: A 12-bit, 1.6 GS/s time-interleaved ADC consumes 37.7 mW, occupies 0.9 mm^2, and achieves 17.8 fJ/conversion.The paper identifies it as a prototype candidate for future 5G UE data conversion.
  • State-of-the-art designs: The cited ADC design enables a theoretical absolute physical data throughput of 10.8 Gb/s for ideal 256-QAM demodulation.This figure is a theoretical converter-level estimate rather than an end-to-end UE throughput result.
  • State-of-the-art designs: Recent converter designs support 200 MHz or even 800 MHz analog bandwidth with FOMW smaller than 50 fJ/conversion.The collected results indicate steady improvement in data-converter figures of merit.

IV. LINK BUDGET CALCULATION AND WIRELESS PERFORMANCE EVALUATION

The performance evaluation estimates downlink and uplink behavior for the proposed DPA-MIMO UE using a 5G channel model and assumed component parameters.

  • Evaluation setup: The evaluation uses a 5G channel model together with stated insertion-loss, noise-figure, and antenna-gain assumptions for link-budget and throughput estimation.The analysis covers both downlink and uplink operation.
  • Evaluation assumptions: The assumed 5G receiver noise figure is around 7 dB, while a single patch antenna element can provide 5 to 7 dBi gain.The paper favors CMOS for cost-effectiveness and widespread use when selecting the receiver noise-figure assumption.

A. Downlink Budget and Data Throughput Analysis

The analysis evaluates downlink and uplink budgets for 28 GHz 8 × 8 MIMO under deployment, distance, antenna, and propagation conditions. Results show that mmWave loss and glass attenuation require sufficient base-station and UE array resources to sustain throughput.

  • Downlink budget: Worst-case UMi Street Canyon NLOS and UMa NLOS models are used for downlink budget and throughput calculations because they have larger path loss and shadowing coefficients.The analysis relates these models to typical microcell and macrocell coverage scales.
  • Downlink budget: Lower SNR can reduce spectral efficiency because the system enables a lower digital modulation order below a threshold.The comparison includes UE architectures with 8 and 16 antenna elements per beamforming module.
  • Downlink throughput: 64 UE antenna elements, 200 MHz bandwidth, and up to 8 MIMO layers define the Fig. 12 peak-downlink-throughput evaluation at 28 and 39 GHz.Regular and infrared reflective glass penetration-loss models are included in the propagation analysis.
  • Uplink budget: 43 dBm is the regulated maximum uplink EIRP for mobile stations, creating a substantial long-distance transmission challenge at 5G mmWave frequencies.The uplink budget is summarized at 28 GHz, while the propagation challenge motivates larger receiving arrays at the base station.
  • Uplink architecture: 8 × 8 MIMO communication delivers 8 streams simultaneously, requiring NBF base-station antenna arrays whose size is determined by link budget and implementation feasibility.With Narray=256 at 28 GHz, each base-station unit is approximately 100 × 100 mm, and eight such units are illustrated.
  • Uplink throughput: A large number of base-station receiver antenna elements is needed to compensate for propagation loss in peak-uplink-throughput evaluations across deployment scenarios.The analysis varies the number of base-station antenna units and examines throughput versus distance.

C. Analysis with Attenuation Models

The attenuation analysis examines infrared reflective glass, multiband operation, and efficiency constraints in practical 5G UE designs. It finds that stronger attenuation can be addressed with additional antennas, while envelope tracking remains bandwidth-challenged.

  • Attenuation models: Infrared reflective glass strongly degrades throughput, but its attenuation can be overcome by embedding more antennas at both the UE and base-station ends.The paper also considers throughput under different penetration-loss models and deployment densities.
  • Multiband operation: Inter-band carrier aggregation can combine 28 GHz, 37/39 GHz, and 64–71 GHz bands, while DPA-MIMO remains applicable to multiband, multimode UE designs.The paper cites dual-band antennas and a multiband Doherty power amplifier as compatible hardware examples.
  • Power efficiency: Envelope tracking is important for improving efficiency and addressing high PAPR, but its modulator must track signal envelopes over stringent bandwidths.The challenge is more pronounced for 5G mmWave bands occupying hundreds of MHz; a 70% efficient 100-MHz X-band envelope modulator is cited as a candidate.
  • Reconfigurability: NANT and NBF can be selected flexibly according to performance targets, practical specifications, and design constraints.The paper gives examples with NANT values of 8 and 16 while NBF remains fixed at 8.

D. Performance Comparison with State-of-the-art

The paper compares DPA-MIMO with prior 5G hardware approaches using state-of-the-art performance summaries and numerical analysis. It reports higher long-distance downlink throughput and handling of human-body blockage, alongside flexible hardware implementation.

  • Prior approaches: Most previous works use conventional beamforming, MIMO, Massive-MIMO, or MU-MIMO as their air-interface technology.These approaches are summarized in the state-of-the-art performance comparison.
  • Performance comparison: DPA-MIMO provides the highest downlink throughput at long distances in the paper’s numerical comparison.The comparison is reported against state-of-the-art 5G works.
  • Performance comparison: DPA-MIMO also addresses human-body blockage, an issue the compared prior 5G research works do not mention or handle.The paper presents this as a distinction of the proposed architecture in the state-of-the-art comparison.
  • Conclusion: Numerical analysis shows increased wireless link budget and enhanced data throughput across different use cases and base-station deployment scenarios with highly flexible reconfigurability.The conclusion states that the architecture can use state-of-the-art circuit, antenna, and system technologies and facilitate peak throughput above 10 Gb/s while maintaining a slim mobile-handset form factor.
Loading 1704.02540v3…