Millimeter Wave in Mobile Networks

Posted on Jan 22, 2025 in Communication

Millimeter Wave Communications for Future Mobile Networks

I. Introduction

Phase 1 of 5G is being standardized in 3GPP. 5G systems are expected to support various usage applications:

  • Enhanced Mobile Broadband (eMBB)
  • Massive Machine Type Communications (mMTC)
  • Ultra Reliable Low Latency Communication (URLLC)

IMT-2020 (5G) is expected to provide the following key performance indicators (KPIs):

  • > 10 Gbit/s peak data rate
  • 100 Mbit/s user-experienced data rate
  • 3x spectrum efficiency
  • > 100 Mbps cell edge rates
  • 10 Mbit/s/km2 area traffic capacity
  • 100x network energy efficiency
  • 1 ms over-the-air latency
  • Support for 500 km/h mobility
  • 106/km2 connection density

Millimeter wave (mmWave) bands, ranging from 30 GHz to 300 GHz, and THz frequency bands are being considered.

mmWave communications are considered crucial for achieving 10 Gbit/s peak data rates due to the large bandwidth available in these bands. Expanding bandwidth is an efficient approach to enhancing system capacity. The channel capacity of an additive white Gaussian noise channel is given by: C = B log2(1 + (P/N0*B)), where P is the signal power and N0 is the noise power spectral density.

II. Key Challenges and Technical Potentials

A. Main Technical Challenges

  1. Path Loss: In free-space transmission, the received signal power (outside the Kirchhoff area) can be determined by the Friis transmission formula: Pr(d) = PtGtGr(λ/4π)2d-n, where Pt is the transmit power, Gt and Gr are the antenna gains of the transmitter and receiver, λ is the wavelength, and d is the distance.
  2. Penetration Loss: High path loss is compounded in non-line-of-sight (NLoS) scenarios.
  3. High Power Consumption: Transmit power needs to increase with bandwidth to maintain signal-to-noise ratio (SNR).
  4. Narrow Beamwidth and Side-Lobes: Directional antennas, MIMO, and beamforming can increase transmission distance for mmWave.
  5. Hardware Impairments and Design Challenges: Practical transceiver hardware is impaired by phase noise (PN), non-linear power amplifiers (PAs), I/Q imbalance, and limited ADC resolution. Nonlinear PAs are a significant challenge in mmWave due to the difficulty of providing linear amplification over a very wide bandwidth.

B. Technical Potentials

  1. Large Continuous Unused Bandwidth: mmWave offers significantly more bandwidth than microwave communications, although wider bandwidth doesn’t always lead to higher rates in the noise-limited region.
  2. Short Wavelength and Narrow Beamwidth: The shorter wavelength of mmWave signals allows for compact antenna arrays. This also leads to higher security against eavesdropping and jamming, and greater resilience against co-user interference.

III. Channel Measurements and Modeling

A. Millimeter Wave Measurement Campaigns

Due to the shorter wavelength of mmWave, radio channel model parameters differ significantly from those in microwave bands below 6 GHz.

  1. Path Loss and Shadowing: These are the two most important large-scale characteristics, reported for various environments in both LoS and NLoS cases.
  2. Power Delay Profile and Delay Spread: The measured power delay profile (PDP) typically appears as a single or superposition of multiple exponentially decaying spectrums.

B. Millimeter Wave Channel Modeling

Channel models are essential for system-level simulations.

  1. 3GPP Spatial Channel Model (SCM) and SCM-Extended: Supports six delay paths with 5 MHz bandwidth in the 2 GHz band under Suburban Macro (SMa), Urban Macro (UMa), and Urban Micro (UMi) scenarios.
  2. WINNER I/II/+ Model: Developed channel models for mobile networks as part of the EU’s 4G project.
  3. 3GPP 3D Model: Defined in the 2 GHz bands at 10 MHz bandwidth.
  4. COST 273/2100 Model: Evolved from the COST 259 model, focusing on mobile broadband using MIMO.
  5. QuaDRiGa Model: Extends the 3GPP-3D model with spatial consistency and multi-cell transmissions.
  6. IEEE 802.11ad Model: Developed for indoor short-range communications at 60 GHz.
  7. MiWEBA Model: Extends the IEEE 802.11ad model to outdoor access, backhaul/fronthaul, and device-to-device (D2D) scenarios.

IV. Access, Backhauling, and Coverage

A. Multiple-Access Technologies

  1. SDMA (Space Division Multiple Access)
  2. NOMA (Non-Orthogonal Multiple Access): Performs multiple access in the power domain.
  3. Random Access: Used for initial access and handover.

B. Backhauling

Ultra-Dense Network (UDN) deployment is considered promising for achieving 5G KPIs. MmWave is desirable for backhauling in UDNs due to:

  • Large Bandwidth: Underutilized mmWave spectrum, including V-band (57-67 GHz) and E-band (71-76 GHz, 81-86 GHz), can provide potential GHz transmission bandwidth.
  • Reduced Interference: E-band coverage is up to several kilometers, while V-band is about 50-700 meters.
  • In-Band Backhauling: Backhaul and user access can potentially use the same frequency band.

C. Coverage and Connectivity

MmWave signals are sensitive to blockage due to high penetration loss.

V. Standardization and Deployment

A. 3GPP’s New Radio at mmWave Band

  1. Vision and Use Cases: 5G aims for significant improvements in data rate, capacity, latency, availability, and reliability.
  2. mmWave and Massive MIMO: Driving fundamental changes in mobile networks, starting at the physical layer (PHY).
  3. Hybrid Beamforming Architecture: Agreed upon for 5G systems at 3GPP RAN1.
  4. PHY Layer Design: Common understandings have been reached for mmWave PHY layer standardization.

B. Prototypes and Deployment Plan

Significant efforts are underway in building mmWave hardware platforms for channel measurement and prototyping. Current measurements focus on:

  1. Basic Throughput Performance: Maximum reported spatial streams is two, with spectral efficiency less than 20 bps/Hz.
  2. NLoS Transmission: Metal and concrete cause high penetration loss; foliage causes severe shadowing; reasonable reflection loss but high diffraction loss.
  3. Coverage: Indoor ≤ 100 m; urban outdoor ≤ 350 m; outdoor-to-indoor ≤ 20 m.
  4. Beam-Tracking: Possible for single users at pedestrian or low speeds; multi-user tracking is lacking.

VI. NOMA in Vehicular Communications

Applicability of NOMA and SM to V2X

NOMA enhances bandwidth efficiency and QoS in V2X, while Spatial Modulation (SM) offers advantages in bandwidth efficiency, cost, and complexity for V2V.

NOMA-SM for Vehicular Communications

This scheme combines NOMA and SM, using NOMA for non-orthogonal resource access and SM for single-RF transmission. A massive transmit configuration further improves bandwidth efficiency.

System Model

Includes vehicle-to-infrastructure (V2I), vehicle-to-vehicle (V2V), and intra-vehicle communications.

Principles of NOMA-SM

  • Applied to both V1-V2 and V1-U links.
  • V1 uses superposition coding for V2 and U.
  • Higher transmit power is allocated to the distant user (V2).
  • ML detection is used at V2 and U.

V2V Massive MIMO Channel Model

Uses a spatiotemporally correlated Rician channel model.

Capacity Analysis

V2 detects its signal directly. Capacity analysis considers signal domain capacity and mutual information. Optimal antenna activation and power allocation are explored.