Polarimetric Massive MIMO Channel Measurements in an Industry 4.0 

Get Complete Project Material File(s) Now! »

Initial Vision: Use-cases for 5G New Radio

With the increasing requirements upon the new 5G communication standards, a new radio (NR) interface and radio access network (RAN) are being developed. 5G NR is the name that the third generation partnership project (3GPP) chose for 5G when Release 15 was announced. NR is the equivalent of LTE for 4G or UMTS technology for 3G technology. 5G NR’s goal is to meet the performance requirements set by the international telecommunication union (ITU) for the year 2020. More specifically, recommendation ITU-T Y.3101 presents distinguishing features and requirements of the international mobile telecommunications 2020 (IMT-2020) for 5G networks. Promising technologies capable of fulfilling the gap from previous generations are sought. An overview of the NR interface standard under development by 3GPP is available in [11] with preliminary specifications for Release 15 approved in December 2017 [12]. Release 16 will provide further specifications for the second phase. The most central use-cases are not final and still being discussed both in ITU, 5G-PPP [13], the METIS project [14] and in 3GPP [12]. The main use cases to be supported span three different dimensions: enhanced mobile broadband (eMBB), massive ma-chine type communications (mMTC) and ultra-reliable low latency communications (URLLC). Additional use-cases may naturally emerge in time with the evolution of the physical layer radio interface [15].

Use-Cases

Enhanced Mobile Broadband (eMBB):
Can be defined as the feature of 5G as the most relevant evolution of 4G. It is a data-driven use-case enabling new applications such as virtual reality (VR). Improved spectral efficiencies, cell-edge data rates and coverage, amongst other requirements, define the shape of eMBB in 5G networks. The relevant 5G requirements are:
• Peak throughput: 20 Gbps in Downlink (DL), 10 Gbps in Uplink (UL).
• Experienced data rates (5th percentile user throughput): 100 Mbps (DL), 50 Mbps (UL).
• Area capacity (e.g. indoor hotspot): 10 Mbps/m2.
• User plane latency: 4 ms
Massive Machine Type Communications (mMTC): Industry 4.0
IoT requires massive connectivity where tens of billions of interconnected low-cost devices and sensors communicate [16]. Recent advancements on machine-to-machine (M2M) communications in 4G networks are presented in [17]. This is being labeled as the fourth industrial revolution or Industry 4.0. There are many advantages brought by 5G cutting edge technologies for industrial automation scenarios in the drive for Industry 4.0 [18]. In [19], challenges and solutions for M2M communications are depicted. Relaxed data rates constraints are sought compared to eMBB but other strict requirements are still to be fulfilled:
• Density: 1 Million devices/km2.
• Wide Coverage: 164 dB Maximum Coupling Loss (MCL).
• Device battery life: 10-15 years.
Ultra-Reliable Low Latency Communications (URLLC)
Critical applications (e.g. Intelligent V2X, remote surgery, smart grids, etc.) define very stringent latency and reliability requirements. For this ultra-reliable and low latency area communications, specific requirements are needed [20, 21]:
• Latency: less than 1 ms.
• Reliability : 99.999%.
• Control plane latency: tens of ms.
• User plane latency: less than 0.5 ms (one-way UL and DL).
• Mobility interruption time: 0 ms.

Evolution of multi-antenna systems with 3GPP

Multiple antennas can increase capacity and reliability but also provide spatial re-solvability, spatial DoF for multiple users to share and higher SE. MIMO systems have evolved lately to include MU-MIMO systems (via the introduction of new transmission modes TM) before the arrival of massive MIMO [63]. This transition was motivated by many factors:
• In the 1-6 GHz range of cellular communication, the number of antennas that can be deployed in compact user terminals is limited.
• The wireless channel to a given terminal can have, in some configurations or scenarios, few contributing paths, limiting the ability to send parallel data streams.
• Advanced signal processing schemes are sometimes needed in point to point MIMO to detect multiple streams.
• For MU-MIMO, users should be spatially well-separated to avoid co-channel interference.
• Small-scale fading can still affect the link reliability.
• Massive MIMO can be a solution to focus, in an efficient manner, the energy towards the intended users.
The following figure illustrates the evolution of multi-antenna systems under the scope of 3GPP standards and releases.

Analog Beamforming (ABF)

ABF [83] is simpler but can accommodate one user (no SDMA). The same signal is fed to each physical antenna element and the signal phases are adjusted in the RF domain using analog phase-shifters to steer the radiating pattern of the array in a given direction. The different copies of the signal from different array elements are constructively summed at the Rx to form the in-beam direction. This is the principle of phased arrays which has been known for a while now. The main difference with DBF is mainly the processing wherein the DBF is applied on the baseband signal (on K data streams) whilst phase shifting in the analog beamformer is applied after digital-to-analog conversion (DAC) for the single stream user.

READ  The context of Tswana pr~se poetry

Table of contents :

List of figures
List of tables
List of publications
1 General Introduction and Motivations 
1.1 Introduction: Overview of The 5th Generation
1.1.1 5G: Evolution or Revolution ?
1.1.2 Initial Vision: Use-cases for 5G New Radio
1.1.2.1 Use-Cases
1.1.2.2 Multi-Layer Spectrum
1.1.3 Gaps and Challenges
1.2 Impacting technologies of 5G
1.2.1 Massive MIMO: Why Now ?
1.3 Multi-antenna System Communications
1.3.1 MIMO Communications
1.3.1.1 Fundamentals and system model
1.3.2 Multi-User MIMO
1.3.2.1 Advantages of MU-MIMO
1.3.3 Evolution of multi-antenna systems with 3GPP
1.4 Massive MIMO: Massive Breakthrough
1.4.1 History and Brief Introduction
1.4.2 General Definitions
1.4.3 Key Features
1.4.4 Main Advantages
1.5 Massive MIMO System Architecture
1.5.1 Digital Beamforming (DBF)
1.5.2 Analog Beamforming (ABF)
1.5.3 No Compromise ?
1.5.4 What is precoding then ?
1.6 Massive MIMO in practice
1.6.1 Real-time Testbeds
1.6.2 Trials and Deployments
1.6.3 Challenges
1.7 Channel Estimation
1.7.1 Time Division Duplexing
1.7.2 Frequency Division Duplexing
1.7.2.1 Coherence Interval
1.7.2.2 5G Frame Structure
1.8 Motivations and Contributions
1.8.1 Special Focus on Industry 4.0
1.8.2 Polarimetric Channel Characteristics and Propagation Conditions
1.8.3 CSI Feedback Overhead Reduction
1.8.4 Antenna Selection Strategies
1.9 Thesis Organization
1.10 Other Contributions
1.11 Summary of Key Points
2 Massive MIMO Channel and System Aspects 
2.1 SISO Wireless Propagation Channel
2.1.1 Characteristics of Propagation Channels
2.1.1.1 Large scale propagation
2.1.1.2 Small scale propagation
2.1.2 Time-Frequency Domain SISO Channel Model
2.1.2.1 Delay Domain Analysis
2.1.2.2 Frequency domain analysis
2.2 Massive MIMO Channel Characteristics
2.2.1 Notations
2.2.2 General Propagation Parameters
2.2.2.1 Average Channel Gain
2.2.2.2 Cross-Polarization Discrimination
2.2.2.3 Ricean Factor
2.2.2.4 Spatial Correlation
2.2.3 The Two Characteristics of Massive MIMO
2.2.3.1 Channel Hardening
2.2.3.2 Favorable Propagation Condition
2.2.4 The Gram Matrix
2.2.4.1 Gram’s matrix Power Ratio
2.3 Massive MIMO Channel Model
2.3.1 Review of Correlation-based Channel Models
2.3.2 Geometrical based Propagation Channel Model
2.3.2.1 Special Case: Rayleigh Channel Model
2.3.2.2 Improving Stochastic Models
2.3.3 Parametric Analysis
2.3.3.1 Channel Hardening
2.3.3.2 Gram’s Power Ratio
2.4 System Model for DL Massive MIMO
2.4.1 System performance: Capacity of MIMO systems
2.4.1.1 Capacity of SU-MIMO
2.4.2 Capacity of MU-MIMO
2.4.2.1 Power Allocation
2.4.2.2 Precoding Strategies
2.4.2.3 Maximum-Ratio-Transmission
2.4.2.4 Zero-Forcing
2.4.2.5 Minimum Mean-Squared Error
2.4.3 Performance Analysis: Simplified System Model
2.4.3.1 Massive MIMO and Linear Processing
2.5 Sum-Rate Capacity Results
2.5.1 Performance in i.i.d
2.5.2 Parametric Analysis with the Geometrical Model
2.6 Conclusion
2.7 Summary of Key Points
3 Polarimetric Massive MIMO Channel Measurements in an Industry 4.0 
3.1 Introduction: Industry 4.0
3.2 Review of Massive MIMO Channel Characterization
3.2.1 Sounding Techniques
3.2.2 Review of Main Results
3.3 Experimental Setup
3.3.1 Radio Channel Sounding
3.3.2 Antennas
3.4 Geometrical Configuration of the Experiments
3.4.1 Multi-User Setup
3.4.2 General Notations
3.4.2.1 Polarimetric Massive MIMO Channel Matrix
3.5 Propagation Channel Characteristics
3.5.1 Channel Transfer Function: Example
3.5.2 Average Received Gain
3.5.3 Coherence BW, Ricean factor and Tx Correlation
3.5.4 Classification
3.5.5 Selected Scenarios
3.5.6 Parameter Cross-Correlation
3.5.7 Polarimetric Channel Characteristics
3.6 Massive MIMO System Evaluation
3.6.1 Does Channel Hardening hold ?
3.6.2 How Favorable is the Propagation ?
3.6.3 Gram’s Power Ratio
3.6.3.1 Influence of the Scenario
3.6.4 Sum-rate Capacity:
3.7 Communication Strategy Using Polarization Diversity
3.7.1 UEs Allocation Algorithms
3.7.2 Results
3.8 Conclusion
3.9 Summary of Key Points
4 Propagation-Based Antenna Selection Strategies 
4.1 CSI Feedback Reduction in FDD mode
4.1.1 Context and Methodologies
4.1.1.1 Related Work
4.1.1.2 Preview of the Method
4.1.1.3 Framework For Channel Estimation
4.1.2 Estimation Procedure
4.1.2.1 Tx Correlation
4.1.2.2 Principle of CSIT Estimation Procedure
4.1.2.3 Determination of the reduced correlation vector
4.1.2.4 Estimation of the channel matrix
4.1.3 Optimization of the Algorithm and Performances
4.1.3.1 Single-User Configuration
4.1.3.2 Multi-User Configuration
4.1.3.3 Quantifying Complexity Reduction
4.1.4 Conclusion
4.2 Antenna Selection Strategies
4.2.1 Context and Methodologies
4.2.1.1 Related Work
4.2.1.2 Antenna selection Procedure
4.2.1.3 Selection criterion
4.2.1.4 Evaluation Algorithm
4.2.1.5 Investigated Scenario
4.2.2 Validation and Results
4.2.2.1 Validation based on Rx correlation
4.2.2.2 Strategy Performance Evaluation and Results
4.2.2.3 Gram’s Power Ratio
4.2.2.4 Parametric Analysis
4.2.2.5 Sum-rate Capacity
4.2.3 Conclusion
4.3 General Conclusion
5 Conclusion 
6 Future Research Directions 
Appendix 156
A List of Notations, Symbols and Acronyms 
A.0.1 Mathematical Notations and Operators
A.0.2 List of Specific Used Symbols
A.0.3 List of Acronyms
B DPC and Waterfilling 
B.1 Dirty Paper Coding
B.2 Waterfilling algorithm
C Ricean Factor Estimation 
D Geometrical Model: Charts and Validation 
E Antennas Characteristics 
F UE Allocation Strategies 
Bibliography

GET THE COMPLETE PROJECT

Related Posts