Source-linked AI summary
Fronthauling for 5G LTE-U Ultra Dense Cloud Small Cell Networks
Haijun Zhang, Yanjie Dong, Julian Cheng, Md. Jahangir Hossain, Victor C. M. Leung
TL;DR
Fronthauling is challenging in 5G LTE-U UDCSNets because dense small-cell deployment and explosive traffic demand high-capacity connectivity. The paper surveys fronthaul requirements and candidate technologies, evaluates mmWave for access and fronthaul, and qualitatively compares alternatives. It reports that mmWave sidelobe interference can confine UDCSNet capacity while optical, unlicensed-spectrum, and FSO approaches involve different advantages and challenges.
Problem
5G LTE-U UDCSNets require fronthaul that can support dense small-cell deployments and explosive data traffic, but their fronthaul challenges and requirements have received limited attention.
Method
The paper surveys fronthaul technologies for licensed and unlicensed spectrum, evaluates mmWave for access and fronthaul links, and provides a qualitative comparison.
Results
MmWave sidelobe interference caused by non-negligible antenna gain can confine UDCSNet system capacity.
Takeaways & Limitations
Optical fiber, unlicensed spectrum, and free-space optical are surveyed as fronthaul candidates with distinct advantages and challenges for 5G UDCSNet.
Abstract
from arXiv · showhide
Ultra dense cloud small cell network (UDCSNet), which combines cloud computing and massive deployment of small cells, is a promising technology for the fifth-generation (5G) LTE-U mobile communications because it can accommodate the anticipated explosive growth of mobile users' data traffic. As a result, fronthauling becomes a challenging problem in 5G LTE-U UDCSNet. In this article, we present an overview of the challenges and requirements of the fronthaul technology in 5G \mbox{LTE-U} UDCSNets. We survey the advantages and challenges for various candidate fronthaul technologies such as optical fiber, millimeter-wave based unlicensed spectrum, Wi-Fi based unlicensed spectrum, sub 6GHz based licensed spectrum, and free-space optical based unlicensed spectrum.
I. INTRODUCTION
UDCSNet combines cloud radio access networks with densely deployed small cells to address 5G traffic growth, but its fronthaul remains challenging and underexplored. The paper surveys fronthaul requirements and candidate technologies for licensed and unlicensed spectrum.
- Motivation: UDN is presented as a promising technique for meeting explosive 5G mobile data traffic requirements.The paper defines UDN using traffic volume per area or user density thresholds.
- Network concept: UDCSNet combines a CRAN with densely deployed small cells, enabling efficient radio resource management and flexible BBU scaling.BBUs can be removed, added, and upgraded easily in UDCSNet.
- Fronthaul problem: Fronthauling connects the BBU pool and RRHs for data transmission, interference management, and mobility handover.The BBU pool centralizes baseband processing while RRHs provide wireless coverage.
- Fronthaul problem: Massive RRH deployment, high optical-fiber cost, and huge traffic volumes create complexity and high-capacity requirements for UDCSNet fronthaul.These challenges are stated as the main reasons fronthauling is difficult in 5G UDCSNet.
- Research gap: Existing literature has addressed related backhaul and fronthaul topics, but fronthauling specifically for 5G UDCSNet has received little attention.Prior work covers wireless backhaul, mmWave and fiber technologies, and fronthaul-constrained CRANs.
- Paper scope: The article describes a CRAN-based ultra dense small-cell network, evaluates mmWave access and fronthaul capacity, and compares unlicensed-spectrum and FSO alternatives.The comparison covers licensed and unlicensed spectrum and includes phantom-cell considerations.
II. 5G ULTRA DENSE CLOUD SMALL CELL NETWORK ARCHITECTURE
The UDCSNet architecture places heterogeneous macro and small RRHs around a centralized BBU pool, combining cloud-based processing with dense, localized radio coverage. MmWave can serve both access and fronthaul links.
- Architecture: UDCSNet consists of a macro RRH and many small RRHs connected to a centralized BBU pool.The architecture supports heterogeneous macro RRHs, small RRHs, and Wi-Fi RRHs.
- Centralized processing: The BBU pool co-locates radio resource control and management for macro and small RRHs instead of distributing those capabilities across basestations.This differs from a traditional heterogeneous network.
- Deployment: Small RRHs can be deployed on building or office floors to improve capacity and coverage, or in hotspots such as stadiums with ultra dense devices.The paper gives indoor and hotspot deployments as examples.
- Link technologies: MmWave spectrum can be used for both access links and fronthaul links in UDCSNet.The next section evaluates performance for both link types.
- Centralized processing: The BBU pool uses optical fiber for backhaul and is enhanced with centralized processing and collaborative functions for heterogeneous RRHs.The BBU pool provides the computing and coordination layer of the architecture.
III. MILLIMETER-WAVE BASED ACCESS LINK AND FRONTHAUL LINK
The paper motivates mmWave for UDCSNet because conventional cellular spectrum is scarce and cannot satisfy ultra-dense small-cell requirements. Its large available bandwidth and narrow directional beams support high capacity and spatial reuse.
- Terminology: The paper distinguishes unlicensed spectrum used by Wi-Fi from licensed or unlicensed spectrum resources considered more generally.This terminology is specified in the article’s footnotes and discussion.
- Spectrum motivation: Current LTE spectrum cannot satisfy the requirements of ultra dense small cell networks.LTE bands cover about 500 MHz of unique spectrum, with many bands shared among LTE, HSPA, and GSM.
- MmWave rationale: Around 60 GHz is considered suitable for 5G because license-free or light-licensed 9 GHz bandwidth is available.The cited availability is around the 60 GHz band.
- Capacity: MmWave can achieve high UDCSNet capacity through large available spectrum and high spatial reuse enabled by narrow directional beams.The paper identifies these properties as the basis for mmWave capacity benefits.
A. Millimeter-wave Access Links in an UDCSNet with Phantom Cells
Phantom cells separate control and user planes across macro and small-cell radio heads, while mmWave access links can provide high capacity under modeled conditions.
- Phantom-cell architecture: Phantom cells use a macro RRH for the C-plane and a serving small RRH for the U-plane.The macro RRH operates in a low-frequency band, while the small RRH uses a high-frequency band for data traffic.
- Phantom-cell architecture: Managing small-cell UEs through the macro RRH reduces RRC handover signaling overhead, especially for high-mobility users.
- Phantom-cell architecture: The phantom-cell split can extend to Wi-Fi RRHs, with licensed-band macro control and unlicensed-band Wi-Fi user-plane service.
- MmWave access capacity: Access-link capacity increases with the number of RRHs because users can associate with better RRHs using maximum-SNR association.The system capacity is higher for a_l = 1.8 than for a_l = 2.2 because the smaller exponent produces lower path loss.
B. Millimeter-wave for Multihop Fronthaul in UDCSNet
Multihop mmWave fronthaul connects distant RRHs to an aggregation node, supporting mesh-like forwarding while exposing spectrum-reuse and interference-related capacity constraints.
- Multihop architecture: RRHs can connect to the aggregation node directly or through multihop relaying, with a maximum of three hops to limit delay and routing complexity.
- Multihop architecture: The aggregation node collects fronthaul traffic from multiple RRHs and forwards it to the BBU pool.The aggregation-node-to-BBU link can use optical fiber in NLOS scenarios or mmWave in LOS scenarios.
- Spectrum reuse: MmWave fronthaul may use in-band sharing with access links or out-of-band operation in a separate spectrum band.In-band operation uses spectrum more efficiently but requires effective spectrum reuse.
- Capacity behavior: Two-hop downlink end-to-end capacity tends toward a stable value when interference between neighboring RRHs and UEs is included.The analysis uses non-negligible sidelobe antenna gain of −5 dB and the log-concave Shannon capacity formula.
- Capacity behavior: System capacity is higher with a_l = 1.8 than with a_l = 2.2 because the smaller exponent corresponds to lower path loss.
A. Fronthauling Using Unlicensed Spectrum Access
The paper identifies Wi-Fi-based unlicensed spectrum as a fronthaul candidate for UDCSNets, motivated by spectrum scarcity and reuse opportunities alongside coexistence requirements.
- Motivation: Low-frequency spectrum is heavily occupied by 2G, 3G, and 4G networks, motivating investigation of additional spectrum for next-generation mobile systems.
- Unlicensed-spectrum context: Wi-Fi uses unlicensed 2.4 GHz and 5 GHz bands, while the 5 GHz unlicensed band provides 500 MHz and is being considered for LTE-U coexistence.3GPP is examining coexistence between Wi-Fi and LTE cellular networks.
- Candidate technology: Wi-Fi-based unlicensed spectrum is presented as a potential UDCSNet fronthaul candidate.
- Candidate technology: Reusing unlicensed spectrum for access and fronthaul can improve spectrum efficiency and avoid purchasing separate fronthaul frequencies.
- Reuse methods: Access and fronthaul links can share unlicensed spectrum through frequency division, time division, or cognitive/opportunistic fronthauling.Cognitive/opportunistic fronthauling uses the spectrum when it is unused or when active-user signals fall below a predefined threshold.
B. Fronthauling Using Free Space Optical
FSO is presented as a high-capacity fronthaul option for RRH–BBU and multihop links, with directional security and rapid deployment. Its main constraint is weather sensitivity, while hybrid FSO/mmWave can improve availability.
- FSO can serve both the fronthaul from RRH to the BBU pool and multihop links in UDCSNet.
- FSO offers wide optical bandwidth and can provide several magnitudes of capacity improvement over mmWave.Combining FSO with wavelength division multiplexing can provide capacities in the Gb/s regime.
- FSO's directional laser beam is inherently secure and robust to electromagnetic interference.
- FSO fronthauling can be rapidly deployed and installed, and its frequency band above 300 GHz is unregulated.
- Weather phenomena, including thick clouds, severe fog, and dust storms, can sharply degrade FSO performance, especially over long distances.Poor weather conditions may result in poor transmission performance, making FSO potentially unsuitable in places such as Beijing during severe dust storms.
- A hybrid FSO/mmWave approach can potentially provide 99.999% link availability and remain robust in rainy or foggy environments.The only weather condition identified as affecting the hybrid system is simultaneous heavy rain and thick fog, which rarely occurs.
V. COMPARISON OF FRONTHAULING CANDIDATES IN 5G UDCSNET
The paper qualitatively compares fronthaul candidates by deployment conditions and cost. LOS settings favor FSO, mmWave, and unlicensed spectrum, whereas NLOS settings require optical fiber or licensed sub-6 GHz.
- Figure 6 presents a qualitative comparison of different fronthaul techniques for UDCSNet.
- Optical fiber and licensed sub-6 GHz can operate in NLOS environments, whereas mmWave and FSO require LOS conditions.
- For LOS fronthaul, FSO, mmWave, and unlicensed spectrum are identified as good choices; mmWave and unlicensed spectrum can also serve access links.
- The comparison defines cost using CAPEX and OPEX, including equipment, installation, and maintenance costs.
- Optical-fiber cost depends on fiber capacity and the distance between the BBU and RRH.
- FSO hardware may be expensive, but its cost can be significantly reduced with appropriate deployment conditions.
VI. CONCLUSION
The article surveys fronthauling challenges and candidate technologies for 5G LTE-U UDCSNets, including mmWave, unlicensed spectrum, and free-space optical approaches. It finds that mmWave sidelobe interference can constrain capacity, while optical approaches avoid interference but may have high infrastructure costs.
- Fronthauling is challenging in LTE-U UDCSNets with massive deployment of CRAN-enabled small cells.
- The article surveys the advantages and challenges of various fronthauling candidates for 5G LTE-U UDCSNets.
- Candidate technologies: MmWave is examined for both access and fronthaul links in UDCSNets.
- Candidate technologies: Interference from non-negligible mmWave sidelobe antenna gain can constrain UDCSNet system capacity.
- Candidate technologies: Unlicensed spectrum and free-space optical technologies are investigated for fronthauling in 5G UDCSNets.
- Trade-offs: Optical fronthauling techniques are free of interference but can have high infrastructure costs, whereas RF-based techniques face co-channel interference.