Source-linked AI summary
Enhanced Inter-Cell Interference Coordination Challenges in Heterogeneous Networks
David Lopez-Perez, Ismail Guvenc, Guillaume de la Roche, Marios Kountouris, Tony Q. S. Quek, Jie Zhang
TL;DR
HetNets address escalating wireless traffic by combining macro and low-power cells to improve proximity, coverage, and spatial reuse, while creating new interference and coordination challenges. The paper surveys these challenges, emphasizes enhanced ICIC, and evaluates eICIC techniques through system-level simulations showing recovered macro-user SINR and eliminated outages in a representative scenario.
Problem
HetNet deployments can improve network performance but introduce cross-tier interference, traffic variability, restricted-access problems, and coordination constraints that threaten proper operation.
Method
The paper presents HetNet concepts and technical challenges, reviews 3GPP ICIC standardization, and evaluates eICIC schemes using realistic system-level simulations.
Results
eICIC restores macro-user SINR and removes outages in the examined pedestrian-user scenario, while ABSF time coordination provides the best macro-user protection.
Takeaways & Limitations
Proper HetNet operation requires enhanced interference coordination, particularly for cross-tier interactions involving macrocells, femtocells, and picocells.
Abstract
from arXiv · showhide
3GPP LTE-Advanced has started a new study item to investigate Heterogeneous Network (HetNet) deployments as a cost effective way to deal with the unrelenting traffic demand. HetNets consist of a mix of macrocells, remote radio heads, and low-power nodes such as picocells, femtocells, and relays. Leveraging network topology, increasing the proximity between the access network and the end-users, has the potential to provide the next significant performance leap in wireless networks, improving spatial spectrum reuse and enhancing indoor coverage. Nevertheless, deployment of a large number of small cells overlaying the macrocells is not without new technical challenges. In this article, we present the concept of heterogeneous networks and also describe the major technical challenges associated with such network architecture. We focus in particular on the standardization activities within the 3GPP related to enhanced inter-cell interference coordination.
I. INTRODUCTION
LTE-Advanced introduces heterogeneous networks as a way to address growing traffic demand by combining diverse cell types and bringing access points closer to users. This architecture can improve coverage and spatial spectrum reuse, but also introduces technical challenges, including interference and traffic variability.
- LTE-Advanced technologies such as carrier aggregation, MIMO, and CoMP provide enhancements, but may offer limited gains under low-SINR conditions.
- Bringing low-power cells closer to users can improve radio link quality, indoor coverage, cell-edge performance, and spatial spectrum reuse.
- HetNets combine macrocells, RRHs, picocells, femtocells, and relays with different capabilities and deployment characteristics.
- HetNet deployments can reduce network overhead and potentially lower energy consumption, OPEX, and CAPEX.
- Femtocells are user-deployed low-power access points that use consumers’ broadband connections and may operate with open or restricted access.
- The architecture also introduces challenges because cross-tier interference and traffic-load variability can hinder successful deployment.
III. TECHNICAL CHALLENGES
HetNets require technical solutions for self-organization because many small cells may be deployed without operator supervision. These solutions must configure, recover, and optimize diverse cells under changing traffic and network conditions.
- Self-organization is a key challenge because user-deployed picocells and femtocells may operate without direct operator supervision.
- Self-configuration automatically prepares newly deployed cells for operation through downloaded software.
- Self-healing enables cells to recover from failures or execute compensation mechanisms automatically.
- Self-optimization continuously monitors network status and adjusts cell settings to improve coverage and reduce interference.
- Deployment is complicated by heterogeneous cells, many network parameters, and random, uneven, time-varying traffic loads.
B. Backhauling
HetNet backhauling and handovers create coordination and overhead challenges because cells use diverse links with different cost, bandwidth, and delay characteristics. These constraints make interference coordination particularly difficult in multi-tier deployments.
- Backhauling: Backhaul design must accommodate heterogeneous cells, including potentially expensive wired infrastructure for picocells and QoS risks from consumer broadband connections serving femtocells.
- Backhauling: A cost-effective, QoS-guaranteed HetNet backhaul may combine wired and wireless links, direct core interfaces, aggregation clusters, and relays.
- Handovers: Handovers support seamless service and traffic balancing but can impose significant overhead because HetNets contain many small cells and diverse backhaul links.
- Interference: Cross-tier and intra-tier interference are difficult because backhaul links have different bandwidth and delay constraints, while restricted access can prevent users from selecting nearby cells.
- Interference: Self-organizing interference mitigation requires continuous radio-environment sensing and monitoring for dynamic adaptation.
- Interference: Frequency reuse can mitigate interference in operator-deployed cells, but banning neighboring subchannels may reduce spatial reuse and motivates universal reuse.
- Interference: The paper reviews LTE-standardized ICIC techniques for interference scenarios created by new cell boundaries and focuses on macrocell interactions with femtocells and picocells.
A. Sources of Interference
HetNet interference arises from many new cell boundaries and from unplanned or restricted-access small cells. These conditions can create severe cross-tier interference when users cannot connect to the nearest suitable cell.
- The large number of new cell boundaries in HetNets exposes users to strong inter-cell interference that can degrade network performance.
- User-deployed femtocells may be moved or switched on and off without operator control, limiting traditional planning and motivating decentralized schemes using local information.
- Restricted-access femtocells can create uplink interference when nonsubscribers transmit at high power toward distant macrocells.
- A restricted-access femtocell can interfere with downlink reception for a macrocell user, making that user a victim of cross-tier interference.
- Open access can reduce downlink interference by connecting users to the strongest cell, but strongest downlink RSS may still produce unsuitable attachments in HetNets.
3) Power difference between nodes:
Large transmit-power differences in HetNets distort cell association and create severe cross-tier interference. The section describes range expansion, interference coordination signaling, and UE-assisted coordination as responses.
- Power difference between nodes: Macrocell users selected by strongest DL RSS can severely interfere with nearby low-power-node uplinks.Associating with a picocell could reduce the user’s uplink power, enabling load balancing and uplink interference mitigation.
- Power difference between nodes: Range expansion adds an offset to picocell or relay RSS, enlarging its DL coverage footprint for users otherwise attracted to macrocells.The approach can mitigate cross-tier uplink interference but may reduce downlink signal quality, with some expanded-region users experiencing DL SINRs below 0 dB.
- Power difference between nodes: Severe interference can cause radio link failure and service outages, especially through unreliable downlink control channels.ICIC therefore needs to address control-channel interference as well as data-channel interference.
- Power difference between nodes: UEs can sense, detect, and report nearby interferers so serving cells coordinate power, frequency, and time resource allocation.This coordination aims to enhance network capacity and mitigate user outages.
- Power difference between nodes: Release 8 ICIC messages exchanged over X2 include RNTP for downlink transmit-power thresholds and OI for per-resource-block uplink interference-plus-noise measurements.These messages support coordination among macrocells, picocells, and relays.
- Power difference between nodes: HII informs neighboring cells about scheduled uplink transmissions by cell-edge users, but femtocells lack these X2 messages.Backhaul exchange, wireless broadcast, or UE relaying are investigated for macro–femto coordination; femtocell backhaul delays remain a concern.
V. STANDARDIZATION FOR HETNET EICIC
Release 8 and Release 9 ICIC do not specifically address dominant HetNet interference scenarios. Release 10 enhanced ICIC work is organized into time-, frequency-, and power-domain techniques, with femtocell coordination receiving priority.
- STANDARDIZATION FOR HETNET EICIC: Release 8 and Release 9 ICIC were not designed specifically for HetNets and may be ineffective against dominant HetNet interference.Release 10 therefore develops enhanced Inter-Cell Interference Coordination techniques.
- STANDARDIZATION FOR HETNET EICIC: Release 10 eICIC techniques are grouped into time-domain, frequency-domain, and power-control categories.The article reviews these three categories for dominant interference scenarios.
- STANDARDIZATION FOR HETNET EICIC: Time-domain eICIC schedules victim users in resources such as subframes or OFDM symbols where interference from other nodes is mitigated.The methods are classified into two categories.
- STANDARDIZATION FOR HETNET EICIC: Aligned macro and home-eNB subframes overlap in control and data channels, motivating control-channel eICIC at femtocells.Almost Blank Subframes transmit only reference signals, with no control or data signals, in coordinated subframes.
1) Subframe Alignment:
Subframe-alignment eICIC mitigates interference by coordinating macrocell and low-power-node transmission timing. Alternatives include shifted subframe boundaries and frequency-domain orthogonalization, while reference signals remain a residual issue.
- Subframe Alignment: For range-expanded picocell users, macrocell Almost Blank Subframes provide lower-interference periods for scheduling their downlink transmissions.Without coordination, these users observe large downlink interference from the macrocell.
- Subframe Alignment: Femtocells still transmit reference signals during coordinated subframes, and those signals may cause severe interference in dominant-interference settings.Reference signals occupy only a limited portion of the whole subframe.
- Subframe Alignment: ABS patterns need not use only even subframes; 3GPP patterns under consideration include duty cycles of 1/8, 2/8, 3/8, and 3/20.Different ABSF configurations are possible.
- Subframe Alignment: Shifting home-eNB subframe boundaries relative to macro-eNB boundaries prevents control-channel overlap but leaves femtocell data-channel interference toward macro control channels.Shared-channel symbol muting and another proposed solution address this residual problem.
- Subframe Alignment: Frequency-domain eICIC schedules control channels and physical signals in reduced bandwidths to achieve orthogonal transmissions across cells.Orthogonalization may be static or dynamic through victim-UE detection, using macrocell reports or home-eNB sensing.
C. Power-Control Techniques
Power-control eICIC adjusts femtocell transmission power to manage dominant interference. The approaches use macrocell received power, path loss, or target SINR objectives, trading femtocell throughput against macro-user protection.
- Power-Control Techniques: Reducing femtocell radiated power can significantly improve victim-MUE performance but reduces total femtocell-user throughput.Power-control design therefore involves a direct trade-off between protecting macro users and serving femtocell users.
- Power-Control Techniques: Power-control formulations use Pmax and Pmin for home-eNB power bounds, PM for strongest co-channel macrocell received power, and α and β as scalar controls.The listed downlink approaches express femtocell transmit power in dBm.
- Power-Control Techniques: A path-loss-based approach sets home-eNB transmit power using the indoor and penetration loss between the home eNB and nearest MUE.The formulation includes minimum and maximum power-offset values.
- Power-Control Techniques: The objective-SINR approach restricts home-eNB users’ received SINRs to a target and reduces femtocell power as needed to achieve it.The expression incorporates detected interference, background noise, target SINR, and path loss between the home eNB and HUE.
- Power-Control Techniques: The objective-SINR-for-MUE approach aims to guarantee a minimum macro-user SINR by setting home-eNB transmit power using the MUE SINR under nearest-femtocell interference.This formulation focuses the power-control objective on macro-user protection.
VI. EICIC PERFORMANCE ANALYSIS
The evaluation simulates downlink eICIC schemes in a residential LTE-Advanced HetNet with a macrocell, picocell, and CSG femtocells. The scenario models pedestrian VoIP users and fully loaded small cells.
- The 300 m × 300 m residential scenario contains 400 houses, 63 CSG femtocells, one macrocell, and one open-access picocell.The macrocell is positioned outside the scenario, while the picocell is deployed at the macrocell edge.
- The simulation tests different downlink eICIC schemes under realistic residential HetNet loading and mobility conditions.
- Eight pedestrian VoIP users follow predefined paths at a mean speed of 1.1 m/s while the picocell and femtocells use all subcarriers.Users are considered in outage when their SINR remains below −4 dB for 200 ms.
A. Macrocell - Femtocell Interaction
Macrocell–femtocell eICIC protects macrocell users by mitigating cross-tier interference, but protection and femtocell throughput trade off. Power-based methods also vary in sensitivity to parameter tuning and signaling overhead.
- Macrocell - Femtocell Interaction: Without femtocell eICIC, cross-tier interference causes pedestrian macrocell-user SINR to fall into outage; eICIC restores SINR and eliminates outages.The macrocell triggers femtocell action when reported user SINR is smaller than -3 dB.
- Macrocell - Femtocell Interaction: ABSF eICIC time provides the best macrocell-user protection, but overlapping ABSFs carry no femtocell data.Power methods provide different victim-user protection levels depending on their design and tuning.
- Macrocell - Femtocell Interaction: Greater average sum throughput for the eight macrocell users corresponds to lower average sum throughput for the femtocell tier.
- Macrocell - Femtocell Interaction: An eICIC method that is not properly tuned to avoid user outages is included only for comparison purposes.
- Macrocell - Femtocell Interaction: Taking no action at femtocells gives the worst macrocell-user protection but the best femtocell throughput performance.
- Macrocell - Femtocell Interaction: Power methods (1) and (3) are highly dependent on tuning α, β, and SINRtar, whereas methods (2) and (4) adapt to each victim user.Methods (2) and (4) can avoid outages and recover maximum throughput at each femtocell, but tailored protection requires more backhaul signaling.
B. Macrocell - Picocell Interaction
Macrocell–picocell interaction is evaluated with range expansion and ABSF-based eICIC. A 10 dB picocell RSS offset expands coverage, while macrocell ABSFs avoid picocell-user outages.
- Macrocell - Picocell Interaction: A 10 dB offset is added to the picocell RSS to increase its downlink footprint and create a range-expanded region.
- Macrocell - Picocell Interaction: Macrocell/picocell handovers occurred 5 times, and picocell-user outages were avoided when the macrocell used ABSFs.Without macrocell eICIC, the handovers resulted in outage.
- Macrocell - Picocell Interaction: The evaluation identifies cross-tier interference coordination as central to proper operation of multi-tier networks and assesses eICIC techniques through system-level simulations.