Load Balancing In Heterogeneous Networks Research Articles

Q-Learning Based Load Balancing in Heterogeneous Networks with Human and Machine Type Communication Co-existence

Q-Learning Based Load Balancing in Heterogeneous Networks with Human and Machine Type Communication Co-existence

User Association for Load Balance in Heterogeneous Networks With Limited CSI Feedback

Conventional user association (UA) scheme based on max-signal-to-interference-plus-noise-ratio (max-SINR) criterion results in a severe load imbalance problem in heterogeneous networks (HetNets), leading to substantial performance loss. The state-of-the-art solutions to load imbalance mostly formulate the UA problem as a complicated global optimization that requires full channel state information (CSI) between every base station-user pair. Therefore, searching for optimal UA scheme imposes extremely high complexity as well as signaling overhead, and thus limits its practical applications. In this letter, the UA problem is reformulated as a local optimization at the macro base station, and is solved with only limited CSI feedback, where each user reports the synchronization signal powers instead of full CSI. After that, a low-complexity successive offloading (SO) scheme is derived. Numerical simulations demonstrate that the SO scheme achieves near optimal performance with much lower complexity.

Interference Mitigation and Dynamic User Association for Load Balancing in Heterogeneous Networks

Viewing the communication system as three-dimensional (3-D) with various tiers cooperating among each other is a new trend to present 5G heterogeneous networks (HetNets). Base stations (BS) in each tier operate with different power levels, access methods, and unique topologies. A proper user (UE) association algorithm for HetNets is a great challenge. We develop a new real-time dynamic user (UE) association algorithm for multi-tier cooperating systems that considers users’ mobility and traffic dynamics considering both overall network load and received signal to interference and noise ratio (SINR). Despite that our proposed UE association algorithm does not depend on an interference mitigation algorithm to improve its performance, we develop a location-based interference mitigation algorithm to mitigate co-tier and cross-tier interferences in the worst case scenario of spectrum sharing among various tier BSs to overcome some of the drawbacks of spectrum partitioning algorithms. Our new algorithms are studied and analyzed through simulation and they are proved to provide the best performance compared to other algorithms.

Towards Throughput Aware and Energy Aware Traffic Load Balancing in Heterogeneous Networks with Hybrid Power Supplies

Green communications has attracted much research attention for its environmental and economic benefits. For a cellular network, base stations (BSs) incur more than 50% of the energy consumption of the whole network. Therefore, BSs can be powered by green energy to reduce its on-grid power consumption. Meanwhile, the effective data rate (EDR) of a user’s flow, which depends on both the user’s channel condition and its BS’s workload status, is an important metric for user performance. From the perspectives of a mobile provider, both the aggregated EDR (sum of EDRs of all users within the coverage area of a macro BS) and on-grid energy consumption should be taken into account in the traffic load balancing process. Therefore, we propose a throughput aware and energy aware (TEA) traffic load balancing scheme for heterogeneous cellular networks to optimize the above two metrics. Since the EDR and energy consumption mutually affect each other, saving on-grid power is at the cost of sacrificing a certain loss of EDR. Thus, we employ an energy-throughput coefficient $\boldsymbol \alpha $ to adjust the tradeoff between the two metrics based on the mobile provider’s practical requirements. Simulation results demonstrate that TEA improves the aggregated EDR and significantly saves on-grid power.

Joint User Association and User Scheduling for Load Balancing in Heterogeneous Networks

This paper investigates joint user association (UA) and user scheduling (US) for load balancing over the downlink of a wireless heterogeneous network by formulating a network-wide utility maximization problem. In order to efficiently solve the problem, we first approximate the nonconvex throughput achieved with US to a concave function, and demonstrate that the gap for such an approximation approaches zero when the number of users is sufficiently large. Then, by exploiting a distributed convex optimization technique known as alternating direction method of multipliers, a joint UA and US algorithm, which can be implemented on each user's side and base station (BS)'s side separately, is proposed to obtain the single-BS association and resource allocation solutions. A remarkable feature of the proposed algorithm is that apart from load balancing, multiuser diversity is exploited in the association process to further improve system performance. We also extend the algorithm design to multi-BS association, whereby a user is associated with multiple BSs. The simulation results show the superior performance of the proposed algorithms and underscore the significant benefits of jointly exploiting multiuser diversity and load balancing.

Cell Selection for Load Balancing in Heterogeneous Networks

The vision of advanced long-term evolution (LTE-A) project is set to ultimate increase of network capacity in heterogeneous networks (HetNets). In HetNets with small cell configuration, a considerable majority of user devices is eventually connected to the macrocell base station (MBS), while small base stations (BSs), such as femtocell access points (FAPs), are still without any user. This results in unbalanced load and reduces the data rate of macrocell user equipment (MUE). In this paper, a method is proposed for load balancing among FAPs, while desired throughput is achieved. The proposed method uses the estimated received signal strength from different BSs and adjusted pilot signals. Under the critical signal to interference plus noise ratio (SINR) condition, a list of candidate FAPs is prepared. The updated candidate list henceforth does not include the least visited FAPs, which in turn leads to lower unnecessary handoffs. Once the BS with the highest number of free RBs and the highest pilot signal power is selected, FAP allocates the RBs with higher SINRs (qualified RBs) to user. In the case of FAP unavailability, the algorithm compels users to connect to the MBS with adequate qualified RBs. The performance of the proposed method was evaluated under a variety of FAPs density, and the number and velocity of users in terms of throughput and Jain’s fairness index. The results evidence affordable improvements in the throughput and Jain’s index in comparison with other methods.

Load Balancing in Heterogeneous Networks Based on Distributed Learning in Near-Potential Games

We present a novel approach for distributed load balancing in heterogeneous networks that use cell range expansion (CRE) for user association and almost blank subframe (ABS) for interference management. First, we formulate the problem as a minimization of an $\alpha$ -fairness objective function with load and outage constraints. Depending on $\alpha$ , different objectives in terms of network performance or fairness can be achieved. Next, we model the interactions among the base stations for load balancing as a near-potential game, in which the potential function is the $\alpha$ -fairness function. The optimal pure Nash equilibrium (PNE) of the game is found by using distributed learning algorithms. We propose log-linear and binary log-linear learning algorithms for complete and partial information settings, respectively. We give a detailed proof of convergence of learning algorithms for a near-potential game. We provide sufficient conditions under which the learning algorithms converge to the optimal PNE. By running extensive simulations, we show that the proposed algorithms converge within few hundreds of iterations. The convergence speed in the case of partial information setting is comparable to that of the complete information setting. Finally, we show that outage can be controlled and a better load balancing can be achieved by introducing ABS.

Low Complexity Distributed Max-Throughput Algorithm for User Association in Heterogeneous Network

This letter presents a distributed user association algorithm for the load balancing in heterogeneous networks. In this new method, we formulate the load balancing problem as a max min throughput problem. To solve this problem, we propose a distributed max-throughput algorithm. The main advantages of this algorithm include low computational complexity and few free parameters. Analysis and simulation results show that the proposed algorithm is very practical due to its simplicity.

Exploring load balancing in heterogeneous networks by rate distribution

In a heterogeneous wireless network (HWN), traffic is distributed primarily by grouping and channelling identical traffic through a particular access point. In this paper, we introduce an effective traffic load balancing scheme for maximising expected throughput in HWNs. Two scenarios are explored. First, we model multiple APs of a HWN as a cascade of independent M/M/1 queues. The expected number of packets in the system is formulated as a convex optimisation problem and solved using a Lagrange multiplier. A recursive approach is used in distributing the traffic to different APs. Then, we model multiple APs as a union of multiple M/M/1 and M/D/1 queues. Here, we minimise the time spent by each job in the system. We use a solver to obtain optimal rate distribution among the networks. Through extensive simulations, we show that distributing the traffic optimally outperforms existing traditional methods of grouping and channelling similar kind of traffic.