This paper proposes a telecom big data based user offloading self-optimisation TBDUOS scheme. Its aim is to assist telecom operators to effectively balancing the load distribution with achieving good service performance and customer management in heterogeneous relay cellular systems. To achieve these objectives, in the cell-level offloaded traffic analysis stage, the optimal offloaded traffic is calculated to minimise the total blocking probability. In the user-level offloading stage, the user portrait is drawn and the K-MEANS algorithm is employed to manage the users clustering in the heavily loaded cell, and finally shifting users to assistant cells. Simulation results show the TBDUOS scheme can effectively reduce the handover failure and call dropping of specific users, especially voice/stream users, high consumption users, high level users. The TBDUOS scheme can also reduce the blocking probability.

Telecom Big Data Based User Offloading Self-Optimisation in Heterogeneous Relay Cellular Systems

Mobility load balancing MLB redistributes the traffic load across the networks to improve the spectrum utilisation. This paper proposes a self-organising cluster-based cooperative load balancing scheme to overcome the problems faced by MLB. The proposed scheme is composed of a cell clustering stage and a cooperative traffic shifting stage. In the cell clustering stage, a user-vote model is proposed to address the virtual partner problem. In the cooperative traffic shifting stage, both inter-cluster and intra-cluster cooperations are developed. A relative load response model is designed as the inter-cluster cooperation mechanism to mitigate the aggravating load problem. Within each cluster, a traffic offloading optimisation algorithm is designed to reduce the hot-spot cell's load and also to minimise its partners' average call blocking probability. Simulation results show that the user-vote-assisted clustering algorithm can select two suitable partners to effectively reduce call blocking probability and decrease the number of handover offset adjustments. The relative load response model can address public partner being heavily loaded through cooperation between clusters. The effectiveness of the traffic offloading optimisation algorithm is both mathematically proven and validated by simulation. Results show that the performance of the proposed cluster-based cooperative load balancing scheme outperforms the conventional MLB. Copyright © 2013 John Wiley & Sons, Ltd.

/pdf/self-organising-cluster-based-cooperative-load-balancing-in-4te2eugv7i.pdf

Self-organising cluster-based cooperative load balancing in OFDMA cellular networks

Mobility load balancing (MLB) is widely used to address the uneven load distribution problem. The basic idea is that a hot-spot cell selects less-loaded neighbouring cells as assistant cells, and thenshiftsitsedge users to assistant cells via the handover region adjustment. However, shifted users receive the reduced signal power after MLB, which may result in the poor link quality problem for shifted users. In order to deal with this problem, this paper proposes a mobility load balancing aware radio resource allocation (MLBRRA) scheme. In the MLBRRA scheme, the assistant cell jointly considers the MLB factor of the shifted user and that of the hot-spot cell, as well as the proportional fairness scheduling factor. More specifically, the assistant cell preferentially allocates radio resources to shifted users, which are suffering poor link quality or previously served by a hot-spot cell with large handover region. Simulation results show that the proposed MLBRRA scheme can effectively deal with the poor link quality problem in terms of handover failure and call dropping. The proposed scheme can also reduce the call blocking probability.

http://ieeexplore.ieee.org/iel7/7394241/7399781/07399952.pdf

Mobility load balancing aware radio resource allocation scheme for LTE-Advanced cellular networks

In this article, we propose a deep reinforcement learning (DRL)-based mobility load balancing (MLB) algorithm along with a two-layer architecture to solve the large-scale load balancing problem for ultradense networks (UDNs). Our contribution is threefold. First, this article proposes a two-layer architecture to solve the large-scale load balancing problem in a self-organized manner. The proposed architecture can alleviate the global traffic variations by dynamically grouping small cells into self-organized clusters according to their historical loads, and further adapt to local traffic variations through intracluster load balancing afterwards. Second, for the intracluster load balancing, this article proposes an off-policy DRL-based MLB algorithm to autonomously learn the optimal MLB policy under an asynchronous parallel learning framework, without any prior knowledge assumed over the underlying UDN environments. Moreover, the algorithm enables joint exploration with multiple behavior policies, such that the traditional MLB methods can be used to guide the learning process thereby improving the learning efficiency and stability. Third, this article proposes an offline-evaluation-based safeguard mechanism to ensure that the online system can always operate with the optimal and well-trained MLB policy, which not only stabilizes the online performance but also enables the exploration beyond current policies to make full use of machine learning in a safe way. Empirical results verify that the proposed framework outperforms the existing MLB methods in general UDN environments featured with irregular network topologies, coupled interferences, and random user movements, in terms of the load balancing performance.

Load Balancing for Ultradense Networks: A Deep Reinforcement Learning-Based Approach

Traffic offloading is a widely used technique to address the unbalanced traffic distribution between pico cells and macro cells in heterogeneous cellular networks. However, the shifted users may result in the macro cell receiving large traffic from multiple pico cells and then becoming heavily loaded. This phenomenon is called the exacerbation problem in this paper. In order to address this problem and balance load effectively, this paper proposes a self-optimised joint traffic offloading (JTO) scheme. The JTO scheme jointly employs two traffic offloading techniques, including cell biasing technique to offload traffic between pico cell and macro cell, and mobility load balancing among macro cells. Simulation results show the JTO scheme can effectively deal with the cell exacerbation problem, in terms of handover failure and call dropping. The JTO scheme can also reduce the call blocking probability.

Self-optimised joint traffic offloading in heterogeneous cellular networks

Mobility Load balance(MLB) is an key technology belonging to self-organization networks(SONs) in Long Term Evolution (LTE) system. In the paper, a high-efficient algorithm is proposed to balance the uneven load between neighboring cells in LTE system. The algorithm adjusts the cell-specific offset between neighboring cells(OCN) adaptively depending on the load difference between neighboring cells. If the load difference exceeds the presetting threshold, then the algorithm will be triggered and adjust the OCN by adding or subtracting an adaptive step-size. Also, a power function is proposed to characterize the adaptive step-size and the step-size will be larger as the load difference between neighboring cells increases. This feature of the power function makes the MLB algorithm more efficient to fit the largely uneven load between neighboring cells. Furthermore, a simulation platform with 7 cell is set to evaluate the algorithm with different power of the function. It demonstrates that the adaptive algorithm can get the lower call blocking rates and handover failure rates because of its high-efficiency in load balancing. It also shows with the small appropriate power, we can get better system performances.

A High-Efficient Algorithm of Mobile Load Balancing in LTE System

In this paper, a new scheme about mobility load balance considering system overhead is proposed. The scheme adjusts the handover parameters, the cell-specific offset between neighboring cells (CSON), adaptively based on the load ratio between neighboring cells. The first order derivation of the CSON on the load difference is defined as the urgency function of the CSON adjustment and the characteristics of the urgency function are analyzed. Besides, a quadratic function is adopted to give an example to the CSON adjustment and the condition of the handover oscillation avoidance is derived. Furthermore, we develop a long-term-evolution(LTE) simulation platform to evaluate the proposed algorithm. Numerical results show that the scheme brings an important gain on call admission rate and the handover failure rate with less handover times.

Pengfei Li

Papers

A High-Efficient Algorithm of Mobile Load Balancing in LTE System

Algorithm about mobility load balance considering system overhead on LTE system