Joint Optimization of Content Placement and User Association in Cache‐Enabled Heterogeneous Cellular Networks Based on Flow‐Level Models

This paper investigates the joint optimization problem of user scheduling, user association, and resource partition in heterogeneous cellular networks (HetNet) with a general concave utility function used as the performance metric. We formulate the joint optimization problem, and decouple the problem into three sub problems. After proving the sub problems belong to the set of problems that maximizes a monotone sub modular set function with mastoid constraint, we solve them by the proposed greedy based algorithms with theoretical approximation factors. Extensive simulation results demonstrate the efficiency of the proposed algorithms in terms of system utility. In addition, we evaluate some assumptions and results in the related work to show their impacts and correctness.

Heterogeneous Networks (HetNets) have gained the attraction of the communication industry recently, due to their promising ability to enhance the performance of future broadband Fifth Generation (5G) networks and are integral parts of 5G systems. They can be viewed in multi-dimensional space where, each slice represents a unique tier that has its own Base Station (BS)s and User Equipment (UE)s. Different tiers cooperate with each other for their mutual benefit. Data can be interactively exchanged among the tiers, and UEs have the flexibility to switch between the tiers. The cells in such a heterogeneous cellular networks have variable sizes, shapes, and coverage regions. However, in HetNets with ultra dense BSs, the distance between them gets very small and, they suffer from very high levels of mutual interference. To improve the performance of HetNets, we have done multiple contributions in this dissertation. First, we have developed analytical derivations for optimizing pilot sequence length which is a very crucial factor in acquiring the Channel State Information (CSI) and the channel estimation process in general. Poisson Point Process (PPP) has been widely used to allocate BSs among various tiers so far. However, BS locations obtained using PPP approach may not be optimum to reduce interference. Therefore, in this dissertation, BSs locations are optimized to reduce the interference and improve the coverage and received signal power. Also, we have derived expressions for static UEs coverage probability and network energy efficiency in HetNets. A proper UE association algorithm for HetNets is a great challenge. The classic max-Signal to Interference and Noise Ratio (SINR) or max-received signal strength (RSS) user association algorithms are inappropriate solutions for HetNets as UEs in this context will tend to connect to the Macro BS, which is the one with the highest signal power. A severe load imbalance and significant inefficiency arises and impacts the performance. The aforementioned algorithms tend to associate UEs to BSs with the best received signal power or signal quality. In HetNets, usually Macro BSs are the ones transmitting the strongest signals; hence most UEs tend to associate with the Macro BS leaving Micro BSs with less load. Also, the conventional max-SINR and max-RSS algorithms do not provide adequate results in multi-tier systems. We suggest two centralized algorithms, LSTD and RTLB, for an even UE association to provide fair load distribution. However RTLB outperforms LSTD in real time scenarios as it easily and quickly adapts to rapid network changes. Furthermore, we consider the mobility of nodes. We derive coverage probability for moving UEs considering both handover and no handover scenarios. Proposed algorithms are fast enough to associate the moving users to different Micro and Macro BSs appropriately in real time. Our algorithms are proved to be feasible and provide a path towards attainable future communication systems.

In ultra-dense networks (UDNs), the dense deployment of base stations (BSs) is facing challenges due to the pronounced unbalanced traffic loads, severe inter-cell interference, and uncertain traffic demands. In this paper, we tame traffic uncertainty for the joint optimization of BS activation and user association in UDNs to mitigate interference and balance traffic loads among BSs. Specifically, we address the traffic uncertainty by using chance constraint programming with the known first- and second-order statistics of the uncertain traffic. We formulate the joint BS activation and user association problem as a mixed integer non-linear programming problem, which is then decomposed into a set of user association sub-problems by modeling the BS states (active or idle) as a Markov chain. We solve the user association sub-problem at each BS state by transforming it into a convex problem over the positive orthant. In particular, at each BS state, the candidate serving BSs that lead to the optimal load balancing performance are identified for each user and parts of the user's traffic are offloaded to the identified BSs. Based on the obtained solutions, we propose a distributed near-optimal BS activation and user association scheme. Numerical results demonstrate that our proposed scheme is more robust to traffic uncertainty and provides better load-balancing performance than the existing schemes.

We introduce a sleep mechanism for energy efficiency in heterogeneous cellular networks, where macro base stations and small base stations coexist in the same frequency band. The strategy is to put high-power macro base stations into sleep mode and offload the users to low-power small base stations or neighboring macro base stations. To do so, we first formulate a joint optimization problem to minimize the total energy consumption while maintaining the QoS of users. Then we decouple it into two subproblems: user association and resource allocation, and macro base station sleep mechanism. We propose a modified many-to-one matching algorithm to solve the first one and a voting-based dynamic sleep mechanism for the second one. Simulation results show that our scheme has both lower energy consumption and lower user blocking ratio compared with existing sleep mechanisms.

In this paper, a joint control method for the user association and almost blank sub-frame (ABSF) based interference coordination (IC) technique in heterogeneous cellular networks is proposed. The optimization objective is simultaneously improving system throughput and guaranteeing proportionally fairness. Usually, user and base station (UE-BS) association is performed before ABSF rate determination, which is suboptimal with respect to UE- BS association. We formulate the joint optimization of UE-BS association and ABSF rate as a combinatorial optimization problem. To deal with it, we proposed an iterative method to discover the ABSF rate, the optimal association and the optimal resource blocks(RBs) allocation that can maximum long-term average throughput. Besides, as we cannot know the interference before user-association and resource allocation, we proposed a graph-based BS clustering method to estimate the interference before user-association and resource allocation. Simulation results show the efficiency of our proposed scheme.

This paper proposes a jointly optimized user association and time/frequency-domain scheduling method that maximizes the downlink system throughput defined based on (p, «)-proportional fairness. A previously reported user association method by members of our research group maximizes the downlink system throughput considering the base station (BS) distribution, user distribution, and the average channel conditions between each of the BSs and user terminals. However, the overall performance of the user association and subsequent time/frequency-domain scheduling conducted at each BS against the connected users taking into account the instantaneous channel variation due to multipath fading is not sufficient, although this approach can be actualized without relying on inter-BS cooperation. The proposed method assumes the centralized radio access network (C-RAN) architecture and achieves joint optimization of the user association and time/frequency-domain scheduling with regard to the achievable system throughput defined based on (p, «)-proportional fairness. Computer simulation results show that the proposed method significantly enhances the achievable system throughput compared to the user association method based on the average channel conditions, especially when the number of pico BSs is large and the system throughput prioritizes the fairness among users.

Traditional cellular networks require the downlink (DL) and uplink (UL) of mobile users (MUs) to be associated with a single base station (BS). However, the power gap between BSs and MUs in different transmission environments results in the BS with the strongest downlink differing from the BS with the strongest uplink. In addition, the significant increase in the number of wireless machine type communication (MTC) devices accessing cellular networks has created a DL/UL traffic imbalance with higher traffic volume on the uplink. In this paper, a joint user association and resource partition framework for downlink-uplink decoupling (DUDe) is developed for a tiered heterogeneous cellular network (HCN). Different from the traditional association rules such as maximal received power and range extension, a coalition game based scheme is proposed for the optimal user association with DUDe. The stability and convergence of this scheme are proven and shown to converge to a Nash equilibrium at a geometric rate. Moreover, the DL and UL optimal bandwidth partition for BSs is derived based on user association considering fairness. Extensive simulation results demonstrate the effectiveness of the proposed scheme, which enhances the sum rate compared with other user association strategies.

A novel multi-stream carrier aggregation (MSCA) enabled coordinated multi-point (CoMP) framework in heterogeneous cellular networks (HetNets) is proposed. Our objective is to maximize the energy efficiency (EE) of the proposed framework by jointly optimizing base station (BS) clustering, user association, sub-carrier assignment and power allocation. The formulated problem is proved to be NP-hard. By analyzing the structure of the original problem, we decouple it into power allocation sub-problem and sub-carrier assignment sub-problem. For the power allocation sub-problem, we propose sequential minimization technique and Zoutendijk feasible direction method to find the optimal power allocation with fixed sub-carrier assignment. For the sub-carrier assignment sub-problem, we propose genetic algorithm (GA) to represent the BS clustering and sub-carrier assignment, and design two GA based joint optimization frameworks. One is using the power allocation algorithm for fitness evaluation and guiding the evolution of GA. Another is optimizing these two sub-problems in an iterative manner, whereas the optimal sub-carrier assignment is updated timely as any change of the power allocation. Numerical results validate the feasibility, fast convergence, and flexibility of the proposed algorithm, and is superior to conventional MSCA or CoMP enabled HetNets.

Heterogeneous network is a novel network architecture proposed in Long-Term-Evolution~(LTE), which highly increases the capacity and coverage compared with the conventional networks. However, in order to provide the best services, appropriate resource management must be applied. In this paper, we consider the joint optimization problem of user association, subchannel allocation, and power allocation for downlink transmission in Multi-cell Multi-association Orthogonal Frequency Division Multiple Access (OFDMA) heterogeneous networks. To solve the optimization problem, we first divide it into two subproblems: 1) user association and subchannel allocation for fixed power allocation; 2) power allocation for fixed user association and subchannel allocation. Subsequently, we obtain a locally optimal solution for the joint optimization problem by solving these two subproblems alternately. For the first subproblem, we derive the globally optimal solution based on graph theory. For the second subproblem, we obtain a Karush-Kuhn-Tucker (KKT) optimal solution by a low complexity algorithm based on the difference of two convex functions approximation (DCA) method. In addition, the multi-antenna receiver case and the proportional fairness case are also discussed. Simulation results demonstrate that the proposed algorithms can significantly enhance the overall network throughput.

SummaryWe investigate the problem of joint downlink wireless backhaul bandwidth (WBB) and power allocation in heterogeneous cellular networks (HCNs). A WBB partitioning scheme is considered, which allocates the whole bandwidth between the macrocell and small cells for data transmission and backhauling. We formulate an optimization problem to maximize the weighted sum logarithmic utility function by jointly optimizing WBB portion and fronthaul power allocation of each base station with consideration of the backhaul capacity limitation on each small cell. In order to solve this joint optimization problem, we propose a hierarchical two‐level approach and decompose the original problem into two independent subproblems: the WBB allocation at the macrocell base station (MBS) and the power allocation at both the MBS and small cell base stations (SBSs). Accordingly, the optimal WBB portion and power allocation solutions are obtained, respectively. Furthermore, we develop a distributed algorithm to implement the joint WBB and power allocation. Numerical results verify the effectiveness of the proposed approach and analyze the impact of the weighted coefficient and backhaul capacity limitation on the network performance. In addition, significant performance gains can be achieved by the proposed approach over the benchmark.

Wireless distributed cache system (WDCS) is a promising technique to offload traffic and improve energy efficiency in content-centric networks. In this paper, the content sharing problem is considered by minimizing the average energy consumption for content caching and delivering in device-to-device (D2D) enabled cellular networks. The problem is formulated as a joint optimization problem of multiple interactive variables, which is NP-hard. In order to solve this problem, we first decompose the original problem into three subproblems, which are content helper (CH) selection, content caching, and content delivering. Specifically, in CH selection, we propose a PageRank based distributed algorithm to estimate the centrality of mobile users (MUs) by jointly analyzing their social relationships and channel conditions. Based on the selected CHs, the content caching and content delivering are formulated into two weighted matching problems. Two belief propagation (BP) based distributed algorithms are then proposed by considering content popularity and channel states to solve the aforementioned problems. Furthermore, based on the correlation of the three subproblems, a heuristic-based alternating iterative optimization method (HAIOM) is proposed to improve the average energy consumption performance of the content delivering of the WDCS. Simulation results show that by jointly optimizing the CH selection, content caching, and content delivering, the proposed scheme is capable of reducing the average energy consumption and the average content delivery delay, as well as of increasing the caching hit ratio.

The mobile edge caching is a promising way to reduce the user-perceived delay and improve the transmission data rates for the wireless networks. However, the cache capacities of base stations (BSs) tend to be limited and users’ interests for the contents are diverse, which makes the content placement decision critical for the network performance optimization. Besides, due to the flexible user-BS association, it is more complicated to optimize the content delivery and placement which are coupled with each other. This paper investigates the content placement and content delivery strategies in the cache-enabled wireless networks. In particular, the effective capacity , which can characterize the end-to-end user-perceived delay and data rates simultaneously, is introduced as the user’s utility metric. As the content caching and content delivery operate in different time-scales, they are investigated separately. For the content caching, a content placement problem is formulated, where both users’ different active levels and diverse content preferences are considered. Due to the NP-hard nature, the problem is decomposed into two sub-problems, and an iterative association-aware content placement algorithm is proposed. For the content delivery, the user-BS association problem is formulated, and a cache-aware user-BS association algorithm is designed. The performance of the proposed algorithms is evaluated based on the simulations. The numerical results show that the proposed algorithms have a better capability to cope with users’ diverse active levels, and achieve a better performance in terms of effective capacity and fairness level, compared with the existing algorithms.

In this paper, we analytically derive the average content delivery delay with delayed packet forwarding in ICN (Information-Centric Networking), which is a mechanism to intentionally delay forwarding of a content request packet (Interest packet) at routers. In ICNs, when a router receives multiple Interest packets requesting the identical content, it can aggregate those Interest packets into a single Interest packet. This makes it possible to reduce the number of Interest packets and Data packets transferred through a network. In this paper, we focus on the delayed packet forwarding that intentionally delays Interest packet forwarding at routers, which aims to increase the aggregation probability at routers. By introducing the delayed packet forwarding to a router, we expect that the aggregation probability is increased, which leads to the reduction in the amount of traffic transfer. In contrast, to intentionally delay packet forwarding at routers might increase content delivery delay between an entity and a repository. In this paper, we therefore derive the average content delivery delay and throughput with and without the delayed packet forwarding in a dumb-bell network topology to quantitatively reveal the effectiveness of delayed packet forwarding. We also investigate the effectiveness of delayed packet forwarding through numerical examples and simulation experiments. Our findings include that, by introduction of delayed packet forwarding, the throughput can be improved by 12% when the output buffer of the repository is small.

The dual connectivity is emerging as a promising solution to boost capacity in heterogeneous networks. However, it is challenging to obtain an optimal user association in heterogeneous networks with dual connectivity, due to its non-convex and combinatorial nature. In this paper, we propose a user association scheme based on deep reinforcement learning to maximize the overall network utility, which takes both throughput and user fairness into account, in the downlink of a heterogeneous network. Particularly, each user associates with the macro base station (BS) and a micro BS. We apply a deep Q-network (DQN) to obtain the nearly optimal policy to associate the users and micro BSs. Simulation results demonstrate that DQN-based user association performs better compared to the conventional user association schemes in heterogeneous networks with dual connectivity, and it behaves good scalability when the environment changes.

In this letter, we investigate an interference-aware user association scheme with cell sleeping technique in heterogeneous cloud cellular networks where the operation state of each base station (BS) is controlled by a cloud computing center. A user association problem which maximizes the user aggregate utility is formulated. Due to high computational complexity of the traversal method, we propose a distributed heuristic algorithm to get the suboptimal solution. At each cycle, the user selects the best BS by imposing a price which is proportional to the interference the BS provides. The BSs without any associated user will be switched off. Based on the varying network topologies, different combinations of active and switch-off BSs can also be obtained during different cycles. Numerical simulation results indicate that the proposed scheme can achieve remarkable throughput gains under the same power consumption, especially in large scale networks.