File Name: design and deployment of small cell networks .zip
Create an Account - Increase your productivity, customize your experience, and engage in information you care about.
- Increasing capacity and coverage where it matters the most
- Why SpiderCloud Small Cell Solutions?
- Financing and Pricing Small Cells in Next-Generation Mobile Networks
How people connect in urban areas will become increasingly important as we densify networks for 5G. Easy to install, low-cost and high performing solutions will be at the forefront of this densification, including small cells and Ericsson's Street portfolio.
Increasing capacity and coverage where it matters the most
The exponential growth of mobile traffic means that operators must upgrade their mobile networks to provide higher capacity to final users. Most of the planning strategies outlined in the literature are aimed at reducing the number of SCs and ignore the impact that the transport segment might have on the total cost of network deployment. In this paper, heuristics are used for the joint planning of radio i. Citation: Araujo W, Fogarolli R, Seruffo M, Cardoso D Deployment of small cells and a transport infrastructure concurrently for next-generation mobile access networks.
This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Competing interests: The authors have declared that no competing interests exist.
Some industrial and academic specialists predict that Global IP traffic will increase nearly threefold over the next 5 years, and will have increased fold in the period from to Allied to this growth, other new types of internet connection are emerging, that have led to new networks like Vehicular Networks VN [ 3 ] and Wireless Sensor Networks WSN [ 4 — 8 ]; it is predicted that these will be combined with others, forming a new paradigm called Internet of Things IoT [ 9 ].
With the development of information network, the popularity of IoT is an irreversible trend and generating new demands that must be assured. Examples are IoT demands [ 10 — 12 ], impacts, and implications on sensors technologies [ 13 — 15 ], big data management [ 16 ][ 17 ], and future Internet design for various IoT use cases, such as smart cities, smart environments [ 18 — 22 ], smart homes, etc.
These new technologies demand even more capacity, not only in terms of throughput but also in latency, and this will require a considerable investment in the mobile network infrastructure. The Fifth-Generation 5G of mobile networks encompasses all the requirements of IoT and has the capacity to interconnect all existing and emerging technologies. However, improvements in the physical layer alone cannot support such a high data rate [ 2 ][ 23 ], and there are many other means of improving the data rate of mobile users in HetNets, e.
The coverage of SCs generally ranges from ten to several hundred meters. The types of SCs include femtocells, picocells and microcells—and these tend to increase in size from femtocells the smallest to microcells the largest. The deployment of SC networks raises several challenges; authors in [ 28 ] investigated their criticalness and according to the findings of a survey for public SC deployments, the backhaul network, electric power, and SC placement strategies, were considered to be the most serious issues.
Backhaul can be regarded as the most crucial factor that affects SC deployments, followed by questions related to scenarios in a coexisting macrocell environment [ 29 ].
Thus, Radio Network Planning RNP is essential to enable operators to deploy wireless cellular networks in a cost-effective manner, and include both radio and transport networks.
There are many possible means of improving the transport network, including the use of copper wire, microwave, and fiber optics. However, most of the current studies related to the base station deployment have focused on radio deployment scenarios without taking account of the transport network. Even when this factor is included, they are concerned with scenarios in which high-speed, fiber-enabled and wired backhaul sites, are available everywhere [ 30 — 32 ], a situation that is considered to be unrealistic by [ 33 ].
The cost incurred by deploying a high-capacity backbone network for SCs can be quite high. Thus, while previous studies have examined serious deployment problems, to the best of our knowledge, none of them has addressed the question of joint deployment of SCs with fiber access for backhaul nodes, which include minimum capacity requirements, interference mitigation techniques and user distribution.
In light of this, this paper seeks to make the following contributions: i a new set of heuristics for SC deployment that includes factors such as minimum Quality of Service QoS , coverage and SC interference; ii a discussion of the applicability of the proposed heuristics and others from the literature, based on the results of simulations. The paper is structured as follows: first, there is an overview of the problem of deployment in SC networks including transport features, which involves carrying out a review of related works on this subject.
Following this, there is an examination of the heuristics, together with the network elements concerned. Finally, the results are analyzed and the conclusions are summarized, along with some suggestions for future works in this area.
The deployment of HetNets comprising of Small Cells with a fiber-based transport system is expected to be a very attractive means of providing coverage and capacity in densely populated areas.
A fiber-based backhaul solution offers the high capacity needed to meet this requirement, but it is costly [ 2 ] and time-consuming to deploy, when not readily available.
Hence, when deploying the infrastructure of next-generation cellular systems, backhaul links should be included in combination with SCs to reduce network costs and optimize performance.
There have been extensive studies of RNP in the literature because of its importance. For example, Guo et al. Cheng et al. Coletti et al. In [ 39 ], a promising strategy was employed to offload a significant amount of data from a macro BS through an SC placement service. This approach was adopted as a means of dimensioning Long Term Evolution LTE cellular networks so that the number of BSs required to cover an area of interest could be determined.
It had to take into account factors such as user density, service subscriptions, resource allocation, and interference mitigation. In [ 40 ], the approach was extended to the use of simulated annealing for HetNets. In [ 41 ] a greedy micro BS deployment strategy was employed over the existing macro cellular network with the aim of maximizing the energy efficiency of the network while meeting the growing demand for capacity.
In [ 43 ], Helou et al. Network-centric and user-centric strategies are set out in [ 44 ], where the authors examine the resource allocation problem by determining the number of resources that must be assigned to the users by each BS. Both strategies involve conducting an analysis of a multihoming approach.
Previous recommendations have made significant contributions to the deployment of SCs. For this reason, this study supplements previous research studies by offering new strategies considering multiple features like interference, Signal-to-Interference-Plus-Noise Ratio, coverage and minimum QoS and comparing them with others in the literature.
SC deployment traditionally calculates coverage on the basis of traffic density. This traffic is difficult to characterize, especially in view of its dynamic nature and the shifting trends in usage patterns and social mobility.
Nonetheless, according to [ 45 ], a great deal of traffic information can be inferred and forecasts made on the basis of the following: i demographics: the distribution of the residential and business community on the basis of demographic data; ii the traffic system: vehicular data based on public transport and the movement patterns of private vehicles; iii fixed line data plans: based on a correlation with fixed line phone call records, most mobile data traffic occurs indoors. In the case of SCs deployment, when there is a set of possible cell-site locations, iterative techniques are usually used to scan the optimal locations.
Optimization methods such as integer programming, simulated annealing, and genetic programming algorithms, can be employed to search for optimal solutions. The mobile SC deployment problem is an NP-hard problem and this can be proved by a reduction from the SCs facility location problem. Moreover, in the same study, it was proved that this is a NP-hard problem. A good solution was found in [ 33 ], where a commercially available CPLEX linear programming solver was used to establish an optimization framework.
However, as pointed out in the paper, the computation time was very significant and depended on the scale of the dataset. If there is a need to plan a SC network for a large region, a heuristic approach may be required to achieve a satisfactory result within a reasonable period. The authors in [ 45 ] state that there is a temptation to deploy SC without articulated radio planning and rely on signal processing techniques to improve the performance.
The danger of adopting this approach is that it is hampered by a lack of effective interference mitigation techniques and also involves a huge increase in the network deployment costs.
For these reasons, this paper both uses and compares heuristics for SC and backhaul deployment, by including real-world factors such as the following: the existence of fiber resources that are sparsely located, interference, costs, coverage and QoS.
The downlink Signal-to-Interference-Plus-Noise Ratio SINR over a given subcarrier n assigned to user k , can be expressed as follows: 1 in which is the received power on subcarrier n assigned to user k by its serving BS b k ; is the thermal noise power; and is the inter-cell interference from neighboring SCs. It was assumed that all the SCs are transmitting with maximum power PS. The received power at user k from b k can be calculated by means of 2 , which relates the received power to a node and is the result of the transmitted power and the fading of the signal calculated by the Stanford University Interim SUI model [ 57 ] This can be expressed as: 2.
The value of is calculated by the three equations shown below: 3 4 5 In which:. By correctly assigning the input parameters, it is possible to simulate urban and suburban environments with shading [ 57 ]. Two techniques were employed for the first group and one for the second. These are outlined below, together with their peculiarities, as well as their benefits and drawbacks, which will be shown in the results section.
Consider a geographical area A in which a number of SCs must be deployed. In the interests of simplicity, each Sf element is termed a node. The deployment variable can be defined as follows:.
The users of the same cell are multiplexed in frequency, and the data of each user are transmitted on a subset of the sub-carriers of an OFDM symbol. Adaptive resource allocation and link adaptation techniques are essential to achieve the challenging targets of spectral efficiency and user throughput targets. In OFDMA systems, resource allocation techniques can make use of the time and frequency variations of the system to optimize the use of the available resources.
They exploit the available Channel State Information CSI at transmitter side so that they can carry out the power allocations and share the subcarriers with the users [ 60 ]. As stated in [ 32 ], it was assumed that N subcarriers were available for downlink transmission and that there was a predefined user distribution in an area of interest.
A simple model was employed for the non-heterogeneous distribution of users, by randomly distributing them over the whole map divided into quadrants. Four dense areas were created to characterize the office spaces, this distribution will be illustrated in the results section. The objective of the heuristics is to find the minimum number of SCs that can still ensure coverage and provide capacity requirements planning for all the users, while at the same time reducing the total cost of deployment, including that of both the wireless and wired infrastructure.
It can be assumed that each user can only be served by exactly one cell and, thus, user demand is indivisible. In formal terms, the problem can be formulated as: 7 8 Where:.
In addition to the objective function, there are some constraints that need to be noted: 10 11 In both cases all the candidate SCs are tested one at a time, and then the SC is removed, to find out if it was dispensable or indispensable.
Calculate the deployment cost of i Using variable Z and Eqs 7 , 8 and 9 ;. Calculate Shannon capacity of k according to Eq 6 ;. After this phase, is calculated the deployment cost of each i St using Eqs 8 , 9 and 10 Z is used as a binary variable, to determine whether the cost of transport will be considered in the cost function , and then St is sorted by cost in descending order, St will serve as input to Algorithm 2.
At the end of this phase, interference between SCs and UEs are calculated and the maximum capacity of each k UE is calculated, considering the resources available in each SCs i. In this algorithm, all i S is tested to determine if this SC is indispensable: the test is done by taking out SC and calculating whether the minimum requirements for coverage and QoS are met. In the last phase, algorithm 1 is called again to recalculate the UEs assignment and Shannon capacity.
After all tests, the remaining SC are deployed and are interconnected by means of Prim's algorithm [ 61 ], which is a greedy algorithm that finds a minimum spanning tree for a weighted undirected graph. This is a standard algorithm used worldwide in the literature for the purpose of making comparisons, although our framework is flexible enough to use any other algorithm. Unlike the previous heuristics, the Type 3 T3 does not include an initial set of SCs to be tested; instead, the users' positions are noted so that the semi-optimal locations to deploy SCs can be found.
A K-means clustering algorithm was used to find these locations [ 62 ]. In the first stage, it is necessary to know the number of centroids in this case, the number of SCs that will be used.
Before carrying out this task, it is essential to find out how many users each SC can serve. Algorithm 3 summarizes the heuristic. The map is divided into several quadrants all of the same size. The tasks of counting the number of users, detecting the number of SCs and calculating the position of the SCs centroids , are carried out locally, in each quadrant. In line 12 is checked if the SCs created attend the minimal coverage and QoS of all users.
In this way, the capacity of the network is increased. An example is given of an application of heuristics in a typical urban area in Stockholm Sweden. The scenario under examination was modeled as a Manhattan street grid, and no macro layer was included. As discussed in the earlier sections, areas with dense SC deployments were added to our simulated network to illustrate some of the key operations of the 5G networks.
Why SpiderCloud Small Cell Solutions?
Driving fiber-deep architectures within the enterprise, our complete and comprehensive solution promotes and advocates for optical convergence. Want to download a copy of the brochure? Open the pdf in the link below and save it to your device or computer. Corning offers products to improve cellular coverage and capacity coverage in the most demanding enterprise environments. Corning solutions include:.
Bringing together world-renowned experts from industry and academia, this is an indispensible volume for researchers, engineers, and systems designers in the.
Financing and Pricing Small Cells in Next-Generation Mobile Networks
Welton V. Araujo 1. Edvar da L. Oliveira 1. Daniel da S.
Pune, Aug. Growing awareness about the advantages of small cell technology will be one of the primary drivers of the growth of this market. To overcome these obstacles, small cells are deployed.
Skip to search form Skip to main content You are currently offline. Some features of the site may not work correctly.
Холодные серые глаза смотрели безжизненно. Живший в ее сознании герой умер, превратился в убийцу. Его руки внезапно снова потянулись к ней в отчаянном порыве. Он целовал ее щеки. - Прости меня, - умолял. Сьюзан пыталась отстраниться, но он не отпускал. ТРАНСТЕКСТ задрожал, как ракета перед стартом.
- Женщина улыбнулась и протянула ему тонкую изящную руку. - Дэвид Беккер. - Он пожал ее руку. - Примите мои поздравления, мистер Беккер. Мне сказали, что вы сегодня отличились.
Да я бы ничего и не взял у умирающего. О небо. Только подумайте.
Черт возьми! - не сдержался Фонтейн, теряя самообладание. - Он должен там. Ищите.
Не было видно даже кнопочных электронных панелей на дверях кабинетов. Когда ее глаза привыкли к темноте, Сьюзан разглядела, что единственным источником слабого света в шифровалке был открытый люк, из которого исходило заметное красноватое сияние ламп, находившихся в подсобном помещении далеко внизу. Она начала двигаться в направлении люка. В воздухе ощущался едва уловимый запах озона. Остановившись у края люка, Сьюзан посмотрела .
Глаза его неотрывно смотрели на Чатрукьяна. - В него попал зараженный файл, сэр. Я абсолютно в этом уверен.
Семь ступеней. Он мысленно прорепетировал предстоящее убийство. Если у входа на площадку взять вправо, можно увидеть самый дальний левый угол площадки, даже еще не выйдя на. Если Беккер окажется там, Халохот сразу же выстрелит. Если нет, он войдет и будет двигаться на восток, держа в поле зрения правый угол, единственное место, где мог находиться Беккер.
- Его доводы звучали волне убедительно. Сьюзан перевела дыхание. Энсей Танкадо умер. Вина ляжет на АНБ. - Мы успеем найти его партнера.
Сьюзан не было дома. Он не мог понять, куда она подевалась. Всякий раз включался автоответчик, но Дэвид молчал.