Evaluation of Rural Broadband Network Based on Broadband Universal Service Management System

[1]. Providing broadband universal service (BUS), similar to education and water, has been labeled as a fundamental human right in several countries (including France, Finland and Spain) and the United Nations [2][3]. In many regions around the world, policymakers have made great efforts to promote broadband network construction. The government of UK announces a universal service obligation of 10 Mbps in 2020 for every home in Britain [4].The European Commission sets up a target of 30 Mbps for all its population by 2020 [5]. In China, the State Council issues the “Broadband China” Strategy and Implementation Plan to facilitate the rollout of broadband service. Ac -cording to the plan, 98% of administrative vil -lages will have access to broadband network by 2020, with the minimum network speed of 12 Mbps [6].Connecting rural impoverished areas to broadband network is an essential step for achieving ubiquitous broadband access [7]. Those areas are often characterized by harsh environment and geographically remote, which bring inevitable technical challenges in network deployments. Even where technology is available, the development of broadband network in impoverished area is usually facing severe economic pressures [8]. Generally, theAbstract: Carrying out pilot project to pro -vide broadband universal service nationwide, especially in rural impoverished areas, is a major policy decision in China. To acceler -ate implementation and ensure quality of the constructed network, it is of great significance to conduct comprehensive and scientific eval -uation of the network status. In this paper, we present the evaluation of the broadband net -work constructed in rural China with several key indicators. It shows that with stepped-up efforts of the telecom industry, broadband networks have been introduced into more and more villages. The average network speed reaches 60 Mbps, which is far exceeding 12 Mbps’ obligation.Keywords: network evaluation; data analysis; broadband service; rural areaI. B ACKGROUNDBroadband network is known to be crucial for stimulating investment, accelerating eco -nomic growth and social development, as well as overcoming obstacles to development. Statics show that every 10 percent increase in the penetration of broadband networks leads to a 1.38 percent increase in the economic growth in low and middle-income countriesReceived: Sep.12, 2018Revised: Dec. 10, 2018Editor: Yuefeng JiEvaluation of Rural Broadband Network Based on Broadband Universal Service Management SystemZejue Wang *, Meimei DangChina Academy of Information and Communications Technology, Beijing 100191, China ∗ The corresponding author, email: wangzejue@C OMMUNICATIONS T HEORIES & S YSTEMSseeable future.To guarantee the project schedule and qual -ity of the constructed network, it is of great significance to conduct comprehensive evalua -tion of the implemented network status. How -ever, to the best of our knowledge, technical approach rarely has been previously proposed. Several countries have established approaches to evaluate broadband network speed, e.g., America, UK, Korea and Japan etc. America and UK delegate a third-party company to per -form network speed test by means of develop -ing friendly-users (no more than 5,000 users), the data of which is obtained in a non-contin -uously and non-automatically manner, and the amount is limited. As most villages without access to broadband network of China locate in the areas with harsh nature condition and inconvenient traffic, the evaluation of broad -band deployment by artificial means is over -whelmingly difficult and barely executable. Therefore, it’s necessary to develop a technical approach to perform scientific evaluation. In this paper, on the basis of the proposed man -agement system, the status of rural broadband network in China is evaluated with the realiza -tion of automatic data acquisition.Specifically, the technical approach and the structure of the proposed management system are introduced in Section II. By analyzing the data collected through the system, Section III provides a demonstration of the broadband network in rural China with several key indi -cators. Conclusion and future works are given in Section IV .II. B RIEF I NTRODUCTION OF BUS M ANAGEMENT S YSTEMA BUS management system is proposed to assess the network quality by technically ob -taining the deployment and running status of the network, on the premise of involving little or no users’ privacy, as well as no influence on the operation and management of the current network.The conceptual diagram of the BUS man -agement system is shown in Figure1. On themore remote and the less populated area, the more it costs in network deployment. Study shows that the cost of broadband construction in the rural area outside a town outperforms that in the town by 80% on average [9]. The technical barrier and economic pressure make the access of high-capacity broadband net -works in rural impoverished areas lags far behind than that in urban and suburban areas.To strengthen communication infrastructure constructions and further improve basic work -ing and living conditions in rural areas, the Ministry of Industry and Information Tech -nology and Ministry of Finance of China have launched a pilot project to promote broadband network construction in administrative villag -es 1, especially those locate in impoverished areas, since the end of 2015 [10]. This proj -ect has established a four-phase construction schedule as follows: (1) the first two phases (from January 2016 to the end of 2017) sup -port the construction of fiber-optic broadband network in almost 100,000 administrative villages, including 30,000 poor villages; (2) the third phase (from June 2017 to the end of 2018) facilitates fiber-optic broadband access in 30,000 more villages; (3) the fourth phase (from June 2018 to the end of 2019) is to support 4G network coverage in rural areas. Along with the widespread deployment of broadband network, the number of villages achieving prosperity through broadband net -work will undoubtedly skyrocket in the fore-1In this paper, a village covered by broadband network refers to that it has access to fiber-optic broadband network or 4G network.Fig. 1. Architecture of the BUS management system.BUS -MPServer NodeDedicated fixed CPEInternetRouterOLTOLTTerminal DeviceWired Transmissionapplication. 2) Otherwise, the dedicated fixed CPE can receive the geographic information from its associated OLT, where the geographic information acquired through the application is already saved in the OLT.The BUS management system is able to collect information about the network status in real-time. On the basis of the massive dataset, we conduct an objective evaluation of the de -ployed network with some key indicators, and present the result in the following section.III. B ROADBAND N ETWORK E VALUATION OF R URAL A REASTo demonstrate the broadband network status, in this section, a detailed analysis is providedfrom perspectives of evaluating the network construction, the network average speed and the users’ liveness. In this paper, 800,000 re -cords are selected to perform the evaluation, which are reported from 150,000 dedicated fixed CPEs widely deployed in the rural areas of China. The time span is from Oct. 2016 to Dec. 2017.3.1 Broadband constructionFigure 2(a) and Figure2(b) present the distri -butions of administrative villages and house -holds 2 that have gain access to broadbandnetworks, respectively. It can be observed thatthe less-developed west and central regionsare the priority regions for broadband network er Premise Equipment (CPE) is deployed by engineer at users’ home, and then converged to the BUS management platform (BUS-MP) by connecting to Internet through Optical Line Terminals (OLTs). The dedicated fixed CPEs are traditional Optical Network Units (ONUs) except for satisfying corresponding industry standards of China.The dedicated fixed CPEs are deployed in broadband subscribers’ houses. The high per -formance BUS-MP is deployed on the cloud, which supports gathering and processing us -ers’ information in real time, and is equipped with abilities of big data analysis, graph pre -sentation, table exporting, etc. These collected data are stored on the servers as records and can be extracted and processed by the BUS-MP. A record refers to a received message at the BUS-MP reported by the dedicated fixed CPE, which can be a registration message, proactively informing message, or passively informing message.• R egistration message. When the dedicated fixed CPE is first installed, it communicates with BUS-MP with a registration message to report its information, including operatorID, vendor ID, serial number, uplink type, etc.• P roactively informing. To inform its active state, the dedicated fixed CPE can com -municate with the BUS-MP with its regis -tration code. Besides, the dedicated fixed CPE is equipped with speed test functions, which are automatically triggered and cor -responding results are sent to the BUS-MP proactively.• P assively informing. To stay up-to-date on the latest state of the CPE, the BUS-MP can also acquire information from the CPE.The dedicated fixed CPE can get the geo -graphic information of its installation location by one of the following two ways: from the connected terminal device or from the associ -ated OLT. 1) By means of a specific applica -tion developed for BUS, the engineer can ob -tain the location of a deployed dedicated fixed Fig. 2. Broadband network constructed in different regions.(a) Distribu ti on of administra ti ve villages that have access to broadband networks.(b) Distribu ti on of households bene fiti ng from broadband networks. 2Households here referto Internet users, notincluding IPTV users.Figure 3 further depicts the ratio of poor villages to administrative villages for which broadband network have been introduced. The purple, green and orange text represent the provinces locate in the east, central and west regions of China, respectively. It shows that the poor villages are the places where commu -nication infrastructures are in great shortage and supports of providing broadband access services are critically needed, especially those locate in the remote west region. For instance, among the total administrative villages that introduced BUS of Tibet Autonomous Region and Yunnan province, the ratios of poor villag -es are up to 99.6% and 73.2%, respectively. Benefiting in promoting economy devel -opment and improving people’s livelihood, strengthening broadband network construction is supported and encouraged by local gov -ernments. Except for deploying broadband network in administrative villages that on the pilot project’s target list (referred as “sched -uled construction”), the local government and relevant enterprises also aggressively expand broadband coverage in the administrative villages out of the target list (referred as “un -scheduled construction”). Figure4 provides the statuses of the scheduled construction and the unscheduled construction. From Figure4(a), it can be observed that the original target of the pilot project in broadband network construc -tion has been fully achieved. Moreover, the relevant enterprises actively take on the con -struction of additional broadband networks, and specifically, they exceed the original con -struction target for up to 21.8%. Figure4(b) provides the distribution of the constructed network that exceeds the target. It can be observed that the intensities of supports for broadband network construction in the west and central regions are greater than that in the east region, e.g. for the unscheduled construc -tions, 45.9% are deployed in the west region and 39.28% are deployed in central region. This phenomenon once again testifies the fact that the less-developed rural areas are the pri -ority areas for providing broadband service.construction. For instance, among all the ad -ministrative villages with broadband network services, those locate in the west and central regions account for 53.44% and 31.18%, re -spectively. As for the households live in the west and central regions that benefit from the constructed network, the ratio is 31.05% and 62.86%, respectively. Statistics indicate that the communication infrastructure of the west region has been greatly improved, given the fact that this region is geographically remote and sparsely populated which bring natural obstacles in network construction.Fig. 3. Ratio of poor villages to administrative villages with broadband network services.Fig. 4. Broadband constructions included in the original schedule versus that not included.0.00%10.00%20.00%30.00%40.00%50.00%60.00%70.00%80.00%90.00%100.00%Xinjiang NingxiaQinghai GansuXizangYunnanGuizhouSichuanGuangxi14.82%39.28%46.29%EastCentralWest21.80%(a)Scheduled construction versus unscheduled construction.(b)Distribution of the unscheduled constructions.China has the highest average broadband net -work speed with 83.77 Mbps, followed by the central region, with 65.78 Mbps. Even the re -mote west areas where plateaus and mountains are often located and bring natural obstacles in communication infrastructure deployments, the average speed is slightly lower than that of the central region, and the difference is no more than 6 Mbps.Figure 6 and Fig.7 depict the distribution of the average value and the temporal trend of the network speed, respectively. It can be observed that the average broadband speed of more than 90% of all users range from 20 Mbps to 100 Mbps, where the speed of the majority (78.3%) ranges from 40 Mbps to 100 Mbps. Fig.7 provides the cumulative distribu -tion function (CDF) curves of the historical average speed and the speed averaged over the recent month. It is shown that the curve of the average speed of the recent month lags behind that of the historical value. The gap between the two curves implies that the network speed is gradually increasing, and the broadband net -work is well maintained since it constructed. Besides, it can be observed that the minimum value is almost 20 Mbps which far exceeds the standard of 12 Mbps.3.3 Broadband users’ livenessIn addition to network speed, which is an im -portant indicator of the network capability to provide service, users’ liveness is also an im -portant factor to evaluate the network whichFig. 5. Broadband network speed over different regions.Fig. 6. Distribution of broadband network speed.Fig. 7. CDF curve of broadband network speed.EastCentralWest83.7765.7859.83Mbps3.21%(20,40] Mbps 13.47%(40,60] Mbps 26.14%(60,80] Mbps 25.27%(80,100] Mbps 26.92%4.98%C D Fconstructed network status. It is shown that the government has given great supports to pro -mote rural broadband network construction, and gain remarkable achievements. Compared with the east and central regions of China, the less-developed west region which usually ac -companied by harsh environment and fragile ecology, is the priority region for broadband network construction. Besides, the average broadband network speed of the west region reaches 60 Mbps. This paper focuses on the first two stages of the pilot project, the future work will cover the third stage data and analy -sis will be provided from more dimensions.References[1]P .Meltzer et al., “The Internet, Cross-Border Data Flows and International Trade,” Asia & the Pacific Policy Studies , vol. 2, no. 1, 2015, pp. 90-102.[2] Z . S. Bischof et al., “ Need, Want, Can Afford:Broadband Markets and the Behavior of Users,” Proc. Internet Measurement Conference , 2014, pp. 73-86.[3] M . Zheleva et al., “Internet Bandwidth Upgrade:Implications on Performance and Usage in Ru-ral Zambia,” Information Technologies & Inter-national Development , vol. 11, no. 2, 2015, pp. 1-18.[4]V .Stocker et al., “Who replies to consultations,and what do they say? The case of broadband universal service in the UK,” Proc. European Re-gional Conference of the International Telecom-munications Society , 2017.[5] E uropean Commission, “A Digital Agenda forEurope,” 2010, from: http://eurlex.europa.eu/le-gal-content/EN/TXT/PDF/?uri=CELEX:52010D-C0245&from=EN.reflects users’ preference and satisfaction about the constructed network. A user who has gotten online at least once in the recent time interval ∆t is identified as an active user. In this paper, the average of liveness of all users is viewed as a benchmark, and relative live -ness is expressed as a difference with respect to the benchmark. Figure8(a) and Figure8(b) presents relative liveness of different regions over different time intervals. It shows that the east region has the highest liveness benefiting from kind geographical environment. In fact, parameter ∆t , up to a certain extent, can reflect users online behaviors. For instance, from Figure8(a) and Figure8(b), it can be observed that longer time interval will bring down the absolute value of the relative liveness, i.e., it reduces the difference of the users’ liveness among different regions, which is mainly be -cause of the fact that users in central and west regions tend to turn off their CPEs when not using them.IV. C ONSLUSION AND F UTURE W ORKBroadband network access is crucial in so -cially and economically empowering rural impoverished areas. To improve broadband infrastructure construction, promote economic development and improve people’s livelihood in rural areas, the pilot project for providing broadband services to rural areas has been implemented in China since 2015. This paper provides a comprehensive evaluation of theFig. 8. User’s liveness over different regions.BiographiesZejue Wang, received PhD de-gree in Signal and Information Processing from University of Chinese Academy of Sciences. She is now an engineer in the Institute of Technology and Standards Research, China Academy of Information andCommunications Technology. Her current research interests include broadband network and wirelessnetwork. Email: wangzejue@.Meimei Dang, received Mas-ter’s degree in Electronic and Telecommunications Engineer-ing from Beijing University of Posts and Telecommunications. Now she is vice chief-engineer in the Institute of Technology and Standards Research, ChinaAcademy of Information and Communications Tech-nology. Her current research interests focus on broadband network, wireless network, Internet of Things, etc. Email: dangmeimei@.[6] N one, “’Broadband China’ strategy: networkand application,” China Communications, vol. 10, no. 9, 2013, pp.8-8.[7] T . Leanne et al., “Enhanced broadband access asa solution to the social and economic problems of the rural digital divide,” Local Economy , vol. 28, no. 6, 2013, pp. 580-595.[8]W .Brian et al., “Broadband’s contribution toeconomic growth in rural areas: Moving to-wards a causal relationship,” Telecommunica-tions Policy , vol. 38, no. 11, 2014, pp. 1011-1023.[9]J .Schneir et al., “A Cost Study of Fixed Broad-band Access Networks for Rural Areas,” Tele-communications Policy , vol. 40, no. 10, 2016, pp. 755-773.[10] “Notice of the Ministry of Finance and theMinistry of Industry and Information Tech-nology on Carrying Out Pilot Project of Pro-viding Broadband Universal Services,” 2015, from the Ministry of Industry and Information Technology of the People’s Repulic of China Web: /zhengwuxinxi/caizhengwengao/wg2016/wg201602/201606/t20160601_2117309.html.。