Jump to: navigation, search

Meeting notes

1. Technology Overview

Principle overview


A detailed discussion of this architecture is found at Nextelco:Technology

Developments in Africa

Africa has jumped from no man telephone land to a wide mobile and wireless voice and telephone services. This is expressed in the fast increase of Tele-density ratio where some countries as Gabon - with a population of about 1,5 Million inhabitants, a total area of 267.667 Km2 , and a GDP of 5.723$- moved only in one decade from 1% of telephone penetration to about 100% of mobile Tele-penetration. If Africa erased the telephone gape as quickly as we see, this is not the case for internet where the continent is far away to cope to the growing data and internet traffic demand (the social media, e-learning, e-banking,..).

In the DRC Congo for example with a population estimated to 70 million, a GDP of about 500$, an area of about 2,5 million Km2, and a mobile Tele-density of about 16%, the country still has an internet penetration rate of less than 1%. The GSM operators continue to leverage -on very demanding & draconian terms – a very poor GPRS service to prepaid end users. The costs of migrating from 2G to 3G and 4 G are restricting GSM operators to update their infrastructures. Classical ISP’s- Vodanet, Microcome- are leveraging monthly postpaid internet service only to corporates and business customers. We can then see that a major part of the population, prepaid users, youth and modest and low revenues, simply said consumers, has been excluded to benefit a reliable internet service.

Unless that the data and internet deficit is differently expressed between the cities and the rural areas, in general the necessity is strongly present in both cases. If it happen that the GSM introduced the 3G, it is for sure that this service will be limited to big towns while in the rural areas the internet situation will remain unchanged.

We see the problem in this case of DRC Congo: On one hand the GSM operators, with service provision based on prepaid voice service, inherit an obsolete infrastructure and technology unable to curry high speed data. On the other hand the classical ISP’s banded to their wireless Ethernet based technology that can deliver high speed data. But the network is not originally planned for retailed billing of single data session.


The proposed solution is a result of deep observation and investigations that bring us to solve the problem as a whole, considering both corporates & business and consumers users in general. In our search to the solution, we consider the global trends and common traits, as below listed:

• Data traffic demand will grow significantly than the data and internet revenues

• The GSM operators are not able to proportionally update their networks

• Big part of the generated data traffic will be mobile and wireless

• 1 of 3 produced mobile handsets will be Wi-Fi capable

The solution consists to equip a wireless network with the retailed and prepaid capabilities and provide broadband access to the existing prepaid mobile users independently of the nowadays African GSM networks. Our solution is a Wi-Fi Public Hotspot resulted from an integration of different technologies: Satellite-Wimax-Wi-Fi. To secure the back end service quality, we use back to back the QoS of each technology and integrate those seamlessly in our Billing and access Control application, the Eprox. The created internet backhaul is then accessed by the end user as a retailed prepaid service through a Wi-Fi Public Hotspot. Depending to the cases, the Wi-fi Public Hotspot can be seen as a single satellite node or Wimax node.

The access device from End user can be a:

• Smarthphone

• Laptop with inbuilt Wi-Fi

• Laptop or PC with external long range dongle

• Tablet

We do believe that the Android based and the Nokia cheap range Smarthphones will dominate the African and emerging economy in general handsets market in the coming years. Having a smathphone in hand is a huge opportunity for the law income to connect to the internet.

Key customer segments

Our customers are divided into two wide categories: Businesses and consumers, but focus is on mobile prepaid users. In order to face the coming paradigm and to tailor products to each segment we have done market research through personal interviews. The interviews were conducted over two trips, each of two weeks, to the D.R. Congo with back to back meetings with potential customers, competitors, government actors and other stakeholders. We will position our product and use the actual and current existing distribution of the GSM Prepaid vouchers. We have not to forget international users who can avoid data roaming and opt for a cost effective Hotspot access to avoid the costly roaming postpaid invoices.

File:NextelcoSystem.pptx File:NextelcoTasks.pptx

Business Model

Nextelco Norway AS owns the internet resource which is controlled from the backbone using the Eprox billing and Control application. We secure the satellite provision and hardware by having a purchase as you grow contract with Vizada Networks in Norway. For Wimax and Wi-Fi we sign with Albentia in Spain.

Nextelco Norway AS partners with local companies that hold the license to provide public internet and frequency for Wimax service. The full service is licensed and leased out through a network of resellers, high-tech service providers who own and operate the Wi-Fi public Hotspots whose are fully paid before delivery.

The all service provided on each level of the distribution chain is a full prepaid service based on whole sales vouchers. Nextelco Norway AS is the root on the distribution chain and is the one who generate the bulk vouchers accounts while having visibility in all chain with multi-floor commission system.


Products and services

To serve the different customer segments Nextelco has introduced two main product lines: private satellite links, and broadband Hotspot services. Private satellite links generally targets large businesses, while the Broadband Public Hotspot product targets small businesses and consumers accessing the internet on their cell phone or laptop at a hotspot (this is a growing business model in the area). We have different values on the vouchers 1$, 5$, 10$ and 25$. The upfront value correspond to a given QoS. The lowest is 256 Kbps down against 64 Kbps Uplink.

2. Technology Vision

2.I. Introduction

WiMAX is a new 3/4G technology that offers Broadband Wireless Access (BWA). WiMAX technology supports broadband data and voice services over the same radio interface. It offers high data rates and network throughput with long transmission distances. WiMAX is a very cost effective technology and is fast to deploy making it a suitable solution for rural areas where difficult terrain limits the application of a wireless cellular network and in developing countries which have limited resources to deploy wireless networks.

WiMAX supports a wide range of services such as VoIP, e-mail and web browsing, FTP, streaming video and audio, and IPTV. It can also serve as a backhaul for Wi-Fi hotspots. In the business plan we have explained in details motivations for the choice on Wimax technology. We also worked out financial aspects for a rolling out of a user driven Wimax Network for a partner company, ISP in the DRC Congo, in the capital of Kinshasa using the Wimax 802.16 2009, the fixed and nomadic variant for outdoor applications. As largely extended on the other parts of the business plan document, the ISP's aim is to provide cheap and reliable internet access to the population of Kinshasa and the tourists visiting temporarily Kinshasa.

Kinshasa has a population approaching 8 million. The town has a surface of about 10.000 km². The pre-study sets a target of installing a Wimax access node each 30 km² on the first phase. This implies that the project will need an average of 300 BTS to cover the city of Kinshasa. We have to take into consideration that the computed area dont fully corresponds to the habitable and densely populated parts of the City. Some of the area may fall into Congo river or mountains surrounded the city of Kinshasa. Following REGIDESO, the national company for water distribution, they are about 400.000 houses registered in the city of Kinshasa. Furthermore we will in long term reach a density of one BS serving about 300 houses in each 120 ̊ sector in order to serve an average of 1.000 houses in full 3 sectors. We must every time thing compromise between the targeted range and the throughput per sector which is a determinant QoS criteria. One site with 3 sectors will provide a capacity estimated to about 3X24 Mbps. The fair bandwidth allocation must consider the criteria of leaving the end user with a minimum of 512/64 Kbps for web browsing and YouTube video and audio streaming.

The large habitable part is built mostly on a homogenous flat terrain. The city is covered ouest to East by a ring of Mountains starting from Mbinza to the Livulu-University. This situation makes the rollout much flexible. The incumbent operator provides 40 high sites for locations included access to the energy. The table 1 gives the GPS locations for all the sites. We can not implement the full network in one phase. Therefore we need to split the project in different phases. Starting with the test phase, the present has aim to demonstrate and test a prototype which is to be fully implemented and tested in Norway and then redeployed on the field in Kinshasa, DRC. If it is not possible in Norway to demonstrate the Demo network, then we will be limited on a simulation on paper and implement in practice the Demo Network only in field in DRC.

2.II. Scope of the demo network

The business plan to launch the above described Wimax service has been finalized and prospects of financing show very positive feedback on the investors and financing side. It has been set as mandatory, before to commit with investors, to test the technology and evaluate the business model in order to secure the investment. We would like to set a demo and prototype network to verify the theoretical assumptions made on the size of the network supposed to cover the entire city of Kinshasa. The network is made :

• The backbone and high speed transport part to be implemented on 10 Ghz microwave link

• A Wimax Wimax backhaul access network to be implemented on the 3,3 Ghz

• Wifi Access points on 2,4 Ghz

Without taking final choice on the equipment vendor, we among other contacted Arbentia, a Wimax vendor who propose solutions on the 802.16 2009, standard d for fixed and nomadic wireless technology. The vendor claims that: its Wimax solution produce the following performances:

• System has a theoretical range up to 20 Km Line of Sight

• A throughput of about 24 Mbps per sector

• That its BT is interoperable with standard Wimax clients from different manufactures

• The vendor claims again that the system provides a toll quality Voip service following the different QoS supported Wimax 802.11 2009 standard: BE, RTPS, nRTPS, eRTPS and UGS at the different layers: 2, 3 and 4.

• The 5,4 Ghz unlicensed link with a speed up to 35 Mbps in the range of 35-40 Km We will then verify and test against the reality the designed network on one side, and on the other verify by practical use the assumptions given by the Wimax vendor.

II. Test description

1.Test Components

We want to set in Kinshasa a first test network. Our Demo network will be constituted of the following elements:

1. Two BTS with with sector directional antennas

2. A Wimax point to point on the 5.4 Ghz.

3. A local server with different large files for test purpose

4. The site are alimented with a 230 V energy source backed with a 5 KVA generator

5. Manageable switch hub at the Wimax End point

6. The 4 Wimax clients are different vendors placed on different locations 1,2,5, 10 Km

7. Three Wimax clients are connected to laptops

8. One Wimax client is connected to a Linksys router providing a Wifi hotspot

2. Methodology of conducting the test and to collect and analyzing data from the different locations

The purpose is not to test the equipments but to demonstrate the network and services. The following will be largely used in the present test measurement:

• Setting a link budget for the different clients locations

• A data capture utility, as Netpersec , will be used to capture data at different layers: 2, 3 and 4

• We have to set locally and out of Internet a large file for test purpose

• The client has to download and upload local file and the internet file

• The test sample shall be repeated at least 100 times for each parameter we are measuring

• We have to connect service as Skype and Voip to test the QoS and QoE

• Make sure the BTS has clear path to 4 clients at the distant locations of 1, 2, 5, 7 and 10 km

• Single client access and simultaneous access and downloading and uploading

• The test is conducted with a team made of 5 guys: A test supervisor, ISP, a RF and radio engineer, an IT man and 2 students in electrical engineering

• All the data are captured as CSSV file plugged into excel to be analyzed and graphically presented

3. Key Performance Indicators to be measured

The performance indicators are key elements in the selection of criterias whose commercially differentiate the product presentations and packing. We will do different tests and measurements in order to isolate and confirm those performance indicators.

1. Local test with local server at the Wimax end point without Internet access via satellite

• The sending and receiving power, signal strength for all sites

• The throughput in different distances 1,2, 5, 7, and 10 km

• Frames lost and dropped packs in a period of time

• The latency, response time and jitter

• Downloading a FTP from a local server

• Uploading from Client to local server at the BTS

• Simultaneously down-and Up- loadind with 4 clients

2. Wifi Access point with Linksys 802.11 g,b,n with PC and Smartphone as connected clients

• The sending and receiving power, signal strength for all sites

• The throughput in different distances 1,2, 5, 7, and 10 km

• Frames lost in a period of time

• The latency and response time

• Downloading a FTP from a local server

• Uploading from Client to local server at the BTS

• The average size of pack transmitted

3. Test in layer 3 and 4 with QoS and QoE with PC and Smartphone

• Total ping time that includes also time to satellite then to the internet

• The throughput in the different distance 2,5,7 and 10 km

• Downloading and Uploading files from the different sites

• Protocols testing and measurements UDP, IP, TCP

• The AAA data capture into text format

• Voip on BE, RTPS, nRTPS, eRTPS and UGS

'III. Demo network requirements Budget

The table below gives the needed hardware and software needed to rollout the projected Wimax demo network and test the proposed Wimax solution. The vendor gives a special offer for 2 BTS with 4 Wimax clients and a Point to Point link on the 5,4 Ghz for a fixed price of 4.000$. The sector antennas are included in prices. The total price for this piece of network is about 15.000$ without transport and customs.

V. Concluding remarks

The presentation and evaluation of the outcome of the test of the demo network will show that:

• The vendor commercial assessments are meeting the practical and test data. If not we ‘ll give what we come up with and comment

• The data measured will be decisive in the coverage analyze

• Due to this, we may reduce the network size while also reducing the throughput or keep the same BTS number. We may have the same range of coverage but add more throughput per sector or group of houses or users.

• If the planned test work, then we start connecting users knowing that the demo site can connect between 200-300 users to Internet when connected to a shared or dedicated satellite link of 2 Mbps Downlink and a 512 Kbps uplink.

• Capturing of the data by the AAA server make it possible to connect a billing application for billing and prepaid management voucher management • We can use Back to back QoS and SLA serving the end users from the hotspot to Internet gateway

• The test is a technology detailed understanding of useful key parameters whose help to understand the product and service to be sold to the end users and compare proposed different solutions

• This participate to the training of local staff to assimilate the Wimax coming industry, use same terminology and procedures for performances

• The positive conclusion is a go signal for investing for the first phase rollout, project implementation and rollout

• We also tacitly get future performance for the fiber backbone access coming in this near future in DRC compared with the long delay satellite backbone access.


1. Signals Research Group HSPA+ and Mobile Wimax Network performance Benchmark Results and Analysis

2. EtherNet/IP performance: Test methodology for Ethernet /IP devices, version 1.0, March 2005

3. Methodology for Wimax performance Optimization using Drive Test Data. Fawzi A. Alghamdi and Ivica Kostanic.

4. IEEE Standard 802.16-2009, Part 16: “Air Interface for Broadband Wireless Access Systems”, May 2009.

5. WiMAX Forum Industry Research Report. Retrieved from

6. WiMAX Application Performance Testing Guidelines Raj Jain,Washington University in Saint Louis

Saint Louis, MO 63130 USA


2.1 Technology status

  • get the technology fixed
  • measure
  • present results (real throughput)

2.2. Infrastructure and Roadmap

  • create an infrastructure overview, consisting of elements like
    • satellite
    • satellite accelerator
    • WiMAX router
    • .....

Generate a CAPEX for the selected infrastructure

3. Customer

How to bring the customer equipment to the market, and establish the OPEX for traffic (based on A6 traffic pattern)

3.1 Customer with power

assume that the customer has a power supply

3.2 Customer with solar cell

assume that the customer has a battery (and or solar cell)

3.3 Customer segment with femto cell

calculate the OPEX costs for an operator access based on a femto cell on our platform.

4 business Model

4.1. Landline versus satellite

what is the cross-over point for a satellite link versus a landline link

  • talk with Nera Networks - Ceragon

4.2. Voice Traffic pattern

establish a traffic pattern containing

  • number of customers
  • traffic/customer
  • "busy hour" versus "low use" hour
  • traffic split:
    • within Lisala
    • traffic within Congo
    • international traffic

A7 Way ahead in partner collaboration

  • Bilateral collaboration with Orange, evtl. trhough Sofia Antipolis
  • Norsk-Fransk Stiftelse
  • EU-ICT-Africa proposal

A8 Nextelco and Mobile Operator

Our value proposition for mobile operators like Orange for going into a collaboration with Nextelco, including

  • a short intro on Nextelco
  • the add-on value for the mobile operator
  • the cost/function for femto