What it takes to build a network

By
Kunal Agrawal
Senior Strategy Associate

When we boil it down, there are four fundamental parts to building a telecom network. To explain this best, let’s pretend you’ve started your own telecom company. Let’s call it… WOMBAT (Worldwide Over-the-air Mobile Broadband and Telephony).

Part 1: Getting your future customers “covered”.

One of the first things you’ll need to figure out is ‘coverage.’ In other words, how you can effectively “cover” an area of land (and all the people that live, work and play there) with your cell towers and the radio waves they emit.

The more covered an area is, the better the stability and speed of a person’s connection. There is a lot of consideration that goes into deciding where to build towers to ensure the best web of radio waves and in turn, the best coverage. Some of the questions you’ll need to answer include:

  • “Where should I build towers to maximize coverage while building the fewest number of towers to keep costs down?”
  • “How much would the rent, electricity and water supply cost to have, power and cool a cell tower here vs there?”
  • “Can we even build a tower at these locations or are there limitations? (e.g. terrain accessibility, land ownership etc.)”

Once you’ve carried out some rich data modeling, analysis and research to answer these questions, you can start to build your towers (or add to one that already exists nearby!).

This evaluation and identification of optimum cell tower locations is one critical part of building a Telecom network.

A typical cell tower can provide service up to ~45 miles(~72 kilometers) away but to maintain strong, stable connection, you’ll often find towers spaced much closer; ~2 miles (~3kilometers) apart in suburban areas and as close as 0.25 miles (~0.4 kilometers) in dense metropolitan areas.

Part 2: Making sure data is relayed swiftly and securely.

While you’re figuring out exactly where you can, and should, build your cell towers, you’ll also need to figure out how your network is actually going to run. Whether it’s a call to a friend, a virtual doctor’s visit, loading a YouTube video, or playing a game of Clash of Clans online, how are you going to ensure your customers have a smooth experience?

This is where the data centers Ayaka Takeuchi mentioned come in. It’s also where buzzwords like cloud and edge computing play an important role.

Let’s take on the elusive cloud first.

In the simplest terms, the “cloud” is centralized computing power and server space that can be accessed anytime, anywhere. Think of it as the computing equivalent of your home’s electricity and water. You get it from a centralized supplier and can access the amount you need, from where you need it, and pay for what you use. Simple.

Services that use the cloud that you might already know about include Dropbox, Google Cloud and Rakuten Drive. Because your information and work are saved on the cloud, if you have your password, you can access it from anywhere, on any device! The same logic applies to your online shopping accounts, social media, other internet-based services, and WOMBAT.

Now, the speed that data is transferred and processed at is ridiculously quick. But the more data you need to send back and forth, the longer it can end up taking. After all, data still travels between cloud data centers and cell towers through physical wires and there is a limit to how much data these wires can transfer effectively (this is what’s known as bandwidth).

So how can we ensure our customers’ data will always be transferred and processed quickly and reliably?

Well, this is where edge computing comes in.

Edge computing is a relatively new advancement in cloud technology and is increasingly being adopted in many tech sectors today. To put it in simple terms, edge computing cuts down the distance data needs to travel. It does that by breaking up the centralized computing power and server storage of one big data center into many smaller pieces.

It places the pieces closer to where they’re generated and needed by users. In other words, the data is processed closer to the “edge” of your new telecom network and the end-users. This way, data is much more accessible, and people get a quicker, smoother and all-around better experience. No more painful loading spinners!  

Did you know, Gartner estimates that by 2025, 75% of data will be processed outside the traditional data center or cloud.

This setup and delivery of cloud and edge computing services is a second critical part of building a telecom network.

Part 3: Stacking the units of functionality in between.

We’ve got our towers and we’ve got our data centers – we’re good to go, right? Well, not quite... Between your tower and your data centers, there’s actually a lot of additional functionality that needs to be in place for a network to run properly. For example, you’ll need things to:

  • Send radio waves signals from your cell towers to your customers’ devices and vice versa.
  • Boost the signal as necessary to make it stronger and reach devices further away.
  • Convert these signals into digital signals that can be transmitted to and from data centers.
  • Carry out this conversion of signals in the most efficient and speedy way possible.
  • Manage all the signals on your network to ensure they don’t get mixed up.

These ‘things’ are called “units” in telecom industry lingo. Each one has a unique unit name and has to be meticulously layered and linked together

  • Send radio wave signals from your towers to your customers’ devices.
    → Antenna
  • Convert radio wave signals into digital signals for transmission.
    → RRU: Remote Radio Unit
  • Synthesize the numerous signal conversions that are taking place at any one given tower.
    → BBU: Baseband Unit
  • Manage all real-time and near real-time processing needs in the most efficient way.
    → CU: Centralized Unit and DU(s): Distributed Unit(s) (as part of the BBU)

This isn’t an all-encompassing list of the U’s of telecom (take a look at our Jargon Dictionary for that!) and in fact, the units above really only scrape the surface of just one part of a network. In any case, to achieve each objective, every unit needs to be paired with its own specialized hardware and/or software. For a smooth WOMBAT network, you’re going to need a heap of different components. There’s good and bad news at this point.

The good news is that you can get most of the hardware from one vendor which simplifies things. The bad news is that because this one-vendor model has been the status quo for decades, there haven’t really been any meaningful strides forward in technological capability. And on top of that, because there isn’t that much choice when it comes to vendors, the prices for this equipment is shockingly high.

What might also surprise you is how hardware dominant the telecom industry is. Whereas other industries have adopted software into every fiber of their tapestry, in telecom, this isn’t the case. It’s nota stretch to say that walking into the telecom industry is a bit like visiting your grandparent’s home and finding you have to use a dial-up internet connection again... Don’t know what that is?

Exactly.

In any case, let’s look on the bright side. Say you’ve got an unlimited budget and you’re going to make the best of what the current limited number of vendors have to offer. The sourcing, layering and linking of the functional units explored above, is a third critical part of building a telecom network.

Part 4: Maintaining and optimizing your network.

Congratulations! You’ve successfully deployed all the infrastructure you need for the launch of the WOMBAT network. And you’ve already got your first million customers? Amazing. Now, it’s all about making sure your network runs in the best way it can for customers, 24/7.This sounds simple enough on the surface, but there’s a lot of work that goes into this too. To be a world-class telecom network you’ll need to be able to:

  • Monitor each of your thousands of towers and ensure the radio waves they’re emitting are strong and stable for all your customers, wherever they are.
  • Identify, assess and fix any coverage or connectivity problems that surface – whether it’s physical damage to a tower, a bug in some code or something else entirely.
  • Safeguard the data being relayed to and from your data centers while also making sure it’s processed quickly and correctly.
  • Manage customer queries, questions and quandaries about their service or account in a timely, professional way.
  • Upgrade the hardware or software your network depends on whenever you or a partner company develops a new cutting-edge piece of tech.

And much, much more.

As you can imagine, all these tasks require a whole host of very different tools: from frequency optimization tools to customer management applications and everything in between. It’s having and deploying this array of tools and applications, that’s a fourth and final critical part of building a telecom network.

But...

...of course, a truly world-class telecom company needs world-class talent and creativity, both technical and non-technical. At Rakuten Symphony, we think we have an exceptional team but are continually looking for more talented people, like you, to join us.

Here’s the next logical question – what exactly is Rakuten Symphony’s place in all of this? Well, to learn about what we’re bringing to telecom and how we’re changing the landscape for good as one global team, have a read of Mamadu 'Basco' Bah’s brief post next!

Jargon dictionary

#
3GPP: 3rd Generation Partnership Project
  • An umbrella term for current seven standard development organization partners. The 3GPP creates protocols for companies (Mobile Broadband Standards) covering cellular telecommunication technologies to ensure qualities.
  • A collaborative project between seven participating organization partners. 3GPP develops standards for mobile telecom systems to ensure industry qualities.
A
AI: Artificial Intelligence
  • AI performs tasks that previously required human intelligence. AI has abilities to recognize patterns and learned information to perform tasks and make decision, which takes human errors and time consumption out of the equation.
  • AI derives from notion of ML (Machine Learning). ML is a concept of feeding computer information and training the algorithm to recognize the data patterns and make predictions based on the knowledge.
API: Application Programing Interface
  • A type of Software that makes between application data sharing possible. When an application has a request for another application, API serves as an automated liaison and determines the method of sharing depending on the application.
  • API represents set of protocols, routines and tools for building a software application. API allows different software applications to communicate and share needed data with each other.
C
CapEx: Capital Expenditure
  • Comes before OpEx, as it includes the non-consumable functions like land, equipment and others. 
CU: Centralized Unit
  • As a component of a 5G network architecture, CU is responsible for planning, allocating resources and securing the RAN. The CU coordinates with the DU on how to stay efficient.
  • In a traditional RAN, the CU and DU are placed together in a Base Band Unit (BBU). Companies sold them as a set, creating limitation in flexibility. However, in an Open RAN setting, the two are places separately, DU typically by the edge and CU in a central location.
D
DU: Distributed Unit
  • A type of computer used in an Open RAN environment (also known as Next Generation Distributed Unit). A DU makes sure the flow between the RAN and the core network (main part of the network) is smooth and efficient.
E
EBPF: Extended Berkeley Packet Filter
  • Extended Berkeley Packet Filter, is a technology that makes it possible to run special programs deep inside the Linux operating system in an isolated way. As it filters data packets from networks and embeds them into the kernel, the BPF also provides a network interface with security layers that ensures the packet data is reliable and accessible.
  • Using this approach, teams can more easily and efficiently collect crucial observability data from Linux applications and network resources.
G
G: Generation
  • Since with the launch of the first generation in 1979, the industry birthed 2G, 3, 4G and now 5G, each representing significant strides in network capabilities.
K
K8S: Kubernetes
  • Usually abbreviated as K8s, this is a type of software that automates some management functions of a containerized applications from deployment to scaling.
  • What is a container? Think of it as Tupperware that has everything you need. The lightweight and standalone container has codes and needed dependencies/elements to run an application.
  • The bigger the scope, more container you need. That’s when K8s comes in to help reduce human led tasks by automating container deployment and storage management and more.
  • Abbreviated as K8s, this software optimizes containerized applications. Containers are lightweight and efficient package that hold necessary code and data and K8s provides a way to manage and orchestrate them.
L
Linux:
  • Built on a Unix server, Linux is a free and open-source operation system that is customizable, flexible and secure. Because it’s open, anyone can access and modify it.
O
OpEx: Operation Expenditure
  • All the costs needed to run and operate a network day-to-day, like wages, energy cost and more.
R
RAN: Radio Access Network
  • RAN is a liaison between networks and user devices. Through radio waves,  RAN enables communication between our devices and the core network.
RIC: RAN Intelligent Controller
  • A computer program that monitors and optimizes the performance of a wireless network. For example, when it detects poor coverage with high demand, the RIC will send more resources to improve the reliability.
S
Small Cell:
  • A miniature, low-power and cost-effective version of a cell tower that enhances network reach and capabilities in a dense urban area, low population rural area or indoors.
Z
Zero Trust:
  • One of many security concepts that protects organizations from cyber threats. Zero trust means literally ZERO trust – that every user, device and application need to be verified and only necessary resources are granted for use.
Scroll To Top

Don't miss
an opportunity

Register for email updates on our job opportunities.

Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.