The demands on each generation of WiFi technology has been different. When it started off, WiFi did not have to address any of the applications that run over it today. It was just going to be a cable replacement for basic enterprise data. If 802.11ac was all about enabling high throughput, 11ax is expected to do a lot more. Customers have been promised higher throughput, lower latency (4x lower!!!) and higher user density. That is a lot to deliver on.
Although there have been announcements galore, there are a very limited number of devices (especially Clients) in the market today with full 802.11ax support. Chip vendors and OEMs are busy delivering products to their announced roadmaps. With no ready to use test equipment, interoperability remains a major challenge. It was being addressed at multiple Plug fests. 802.11ax is fundamentally more complex than earlier WiFi technologies, imposing many technical challenges for all participants in the ecosystem. Current testing is focused mostly on functionality using very specific MATLAB Models for 802.11 ax.
With this landscape in the backdrop, how does one start preparing and understanding what it will take to test a full-fledged 802.11ax solution? How is it possible to model what the user is going to experience. We being a Test & Measurement company, asked this question to ourselves, a year back. I would like to share our thought process here.
We need to go back to basics and ask what is unique in 802.11 ax? Unique does not mean just the features like OFDMA based physical layer multiplexing the users, higher MIMO or higher QAMs. We need to understand the User Experience aspect of AX solution, rather than focusing merely on technology features that are tangible. For user experience testing, basics are different. We need to predict the usage and try to create a similar environment and measure acceptable KPIs.
High Density WLAN with current technology
High-density wireless LAN network will have a large number of clients and more number of Access points within a specific area. Big campuses with multiple floors, distributed buildings/ large apartments, enterprises, and large event spaces are considered high density due to the number of access points and devices connected to the network
The most common examples of high density environments include:
- Stadiums, Arenas and Concert Halls
- Conference Halls, Exhibition Halls
- Campuses & lecture halls
- Transport hubs airports, railway stations
- Transport systems metros, trains, aircrafts, cruise liners
High-density environments with more number of devices and Access Points always create challenges for designing, deploying and testing the network. Connectivity has become as indispensable to our current generation as the air we breathe. Failing to ensure strong WiFi in a high-density environment can seriously impact service objectives and overall image. WiFi users will expect seamless availability and superior experience all the time.
The number of devices accessing the network and number of access points in dense deployments make the infrastructure truly unique and demanding. The density along with applications running on these devices are driving the demand for high reliability, and high-performance connectivity. Bandwidth requirements for some popular high bandwidth consuming applications are as follows:
|Streaming Audio||128-320 kbps|
|Video Streaming||728 kbps|
|Video Conferencing||1.5 Mbps|
|HD video streaming||768kbps 8 Mbps|
|4K video streaming||8 Mbps 20 Mbps|
For example, let us consider a large hotel conference hall that accommodates 1000 clients. Assume that there is a High-density WiFi network that is capable of handling 1000+ with an SLA of 2Mbps per user. When it comes to the real world, can you assure that all 1000 users can simultaneously stream a 2Mbps demo video without interruptions and artifacts? How do you verify the same before you give access to users? Do you think it will be easy to replicate the scenario with 1000 real devices?
How do you test User Experience
WLAN Installers focus on coverage rather than End-User Experience. To the end user, network experience matters irrespective of the technology deployed. You can not just rely on theoretical numbers and believe that network would work fine with 1000s of users. How can you then guarantee the user experience in dense deployments like Stadiums, Auditoriums, and Airports?
How to prepare for 802.11 ax?
HEW is also working towards developing and understanding High Density deployment scenarios. We also want to understand 802.11ax in the following stages (From HEW group):
- Trends What are the trends that are redefining WiFi usage?
- Usage Models Based on trends, what are the new usage models, that drive goals/directions.
- Goals of the technology Based on Usage Models, what goals are defined for 802.11ax. What is unique in 802.11ax compared to previous standards?
Understanding these, we will be able to get clarity on the real world scenarios from the 802.11ax perspective and replicate the same in a LAB to verify the goals
1. Trends: There are three major trends
- Increased usage of mobile devices most environments will involve high density of STAs and/or Access Points
- User demand for Wi-Fi anytime and anywhere has created demand in locations e.g. airports.
- Increased usage of WiFi leads to an increased density in traditional Wi-Fi environments as well e.g. apartments and enterprises
- Growing use of WiFi outdoors e.g. parks, campus, streets and events
- 2.4Ghz is congested and 5Ghz will follow soon as WiFi by design is not meant for high density
- Evolution of WiFi Applications
- per user average real-world throughput follows a regular increase (pushed by higher resolution video for instance)
- cloud services generating more Uplink traffic
- more and more peer-to-peer applications
- Advent of the Internet Of Things
- By around 2020, 50 Billion devices will hit wifi networks. They will demand smaller but regular packet transfer capabilities, again a super High-Density scenario.
2. Usage Models Let us delve deeper and understand components, classification and requirements posed by usage models derived from the above trends.
Usage scenario can be broken into a few components as below
- Pre-condition Initial conditions before the use case begins
- Environment Vertical or location for the network
- On Desk or in cubicle (short range, line of sight)
- Conference Room (medium range, line of sight)
- High density
- Hotspot in Public Places
- Airports, Bus Stations, Train Stations
- Restaurants, Exhibitions Halls, Shopping Malls
- Outdoor Hotspots
- Stadiums, Recreational places like Parks, Theatres
- Auditoriums/lecture halls, classrooms in schools, colleges
- Remote medical assistance in hospitals
- Flights, Trains, Buses, Ships
- Traffic Condition Background traffic or interference that is expected during the scenario e.g. interference from bluetooth
- Applications A few applications to consider are Video Conferencing, Virtual Desktop Infrastructure, Progressive Streaming, Interactive Multimedia & Gaming and more
- User Actions Actions by user or device.
Usage Model predicted going forward can be categorized as follows
- High density of APs and High density of wireless clients per AP
- Ex: Shopping malls, airport, exhibition halls
- High density of clients Indoor
- Ex: Auditoriums, buses, wireless office
- High density of APs & low / medium density of clients per AP
- Ex: Dense apartment building, community WiFi
- High density of APs and high density of clients per AP Outdoor
- Ex: Dense urban street, pico-cell street deployment, Macro-cell street deployment
- Throughput demanding applications
- Ex: Healthcare, production in the stadium, smart car
3. Goals Here they are
- Improve Spectral Efficiency and Area throughput
- Improving real-world performance in indoor and outdoor deployments
- In the presence of interfering sources, in indoor and outdoor deployments
- In moderate to heavy user loaded wifi Access Points.
Till now, technology chased high peak throughput in ideal condition. End to that era with 11ax. The goals of 11ax are efficiency, ability to work with interference, the ability to work in density and so on. There is an acceptance that the real world is far from ideal, so what can be achieved in ideal condition can at best maybe a lab test case.
The uniqueness of Goals compared to previous technologies
The uniqueness of 802.11ax compared to 11ac/b/g/n is categorized with respect to four elements as follows:
- With respect to the environment
- Hotspots for public access and cellular offload in dense areas for indoor & outdoor like street deployment use cases are unique
- With respect to applications
- 802.11ax focus on the user experience of all the clients in high dense scenarios (with a large number of clients and a large number of applications)
- With respect to objectives
- unlike 802.11ac which improved theoretical peak throughput, the target of 802.11ax will be to increase Wi-Fi efficiency by improving user experience by a fair and efficient repartition of throughput among all users/applications in real-world scenarios and to increase overall capacity per area
- With respect to metrics
- The metrics that are in consideration are per user average Throughput, Area Throughput (taking MAC and PHY overheads)
Aletheas approach to testing at scale
Alethea offers a solution, Distributed Device Architecture (D2A) to validate high-density deployments. D2A helps to validate the WiFi infrastructure end-to-end effectively. With a flexible, distributed & scalable architecture, ability to emulate more thanÂ 2000 stateful clients, real traffic generation and much more, the tool is in use by WLAN equipment manufacturers, chip vendors, service providers, and enterprise users to make sure that the products and networks they offer meet customer expectations.
Currently, D2A can emulate 802.11 a/b/g/n/ac clients and supports maximum throughput up to 1.3 Gbps / Radio Head.Â 802.11AX is coming soon and we will continue to support all the latest and legacy technology standards to test the scenarios with a mix of different wireless devices.
Alethea strives towards Perfecting Broadband and help the WiFi Ecosystem meet end users expectations.