In my last post, WLAN Design for IoT is all in the prep, I summarised the importance of preparation and scope of works for a project. In this article, I discuss the importance of designing the WLAN for capacity.
Vint Cerf, one of the founding fathers of the internet, once said. “There’s nothing special about wireless networks except that wireless capacity is sometimes less than what you can get, for example, from optical fiber”. While part of this quote holds true still to this day I don’t think Vint Cerf realised that there will be approximately 26 billion devices on the IoT by 2020. With all this in mind ensuring when designing a wireless network, the key is adopting the right approach. With today’s networks having to deal with the proliferation of all mobile devices and other IoT devices, while still supporting the normal business applications and use cases of old. In the past, many wireless networks had been designed to provide basic connectivity for casual internet browsing and other low bandwidth applications within a defined service area. This coverage-based design model has previously worked in providing basic wireless access, however a different approach is required to support the explosion of IoT. So what am I telling you? In essence we need to start thinking about and accommodating for the current and also anticipated capacity and performance levels you require rather than just focusing on providing basic coverage (unless this is all that is required).
So why do we need to change the approach? When using a coverage-based design model, we are only taking into consideration one metric, signal strength, to define acceptable user experience which has demonstrated to be a poor indicator of network success. A coverage-based design in most cases will not take into consideration many critical variables required to meet both capacity and performance needs, such as:
- Reducing (where possible) co-channel interference
- Meeting bandwidth, latency and jitter requirements of specific applications
- Understanding device capabilities and quantities
- Load balancing between APs/radios (based on airtime/load)
- Providing spectral capacity
- Provide QoS end-to-end
So is there a better approach to take? Of course the answer is yes! A better approach is to take a capacity-based design approach which takes into account high-capacity planning and design. Also it is key to use networking equipment that is high performance and has smart features built in to assist in optimising the RF (Aerohive recently released a product that can provide Dual 5GHz capabilities). So only through the use of good requirements gathering, network design, configuration and optimisation can wireless networks meet the demands of dense user and device populations. As we have seen in the previous blogs through the process of capturing business and technical requirements we can identify network design objectives and design the network accordingly to provide the desired performance levels. The key to getting this right is understanding the following:
- The client devices being used on the network
- The capabilities of these devices (A great source of device capabilities here)
- Client device density per coverage area
- Applications required per device
So I now understand my device capabilities, what comes next? As we have already eluded to network design involves a mixture of identifying requirements into a network architecture that meets the capacity and performance requirements. This would need to be backed up by understanding and adopting design principles for capacity-based design over the traditional coverage-based design method. This process is will allow you to determine access points required and radio capacity needed to support the total client and application requirement, as previously defined. This alongside Wi-Fi best practices of selecting the right placement for the access points, antennas and other required accessories for optimal coverage and importantly performance. As we have already seen in the previous blog post Wi-Fi surveys and WLAN preparation is key to this. Once you have completed a survey it will translate into the network configuration, this is from a combination of the technical requirements and from any conducted site survey and capacity planning exercise (We will cover this stage in the next blog post). Once the network has been configured as required there is always a need to optimise and adapt to changes, network management platforms can assist in providing this type of statistics and metrics. Aerohive’s network management system (HiveManager NG) provides a facility to view statistical information and reporting on key performance metrics which are relevant to successful operation of the Wi-Fi network, which will allow you to identify and change any design or performance related issues. Features such as Health View and the Troubleshooting Dashboard provide a great facility to view these metrics.
Next in the series we will look into WLAN configuration. This is an important stage in defining how the WLAN network should be configured. It will detail configurations for security, QoS, application control, rate limiting, radio setting, SSID settings and so on. Stay tuned…..