Troubleshooting WiFi – How to Resolve Poor WiFi Experiences in Enterprise Networks

All IT professionals are familiar with variations of the “WiFi isn’t working” complaint. Determining the root cause of network connectivity issues can be time consuming, even more so if the initial reported problem cannot be replicated. This guide will first walk through the traditional methods used to troubleshoot WiFi issues and then how to troubleshoot the same issue using Wyebot’s Wireless Intelligence Platform ™ (WIP). 

Challenge Number One: Capturing the Problem at the Right Time

Wireless users face problems while running typical cloud applications, transferring files, streaming and other normal daily activities. When a problem occurs, ideally IT staff is available to collect data immediately for analysis, problem identification and resolution. However, we know this rarely the case. For some IT professionals, the more common scenario is that the problem is reported after hours or on the weekend. Higher education campuses report peak network usage is between 10pm – 2am. Different time zones can cause problems for IT teams responsible for multiple national or international offices. The inability to capture data at the time of a disruption greatly complicates eventual resolution. In some instances, problem replication might be impossible. While this could mean the issue has resolved itself, it can be frustrating for the client and IT because there’s no way to be sure the problem will not reoccur.

Even in situations where the problem could not be replicated, it is good standard operating procedure to gather basic, defining information surrounding the issue:

• Number of clients experiencing the problem

• Device types, hardware and drivers

• Specific location and APs affected

This will be useful if the situation does repeat itself. For ongoing connectivity problems, use the information to help plan the next steps.

Challenge Number Two: Analysis and Resolution

Successful troubleshooting requires data capture and analysis. Use all available captured data to begin analysis, while always remembering that gathering more information might be required to diagnose the issue. There’s no one-size-fits-all answer as to what is wrong with a WiFi network. Problems can even be due to interference from a client or device in an adjacent building. Data should be analyzed with regard to several different scenarios:

Scenario One: Interference and/or Distance from an Access Point

If WiFi devices can’t hear each other, they can’t communicate. Non-WiFi devices, like microwaves, other WiFi devices, or other wireless networks can all cause interference. A proper site survey, done at the time of AP installation, should guarantee consistent coverage in desired areas. However, office layouts change over time and more users may be on the network. The explosion of IoT devices – many of which operate only on the congested 2.4GHz spectrum – combined with the new norm of BYOD can all change an office and impact the behavior of the wireless network.

If interference or distance from an AP are the presumed issues, check the device’s signal strength on the WLAN controller. Many controllers do not contain a history of the signal strength so, this information needs to be reviewed immediately. Attempting to diagnose WiFi issues under these circumstances requires analysis of packet traces. If done manually, this process is time-consuming and tedious.

There are a number of apps on the market designed to give basic WLAN information. For Windows users, WiFi Analyzer is a popular one. It displays a real-time graph of the signal strength of any nearby wireless networks, and suggests which channels to use. It also gives a few details about the currently connected network: what channel it uses, its bandwidth and its protocol. WiFi Explorer is a similar macOS app. It displays every nearby wireless network on two graphs, one for the 2.4GHz band and one for the 5GHz band. However, these types of apps usually only report information on the APs and not the clients. 

Scenario Two: Roaming Issues

Since the inception of wireless networks, roaming has always caused problems. On most WiFi networks, the decision to roam to a new AP is made by the client. This presents a problem for administrators whose networks are blamed for roaming issues that are really caused by the client. The most common roaming problem is the “sticky client” where a client hangs on to its old AP even though an AP with a stronger signal is available. This can sometimes be alleviated by changing settings on the client or making changes to the wireless network.

The IEEE has made standards such as 802.11k, 802.11v, and 802.11r that are designed to create a more seamless roaming experience for wireless clients. They each have their own defining elements, restrictions and troubleshooting requirements. If a client attempts Fast Transition (FT) Roaming using 802.11r, and the new association fails, was it because of a failed FT authentication request? Were the messages exchanged using over-the-air FT Roaming, or over-the-DS FT roaming?

Clients using assisted roaming with 802.11k use neighbor reports from APs that list nearby APs and which channel they are using. This tells the client which APs are the best candidates for roaming. This feature, like all roaming features, are reliant on support from both the client and AP infrastructure. Interoperability problems between devices can end up causing even more problems. It’s also possible that some AP vendors have made proprietary enhancements to their software.

Scenario Three: Software, Drivers and OS Upgrades

The software of the WLAN infrastructure should always be kept updated. Unlike other network equipment – like wired switches – the software on WLAN equipment is constantly changing. The WiFi industry is very dynamic and vendors must keep up with new standards and certifications.

Client devices should also keep their software updated. This is difficult for administrators to enforce unless they are using some type of mobile device management (MDM) software. Furthermore, some mobile phone service providers prevent the device software from being updated. Issues with outdated client software show why it’s important when troubleshooting WiFi problems to figure out whether the client or the WLAN is at fault.

Scenario Four: Channel Overlap

Overly crowded wireless channels can lead to poor wireless performance. Due to high-density work and school environments, combined with the number of devices everyone carries, APs need to be placed even closer together. This leads to a greater chance of channel overlap and interference. While this has always been a problem in the 2.4GHz spectrum – with only 3 non-overlapping channels available – it’s also now become a problem in 5GHz with the introduction of 80MHz channel widths as part of 802.11ac standard. 

Administrators must use techniques such as reducing AP radio power, removing lower data rates and disabling some radios on 2.4GHz. On 5GHz, channel widths should be reduced to 40 or 20MHz. DFS channels should also be used to ensure enough unique channels. Bandsteering should be enabled to move clients onto 5GHz radios. It may be tempting to remove networks from 2.4GHz altogether, but as previously mentioned there are still many IoT and wearable devices that only support 2.4GHz. Separate SSIDs can be created for these devices and the main production SSIDs can be on 5GHz only.

All WLAN vendors support some type of automatic channel selection (ACS) that allows APs to dynamically choose which channels to use. The problem is ACS doesn’t always make the best decisions. Different vendors implement this feature differently and most do not dynamically change the channel width being used. Channel planning is critical during the installation of the network, but it’s also important to monitor it on an ongoing basis.

Wyebot’s Wireless Intelligence Platform ™

Rather than combing through packet data and eliminating scenarios one-by-one, it would be far more efficient – and far less frustrating – if the cause of the problem were known immediately. This would lead to faster resolutions, improved user experience and allow IT staff to move on to other tasks. With Wyebot’s WIP, there’s only one troubleshooting scenario regardless of the problem:

Step 1: Review dashboard interface for any issues

Step 2: Review detailed analysis and suggested action steps for resolution

Step 3: Resolve the issue

Step 4: Setup scheduled network tests to automatically test and monitor the WiFi network

Wyebot’s vendor-agnostic platform works with any WLAN controller and APs. WIP’s sensors work 24/7 to capture and analyze all traffic using an Artificial Intelligence engine. It is the only existing product providing full packet capture and automatic analysis. The AI-based engine produces an easy and human-friendly summary of all data, in real-time. WIP’s dashboard interface displays any connectivity issue, the cause of the issue, and suggested actions to take for quick resolution. If problems arise after-hours, WIP is still working, capturing and analyzing data. IT personnel receive everything they need as soon as they log in.

WIP stores historical data, making it easy to check past issues and track client and AP network usage. It creates detailed fingerprints for all client devices and access points in the network. With this data, WIP proactively detects any change in the network and automatically reports it. This gives IT the capability to stop problems before they happen, thereby eliminating network downtime and improving performance. Wyebot is the leader in WiFi Assurance. Use it today to optimize your WiFi network.