Reliable connectivity is about more than having a fast internet connection.
For many organisations, particularly those operating in mobile, remote or mission-critical environments, the consistency of a network connection can be just as important as its headline bandwidth.
One of the factors that can undermine that consistency is network jitter.
Jitter can cause broken-up voice calls, unstable video, delayed data and unpredictable application performance. In environments such as emergency services, transport, defence, maritime operations and remote infrastructure, those interruptions can have a much greater impact than an occasional frozen video call.
Understanding network jitter, what causes it and how modern networking technologies can reduce its impact is therefore an important part of building resilient communications.
What is network jitter?
Network jitter is the variation in the time it takes individual data packets to travel across a network.
Information sent over an IP network is divided into packets. Ideally, those packets reach their destination at reasonably consistent intervals.
For example, packets might arrive approximately every 20 milliseconds. If network conditions change, however, one packet might arrive after 18 milliseconds, another after 35 milliseconds and another after 22 milliseconds.
That variation is jitter.
It is particularly noticeable with real-time applications, where information has to be received and processed continuously. These include:
- Voice over IP (VoIP)
- Video conferencing
- Live video streaming
- Remote monitoring
- Telemedicine
- Vehicle and vessel connectivity
- Command-and-control applications
- Real-time sensor and telemetry data
Downloading a file can often tolerate variations in packet arrival times because the application can simply wait for the missing information. A live voice or video application has far less flexibility.
Jitter vs latency: what is the difference?
Jitter and latency are related, but they are not the same thing.
Latency describes the time it takes data to travel from one point to another.
Jitter describes how much that time varies between packets.
A connection could therefore have relatively high latency but low jitter. A satellite connection, for example, may have a noticeable transmission delay but still deliver packets at reasonably predictable intervals.
Conversely, a network could have relatively low average latency but significant jitter. Most packets may travel quickly, but if their arrival times vary considerably, a real-time application can still perform badly.
For critical applications, consistency can therefore matter just as much as raw speed.
Jitter vs packet loss
Packet loss is another network-performance problem that is sometimes confused with jitter.
With packet loss, some packets never successfully reach their destination.
With jitter, the packets may arrive, but their timing is inconsistent.
The two can also occur together. Congestion, poor wireless conditions or an unstable network path may increase latency variation while simultaneously causing packets to be dropped.
This is one reason network performance cannot be assessed using bandwidth alone.
A connection advertised as providing hundreds of megabits per second can still deliver a poor experience if it suffers from high latency, packet loss or excessive jitter.
What causes network jitter?
Jitter can have many causes, and the source may exist anywhere between the application and its destination.
Network congestion
When a network becomes congested, routers and other network devices may have to queue packets before forwarding them.
As traffic levels fluctuate, the amount of time packets spend waiting also changes. Some packets therefore pass through quickly while others are delayed.
This variation creates jitter.
Wireless network conditions
Mobile communications introduce additional variables.
Signal strength can change as a vehicle moves. Different users compete for capacity. Handover between cells can affect performance, and environmental conditions can influence the quality of wireless links.
A 4G or 5G connection may consequently provide excellent performance at one moment and deteriorate shortly afterwards.
Changing network routes
Packets do not necessarily follow exactly the same route across a network.
Routing decisions, congestion and network failures can cause traffic to take different paths. If those paths have different characteristics, packet arrival times can vary.
Network equipment
Routers, switches, firewalls and other network devices can contribute to jitter if they become overloaded or introduce inconsistent processing delays.
Buffering and queue management
Buffers temporarily hold packets when a network cannot immediately forward them.
Some buffering is necessary, but excessive buffering can dramatically increase latency and produce poor real-time performance. This is closely associated with the network phenomenon known as bufferbloat.
Competing applications
Not every application has the same network requirements.
A large file transfer, cloud backup or software update may consume substantial bandwidth at exactly the same time that another user requires a stable connection for a voice call or live video stream.
Without intelligent traffic management, those applications can compete for the same network resources.
How does jitter affect voice communications?
Voice is particularly sensitive to inconsistent packet delivery.
During a VoIP conversation, audio is continuously converted into packets and transmitted across the network. Those packets need to arrive quickly and in approximately the correct sequence for the conversation to sound natural.
Excessive jitter can result in:
- distorted or robotic speech
- short gaps in conversations
- delayed audio
- words or syllables being lost
- speakers talking over one another
- calls becoming difficult to understand
Many voice systems use a jitter buffer to compensate.
The buffer temporarily stores incoming packets before playing them in the correct sequence. This can smooth out relatively small variations in arrival times.
However, there is a trade-off. A larger buffer can accommodate greater jitter, but it also introduces additional delay.
Jitter buffers can therefore mitigate the symptoms of jitter, but they cannot compensate indefinitely for a highly unstable network.
How does jitter affect video?
Live video places even greater demands on connectivity because large amounts of data must be delivered continuously.
Jitter can contribute to:
- buffering
- frozen images
- reduced picture quality
- audio and video becoming unsynchronised
- dropped frames
- interruptions to live streams
For a consumer watching entertainment, this may simply be frustrating.
The consequences are potentially much greater when video is being used for remote inspection, situational awareness, emergency response, security, broadcasting or operational decision-making.
A video stream that works perfectly most of the time may not be sufficient if it becomes unstable precisely when the network is under pressure.
Why can mobile networks experience jitter?
Mobile environments are inherently dynamic.
A fixed office connection may operate over the same physical infrastructure continuously. A moving ambulance, train, drone, vessel or connected vehicle may pass between different cells, coverage areas and network conditions during a single session.
Available bandwidth, latency and signal quality can change constantly.
This is why relying exclusively on a single cellular connection can introduce a point of vulnerability.
5G provides significant improvements in capacity and latency, but it does not remove every cause of network degradation. Coverage varies, cells become congested and mobile devices still need to move between network environments.
Resilience therefore requires more than simply selecting the fastest available bearer.
Satellite connectivity and jitter
Satellite has become an increasingly important part of hybrid connectivity.
Modern satellite services can provide high-speed connectivity in locations where terrestrial communications are unavailable or unreliable. Different satellite technologies and orbital architectures, however, have different latency and performance characteristics.
When satellite is combined with cellular, Wi-Fi or other terrestrial connections, the available network paths may therefore behave very differently.
Simply distributing packets across multiple connections without understanding those differences can itself create performance problems.
An intelligent networking platform must be able to respond to the characteristics of each available connection rather than treating every bearer as interchangeable.
What is acceptable network jitter?
There is no single jitter figure that is appropriate for every network or application.
Tolerance depends on factors including the application, codec, buffering strategy and underlying network.
A file transfer may be largely insensitive to jitter, whereas interactive voice, real-time video or remote-control traffic can be much more sensitive.
This is why network performance should be considered in the context of the application being delivered, rather than judged solely by a generic speed test.
For organisations running critical communications, the more useful question is often not simply “How fast is the connection?” but:
Can the network consistently deliver the performance required by the application?
How can SD-WAN help manage network jitter?
Software-defined wide area networking provides a more intelligent approach to managing connectivity.
Rather than treating the WAN as a single fixed connection, an SD-WAN platform can monitor available network paths and make decisions about how traffic should be transported.
Depending on the implementation, these decisions can take account of factors such as:
- latency
- packet loss
- jitter
- bandwidth
- link availability
- application requirements
This enables traffic to be moved away from a deteriorating connection or prioritised according to operational importance.
However, SD-WAN, failover and bonding are not automatically the same thing.
Traditional failover typically waits for one connection to fail before switching to another. Load balancing distributes traffic between connections. More advanced bonding technologies can use multiple available links together and dynamically manage traffic across them.
For highly mobile and critical applications, those distinctions matter.
Can network bonding reduce the impact of jitter?
Network bonding can help improve resilience by removing dependence on a single network path.
Instead of relying solely on one cellular, satellite or fixed connection, multiple bearers can be combined and intelligently managed.
If the performance of one connection deteriorates, traffic can potentially be moved to or shared across alternative connections.
The important word is intelligently.
Different links may have very different latency, jitter, bandwidth and reliability characteristics. Effective bonding therefore requires more than simply combining their theoretical bandwidth.
The network needs to understand how individual paths are performing and make appropriate decisions about the traffic using them.
Why application-aware networking matters
Not all network traffic is equally important.
Imagine an emergency response vehicle transmitting:
- live video from the scene
- voice communications
- patient or incident data
- GPS and telemetry
- cloud synchronisation
- routine software updates
Treating all of that traffic identically makes little sense.
An application-aware network can identify different types of traffic and apply policies according to their requirements and priority.
A live operational video stream may need low jitter and consistent throughput. Voice communications may require low latency and predictable packet delivery. A background file synchronisation task can usually wait.
This makes it possible to use available connectivity more effectively rather than allowing lower-priority traffic to compromise critical applications.
Why jitter matters for mission-critical communications
For conventional business connectivity, intermittent network degradation can reduce productivity.
For mission-critical operations, the consequences can be significantly more serious.
Emergency services may depend on voice, video and operational data while vehicles are moving between coverage areas.
Transport systems increasingly depend on continuous communications for telemetry, monitoring and passenger services.
Maritime operators may combine cellular and satellite networks as vessels move between coastal and offshore environments.
Autonomous and remotely operated systems require reliable communications for command, control and situational awareness.
In each case, the challenge is not merely achieving connectivity.
It is maintaining usable, consistent connectivity as network conditions change.
That requires visibility of the underlying network performance and the ability to react when an individual bearer begins to deteriorate.
Building a more resilient network
There is no single technology that can eliminate every source of network jitter.
Instead, resilient connectivity comes from combining several approaches.
These can include using diverse network bearers, continuously monitoring link quality, prioritising important applications, intelligently selecting network paths, managing congestion and maintaining alternative routes when individual connections deteriorate.
This is particularly important in mobile environments where the “best” network can change from one moment to the next.
Livewire’s RazorLink® technology is designed around this principle. Rather than depending on a single connection, RazorLink can bring together available connectivity such as cellular, satellite, Wi-Fi and fixed networks into an intelligently managed service.
The objective is not simply to aggregate bandwidth. It is to help applications maintain the connectivity they need as the performance and availability of the underlying networks change.
From faster connectivity to more consistent connectivity
Bandwidth will always matter, but it is only one measure of network performance.
For real-time and mission-critical applications, organisations also need to consider latency, packet loss, jitter and the resilience of the overall connection.
A network that is exceptionally fast for 95% of the time but unpredictable for the remaining 5% may be less useful operationally than one that provides consistent performance throughout a mission.
As organisations increasingly depend on live video, cloud applications, connected vehicles, remote operations and real-time data, that distinction becomes increasingly important.
The future of resilient communications is therefore not simply about finding the fastest network.
It is about intelligently combining, monitoring and managing multiple networks so that critical applications remain connected when conditions change.
Frequently Asked Questions
What is network jitter in simple terms?
Network jitter is the variation in the amount of time it takes data packets to travel across a network. High jitter means packets are arriving at inconsistent intervals, which can affect real-time applications such as voice and video.
What is the difference between latency and jitter?
Latency is the time taken for data to travel across a network. Jitter is the variation in that time between individual packets. A network can therefore have relatively low average latency but still suffer from high jitter.
What causes network jitter?
Common causes include network congestion, wireless signal variation, changing network routes, overloaded network equipment, excessive buffering and competition between applications for available bandwidth.
Does jitter affect video calls?
Yes. Excessive jitter can contribute to frozen video, dropped frames, poor audio, buffering and audio/video synchronisation problems.
Can 4G and 5G networks experience jitter?
Yes. Cellular performance can change because of signal strength, congestion, movement between cells and other network conditions. 5G can provide very low latency and high bandwidth, but it does not eliminate jitter.
Can network bonding reduce jitter?
Intelligent network bonding can reduce the impact of jitter by providing multiple network paths and allowing traffic to be managed according to the performance of the available connections. It cannot eliminate every possible source of jitter, but it can help prevent applications from depending entirely on a deteriorating link.
How does application-aware bonding help?
Application-aware bonding considers the requirements of different types of traffic. Critical voice, video or operational data can be prioritised differently from less time-sensitive traffic, helping maintain application performance when network conditions change.
Why is network jitter important for mission-critical communications?
Mission-critical applications often depend on continuous real-time information. Inconsistent packet delivery can affect voice, video, telemetry and operational data even when a network still appears to be connected. Monitoring and managing jitter is therefore an important part of maintaining resilient communications.
