Solution for congested networks

Solution for congested networks

Solution for congested network especially for Safety Critical Networking.

Prioritising/Optimising Packet Flow in a congested Network

Safety Critical environments such as in airports have communication systems that are crucial to its operation. This may include air traffic control data from radar (aircraft location, height and speed). An error in the Radar information has a potential for having catastrophic consequences. The network algorithm that we have developed and patented, focuses on designing an airport critical network for supporting critical (error-free) data but also, time critical with low delay tolerance.

Ultra-low latency (1ms) in high network utilization (above 50%), high throughput (3x quicker compare to TCP/IP)

Suitable for private (1ms latency)/public network (Reduce by 1/3 on standard time)

Work on existing TCP/IP stack

Suitable for high volume (high data size), high frequency (low interval)

Has innate traffic shaping, filtering feature to improve the discovery of unauthorised traffic, and load balancing to protect from denial of service

Has 10x improve in latency, response and lower packet loss in Wireless communication especially in public 4G/5G network

The Network Traffic Oscillator (NTO) is a concept that combines network switching and routing seamlessly together. These two concepts have fundamentally different operations. The former refers to packet transmission control using network addresses such as Internet Protocol (IP) to send data from source to destination. On the other hand, the latter refers to external factors concerning the steering of data transmission via the best route on a network. Thus, network switching focuses on the delivery of traffic whilst network routing focuses on managing network capacity.

Custom applications that use multi-purposed packet protocols such as User Datagram Protocol (UDP), IP, and Ethernet, have no network troubleshooting tools or responses to network faults. The only defence against missing data is left to the transport or the application layer, which are ineffective when the missing data are lost as a result of networking issues.

We are currently in the process of safety critical networking progressing from small network to large ones with hundreds of routes and nodes. In this scenario, greater delays are introduced from switching, routing and applications rather than from the physical constraints of the communication medium. In the air traffic context, the introduction of SWIM has leads to many modern communication problems, particular in network switching, routing and application load traffic management.

While tools such as EIGRP have addressed these issues with traffic management by configuring an active routing plan (via the K-values and resultant vector metric) to stave off transmission disasters, these tools are poorly utilised given the modern plethora of application traffic types. Our algorithm shows how to increase data throughput by considering additional parameters such as the application bandwidth usage, and the synchronisation between receiver and transmitter response time.

In comparison with EIGRP, this system offers a reduction of tens of percent in the delay time coupled with a narrow spread in this response. Moreover, as a result only 1% of packets experience problematic transmission delays compared to 10% using EIGRP. This is a great improvement QoS via the improved response time and removal of unnecessary bandwidth wastage from delays and retransmission.

Network Traffic Oscillator

In summary, the NTO approach offers not only reliable traffic delivery but also high QoS with minimal delayed or lost packets. It removes the need to add ever greater overheads to network packets and facilitates real-time deterministic Ethernet performance

The Network Traffic Oscillator (NTO) was designed to provide to find a means to considerably improve the response time of a safety critical application. The network is further improved by effective traffic management in network routing and switching.

Congested Network Graph Congested Network Graph

Packet response times for: (a) the NTO-based system; (b) packet- based EIGRP

Congested Network Graph

Simulation results for application packets affecting the flow management of the routers 1-3. Packet based EIGRP with ad-hoc traffic congestion experiences additional delay of the transmissions and uses all three links. Critical networking uses only routers 2 and 3 with delay affecting only a negligible proportion of the packet transmissions

Do you have any questions?

For further information please contact FTI.
You can ask us a question using the form or please call us on
+44 (0) 24 7641 7070.

Call by phone