End-to-end Synchronization, Adaptive Link Resource Reservation and Data Tunnelling System


Random and excessive communication latency has become a major constraint for performance improvement of complex ICT (information and communication technology) systems, where a large number of computing and communication facilities are interconnected to enable end-to-end communication. Such ICT systems and related applications include: data center facilities using multiple servers, Internet of Things (IoT), networked controllers/actuators (such as those used in some manufacturing facilities), interactive applications over the Internet and machine to machine (M2M) communications.

Long communication latency impacts an application's performance mainly by forcing an ICT application to stall while waiting for the arrival of the expected data before it can continue to the next step in its processing. On the other hand, a user's experience, interest and productivity could also be adversely affected by long and random latency in interactive applications.

The following ICT applications are more susceptible to performance degradation due to network latency:

• Applications that require real-time data or very low latency, such as data centre/cloud centre and remote control
• Interactive applications that depend on frequent exchange of data, as opposed to the transfer of large quantities of data, such as online gaming and remote desktop access.

There are several major factors contributing to long and varying end-to-end latency in a complex ICT system. They include: a) the signal propagation time over the actual communication medium for each link segment, b) the processing required for the transmission and reception of the data for each link segment between each pair of neighboring nodes along the whole end-to-end communication path, c) the delays from the use of layered communication protocols enabling point-to-point data exchange for each link segment, due to the availability and conditions of the communication resources as well as any potential repeated transmission due to the requirements of Quality-of-Service (QoS), and d) the delays in switches and routers to support end-to-end communications involving multiple communication link segments over ICT systems.

Description of the Invention:

The fundamental purpose of this invention is to reduce the end-to-end latency of communication in a complex ICT system involving multiple communication links, switches and routers. Through the proposed adaptive reservation of the physical resources of intermediate links and nodes, the processing delays from the routers, protocol stack and buffers are minimized. The proposed method for end-to-end latency reduction involves adaptive reservation of the physical resources from the intermediate routers and switches, such that the communication signal originating at the source node can be propagated to the destination node with minimal latency-caused redundant processing from the current protocol stack and buffer. As a result, a direct signal propagation or data transmission tunnel between the source and destination can be established based on the coordination of all network nodes involved in the communication process. In addition, with the accumulation of the impact and noise from multiple communication links, the transmission parameters at the source node will be adapted based on the end-to-end channel quality and conditions.

Potential Applications:

• High performance computing and data centres:
As the computing speed of single microprocessors becomes increasingly difficult to increase further, using a large number of interconnected processors and servers has becoming the mainstream of today's supercomputing technology. However, the communication latency experienced by the interconnected processors plays an important role in the overall performance of the computing facilities, due to the highly interactive and distributed nature of the overall computing process. As a result, this invention can be applied to multi-tiered data centre networks in order to reduce overall end-to-end latency and improve computational performance.

• 5G Radio Access Networks and Heterogeneous Networks:
Communication latency reduction in 5G cellular communication networks will enable new business and applications that are otherwise "delay sensitive". Emerging related applications, including transportation safety, smart grid or networked control systems have very stringent requirements on latency.

• Networked Control Systems:
Networked Control Systems (NCSs) are real-time control systems that feature spatially distributed sensors and are controlled through a digital communication network. NCSs have been widely used in factory automation, environment monitoring and smart homes etc. This type of system exhibits several characteristics that make it unique from a control perspective. Among the many problems in networked control systems is the need to reduce communication latency to improve the control effectiveness.

• Interactive Internet Applications:
Latency within Internet data flow directly affects a user's experience, interest and productivity when using Internet enabled applications, such as  Internet gaming, cloud computing and remote access for business applications.

Patent Status

• US provisional patent application has been filed.

Patent Information:

newsletter sign up

Stay informed of our latest innovations. Subscribe to our newsletter today!

a leader in turning research and innovations into profitable consumer-ready products...