Sub-rate Codes in 10GBASE-T Frame Structure
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing 10GBASE-T Ethernet standard lacks flexibility in scaling data rates, often reducing transmission rates unnecessarily to lower levels than what the communication link can support, especially in cases where the link quality is impaired or in low-power applications.
Innovation Solution
A method is introduced that allows for autonegotiation and training processes to determine suitable sub-rate modes of operation, selecting from various symbol constellations and reducing the number of lanes to adapt data rates without restarting the autonegotiation process, thereby enabling communication at higher rates than conventional standards allow.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional 10GBASE-T standard is used, then backward compatibility with legacy Ethernet standards is maintained, but data rate flexibility is lost and transmission rate must be reduced to legacy levels
Solution Approach 1:
The patent implements dynamic data rate adaptation by allowing the transmission rate to vary continuously between 10 Gbps and sub-rate levels (e.g., 2.5 Gbps, 1 Gbps) based on link quality conditions. The system transitions from a static rate selection to a dynamic rate adjustment mechanism that responds to signal-to-noise margin measurements, enabling optimal performance across varying channel conditions without being locked to fixed legacy rates
Solution Approach 2:
The patent changes the operational parameters by introducing sub-rate modes with different modulation schemes (e.g., changing from 128-DSQ to 64-SQ or 32-DSQ constellations) and adjusting the number of active lanes. This allows the system to modify transmission characteristics continuously rather than jumping between discrete legacy rates, providing fine-grained control over data rate adaptation while maintaining compatibility with existing infrastructure
2Adaptability or versatility
If data rate is reduced to legacy levels, then compatibility with older Ethernet standards is achieved, but unnecessary rate scaling occurs and high link performance is wasted
Solution Approach 1:
The patent applies partial action by implementing rate reduction only when and to the extent necessary. Instead of always falling back to full legacy rate reduction, the system reduces rate partially based on measured link conditions (signal-to-noise margin). When link quality is good, the system maintains high rates (10 Gbps or sub-rates above 1 Gbps); when link quality degrades, it reduces rate proportionally to the impairment, avoiding unnecessary scaling to legacy levels unless absolutely required
3Reliability
If autonegotiation process is restarted for sub-rate modes, then proper rate selection is achieved, but time loss occurs and training efficiency is reduced
Solution Approach 1:
The patent performs preliminary actions by pre-configuring multiple sub-rate modes and training sequences before actual data transmission begins. The system prepares training sequences for different sub-rates (2.5 Gbps, 1 Gbps, etc.) in advance and stores them for quick selection. When rate adaptation is needed, the system can immediately switch between pre-prepared training sequences without restarting the entire autonegotiation process, significantly reducing training time while maintaining accurate rate selection
Data Source
AI summary
A method of operation in a high-speed Ethernet transceiver is disclosed. The method includes engaging in an autonegotiation process with a link partner transceiver to indicate whether one or more sub-rate modes of operation are supported. Each sub-rate mode of operation corresponds to a sub-data rate that is less than a maximum data rate. The autonegotiation process is terminated. The transceiver then participates in a training process. The training process includes receiving a first training sequence corresponding to a first sub-data rate that is less than the maximum data rate. A signal quality parameter for the received first training sequence is measured. The training sequence is terminated based on the measured signal quality parameter failing a predetermined criteria. A second training sequence is then initiated that corresponds to a second sub-data rate that is less than the first sub-data rate without starting a second autonegotiation process.