PDCP Flow Segmentation for Wireless Packet Delivery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In wireless communication systems, packet loss in a single Data Radio Bearer (DRB) can negatively impact all flows mapped to it, leading to delayed packet delivery and triggering error recovery mechanisms for independent data streams, affecting throughput and user experience.
Innovation Solution
A method and system that manage data flow at the Packet Data Convergence Protocol (PDCP) layer by identifying and segregating packet sequences based on flow identifiers and sequence numbers, allowing in-order packets to be transferred to upper layers while waiting for missing packets, thus avoiding unnecessary retransmissions and buffer delays.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If packets are buffered at PDCP layer for in-sequence delivery, then packet delivery order is maintained, but packet loss impact propagates to all flows and delays throughput
Solution Approach 1:
The patent segments the single DRB into multiple virtual flows by introducing flow identifiers (FLOW_ID) and per-flow sequence numbers (PDCP_FLOW_SN). This allows the PDCP layer to track and manage sequence numbers independently for each flow, enabling selective buffering and delivery per flow rather than blocking all flows due to packet loss in a single flow.
Solution Approach 2:
The patent adds a new dimension of flow identification by introducing FLOW_ID as an additional parameter alongside the traditional PDCP sequence number. This dimensional expansion allows the system to distinguish between packets belonging to different flows, enabling independent reordering and delivery decisions for each flow while maintaining overall DRB management.
2Reliability
If reordering timer is triggered for out-of-order packets, then in-sequence delivery is ensured, but all packets wait in reordering window causing delays
Solution Approach 1:
The patent segments the reordering process by flow, introducing per-flow sequence numbers (PDCP_FLOW_SN) that are independent for each FLOW_ID. This allows the receiver to maintain separate reordering tracking for each flow, so that out-of-order packets in one flow do not trigger reordering timers that would delay packets in other flows.
Solution Approach 2:
The patent applies reordering and buffering selectively only to packets that are actually out of order within their specific flow, rather than applying a blanket reordering mechanism to all packets in the DRB. This partial action approach minimizes unnecessary waiting time for packets that are already in order.
3Productivity
If multiple flows are mapped to single DRB, then resource efficiency is improved, but packet loss in one flow affects all other flows
Solution Approach 1:
The patent introduces flow identifiers (FLOW_ID) and per-flow sequence numbers (PDCP_FLOW_SN) to segment the DRB into independent virtual flows. This segmentation allows the system to maintain resource efficiency by mapping multiple flows to a single DRB while simultaneously ensuring flow independence through separate sequence number tracking and reordering per flow.
Solution Approach 2:
The patent applies different quality of service handling to different flows within the same DRB by tracking sequence numbers and reordering behavior locally per flow rather than globally for the entire DRB. This local quality approach allows packet loss in one flow to be isolated and handled independently without affecting the delivery of packets in other flows.
4Reliability
If transport layer error recovery mechanisms are triggered, then packet loss is recovered, but throughput decreases due to retransmissions
Solution Approach 1:
The patent performs preliminary reordering and sequence verification at the PDCP layer before packets reach the transport layer. By using per-flow sequence numbers to identify and reorder packets early in the protocol stack, the system prevents out-of-order delivery from triggering unnecessary transport layer error recovery mechanisms and retransmissions, thereby maintaining higher throughput.
Data Source
AI summary
A method for management of data flow at a receiver includes receiving, in a reordering window of a PDCP layer, a plurality of data packets each including header information, and determining, based on the header information, packet sequence numbers of the received data packets, flow IDs each associated with a corresponding data packet of the received data packets, and at least one packet flow sequence number of each of the flow IDs, determining, based on the determined flow IDs and the at least one packet flow sequence number of each of the determined flow IDs, a set of data packets of which associated packet flow sequence numbers are in order with respect to the flow IDs, and transferring the determined set of data packets to an upper layer based on determining that the reorder timer of the PDCP layer is running.