Wireless IP Packet Clustering to Reduce Lower-Layer Processing

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing wireless communication systems face inefficiencies in handling IP packets due to variable sizes and grant-dependent concatenation, leading to heavy segmentation and increased processing load, particularly in 5G and beyond networks.

Innovation Solution

Implementing a packet blender layer that clusters IP packets into a predetermined maximum size before transmission to the lower layers, minimizing the number of packets handled and adapting to channel variations and grant sizes.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If IP packets are handled individually through traditional layering (PDCP-RLC-MAC), then each packet can be processed with proper header handling and sequence numbering, but the processing load increases significantly and throughput decreases due to heavy segmentation and grant-dependent concatenation

Engineering Contradiction:
Improvedata plane processing throughputVSAvoidprocessing complexity at each layer
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent merges multiple IP packets into a single concatenated payload at the PDCP layer before passing to RLC. This combining approach reduces the number of individual packet processing operations across PDCP, RLC, and MAC layers, directly decreasing processing complexity while maintaining throughput efficiency

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent performs preliminary concatenation of IP packets at the PDCP layer before they reach the RLC and MAC layers. This advance preparation of packet bundles reduces the processing burden on subsequent layers, as they handle fewer, larger units rather than multiple small packets individually

Inventive Principle:
Principle #10Preliminary action

2Productivity

If concatenation is performed at the RLC layer dependent on grant size, then packets can be efficiently packed according to available resources, but the variable number of SDUs per PDU creates variable header sizes and increases processing overhead

Engineering Contradiction:
Improvepacket packing efficiencyVSAvoidvariable header processing
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent performs preliminary concatenation of IP packets at the PDCP layer before they reach the RLC and MAC layers. This advance preparation of packet bundles reduces the processing burden on subsequent layers, as they handle fewer, larger units rather than multiple small packets individually

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent changes the fundamental parameter of packet handling from individual IP packets to concatenated packet bundles. By transforming the data unit structure at PDCP layer, it enables fixed-size header processing while maintaining efficient resource utilization at lower layers

Inventive Principle:
Principle #35Parameter changes

3Ease of manufacture

If RLC SN is allocated to every RLC SDU in 5G NR system, then pre-processing becomes possible with fixed size RLC header, but the number of sequence numbers increases from 1024 to over 262144, increasing memory and processing requirements

Engineering Contradiction:
Improvepre-processing capabilityVSAvoidnumber of sequence numbers
Core Design Contradiction:
Ease of manufactureVSQuantity of substance

Solution Approach 1:

The patent performs preliminary concatenation of IP packets at the PDCP layer before they reach the RLC and MAC layers. This advance preparation of packet bundles reduces the processing burden on subsequent layers, as they handle fewer, larger units rather than multiple small packets individually

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent merges multiple IP packets into a single concatenated payload at the PDCP layer before passing to RLC. This combining approach reduces the number of individual packet processing operations across PDCP, RLC, and MAC layers, directly decreasing processing complexity while maintaining throughput efficiency

Inventive Principle:
Principle #5Merging (Combining)

Data Source

PatentUS12389269B2Method and apparatus for clustering IP packets in wireless network
Publication Date: 2025.08.12 SAMSUNG ELECTRONICS CO LTD
  • US12389269B2 patent drawing
  • US12389269B2 patent drawing
  • US12389269B2 patent drawing

AI summary

A method and an apparatus for clustering Internet protocol (IP) packets in a wireless network are provided. The method includes receiving, by a network device in the wireless network, a plurality of IP packets from an upper layer of a user equipment (UE) in the wireless network or vice versa, initiating a timer on receiving the plurality of IP packets from the upper layer, forming a cluster of IP packets having a predetermined maximum size while the timer is running, stopping the timer once the formation of the cluster is completed, and transmitting the cluster of IP packets to a lower layer of the UE, where the lower layer treats the cluster of IP packets as a single payload to minimize the IP packets handled at the lower layer of the UE.