Local Call Local Switch

Description

Local Call Local Switch (LCLS) is a network optimization feature defined in 3GPP standards that allows for the direct routing of user plane traffic between two user equipment (UE) devices when they are connected to the same radio access node, such as an eNodeB in LTE or a gNB in 5G NR. The primary architectural principle involves the access node performing local switching functions, which traditionally were handled by core network entities like the Serving Gateway (SGW) and Packet Data Network Gateway (PGW) in the Evolved Packet Core (EPC) or the User Plane Function (UPF) in the 5G Core (5GC). When two UEs establish a communication session and are both served by the same cell or a common local gateway, the access node identifies this condition and establishes a direct local bearer between the UEs. This local bearer bypasses the core network user plane path entirely for the traffic between these two endpoints.

The operation of LCLS is typically triggered during the establishment or modification of a dedicated bearer. The access node, in coordination with the core network control plane (e.g., MME in EPC or AMF/SMF in 5GC), evaluates the location of the UEs. If they are determined to be under its service area for the relevant Packet Data Network (PDN) connection or PDU Session, the access node configures its internal switching fabric to route packets directly between the corresponding radio bearers of the two UEs. The control plane signaling (e.g., session management, mobility management, charging) still traverses the core network to maintain subscriber management, policy enforcement, and lawful interception capabilities. The access node may also handle necessary quality of service (QoS) enforcement and packet marking on the locally switched path.

Key components enabling LCLS include the enhanced radio access node (e.g., eNodeB with LCLS functionality), the core network control plane entities that authorize and instruct the local switch, and potentially a Local Gateway (L-GW) collocated with the access node in some architectures. Its role in the network is that of a traffic offload and optimization mechanism. It is particularly significant in scenarios with high local communication density, such as stadiums, factories, or campuses, where it improves performance by reducing end-to-end latency and jitter while conserving valuable backhaul and core network resources.

Purpose & Motivation

LCLS was created to address the inefficiency of routing all user plane traffic through centralized core network gateways, especially for communication between devices that are geographically proximate and connected to the same access point. Prior to LCLS, even a call between two phones in the same building would have its traffic routed potentially hundreds of kilometers to a core data center and back, incurring unnecessary latency, consuming backhaul bandwidth, and loading core network nodes. This 'tromboning' or 'hairpinning' of traffic is wasteful and detrimental to performance-sensitive applications.

The historical context for LCLS includes the rise of localized high-density communication needs and the push for network efficiency in 3GPP standards around Release 9. It supports the broader objectives of proximity-based services (ProSe) and network offload. By solving the local traffic routing problem, LCLS enables low-latency local services, reduces operational costs for operators by saving transport resources, and improves the overall scalability of the network for machine-type communication and Internet of Things (IoT) scenarios where devices in a local area communicate frequently with each other.

Furthermore, LCLS laid groundwork for later enhancements in mobile edge computing and local breakout architectures. It addresses the limitation of a purely centralized core network model by distributing switching intelligence to the network edge, aligning with trends towards decentralized and cloud-native network architectures in later 5G releases.

Key Features

• Direct user plane path establishment between UEs at the radio access node
• Bypasses core network gateways (SGW/PGW, UPF) for local traffic
• Reduces end-to-end latency and packet delay variation (jitter)
• Conserves backhaul and core network transport bandwidth
• Maintains core network control for signaling, policy, and charging
• Applicable to both LTE/EPC and 5G NR/5GC architectures

Evolution Across Releases

Defining Specifications