Introduction
The era of cellular and digital telecommunications began in the 1990s with second-generation (2G) mobile networks, based on time-division multiple access (TDMA).
In the 2000s, third-generation (3G) networks were developed on the principle of wideband code-division multiple access (WCDMA). Although the third generation has dominated the market thanks to the increase in data throughput, it has never completely replaced the second generation.
The early 2010s saw the start of fourth-generation (4G) networks using orthogonal frequency-division multiple access (OFDMA) for the downlink and single-carrier frequency-division multiple access (SC-FDMA) for the uplink.
The development of 4G networks followed three steps identified by the releases of 3GPP (3rd Generation Partnership Project) standard:
- – releases 8 and 9 are the basis of LTE (Long-Term Evolution) standard;
- – releases 10, 11 and 12 are the basis of LTE Advanced standard;
- – releases 13 and 14 are the basis of LTE Advanced Pro standard.
The 3GPP standardization body specified service models corresponding to specific use cases and requirements:
- – MBB (Mobile Broadband) service corresponds to applications and services that require faster connection, which make it possible, for example, to watch videos in ultra-high definition or use virtual or augmented reality applications;
- – LLC (Low Latency Communication) service groups together all the applications requiring extremely high reactivity as well as reliability of the data transmission service, for example civil security for critical missions;
- – MTC (Machine Type Communication) service mainly groups applications linked to the Internet of Things (IoT). These services do not require very high bit rates, but require more extensive coverage and lower energy consumption.
I.1. LTE standard
Release 8 defines the evolved packet system (EPS) consisting of a new evolved packet core (EPC) coupled to a new evolved universal terrestrial radio access network (E-UTRAN).
Release 8 defines a new radio interface based on orthogonal frequency-division multiplexing (OFDM) and four-channel spatial multiplexing (MIMO (Multiple Input Multiple Output) 4×4). The MIMO function relies on the availability of the cell-specific reference signal (CRS).
Category-4 mobiles are able to achieve up to 150 Mbps for the downlink and up to 50 Mbps for the uplink, with the following characteristics for the radio interface:
- – a radio channel bandwidth of 20 MHz;
- – 64-QAM (Quadrature Amplitude Modulation) for the downlink and 16-QAM for the uplink;
- – two-channel spatial multiplexing (MIMO 2×2) for the downlink.
LTE standard only offers services based on packet-switching (PS), and as such, only allows the transport of IP (Internet Protocol) packets. In release 9, the telephone service VoLTE (Voice over LTE) is therefore provided by the network IMS (IP Multimedia Sub-system). If the VoLTE is not deployed, the mechanism CSFB (Circuit-Switched Fallback) is used to redirect the mobile to 2G/3G networks in the CS mode in the case of an incoming or outgoing telephone call.
I.2. LTE Advanced standard
Release 10 provides throughput enhancement through carrier aggregation (CA), which increases the overall bandwidth of the radio channel.
The throughput improvement is also achieved by increasing the number of channels spatially multiplexed (MIMO 8×8). Additional resources are specifically allocated to each mobile for the channel state information reference signal (CSI-RS).
The modulation scheme has been increased from 64-QAM to 256-QAM, allowing an increase in the downlink bit rate.
Release 11 introduces new features to improve data throughput and edge coverage, with enhanced inter-cell interference coordination (eICIC) and coordinated multipoint (CoMP) transmission.
Release 12 defines a new MTC architecture that takes into account connected objects. A new category of mobile (category 0) is introduced, allowing a reduced energy consumption in return for a lower data rate.
LTE Advanced standard also defines the evolved Multimedia Broadcast/Multicast Service (eMBMS) in order to broadcast content shared between multiple mobiles. In addition, in the areas of public safety and critical communications, the eMBMS network improves the efficiency of the MCPTT (Mission Critical Push-To-Talk) service that enables the transmission of voice to all participants of a group.
In addition, release 12 introduces proximity services, from device to device (D2D), to obtain a reduced latency for the time of both communication establishment and the voice transport.
I.3. LTE Advanced Pro standard
The goal of LTE Advanced Pro standard is to increase the throughput for mobiles to reach the value of Gigabit/s, to bring new functionalities to EPS, MTC and eMBMS networks, and to introduce new proximity services, namely vehicle to everything (V2X).
I.3.1. MBB service
I.3.1.1. Network architecture
The control and user plane separation (CUPS) aims to define a more flexible distributed architecture, taking advantage of the evolution towards software-defined networking (SDN) implementations.
The CUPS architecture is based on the separation between the user plane and the control plane for the serving gateway (SGW) and the PDN (Packet Data Network) gateway (PGW). This architecture enables mobile edge computing (MEC) deployments that leverage a distributed user plane, collocated with the evolved node base station (eNB), and a centralized control plane.
Dual connectivity (DC) introduced in release 12 improves the downlink throughput. IP packets are simultaneously transmitted from two eNB entities, the master eNB (MeNB) and the secondary eNB (SeNB).
Release 13 introduces the traffic transfer to two radio stations for the uplink, according to two parameters: a primary link and a threshold value. When the mobile buffer is below the threshold, the mobile only sends data on the primary link. When the amount of buffered data exceeds the threshold, the mobile can send data to both the MeNB and the SeNB.
I.3.1.2. Spatial multiplexing
A significant improvement in release 13 is the introduction of active antenna system (AAS), with antenna elements ranging from 8 to 64, which is relevant for frequencies above 3.5 GHz.
The FD-MIMO (Full-Dimension MIMO) mechanism enables beamforming in the horizontal and vertical directions and the generation of three-dimensional spatial links.
Associated with the FD-MIMO mechanism, two methods for using the CSI-RS are defined:
- – for the method Class A, the CSI-RS is associated with an antenna element, their number being limited to 16;
- – for the method Class B, the eNB entity can configure up to eight beams per mobile, each beam being formed from a CSI-RS.
Release 14 improves the FD-MIMO mechanism, for the method Class A, by increasing the number of CSI-RS up to 32 and decreasing the density of the CSI-RS. For the method Class B, the improvement concerns the efficiency of the CSI-RS.
I.3.1.3. Channel aggregation
Channel aggregation has increased to 32 (the number of aggregated components). In order to meet growing data traffic, LTE Advanced Pro standard has also introduced new aggregation techniques: LAA (License Assisted Access), LWA (LTE–Wi-Fi Aggregation) and LWIP (LTE/WLAN radio level integration with IPsec tunnel).
LAA is an extension of LTE aggregation. Transmission is carried out on licensed (LTE) and unlicensed frequency bands (Wi-Fi at 5 GHz U-NII band), between the mobile and the eNB entity, without an intermediate access point. The eNB entity is the anchor point for channel aggregation.
LAA is similar to dual connectivity, for which LTE transmission takes place on the MeNB station and Wi-Fi transmission on the SeNB station.
In release 13, transmission on the unlicensed frequency band occurs only for the downlink. The transmission for the uplink exists in release 14.
LWA and LWIP use LTE and Wi-Fi frequency bands. The transmission on the Wi-Fi radio channel occurs between the mobile and the access point (AP) in accordance with the 802.11 standard. The eNB entity is the anchor point for channel aggregation.
Release 14 brings the following enhancements to the LWA features:
- – transmission of data for the uplink on the Wi-Fi network;
- – support for new 60 GHz frequency bands and 802.11a, 802.11ad and 802.11ay interfaces;
- – collection of information for available capacity on the Wi-Fi network;
- – discovery of neighboring Wi-Fi networks under the coverage of eNB entities.
LWIP uses an IPSec tunnel to transport IP packets between the eNB entity and the Wi-Fi access point. Unlike LWA, LWIP aggregation does not require any modification for Wi-Fi transmission.
I.3.2. LLC service
For D2D communication, the main improvement lies in the support of relaying by mobile. This allows, for public safety, mobiles out of coverage to communicate with the network via mobiles under the radio coverage.
The SC-PTM (Single-Cell Point-To-Multipoint) feature was introduced in release 13 to improve the efficiency of the radio interface of the eMBMS network, by supporting, on one cell, the broadcast service using specific radio resources.
Prior to release 13, the 3GPP organization standardized functionality for use as an enabler for mission-critical services. For example, an MCPTT group voice call must have a bearer already established for immediate use due to the time required to establish the bearer on the eMBMS network.
Release 13 defines different application services for the MCPTT function: user authentication, group affiliation, group calls and private calls, and floor control.
Release 14 completes the MCPTT function with various management services: configuration management, group management, identity management and key management.
Release 14 introduces vehicle-to-everything (V2X) communication, which comes in four applications depending on the different types of device to which the vehicle connects:
- – vehicle-to-vehicle (V2V) communication;
- – vehicle-to-infrastructure (V2I) communication;
- – vehicle-to-pedestrian (V2P) communication;
- – vehicle-to-network (V2N) communication.
I.3.3. MTC service
Release 13 changes the architecture of the network to optimize the data transfer using different planes:
- – the control plane, to reduce the number of messages during the processing of a session establishment procedure;
- – the user plane, for which the management of the connection avoids deleting the context when the terminal has no longer data to transmit.
The architecture enhancements for service capability exposure (AESE) is used to expose network services and capabilities to third parties, and to provide access to network capabilities:
- – high latency communication, to support the scenario in which applications communicate with temporarily inaccessible terminals;
- – point-to-multipoint communication;
- – increase in the discontinuous reception (DRX) cycle;
- – event monitoring affecting the terminal operation.
Terminals Cat. 1 and Cat. 0 were introduced in releases 8 and 12 respectively for MTC service. These terminals have reduced functionality but can operate in a bandwidth of 20 MHz.
To reduce the complexity of the terminal and improve the battery life and radio coverage, release 13 introduces two new technologies for the radio interface:
- – LTE-M operating in a bandwidth of 1.4 MHz, with terminals Cat. M1;
- – NB-IoT (NarrowBand Internet of Things) operating in a bandwidth of 180 kHz, with terminals Cat. NB1.
In order to increase the throughput on the radio interface, release 14 introduces two new categories of terminals: Cat. M2 for LTE-M and Cat. NB2 for NB-IoT.
I.4. Wi-Fi integration
Release 8 defines the integration of the Wi-Fi radio access network at the EPC, addressing all aspects of interworking: mobility between Wi-Fi and LTE access and security (authentication, protection of data). However, release 8 does not allow simultaneous connections to multiple access networks. In addition, release 8 specifies the access network discovery and selection function (ANDSF).
Several access architectures connected to the EPC are defined:
- – the architecture based on the S2a interface, for which the Wi-Fi radio access network is trusted and the mobility is managed by the network;
- – the architecture based on the S2b interface, for which the Wi-Fi radio access network is untrusted and the mobility is managed by the network;
- – the architecture based on the S2c interface, for which the mobility is managed by the mobile and the Wi-Fi radio access network can be trusted or untrusted.
Release 9 improves the ANDSF feature that provides access network discovery and selection information for roaming scenarios.
Release 10 introduces simultaneous connections to several radio access technologies.
The NSWO (Non-Seamless WLAN Offload) feature allows traffic to be directly routed to the Internet network without crossing the EPC.
The MAPCON (Multi-Access PDN Connectivity) function supports various connections to the PDN transiting either via the LTE interface (e.g. telephone service) or via the Wi-Fi interface (e.g. Internet service), depending on the operator policy.
Release 12 improves the S2a solution using three modes of operation of offloading traffic:
- – single-connection mode (SCM) supports the mobility between LTE and Wi-Fi access and NSWO via Wi-Fi access;
- – multi-connection mode (MCM) supports one or more PDN connections and NSWO via Wi-Fi access at the same time;
- – transparent single-connection (TSC) mode provides a single connection (LTE or Wi-Fi) and does not support the mobility between LTE and Wi-Fi access.
The discovery and selection functions are also defined in the 802.11u specification of the Institute of Electrical and Electronics Engineers (IEEE), integrated and supplemented by the Wi-Fi Alliance in the Hotspot 2.0 specification. Release 12 helped to align the ANDSF with Hotspot 2.0 features.
Release 13 completes the transfer modes for IP packets by introducing the IFOM (IP Flow Mobility) function for routing the different IP streams of a PDN connection, which corresponds to an access point name (APN), through both LTE and Wi-Fi interfaces.
I.5. 5G integration
The fifth generation (5G) of mobile networks, the first phase of which is defined in release 15, must be more flexible and scalable to allow a wider range of services. It will move to an architecture based on network function virtualization (NFV) where network elements are hosted in virtual environments, and network slicing allows adaptation to different requirements.
As in previous generations, the fifth generation defines a core network (5GC) and a radio access network (5G NR (new radio)). Unlike previous generations that required the deployment of the core and radio access network of the same generation, the fifth generation makes it possible to integrate elements of different generations in different configurations.
The 5G NR access network is defined to support two operational modes, namely standalone (SA) and non-standalone (NSA):
- – in the NSA mode, the 5G NR connects to the 4G core network only for the user plane, the control plane being processed by the 4G radio access network;
- – in the SA mode, the 5G NR connects to the 5G core network for the user plane and control plane data.
Although both of the operation modes should have coexisted from the beginning, a consensus has emerged to prioritize the NSA mode in order to quickly respond to the need for throughput. This mode exploits existing 4G deployments by combining LTE and NR radio resources with 4G network cores.
Release 15 does not provide a fundamental technological breakthrough on the 5G NR interface with respect to the 4G LTE radio interface, the multiple access mode being identical, but rather some adjustments with respect to time-division multiplexing structures, frequency-division multiplexing and error correction codes.