Cellular Networks: An Evolution from 1G to 4G
A cellular network is a wireless communication network which provides services to mobile users covered by a region consisting of multiple cells. Frequency reuse is a fundamental concept in cellular networks.
Prior to the invention of cellular technologies, there were few mobile telephone systems in the late 1940s in the USA and in the early 1950s in Europe, such as car-based telephone systems. The push-to-talk technique was used in these systems, and communication was performed over a single channel. With a single channel and a single antenna at each device, the transmission and reception of signals were performed in a half-duplex manner. Due to low capacity, restricted mobility support, and poor service quality, those systems were not efficient. Later on, mobile telephone system (MTS) and improved MTS (IMTS) were introduced in order to support a larger number of mobile stations. IMTS adopted two channels, one for sending and one for receiving, allowing mobile communication to work in a full-duplex mode. However, the concept of cells did not apply to these pre-1G mobile communication systems.
The breakthrough for increasing capacity appeared in the 1970s when the technique for frequency reuse was developed by dividing a coverage region into multiple cells and reusing the same frequency at different cells which are sufficiently far away from each other. In 1979, the first 1G cellular network was launched in Japan. In 1981, the first international cellular network, Nordic Mobile Telephone (NMT) systems, came to operation in Nordic countries. In 1983, two other 1G systems, Advanced Mobile Phone System (AMPS) and Total Access Communication System (TACS), were introduced in the US and other European countries including the UK and Italy, respectively.
Through a journey which started about 40 years ago, cellular communications have experienced dramatic changes and upgrades from 1G to 4G. Those transitions include the system change from analog to digital, the core network evolution from circuit-switched to packet-switched technologies, and its further transfer toward all Internet Protocol (IP) networks.
First-Generation Cellular Systems
The 1G mobile systems were designed for providing voice services only and were developed based on analog technologies. For medium access of multiple users, frequency-division multiple access (FDMA) schemes were adopted. For carrying voice traffic, frequency modulation (FM) was employed as the modulation scheme.
Nordic Mobile Telephone Systems
NMT is a mobile telephone network covering first the five Nordic countries, Denmark, Finland, Norway, Sweden, and Iceland, and later quite a few other countries. NMT has two variants based on the operational frequency bands, known as NMT-450 and NMT-900, respectively. NMT-450 was developed in order to establish a compatible telephone system in the Nordic countries (Nordic Mobile Telephone Group, 1995). Initially NMT-450 targeted at deploying macro cells at the 450 MHz band to provide larger cell coverage, and later it was modified to operate in the 900 MHz band by considering the size and transmission power constraint of handsets. NMT systems were initially launched in Norway and Sweden as a national service, and later on it was enhanced with roaming services across countries. NMT-900 bears more channels than the NMT-450 network, able to serve a higher number of subscribers.
Advanced Mobile Phone System
The AMPS systems were deployed in North America. It has 2 × 20 MHz bandwidth within the 800–900 MHz frequency band (825–845 MHz for uplink and 870–890 MHz for downlink traffic respectively) allocated by the Federal Communications Commission (FCC). With 30 kHz bandwidth for each channel, it provides 832 duplex channels. This system employs seven-cell clusters for frequency reuse and uses mainly 120∘ sector antennas. Three sectors for a single AMPS cell site were designed to achieve a carrier-to-interference ratio of 18 dB with satisfactory voice quality. In 1991, an improved version of AMPS named as narrowband AMPS (N-AMPS) was developed to further increase AMPS capacity with additional advanced features such as authentication and caller ID.
Total Access Communication Systems
TACS was the first standard that used the 900 MHz band and was deployed in other European countries (Goldsmith, 2005). TACS was operated at a higher frequency than AMPS and adopted narrower bandwidth per channel (25 kHz in TACS versus 30 kHz in AMPS). With narrower bandwidth for each channel, the total number of channels is increased for a band with fixed bandwidth. Operated at a higher frequency, the coverage for each cell will be reduced given the same transmission power and channel condition. In other words, this system was designed for having higher capacity rather than coverage by deploying a larger number of cells and allowing lower transmission power for mobile stations. Indeed, TACS was proved to be efficient and economical for highly dense urban areas. A variant of TACS, J-TACS, was also adopted in Japan.
Second-Generation Cellular Systems
The core network of both 1G and 2G cellular networks was built based on circuit-switched technologies, and the service was mainly targeted at voice traffic. However, unlike 1G systems, 2G employed all-digital transmission technologies for both control signaling and data traffic. The introduction of digital communication provides a series of important features such as the support of advanced source and channel coding, more efficient spectrum utilization, and a high degree of resistance against interference and channel fading. In addition, the handling of control information is more efficient in digital systems.
2G cellular networks were deployed worldwide. They are represented by four major standards, i.e., Global System for Mobile communications (GSM), Interim Standard (IS)-136 or Digital AMPS (D-AMPS), IS-95 or cdmaOne, and Personal Digital Cellular (PDC). Except cdmaOne, the other three systems are based on time-division multiple access (TDMA) mechanisms for medium access. TDMA allows multiple users to share the same channel in the frequency domain by allocating a specific short period of time, known as time slot, to each user for their channel access. Among these four standards, GSM is the most popular 2G technology. Given the fact that 3G and 4G are already widely deployed in the world as of 2017, GSM still occupies approximately 39% of the global mobile market share.
Global System for Mobile Communications
In 1982, a committee called Groupe Spécial Mobile was established in order to design a Pan-European digital cellular standard for mobile communications which would replace the incompatible 1G analog systems. The European Telecommunications Standards Institute (ETSI) initiated the development of the first version of GSM, and the first GSM call was made in 1991. The success of GSM is represented by its over 90% market share in the 2G world, covering globally around 200 countries and territories.
GSM is operated mainly in three frequency bands, i.e., GSM-900, GSM-1800, and GSM-1900 with 124, 374, and 299 radio channels, respectively. The original GSM network was developed for voice communication supporting a variety of voice codecs. The bandwidth for each GSM channel is 200 kHz, and there are eight time slots in each frame which lasts for 4.615 ms. The raw data transmission rate achieved at each channel is 270.833 kbps, and it is shared by eight users. With the combination of TDMA and FDMA, GSM provides the capability of simultaneous conversations at the same frequency through different time slots. Other salient features of GSM include data encryption, subscriber identity module (SIM) card which gives a unique identity to each mobile station, and global roaming (Eberspächer et al., 2009). Furthermore, GSM is enhanced to provide short message service (SMS), however, still based on circuit-switched technologies.
GPRS and EDGE
To support higher data transmission rates in GSM networks and provide IP services without replacing the network infrastructure, developments were made to upgrade GSM networks into General Packet Radio Service (GPRS) in year 2000. GPRS is also known as 2.5G, and there are two major enhancements for the evolution from GSM to GPRS. At the radio access network, a user is allowed to occupy up to five time slots so that 114 and 20 kbps are achieved for downlink and uplink traffic, respectively. At the core network, two entities, Serving GPRS Support Node (SGSN) and Gateway GPRS Support Node (GGSN), are introduced. With the support of Packet Data Protocol (PDP) and GPRS Tunneling Protocol (GTP), end-to-end IP services can be provided.
In 2003, there was another improvement in GSM systems with the deployment of Enhanced Data rates for GSM Evolution (EDGE) networks. EDGE is regarded as an extension of the GPRS radio access network through advanced modulation schemes. It increased the data rates to 384 kbps for downlink and 60 kbps for uplink, respectively.
Third-Generation Cellular Systems
3GPP release dates and features
Summary of key features
First release of the UMTS standard
This release added features including an all-IP core network. It was originally referred to as Release 2000
IP Multimedia Subsystem and High-Speed Downlink Packet Access (HSDPA)
Integrate the operation of UMTS with wireless LAN networks and added enhancements to IMS (including push-to-talk over cellular), and it added High-Speed Uplink Packet Access (HSUPA)
Detailed improvements to QoS for applications such as VoIP and upgrades for High-Speed Packet Access (HSPA+) Evolution as well as changes for EDGE
Provide details for the LTE System Architecture Evolution, SAE, an all-IP flat network architecture providing the capacity and low latency required for Long-Term Evolution (LTE) and future evolutions
Further enhancements to the SAE as well as allowing for WiMAX and LTE/UMTS interoperability
Up to 3 Gbps downlink and 1.5 Gbps uplink, carrier aggregation (CA), relay nodes to support heterogeneous networks, higher-order MIMO antenna configurations
Enhancements to carrier aggregation, MIMO, relay nodes, coordinated multipoint transmission and reception to enable simultaneous communication with multiple cells, introduction of new frequency bands
Enhanced small cells for LTE, inter-site carrier aggregation, interworking between LTE and Wi-Fi or HSDPA
LTE-U, LTE for machine-type communication (MTC), full-dimension MIMO, LTE Advanced Pro
Energy efficiency, location services, mission-critical data and video, massive IoT
5G Phase 1 (new radio)
5G Phase 2
The key mechanism used for medium access in 3G is CDMA which allows multiple mobile stations to transmit at the same time and in the same frequency band. In a CDMA cell, each subscriber is assigned a unique code, and the codes assigned to different stations are orthogonal to each other. As such, the number of simultaneous calls in a CDMA cell is soft limited (based on the interference level), not hard limited as in GSM.
Wideband CDMA (WCDMA) was the air interface for the UMTS standard originally proposed by ETSI in 1998. To provide peak data rates from 384 kbps to 2.048 Mbps, the WCDMA system operates on wider channels each with 5 MHz bandwidth. However, the core network architecture of UMTS remains the same as the existing GSM/GPRS networks.
WCDMA is a direct-sequence spread spectrum (DSSS) system which initially operates in the 1885–2025 MHz and 2110–2200 MHz frequency bands for uplink and downlink, respectively. It supports operation modes of both frequency-division duplex (FDD) when symmetric uplink/downlink channels are available and time division duplex (TDD) when only asymmetric spectrum is available. In addition to achieving higher data rate, the support for multi-code operation, larger number of spreading factors, and enhanced transmission diversity are the key factors which lead WCDMA to the most popular 3G technology. Moreover, for the purpose of leveraging the GSM coverage for WCDMA, seamless handovers between GSM and WCDMA are also supported as well as dual-mode handsets.
High-Speed Packet Access (HSPA)
HSPA includes two phases as the beyond UMTS WCDMA enhancements by 3GPP, i.e., HSDPA in R5 and HSUPA in R6. The main idea of HSDPA is to increase downlink data rate through techniques including adaptive modulation and coding (AMC), hybrid automatic repeat request (HARQ), and fast packet scheduling. Unlike in the previous standards, the medium access control (MAC) layer of HSDPA systems is installed at the base station, i.e., NodeB. Thus, retransmissions can be controlled directly by NodeB, leading to faster retransmission and accordingly shorter delay with packet data operation (Holm and Toskala, 2007). HSDPA is capable of supporting up to 14.4 Mbps peak theoretical throughput. Similarly HSUPA is developed to support enhanced packet data throughput for uplink. HSUPA enables additional features such as fast power control and variable spreading factors which are disabled in HSDPA. However, HSUPA does not support AMC. It is capable of providing up to 5 Mbps peak uplink throughput.
Fourth-Generation Cellular Systems
The data rate provided by 3G networks was not able to meet the growing demand for Internet access via mobile phones. From 2008, ITU’s Radiocommunications Sector (ITU-R) initiated the process for developing a new system known as IMT-Advanced. The main requirements for this new system include supporting 100 Mbps and 1 Gbps peak data rate for high- and low-mobility scenarios, respectively, bandwidth scalability up to 100 MHz, mobility support with up to 350 km/h, 10 ms user plane latency and 100 ms control plane latency, worldwide roaming capability, inter-networking with other 2G and 3G systems, and improved spectral efficiency (Korhonen, 2014). Radio technologies that could meet these requirements are termed as 4G systems. LTE Advanced (LTE-A) developed by 3GPP is regarded as the de facto 4G mobile communications system from a global perspective. Although LTE does not meet the requirements for 4G as specified by ITU, it is often marketed as a 4G technology.
LTE is presented here under the 4G umbrella considering the fact that both LTE and, the true 4G technology, LTE-A, are based on orthogonal frequency-division multiplexing (OFDM)/OFDM access (OFDMA) mechanisms for medium access. LTE supports both FDD and TDD operations and provides flexible operations in both symmetric and asymmetric spectra. To enhance uplink power efficiency, LTE adopts single-carrier frequency-division multiple access (SC-FDMA). To achieve higher data rate, LTE offers flexible bandwidth allocation up to 20 MHz. Moreover, much shorter frame sizes (10 ms frames and 1 ms sub-frames) are introduced. With a configuration of 20 MHz spectrum and 4 × 4 MIMO, 326 Mbps on the downlink and 86 Mbps on the uplink can be achieved in LTE radio networks.
To upgrade LTE core networks, 3GPP R8 introduced Evolved Packet Core (EPC) which was designed to provide higher capacity, all-IP support, and reduced latency. Unlike the hierarchical architecture used in GPRS and UMTS core networks, the main design principle in EPC was to keep the architecture simple and flat.
The LTE-A standard ratified by 3GPP R10 is the major standard for 4G. Many of the existing features in R8 are inherently supported in LTE-A. In addition, carrier aggregation up to five component carriers, use of relays, higher-order MIMO, and enhanced inter-cell interference coordination (eICIC) served as new features in LTE-A. Furthermore, in R11, coordinated multipoint (CoMP) transmission and reception and enhanced self-organizing network (SON) are designed as parts of the LTE-A networks.
The same as in LTE, the medium access mechanism in LTE-A adopts OFDMA for downlink and SC-FDMA for uplink. The main advantages of OFDMA include the ability to gain much frequency diversity through randomly distributed subcarriers, multiuser diversity via assigning contiguous sets of subcarriers, and adaptive bandwidth allocation. To increase transmission power efficiency and reduce the cost of power amplifiers for mobile stations, SC-FDMA is employed thanks to its low peak-to-average power ratio. At the physical layer, adaptive coding and modulation (ACM) is adopted in 4G systems. With ACM, the modulation order and the coding rate are fine-tuned based on the channel state information to gain the full use of radio channels.
Another technique for data rate enhancement is channel aggregation which assembles multiple channels together to perform data transmission over wider bandwidth. Furthermore, LTE-A continues to improve capacity and reliability through more advanced MIMO technologies. While single-site MIMO introduces beamforming, spatial multiplexing, and transmit diversity into the system, cooperative MIMO facilitates CoMP transmission and reception. Under CoMP, a mobile station is able to receive signals from multiple base stations and likewise for the uplink transmission. By adopting a proper coordination scheme, the received signal quality is greatly improved.
Due to the fact that mobile stations cannot support MIMO with many antennas, cell-edge users may not obtain 4G-level quality of service, although a large number of antennas can be installed at the base station. To improve the performance of these users, relaying is adopted as a cooperative communication technique where single-antenna mobile stations transmit their signals to the base station via a relay station that is located much closer to the cell edge (Sesia et al., 2011). Another critical requirement for IMT-Advanced is effective power management. The aforementioned techniques including SC-FDMA, advanced multi-antenna design, and relaying collectively contribute to significant reduction of mobile station power consumption.
On its journey from 1G to 4G, cellular networks experienced FDMA-based analog systems (NMT, AMPS, TACS), TDMA-based digital systems (GSM, GPRS), CDMA-based IP-enabled systems (WCDMA, HSPA), and OFDM-based broadband mobile systems (LTE/LTE-A). The core network of cellular systems has gradually evolved from a circuit-switched telephone network to a packet-switched mobile network with an all-IP capability. Another trend along with this evolution is that nowadays cellular technologies are being developed as a global common standard in lieu of multiple regional standards developed in the early years which were not compatible with each other.
- Eberspächer J, Vögel HJ, Bettstetter C, C Hartmann C (2009) GSM architecture, protocols and services. Wiley, West SussexGoogle Scholar
- Korhonen J (2014) Introduction to 4G mobile communications. Artech House, LondonGoogle Scholar
- Nordic Mobile Telephone Group (1995) Nordic mobile telephone. StockholmGoogle Scholar