Non-orthogonal multiple access (NOMA) has been recently proposed as a technique to increase the network throughput and to support massive connectivity, which are major requirements in the fifth generation (5G) communication systems. The NOMA can be realized through two different approaches, namely, in (a) power-domain, and (b) code-domain. In the power-domain NOMA (PD-NOMA), multiple users are assigned different power levels – based on their individual channel quality information – over the same orthogonal resources. The functionality of PD-NOMA comprises of two main techniques, namely, superposition coding at the transmitter and successive interference cancellation (SIC) at the receiver. An efficient implementation of SIC would facilitate to remove interference across the users. The SIC is carried out at users with the best channel conditions and is performed in descending order of the channel. On the other hand, in the code-domain NOMA (CD-NOMA), multiplexing is carried out using low-density spreading sequences for each user, similar to the code division multiple access (CDMA) technology. In this article, we provide an introduction to NOMA and present the details on the working principle of NOMA systems. Later, we discuss the different types of NOMA schemes under PD- and CD-domains, and investigate the related applications in the context of 5G communication systems. Additionally, we discuss the integration of NOMA with other technologies related to 5G such as cognitive radio and massive MIMO, and discuss some future research challenges.
Index Terms—Fifth generation (5G) communication systems, interference mitigation, non-orthogonal multiple access (NOMA), successive interference cancellation (SIC), superposition coding (SC).
1 Wireless Communication Systems
The success story of the wireless communication technology is unprecedented. No other technology – neither the radio, television nor personal computers; not even the internet – has managed to attract billions of users in such a short time. Ever since the development of the first analog wireless communication system in the 1980s, a new generation of mobile communication system has been introduced in every decade. Each generation has received a considerable research attention in terms of the key innovations, wireless service, regulations and standards, from both the academia and the industry. The term innovations refers to the factors related to the underlying technology, while the term service represents the fundamental applications driving a particular generation – such as voice calling, messaging service, internet, etc.
In the following, we provide a brief overview on each generation of the mobile communication systems, in terms of innovations, services and standards.
1.1 First Generation (1G)
The spectrum during 1G was given free of cost to state-owned operators, and the user equipment, subscription, and the tariff were much higher as compared to the most expensive handsets and calling costs today. The basic service offered was high quality voice calling with excellent network coverage. Additionally, data communication such as facsimile was possible. The wireless standard used in the US was the advanced mobile phone system (AMPS); the Nordic mobile radio (NMR) system was used in most of European countries, and C-network – which marked the start of the subscriber identity module (SIM) – was used in Germany. The cellular handover across the European countries was not possible.
1.2 Second Generation (2G)
The groupe speciale mobile (GSM) was the dominant standard in 2G, which quickly gained worldwide popularity in the early 1990s. The other contemporary systems such as the IS-95 (in the US), and the Japanese personal digital (JPD) were proved to be no match to the growth of GSM. The monopoly of earlier operators was broken, which attracted competition from several operators in each country. This new regulation and the resulting competition is often credited for the massive success of GSM. Apart from voice calls, the short messaging system (SMS) was introduced; which not only became hugely popular, but also paved way for modern chatting applications such as WhatsApp. Digital data services were first introduced in GSM with a speed of $9.2$ kbps, which was further enhanced to tens of kbps to few hundreds of kbps through the general packet radio service (GPRS) and enhanced data rates for GSM evolution (EDGE) technologies, respectively. In terms of innovations, the user equipment became smaller and lighter for good, with extended battery lifetime lasting upto a few hours.
1.3 Third Generation (3G)
The 3G era introduced the spectrum allotment via an open-market auction in some European countries such as the UK and Germany. Despite the high quoted prices, a few licenses were successfully auctioned. Subsequently, cellphones following the universal mobile telecommunication system (UMTS) were introduced. The selling point of UMTS was a high data rate of about $2$ Mbps. The initial models of UMTS phones were bulky with low battery life and attracted very less user attention, until phones comparable to the GSM-type models were manufactured. The auction-based regulation and the non user-friendly mobile phones were among several reasons which led to several industries filing bankruptcy and quitting the cellular business. The UMTS has largely turned out to be a superhype, and is proved to be a failure. Additionally, UMTS also faced a severe competition from the wireless local area network (WLAN) technology, which provided good connectivity with high data rates for a cheaper price.
1.4 Fourth Generation (4G)
The failure of UMTS and success of WLAN led to the introduction of the long term evolution (LTE) in 4G, which is essentially a cleverly modified version of WLAN. Additionally, the operators got the spectrum licenses for significantly less money. The simultaneously introduced elimination of roaming charges for services such as voice calling, SMS and data across the countries in parts of Europe and Asia helped in the success of LTE. The maximum speed promised by LTE is about $300$ Mbps in downlink and $50$ Mbps in uplink. Apart from the usual voice and data services, a technology similar to the voice over IP termed as the voice over LTE (VoLTE) has seen improved speech codec rates and voice quality. Currently, LTE is being improved with technologies such as massive multiple-input, multiple-output (MIMO) systems and device-to-device (D2D) communications.
1.5 Fifth Generation (5G)
The 5G systems are envisioned to start functioning from 2020 onwards, with several promises such as significant improvements in data rates (about $10-20$ Gbps), latency (about $1$ ms), and spectral efficiency, as compared to LTE. Additionally, D2D and machine-to-machine (M2M) communications leading to the successful implementation of internet-of-things (IoT) is expected to be a reality very soon. The services offered in 5G are expected to find applications across various scenarios including vehicular networks, e-health, education, and industrial IoT. Some of the key innovations that drive 5G include massive MIMO, cognitive radios, millimeter wave communications, network virtualization, and software defined networking, to name a few. Although research surrounding 5G seems promising and exciting, several stake holders believe that 5G is superhyped, similar to the UMTS during 3G era. The services and promises surrounding 5G are strikingly similar to what was envisioned during 3G era, which includes applications in vehicular networks and IoT. However, the debate on whether this claim is true, and if not, how should the technologies driving 5G are expected to compete against LTE and LTE advanced are relevant topics of discussion for another study.
A summary of key comparative aspects in 1G – 4G technologies are provided in Table 1.1. One of the “revolutionary” features in the above mentioned generation of wireless communications are their respective multiple access techniques. The multiple access technologies used in 1G, 2G, 3G and 4G are frequency division multiple access (FDMA), time division multiple access (TDMA), code division multiple access (CDMA) and orthogonal frequency division multiple access (OFDMA), respectively. In the next section, we provide a brief description of these techniques.
|First Commercialization||USA||Finland||Japan||South Korea|
|Technology||AMPS, NMT||IS95, GSM||IMT2000, WCDMA||LTE, WiMax|
|Switching||Circuit Switching||Circuit/Packet Switching||Packet Switching||Packet Switching|
|Data Rates||2.4 – 14.4 kbps||14.4 kbps||3.1 Mbps||100 Mbps|
|Special Characteristics||First Wireless||Digitalized 1G||Broadband||All IP, high speed|
|Features||Voice calls||Multiple users voice calls||Multimedia||Live streaming|
|Supports||Voice||Voice and Data||Voice and Data||Voice and Data|
|Internet Services||Nil||Narrowband||Broadband||Ultra Broadband|
|Bandwidth||Analog||25 MHz||25 MHz||100 MHz|
|Operating Frequencies||800 MHz||900/1800 MHz||2100 MHz||850.1800 MHz|
|Band Type||Narrowband||Narrowband||Wideband||Ultra Wideband|
|Carrier Frequency||30 kHz||200 kHz||5 MHz||15 MHz|
|Advantages||Simple||SMS/MMS, Internet access||High security, Roaming||Speed, MIMO, Global mobility|
|Disadvantages||Limited capacity||Limited network range||High power consumption||Hardware complexity|
|Applications||Voice calls||Voice calls, SMS, Browsing||Video conference, mobile TV||High speed applications|
2 Orthogonal Multiple Access (OMA)
From the theoretical and design principle point of view, the TDMA, FDMA, CDMA and OFDMA belong to the class of orthogonal multiple access (OMA) techniques. The techniques in OMA all share a set of resources across users, which are orthogonal to each other. This enables successful separation of information-bearing signals intended for each user, by employing optimal, low-complexity and cost-efficient receiver structures. These orthogonal resources are time, frequency, code and sub-carriers in case of TDMA, FDMA, CDMA and OFDMA, respectively. Next, we provide a brief explanation on all these schemes, in the light of earlier generation of mobile communication systems. A schematic representation of FDMA, TDMA and CDMA is shown in Figure 1.
2.1 Frequency Division Multiple Access
In FDMA, each user is allocated different bandwidths, which are wide enough to carry the information-bearing signal spectrum. This OMA technique was widely used in classical wired telephone systems and subsequently in the 1G analog wireless systems. Other than that, FDMA is also used in digital TV cable television, fiber optics, and aerospace telemetry. Early satellite systems also used FDMA.
2.2 Time Division Multiple Access
The TDMA was an integral part of the 2G communication system, and is used by the celebrated GSM technology. In TDMA, every channel/bandwidth is divided into time slots and each user sends its information over different time slots, sequentially. Ideal for the relatively slowly varying voice signals – the backbone of GSM, the TDMA finds less utility in transmission of high-speed data. In GSM, the spectrum is divided into eight time slots of 200 kHz band each, where each slot is transmitted at a rate of 270 kbps using the Gaussian minimum shift keying (GMSK) modulation.
2.3 Code Division Multiple Access
One of the earliest forms of the direct sequence spread spectrum technique, CDMA spreads the data over the entire bandwidth with a lower power level. The CDMA is the dominant multiple access technology in 3G communication systems. Each user is assigned a sequence of spreading codes, which are orthonormal with each other. This technique enables the users to use the entire available bandwidth at the same time, without inter-user interference. In the IS-95 standard, CDMA is used with a digitally compressed voice at 13 kbps, which is spread using a 1.2288 Mbps chip sequence derived from a pseudo random code generator. As a result, the voice signal is spread over a bandwidth of 1.25 MHz. At the receiver, correlator circuit is used to separate out the intended signal from the rest. The wideband CDMA (W-CDMA) uses CDMA with 3.84 Mbps chip sequences over a 5 MHz wideband channel.