A review on activities of fifth generation mobile communication system

1. Introduction

In the year 2020, the data consumption is expected to increase by 30% which cannot be supported by the current technologies such as 3G and 4G. Hence there is a need of next generation wireless communication system. Now, the rollout of 5G wireless communication system is taking place all around the world. 5G is expected to be commercially available in the year 2020. At a same time there is huge increasing demand from industries, health sectors and educational sectors to utilize the advantage of wireless communication. Such kind of innovation will give a motivation to Internet of things (IOT) [1]. Till now, 5G is not defined; however, it may be the integration of several wireless techniques. Some of the technical requirement of 5G is given below [2]:

1.
Thousands time higher mobile data as compared to 4G.
2.
User data-rate greater than 1 Gbps.
3.
Ten to hundred times numbers of connected devices as compared to 4G.
4.
More battery life.
5.
Five times reduced latency as compared to 4G.

Currently, CDMA and OFDM are the modulation techniques used in 3G and 4G mobile communication system. ISI (Inter-Symbol-Interference) and High power consumption were certain disadvantages of CDMA [3]. CDMA was used in 3G system which was replaced by OFDM due to several advantages of OFDM like ease of implementation, immunity to interference, high data-rate etc. But OFDM also possess certain disadvantages like use of Cyclic Prefix (CP), large side lobe which limits the utilization of spectrum. Additionally, PAPR is also considered to be one of the biggest hurdles in OFDM which greatly reduced the performance, efficiency of non-linear OFDM amplifier. For instance, the loss of bandwidth due to the cyclic prefix is about 9% in 4G Wi-max. Therefore, OFDM is not likely to be considered for next generation mobile communication system. Hence, Many Researchers, Scientist are exploring better modulation techniques that can be a suitable for 5G. Hence, among many modulation techniques, FBMC is considered as most suitable candidate for next generation mobile communication system. Basically, FBMC is an advanced technique of OFDM where no cyclic prefix is used and even gives a best performance and efficiency as compared to OFDM. Hence, the use of FBMC results in greater spectral efficiency and increase in capacity of the system. The details of FBMC are discussed in latter part of article. The integration of cognitive radio in 5G mobile communication is considered to be latest technique that makes the next generation wireless communication system to be more intelligent. Most of the researchers and academicians believe that the integration of 5G with cognitive radio will achieve the success of accessing the communication at anytime and anywhere and by anybody [4]. In addition, cognitive radio also provides a new approach of spectrum sharing between licensed and un-licensed users also known as dynamic spectrum sharing technique which helps to overcome the crisis of spectrum which most of the country is facing. One of the most challenging issues is to provide a better wireless connectivity in rural areas. It is now most discussed topic in industries and academics which is still not fulfilled as in many rural areas no wireless accessing is possible and where wireless access is there it is not good. In true sense, the success of 5G is to provide a real wireless world free from existing wireless hurdles. The paper is organized in the following manner: Section 1 describes the introduction, Section 2 presents the features of 5G, Section 3 describes the review work, Section 4 discusses the project going on in 5G, Section 6 presents the key technologies in 5G and Section 7 illustrates the issue related to spectrum.

2. Features of 5G [5]

1.
Fast Network: The user data-rate of 4G wireless communication system is 100 Mbps which is fast but not so—fast that satisfies the ever increasing demands of subscribers, industries, etc. The user in the year 2020 will experience a data-rate greater or equal to 1 Gbps [5].
2.
Reliable service in crowd areas: Due to a huge traffic, Users experiences denial of service due to overloading of network. Hence, 5G aimed to give a better service and connectivity in crowd place such as shopping malls, metro station [5].
3.
Service in Remote Place: Some of the application for remote place includes remote meter reading for billing purpose, e-health like telemedicine, smart city, and video surveillance. 5G aimed to improve this services in remote place [6].
4.
Integration of numbers of low power devices: Already 4G supports huge numbers of low power devices but still for some application 4G does not meet the requirements. Hence 5G aimed to supports huge number devices consuming low power and such devices will be seamlessly integrated in commercial 5G mobile [6].
5.
Intelligent Handover: Handover means a switching of call from one network to network or switching within the cell of same network. Present scenario of handover is quite complicated since the delay occurs during handover is large which results in call dropping. Hence in 5G, an intelligent handover is expected with a least delay during the switching of the network.
6.
Pseudo Outdoor Communication: Research has proved that more than 50% of voice traffic and 70% of data traffic originate from indoor areas but network coverage and service in indoor area are not so good as compared to outdoor area. Hence, next generation mobile communication system is aimed at pseudo outdoor communication where network coverage, data-rate and other services in indoor area are equivalent to outdoor area.
7.
Utilization of White Spectrum: White band utilization should be one of the important aims of 5G because at present white band spectrum is un-utilized and its utilization solves the issue of spectrum crisis in maximum possible extent.
8.
High Capacity: In the year 2020, consumption in wireless traffic is expected to increase by 30%; hence, 5G network should accommodate the increasing numbers of users with best quality of service.

3. Review

The work done by researchers and academician in the field of fifth generation mobile communication system is highlighted. Fifth generation mobile communication is one of the emerging technologies which will change the face of engineering communication system. So it is essential to discuss its requirements, challenges, benefits, disadvantages, etc., for the successful implementation of 5G. 5G challenges including hardware and software are described and discussed [7], [8], [9], [10]. Still 5G is an undefined standard so the future technology use in 5G is open due to which a lot of innovation for 5G is in progress. The key technologies used in 5G are described [11], [12], [13], [14], [15]. In this work, a complete hardware, millimeter wave is designed and developed for 28 GHz and 38 GHz frequency which can be utilized by steerable directional antennas at the base and mobile station [16]. This work describes role of cloud computing technology to achieve the flexible 5G radio access network. It is regarded as one of the most complicated problems due to the increasing numbers of wireless device, sensors, etc [17]. In this work an efficient 5G network is designed and developed by using a combination of green communication and software focused mainly in energy efficient design, cognitive signaling, invisible base station and full duplex radio [18]. The author in this work has explored the different types of network and devices that contribute to the success of 5G mobile communication system [19]. In this work, the authors have proposed a separate network architecture for indoor and outdoor communication system. Some of the major technologies such as Massive MIMO and visible light communication are also discussed [20]. An article in communication magazine has described the impact and potential of five technologies and they are device centric architecture, millimeter wave, massive MIMO, smart device and M2M communication that could bring a revolutionary impact on design and concept of 5G [21].

4. Project in 5G

This section reveals the project going on for implementation of 5G.

1.
METIS: In METIS project, the 5G scenarios are explained and described. The project gives the details of challenges of 5G such as greater than 1 Gbps, accessibility, mobility, reliability. Metis have also carried out the successful researches in technology component such as MIMO, Multi nodes, spectrum, rat [22].
2.
5G Now: This project focuses on non-orthogonal wave for asynchronous signaling. It introduced an efficient air interface technique that follows a strict orthogonality and synchronization. Some of the possible candidates for 5G waveform are universal filter multi carrier (UFMC), filter band multi carrier (FBMC), generalized frequency division multiplexing (GFDM) [23].
3.
EMPHATIC: Enhanced multi carrier technology for professional Ad-hoc and cell based communication has designed and developed efficient filter bankprocessor, equalizer, etc [24].
4.
E3NETWORK: Energy efficient E-band transceiver for back haul of future network has utilized SiGe BICMOS advanced technology and digital multilevel modulation to implement an energy efficient and high speed transceiver [25].
5.
PHYLAWS: Physical layer wireless security aimed at designing of secured wireless communication by using secrecy coding approaches [26].
6.
DUPLO: Full duplex radio for local access designed an efficient transmitter and receiver that can provide a high capacity and efficiently utilized the bandwidth [27].
7.
CROWD Project: It focuses to build a heterogeneous network that can be integrated to next generation wireless communication system [28].
8.
MAMMOET: Massive MIMO for efficient transmission project aimed at efficient designing of massive MIMO for 5G mobile communication [29].
9.
LEXNET: Low EMF exposure network focused to reduce electromagnetic field up to 50% without compromising the quality of service [30].
10.
Tejas Network: The IIT-Hyderabad, India, working with Tejas network to implement a 5G cloud radio access network (CRAN) [31].
11.
5G project partnership: It aimed at developing an efficient multi service air interface techniques, high capacity network, green network and it is also working on validation of different developed concepts.
12.
Samsung: In the year 2012, Samsung claimed to be the first one to design a 5G millimeter wave band at the frequency of 28 GHz with the speed of 1.056 Gbps to a distance up to 2 km [32].

5. Key technologies in 5G

In this section the various future and key technologies are discussed which will play a very important role in development of 5G mobile communication system.

5.1. OFDM (Orthogonal frequency division multiplexing

OFDM stands for the orthogonal frequency division multiplexing where bandwidth is divided into number of subcarriers which are orthogonal to each other. It is implemented by using FFT (Fast Fourier Transform) and IFFT (Inverse Fast Fourier Transform) at transmitter and receiver. The advantage of this technique is increased in data-rate, high capacity, and immune to Inter Symbol Interference (ISI). While dividing the subcarriers, the response of the channel is flat which makes this system more efficient and hence it is preferred in 4G mobile also known as LTE (long-term evolution). But when it comes for 5G, OFDM is not a suitable candidate due to several disadvantages such as cyclic prefix (CP) and PAPR (Peak Average Power Ratio). When the symbols are so close to each other, it may cause a noise; hence, CP is inserted between the symbols due to which ISI is reduced. But addition of CP also results in loss of bandwidth and about 9% of bandwidth is lost due to CP due to which spectral efficiency will reduce. PAPR also significantly reduced the performance of the system which results in addition of data in subcarriers. If PAPR of the system is 10 dB then it means that in order to transmit it to 1 dB of signal, and it needs a 10 watt of power which efficiently reduces the performance of the system. Hence, OFDM is not suitable modulation technique for 5G. Therefore, researchers around the world are looking for the new modulation technique that satisfies the need of 5G [33], [34], [35].

5.2. Filter band multi carrier (FBMC)

It is an advanced technique of OFDM which does not use CP but uses arrays of filters at transmitter and receiver. The advantage of this technique is that without using the CP, it can give an efficient and better performance than OFDM. Hence, it is one of the most promising modulation techniques for 5G. In this technique, a bank of filter is used through which a set of parallel data is transmitted. The adjacent leakage and localization of frequency can be controlled by using an appropriate prototype filter. Due to the flexibility in the frequency domain of FBMC it is better suited for TVWS (Television White Spectrum) and spectrum usage. The channel delay spread can be easily handled by FBMC and also fragmented spectrum accessing is achievable [36], [37].

5.3. Universal filter multi carrier (UFMC)

UFMC is based on a filter bank multi carrier and OFDM which is considered to be best and novel modulation scheme for 5G. In this technique a filtering operation is performed on a group of sub-carriers unlike FBMC where filtering is applied to each subcarriers. The technique efficiently reduces the side lobewhich increases the performance of the system. It is usually for short burst communication because it utilized a very short length of filter. In this technique, the bandwidth is divided into numbers of sub bands and is allocated to the number of sub-carriers. At the transmitter N-point IFFT operation is performed which converts the time domain of the signal to a frequency domain. At receiver, FFT is performed which converts frequency domain to a time domain [38].

5.3.1. GFDM (Generalized frequency domain multiplexing)

It is non-orthogonal in nature and is first contender for non-orthogonal waveform for 5G mobile communication system. In this technique, the modulation is based on a Balian low theorem and transmission is achieved by using a time and frequency localized pulse [39]. In this technique, a filtering is used for each subcarrier which reduces the side lobe, PAPR, etc. At receiver, it utilized the Poisson summation algorithm for each symbols [40].

5.4. OFBMC (orthogonal frequency band multi carrier)

It can result in better utilization of available bandwidth and provide more robustness to the system. In this technique alike OFDM, the available bandwidth is divided into number of subcarriers which are orthogonal with each another and further it does not use cyclic prefix. This technique can be utilized by combining with spatial division multiplexing (SDM), space time block codes (STBC) and space time trellis codes (SPTC) which increases the performance by reducing the ISI and other impairments. Co-Channel interference effect is more with MIMO techniques; hence, OFBMC is the further interest of researchers [41].

5.5. Faster than Nyquist rate

In this technique the capacity is increased by sending more data in the time domain. It utilizes linear model techniques where a pulse is sent at a faster rate in a time domain which results in ISI, loss of orthogonality, etc. but still the signal can be recovered by employing an advanced detection techniques [42].

5.6. Cognitive radio

A CR is an intelligent communication system that is aware of its surrounding environment and utilizes the methodology of understanding the environment and takes the decision. The CR can be integrated with the wireless communication system. 5G also known as WISDOM assimilates and interrelates all the radio technologies, and CR acclimates and works with all the radio technologies. CR efficiently utilized spectrum efficiency, by unlicensed (CR users) using the free spectrum when primary user is ideal without creating any interference. The definition of Cognitive radio may vary from concept to concept. It may be defined as the radio that studies the environment and takes the decision accordingly; basically, it is adding intelligence in the network. The Cr depends upon a several parameters such as idle channel, types of data to be transmitted, channel occupancy and type of modulation schemes to be practiced. Hence, it is necessary to use the software that re-configure itself to take on the requirements or various requirements of the user. Cr utilizes a software defined radio (SDR) [43], [44].

5.7. Energy density spectrum sensing

It is one of the simplest methods for spectrum sensing, since it do not require the prior information of primary user or estimation of channel and its mathematical model is also simple to implement not involving any complexity. In this method, the energy of the received signal is estimated and compared with the threshold. If the energy of the received signal is greater than the threshold value then it is assumed that the signal detection is assumed and if the energy of the received signal is smaller than threshold value then the signal is absent. The energy detection method should be intelligent or sensitive enough to compare the difference between noise and signal. The energy detector consists of a square law device whose output is given to integrator. The output of integrator at any time is energy of input to the squaring device over the interval t [45].

5.7.1. Hypothesis for energy detection spectrum sensing

On the basis of idle and busy state of primary user with the addition of noise, the presence of signal detection at the secondary user can be modeled as hypothesis for energy detection spectrum sensing given as follows [46]:

Hypothesis 0

Signal is absent.

Hypothesis 1

Signal is present.

The original signal is complex component which has a real component and complex component. The received signal is $Y (n)$ is given as follows: $Y (n) = w (n) : Signal is absent .$ $a (n) + r (n) : signal is present .$ where $a (n$ ) is given as $h \cdot x (n)$ and h is the fading coefficient, $r (n) = rr (n) + jri (n)$ is a noise sample which is a complex Gaussian variable with mean zero (µ = 0) and variance is unity (σ = 1).

5.7.2. Dynamic energy detection

In order to improve the traditional energy detection technique, dynamic energy detection method is used. In this method the detected signals are squared and the square detected signals are subtracted with square value of current symbol. Mathematically it is given as follows: $S = {ai}^{2} - ({ai}^{2} - 1), where i = 1, 2, 3 \dots m.$

5.7.3. Power spectral density

Power Spectral Density (PSD) may be defined as the strength of variation of energy as a function of frequency. It also gives the information that when frequency variation is strong and when it is weak. The PSD of a function can be obtained by computing autocorrelation function in a Fourier space.

The PSD and auto correlation function are given by the following equations: $Yx (f) = \int_{- \infty}^{\infty} Rx (τ) e^{- j 2 π f τ} d τ$ $Rx (f) = \int_{- \infty}^{\infty} Sx (τ) e^{- j 2 π f τ} df$

The power spectral density with probability density function is given by the following: $Px (f) = \frac{Sx (f)}{\int_{- \infty}^{\infty} Sx (f) dt}$

The input and output random process correlate with power spectral density is given by the following: $Y (f) = mod (H (f^{2}) * X (f)$ where H(f) is the frequency response of the spectrum, X(f) is PSD input random process and Y(f) id PSD of output random process [47].

5.8. Matched filter detection

For Gaussian case Matched Filter detection is optimal because it maximizes the SNR of received signal and makes it apt for detection. But it is not optimal for non-Gaussian case. In general, the matched filter applies the greatest weighting to spectral components that have the greatest signal-to-noise ratio. This requires a non-flat frequency response, the associated distortion is not significant in situations such as radar and digital communications, where the original waveform is known and the objective was to detect the presence of this signal against the background noise. But the worst part of matched filter is that signal is being detected should be known; otherwise, it is not worth. A match Filter detection is the linear filter which increases the signal t noise ratio (SNR) for a given input signal. The matched filter is used when the secondary user are aware of channel information of secondary user. The work of matched filter detection is similar to the co-relation in which unknown signal is convolved with filter whose impulse response is mirror and time shifted version of reference signal [48]. Mathematically, it may be defined by the following equation: $Y (n) = \sum_{k = - \infty}^{\infty} h (n - k) x (k) .$

The coefficient of matched filter detection is defined by complex conjugate reverse signal. Two types of coherent matched filter detection are used. Coherent receiver is used when the amplitude and phase of the primary signal are known by secondary user and for non-coherent receiver, the received signal is Xerox of original signal with random phase error. For non-coherent receiver, matched filter detection is based on power or magnitude of signal. One of the reasons for the design of matched filter detection is to analyze the Pu signal in a given spectrum over a time. The primary matched filter is most suited for wireless communication, radar, sonar, intelligent radio system, etc. [49].

5.8.1. System model

Let X (t) be the original signal. The received signal for matched filter is given by $Y (r) = x (t) + n (t)$ where n (t) be the additive white Gaussian noise in a time domain. The output of matched filter is convoluted with the impulse response of a matched filter h(t). Hence, the received signal is given by the following: $Y (r) = h (t) * x (t) + h (t) * n (t)$ $Y (r) = h (t) * x (t) + ω (n)$ where $ω (n)$ is noise with µ = 0 and σ = 1.

The convolution function will match the received signal with the signal of Pu. The method of maximizing the SNR of the signal is not popular because it demands the prior information of primary user. Match Filter is basically an integrator which extracts the energy of the signal. With noise being random obviously the energy due to noise over one bit duration will be very much lower the signal energy, and thus improved SNR.

5.9. Cyclo-stationary detection

A signal is said to be stationary if its frequency or spectrum is not changing with respect to time. The frequency is constant because the function generator or any software device uses to generate a sin wave where a constant frequency is selected. This method is also known as interleaved stationary process which is not periodic function of time but its statistical features exhibit a periodicity. For this technique, the mean value of the signal and its autocorrelation function exhibit a periodicity. This method deals with the first order and second order transformation of a function and its spectral representation. This method gives better results as compared to other detection techniques at low SNR because of its ability to rejecting the noise. This method utilized the periodicity property of a primary signal to identify the presence of signal. This method uses a Cyclic Spectral Correlation Function (CSCF) for detection of primary signal. SCF uses a two dimensional spectral correlation method to identify the periodic characteristics of Primary user. It uses signals that are periodic at time t. The periodic auto-correlation function is given by the following equation: $Ry (t + τ) = Ry (t + To, τ),$

Now taking the FFT of autocorrelation function is given by following equation: ${RY}^{α} (τ) = \lim_{t \to \infty} \int_{t}^{t} y (t + \frac{τ}{2}) y (t - \frac{τ}{2}) e^{- j 2 π xtdt}$ where α is fundamental cyclic and ${RY}^{α}$ is cyclic autocorrelation function. From above, SCF may be defining following equation: ${Sy}^{α (f)} = lim_{Δ t \to at,} \frac{1}{α Δ t} \frac{1}{t} \int_{Δ t / 2}^{- Δ t / 2} YT (t,f + \frac{α}{2}) Y {(t)}^{*} (t,f - \frac{α}{2})$

The SCF determines the occupancy status of the spectrum that needs to be detected. Fresh filters can be also employed for the detection of signal which consists of array of branches that consist of frequency shifter followed by invariant filter. It is also known as frequency shift filter which exploits the spectral coherence in the signal. In this process both conjugate and non-conjugate cycliced power also imply reduction in interference toward other users [50].

5.10. Challenges in cognitive radio

1.
Spectrum sensing challenge:

Despite advanced technique introduced and proposed for accessing the spectrum, still the accessing of spectrum is difficult due to the following process:

a.
The detection of ideal spectrum is considered to be one of the most difficult tasks. Hence, a very sensitive detector is needed to sense the signal more precisely.
b.
It is very difficult to detect the primary user operating in frequency bands where transmission and reception are dynamic.
c.
The False alarm detection occurs when the noise is misinterpreted as a signal and it is most common in real world environment.
d.
Detection of primary user/ideal spectrum at a minimum interval of time is considered as one of the critical issues in Cr.
e.
The hidden terminal problem is also considered as one of the important issues in spectrum sensing.

The introduction of Cognitive radio will solve the many problems of wireless communication system. The future of Cognitive technology plays an important role to solve the bandwidth utilization problem by introducing the dynamic spectrum sharing techniques. It is expected that for next generation wireless communication, the cognitive radio is essential and it should be integrated with each next generation wireless communication system. The rollout of 5G is going everywhere which aims to increase the speed up to 1GBps, capacity to be increased by 10 times, everybody connected at anytime, anywhere. However, the motto of communication system cannot be fulfilling without introducing the cognitive radio. However, in this chapter, the basic theory, implementation of CR, problems in Cr, and possible area of future research are discussed. The techniques such as Energy detection spectrum sensing, Matched Filter spectrum sensing, Cyclo-Stationary detection are also discussed with its implementation, advantages and disadvantages [51], [52], [53].