Haneet Kour, Rakesh K. Jha, Sanjeev Jain, Shubha Jain
©SHUTTERSTOCK.COM/SOLARSEVEN
The 5G New Radio (NR) technology is under a standardization process by the 3rd Generation Partnership Project to provide an outline for a new radio interface for the next generation of cellular networks. The aims of 5G networks include not only providing greater capacity and coverage but also supporting advanced services, such as enhanced mobile broadband, ultrareliable low-latency communication (URLLC), and massive machine type communication.
Key features of NR include ultralean carrier design to minimize the power consumption by limiting the “always-on” signal transmissions and reduce interference in neighboring cells (Parkvall et al., 2017). Another feature is the use of a massive number of antennas for the transmission as well as reception of signals. This rise in the number of antennas to provide greater coverage brings about various challenges and impacts in the system.
With the increase in investigations in the millimeter-wave frequencies, there is a need to investigate the health hazards they have on the human body and environment at large. This article intends to provide an insight into the harmful impacts of radio-frequency (RF) fields. The radiation metrics used to study the RF impact are the power density for the far field and specific absorption rate (SAR) for the near field These are the two main electromagnetic (EM) radiation metrics employed to find out the exposure due to uplink (UL) and downlink (DL) phenomena in mobile communications.
Mobile communication systems are addressed particularly to discuss the EM radiation impact as smart phones are used in close proximity to the body. A proposal in the form of a “thermal radiation” (TR) mode is given to reduce the radiation emitted from a mobile phone. The performance of the proposed mode is validated from the results by achieving reduced power density, complexity, and exposure ratio (ER).
The adverse impacts caused by EM waves are studied by standards organizations, such as the International Commission on Non-Ionizing Radiation Protection (ICNIRP) (2018) and IEEE Standard C95 (IEEE, 2006). The restrictions presented in these documents are based on scientific data and studied by measuring the effects in terms of the electric and magnetic fields.
Rising power density levels disturb the photosynthesis of plants and trees, which leads to their poisoning and hinders growth and physiological processes (Tang et al., 2018). As birds and animals can detect magnetic stimuli, rising EM radiation has led to a decline in the population of birds, such as sparrows (Cucurachi et al., 2013). Studies in the literature have validated a reduction in the growth rate in animals due to EM radiation (Cucurachi et al., 2013). Also, extreme weather conditions are on the rise every year due to an increase in the emission of harmful gases, such as CO2 [179 Mt of CO2 emission, leading to a greater greenhouse effect (The Climate Group, 2020)].
In humans, initial symptoms, such as headache; eye and skin problems; and, in the worst case, traces of carcinogenicity and other diseases, such as Alzheimer’s and Parkinson’s diseases, can also occur (Kshetrimayum, 2008; Wu et al., 2015). Figure 1 is a diagrammatic representation of the findings from Kshetrimayum (2008), Wu et al. (2015), Jamshed et al. (2020), and the World Health Organization International EM Field (EMF) Project for diseases and effects on human beings. The effects on animals, birds, and insects are based on research from Cucurachi et al. (2013), and the destruction of nature by EM pollution is from an investigation by Tang et al. (2018). The increase in the CO2 footprint due to RF waves is based on findings from The Climate Group (2020).
Fig 1 The general architecture corroborating the EM radiation impact. CFCs: chlorofluorocarbons; HFCs: hydrofluorocarbons.
Various works discuss new opportunities for reducing EM exposure and planning future networks in a way so that the EMF limits and constraints are abided by (Chiaraviglio et al., 2018). There is a requirement to update the safety guidelines and EMF evaluation framework while keeping in view the impact of the wireless communication industry on EMF exposure (Jamshed et al., 2020).
For future-generation networks, the heterogeneity increases as we move toward ultradense networks (UDNs) with incorporation of newer technologies. There will be a significant role of the EM radiation that will be emitted with these networks and the effects produced by them. We discuss some techniques that can be incorporated to reduce this impact (Sambo et al., 2014):
It has still been found that the expected percentage of the CO2 footprint has not reduced and is predicted to rise considerably in the years to come. This calls for more enhanced techniques or proposals to be introduced for the upcoming 5G NR scenario.
The large amount of interference caused in UDNs increases the requirement for self-interference cancellation techniques, such as nonorthogonal multiple access schemes, and so on. Various challenges in a full-duplex system include UL-to-DL interference among different users in a single-cell scenario and a complex version of it in a multicell situation, where there can be DL-to-DL interference between user equipment and UL-to-UL interference at the base stations. They have an adverse impact on the overall performance of a system.
In Fig. 2, a proposal is presented to reduce the EM radiation impact and improve the signal-to-interference-plus-noise ratio (SINR) for mobile communication systems. The scenario presents the conventional and proposed approaches. There is consideration of 50 users in the active mode (AM), and, in the TR mode region, 30 users are assumed to be in the AM and 20 in the TR mode. Nonstationary and wideband channels are assumed considering the Rayleigh fading distribution (Jha and Kour 2020).
Fig 2 The proposed TR mode. AP: access point; API: application programming interface; U: user.
Catering to the issue of “always-on” signals in current mobile communication systems, the TR mode switches the communication mode from full to half duplex. When the “TR mode” is switched On in the network, only the DL information transfer signals remain active, and the UL ones are disabled. Only the reference signals required for DL information transfer and to maintain a stable connection with the base station remain active; i.e., the reference signals are on only when there are data to be transmitted, and the base station detection signal is active (Fig. 2).
The proposed methodology for a device transitioning in the TR mode is based on an adaptive switching operation. The switching occurs when the mobile phone does not have adequate signal strength to support applications requiring high bandwidth. We consider a mobile device communicating with “always-on” signals in a cell. The received signal strength changes in accordance with the channel conditions between the base station and mobile user. When the received signal strength degrades below a threshold, the device cannot support applications requiring a high data rate.
We propose adapting switching in the mode of the device from “active” to “TR.” With this mode, the device supports only applications requiring a low data rate, such as voice calls and regular text messages. All applications requiring a high bandwidth cannot be supported until the signal strength improves. Whenever the channel conditions are favorable, and the received signal strength is greater than the threshold, the device automatically transitions to AM, thereby supporting all of the applications with “always-on” signals.
This mode improves the SINR of the entire network. With respect to near-field communication, the thermal heating produced in the device is reduced, which, in turn, decreases the SAR. Regarding the far-field communication, the power density in the network is lower; therefore, both factors collaboratively help in reducing the EM radiation exposure produced in the network.
The waveforms supported in 5G NR are similar to those in LTE, which include cyclic prefix-orthogonal frequency-division multiplexing (OFDM) (UL/DL) with discrete Fourier transform-spread-OFDM (UL). There is flexible numeric support in 5G NR with 15 kHz × 2n subcarrier spacing. The proposed frame structure for 5G NR is depicted in Fig. 3, keeping the LTE frame format as the basis, along with the supported duplex schemes for the spectrum allocation.
Fig 3 The frame format for 5G NR (the FDD and TDD frame types). D: downlink; DwPTS: downlink pilot time slot; FDD: frequency-division multiplex; GP: guard period; TDD: time-division multiplex; U: uplink; UpPTS: uplink pilot time slot.
One radio frame consists of 10 subframes, each of 1-ms duration. The duplex schemes supported are similar to LTE; i.e., frequency-division duplex (FDD) and time-division duplex (TDD). As depicted for the full-duplex frame type, both the UL and DL frames are of 10-ms duration each and transmitted simultaneously, as they are separated by different UL and DL frequencies. For the TDD frame type, the transmission frequency remains the same, and the multiplexing takes place in the time domain. For the FDD one, on activation of the TR mode, the mobile device receives only DL information and does not transmit in the UL.
A frequency-switching subframe has been added in the UL frame, which, on activation of the TR mode, changes its state from zero to one; i.e., it stops the UL transmission. On activation of the TR mode in the handset, slot 1 is active—that is, the frequency switching occurs; otherwise, the UL frame continues its transmission, and slot 0 is active.
For the proposed TDD frame type, a subframe has been added in the previously termed special subframe, which is now called the superframe (TR mode). On activation of the TR mode, the added subframe goes in the “hold” mode; i.e., it avoids the switching of the subframe from DL to UL and stops all of the UL transmissions in the TDD frame. When the TR mode is deactivated, it is released, and the TDD frame continues its transmissions in the conventional way. The hold/release modes are allocated one slot each in a subframe for the proposed mode.
According to the standardization for 5G NR, there are three protocol states that exist in the radio resource control (RRC) state machine: RRC_Idle, RRC_Inactive, and RRC_Connected. RRC_Idle is optimized for a lesser consumption of power as well as resources in the network. RRC_Connected is for high activity of the user equipment. RRC_Inactive is a state that reduces/lightens the transition procedure. We propose an RRC_Energy Efficient (RRC_EE) state for the proposed TR mode, which is a low-activity state in the transition model.
In the proposed state transition model for the TR mode, the state machine consists of RRC_Idle, RRC_Active, RRC_Inactive, and the novel RRC_EE, as depicted in Fig. 4. The proposed state—i.e., RRC_EE—is the one corresponding to the TR mode, that is, a low-activity state. The characteristic of RRC_EE is that the 5G Core-NG‐Radio Access Network (5GC-NG-RAN) connection is kept for the user equipment only for DL, and not UL, information transfer. This reduces the overhead signaling in the air interface.
Fig 4 The potential state transition model for the TR mode in 5G.
The transitions among the idle, connected, and inactive states are expected to follow the LTE procedure. The state transition to RRC_EE is initiated when the user equipment is in the TR mode. Whenever there is a UL information transfer to take place, the mode is changed to active, and AM communication resumes. This state has been configured for quick transitions and to incorporate URLLC services.
The performance of the proposed mode is presented in terms of reduced power density, ER, and complexity of the system. As we move on to the higher generations of wireless communications, the power density levels also increase correspondingly in the atmosphere. Figure 5 depicts the power density as a function of rising generations of wireless communication, with the highest value being obtained for 5G [IEEE Standard C95 (IEEE, 2006)]. The comparison of the power density values is made with the TR mode. As the TR mode operation suspends the UL transfer of information, there is less overhead signaling involved. This decreases the transmitted power from each device and, therefore, the overall power density in the network. The initial and new values obtained are given in the graph in Fig. 5.
Fig 5 The evolution of wireless generations versus the power density.
The TR mode in a device reduces the ER obtained, which is a ratio of the electric field produced by the device and the reference level of electric field or maximum permissible exposure allowed in that area. A comparison of the ER is made for 1G to 5G with the ICNIRP and IEEE C95 standards, as depicted in Table 1. It is evident that the TR mode reduces the ER in both of the scenarios for the ICNIRP standard as well as IEEE Standard C95.
Table 1. A comparison of the ER values for wireless generations with the ICNIRP and IEEE C95 standards.
With the suspension of the UL information transfer signals, the EM radiation produced from the device on activation of the TR mode is also limited. Therefore, the interference produced due to each device operating in the TR mode reduces with the suspension of some signals temporarily. A 3D pattern for rising ERs for AM and TR mode is obtained. The obtained ER values increase as we move from 1G to 5G and are evidently lower for the TR mode in comparison to the AM (Fig. 6).
Fig 6 The 3D patterns for the ER rise in the (a) AM and (b) TR mode for wireless generations (IEEE Standard C95).
As we have considered a nonstationary and wideband channel, there is fluctuation in the received signal-to-noise (SNR) ratio due to fading, multipath, and so on. The outage probability has been obtained for devices communicating in the AM and TR mode. Those communicating in the TR mode produce less interference in the network, which improves the associated SNR. This helps in achieving low EM radiation exposure without compromising the target data rate. The outage probability graph in Fig. 7 denotes the low-SNR region as a zone of high EM radiation, and, when the device transitions to the TR mode, there is an improved SNR obtained with less EM radiation emission.
Fig 7 The outage probability versus the signal-to-noise ratio (SNR) for the AM and TR mode.
The reduction in interference caused due to adjacent cells in the TR mode decreases the complexity of the entire system. The graph in Fig. 8 depicts a comparison of the complexity in the network as a function of the interference power for active devices in the AM and TR mode. It is evident from the graph that there is reduced complexity in the network for the TR mode in comparison to the AM.
Fig 8 The complexity versus the number of active users.
This article aims to provide insight into the 5G NR interface standard that will be operating at a very wide spectrum range, utilizing high-frequency bands to incorporate a large number of devices and high-bandwidth-demanding applications. The concern of high EM radiation exposure that arises with the growing wireless communications is addressed. Rising EM radiation is responsible for increasing power density levels in the environment.
A proposal has been given for mobile communications with a proposed mode (the TR mode) to reduce the radiation levels in the atmosphere and also improve the biological safety of human exposure. For this, a frame structure design and RRC state handling procedures were also presented for 5G NR. The performance of the proposed mode can be seen in terms of reductions in the power density, complexity, and ER produced.
The authors gratefully acknowledge the support provided by 5G and the Internet of Things Lab, Department of Electronics and Communication, and the Technology Business Incubation Center, Shri Mata Vaishno Devi University, Katra, Jammu.
Haneet Kour (hani.kpds@gmail.com) earned her B.E. degree in electrical and computer engineering from Jammu University, Jammu and Kashmir, India, in 2015, and her M.Tech. degree from Shri Mata Vaishno Devi University (SMVDU), Katra, Jammu and Kashmir, 182320, India, in 2017. She is earning her Ph.D. degree at SMVDU. Her research interests include the emerging technologies of the 5G wireless communication network. Currently, she is investigating power optimization in next-generation networks, and she is working on MATLAB tools for wireless communication. She is a Student Member of IEEE.
Rakesh K. Jha (jharakesh.45@gmail.com) is an associate professor with the School of Electronics and Communication, Indian Institute of Information Technology, Design and Manufacturing, Jabalpur, 482005, India. He has published several journal articles on scalable coherent interfaces, including in IEEE transactions, journals, and IEEE magazines. His research interests include wireless communication, optical fiber communication, and security issues. He received an Asia Pacific Advanced Network fellowship in 2011, 2012, 2017, and 2018. He is also a member of the Association for Computing Machinery and the Computer Security Institute. He holds many patents and has more than 2,900 citations to his credit. He is a Senior Member of IEEE.
Sanjeev Jain (dr_sanjeevjain@yahoo.com) earned his postgraduate and doctorate degrees in computer science (CS) from the Indian Institute of Technology, Delhi. He has more than 24 years of experience in teaching and research. He has served as director of the Madhav Institute of Technology and Science, Gwalior. Presently, he is vice-chancellor of the Central University of Jammu, Jammu and Kashmir, 180001, India. He has the credit of making a significant contribution to R&D in the areas of image processing and mobile ad hoc networks. He is a Member of IEEE.
Shubha Jain (shubhajain1203@gmail.com) earned her bachelor’s degree in electronics and telecommunications from Shri Govindram Seksaria Institute of Technology and Science, Indore, India, and her master’s degree in telecommunications from the University of Maryland, College Park, Maryland, USA. Currently, she is working with Amazon, North Seattle, Washington, 98109, USA. She is a Student Member of IEEE.
Digital Object Identifier 10.1109/MPOT.2021.3091077