Radio Technologies for Environment-Aware Wireless Communications

TOMAŽ JAVORNIK, ANDREJ HROVAT, ALEŠ ŠVIGELJ

Department of Communication Systems,

Jožef Stefan Institute,

Jamova cesta 39, SI-1000 Ljubljana,

SLOVENIA

Abstract - The contemporary wireless transmitter in addition to information symbols transmits also training

symbols in order to help the receivers in the estimation of the information symbols by estimating the channel

state information (CSI). In this paper, we look at existing wireless communication technologies in light of

environment-aware wireless communications, which is a new concept of wireless communications that queries

the time-invariant CSI from the local or global database, using information about the transmitter and receiver

location. Thus, this study is the first critical review of the potential of today’s terrestrial wireless

communication systems including wireless cellular technologies (GSM, UMTS, LTE, NR), wireless local area

networks (WLANs), and wireless sensor networks (WSNs), for estimating CSI, the ratio between training and

information symbols and the rate of channel variation, and the potential use of time invariable CSI in

environment aware wireless communications. The research reveals, that early communication systems provide

means for narrowband channel estimation and the CSI is only available as channel attenuation based on signal

level measurements. By increasing the frequency bandwidth of communications, the CSI is estimated in some

form of channel impulse response (CIR) in almost all currently used radio technologies, but this information is

generally not available outside the communication systems. Also, the CSI is estimated only for the channel with

active communications. The new radio technology (NR) offers the possibility of estimating the CIR for non-

active channels as well, and thus the possibility of initiating environmentally aware wireless communications.

Key-Words: - environment-aware wireless communications, wireless cellular communication systems, wireless

local area network, wireless channel estimation, wireless sensor network, channel state information (CSI),

channel impulse response (CIR).

Received: October 15, 2021. Revised: October 21, 2022. Accepted: November 12, 2022. Published: December 31, 2022.

1 Introduction

Contemporary wireless communication systems

exploit channel state information (CSI) to mitigate

wireless channel impairments, increase wireless link

reliability, or increase throughput. The standard

approach to CSI estimation is to insert the training

symbols into the transmitted signal, which leads to a

decrease in the net throughput of the wireless link.

Moreover, the ratio of training symbols to

information symbols increases as the bandwidth of

the wireless signal, the number of transmitter and

receiver antennas, and the power of the spectral

efficiency of the communication system increase.

The problem becomes critical in broadband

communication systems and in communication

systems that apply multiple antennas at the

transmitter and/or at the receiver, where the number

of training symbols becomes comparable to the

number of information symbols.

The ratio between symbols conducting

information and symbols for training decreases with

the generation of communication systems. The first

generation of wireless cellular communication

systems (1G) Nordic Mobile Telephone (NMT) is

based on analog narrowband Ultra High Frequency

(UHF) communication in 25 kHz radio channels.

The NMT voice channel is transmitted with

frequency modulation (FM). NMT supports

handover between cells, which requires signaling

and channel quality monitoring. NMT signaling

uses Fast Frequency Shift Keying (FFSK) digital

modulation. Signaling between the base station (BS)

and the mobile station occurs over the same RF

channel used for audio communication.

Consequently, users are disturbed by periodic short

noise pulses during communication, [1]. The NMT

applies frequency duplex for uplink and downlink

communication and frequency division multiplex to

allow multiple users to communicate simultaneously

in the same cell. The link quality is tested during the

WSEAS TRANSACTIONS on COMMUNICATIONS

DOI: 10.37394/23204.2022.21.30

Tomaž Javornik, Andrej Hrovat, Aleš Švigelj

E-ISSN: 2224-2864

250

Volume 21, 2022

call by the BS sending a control signal that is

returned by the mobile station. Based on the

characteristics of the returned signal, BS estimates

the signal-to-noise ratio in the downlink and uplink

channels. The amount of signaling was negligible

because the signal quality is estimated based on the

received signal strength.

Second-generation (2G), [2], wireless

communication systems benefit from the use of

wider radio channels and digital technology. The

GSM system uses 200 kHz, where selective

frequency fading can be caused by multipath

propagation. To cope with multipath propagation,

the training sequence is inserted in the middle of the

time slot. In 2G systems, approximately one training

symbol is transmitted per four information symbols,

while this ratio is increased in the later wireless

cellular systems (3G, 4G, and 5G). Since the main

goal in designing a wireless communication system

is to achieve high system spectral efficiency, the

number of training symbols must be reduced, which

may lead to erroneous estimation of CSI. However,

the static parameters of the wireless channel depend

only on the position of the terminal in a static

environment, so CSI can be retrieved from the CSIs

of previous communications stored in the central or

local database, [3]. Information about CSI is

retrieved from the database when the

communication is initialized, reducing the number

of training symbols required to estimate CSI. This

type of communication is referred to as

environment-aware communication. There is a

plethora of wireless technologies with technology-

specific quality of extracted CSI. In this context, we

will provide an overview of the wireless

technology-dependent quality of CSI extraction and

explore the potential of using CSI in environment-

aware wireless communications.

The contributions of this paper can be

summarized in the following points:

• the survey of the CSI estimation for the existing

terrestrial wireless technologies,

• the study of the existing terrestrial wireless

cellular technologies for environment-aware

wireless communications,

• the study of the existing wireless sensor

networks and wireless local area network

technologies for environment-aware wireless

communications, and

• the guidelines for future wireless technologies

to be suitable for the environment-aware

wireless communications.

The paper is organized as follows. This

introduction is followed by the definition of channel

impulse response (CIR) and channel state

information. After that, an overview of the potential

of wireless cellular technologies for estimating CSI

is given. The next section takes a closer look at CSI

estimation in wireless local area networks

(WLANs). We then explore the potential of channel

estimation in wireless sensor networks (WSNs). We

then discuss the impact of channel variations due to

either terminal movement or environmental

changes. Discussion and conclusions are provided in

the last two sections.

2 Channel State Information and

Channel Impulse Response

CSI describes the properties of the communication

channel. It describes how radio waves propagate

from the transmitter to the receiver. The content of

CSI depends on the type of communication system.

While channel attenuation is sufficient for

narrowband, flat-fading communication channels,

channel impulse response (CIR) is desired for

broadband communication systems. The CSI can be

instantaneous or statistical. The estimation of the

CSI can be data aided, where the training symbol is

added to the information signals, or blind, where the

CSI is estimated from the statistical properties of the

transmitted signals. Modern wireless

communication systems use data assisted CSI

estimation, [4], due to its fast and accurate CSI

estimation. In addition, wireless channel attenuation,

the level of the received signal or Received Signal

Strength Indicator (RSSI), and Reference Signal

Received Power (RSRP) or signal-to-interference

and noise ratio are often applied for coarse

characterization of the CSI. The CIR or Power

Delay Profile is often applied for the

characterization of the frequency-selective wireless

channels.

The basic representation of the wireless channel is

given by CIR h(τ). It can be considered as a linear

filter, [5]. If the interfering objects, the transmitter

and the receiver do not change their position or

orientation, the wireless channel or its

representation of CIR is static. However, in modern

wireless communication, the position and

orientation of the transmitter and receiver change,

and interfering objects such as vehicles, leaves,

people, or doors and windows may also change their

position or orientation, which may affect the CIR. In

WSEAS TRANSACTIONS on COMMUNICATIONS

DOI: 10.37394/23204.2022.21.30

Tomaž Javornik, Andrej Hrovat, Aleš Švigelj

E-ISSN: 2224-2864

251

Volume 21, 2022

this context, the CIR is described by a time-varying

function h(t,τ), where t is the instantaneous

(absolute) time and τ is the propagation delay. The

theory of linear time-variant systems can be applied

to the theoretical analysis of real wireless channels.

The wireless channel can be represented as an

algebraic function in the time domain as CIR h(t, τ)

or in the frequency domain as the channel transfer

function H(t, f). The channel transfer function and

CIR are related by the Fourier transform. However,

the algebraic representation is not suitable to be

stored in the database, so the discrete function

representation is often used

for n = 0, 1, 2, ··· , N − 1 and k = 0, 1, 2, ··· , N − 1.

The discrete representation of CIR is expressed for a

given sampling frequency

where T is the sampling interval. The bandwidth of

the channel transfer function H depends on the

sampling frequency fs and lies in the frequency

interval [−fs/2, fs/2], while the frequency step ∆f

depends on the number of samples

To provide complete information about the

channel transfer function H(n) or the CIR h(k), the

information about the sampling interval T or the

sampling frequency fs should be stored in the

database. The CSI is often specified as a power

delay profile (PDP), which gives the intensity of a

signal received over a multipath channel as a

function of time delay and can be expressed as CIR

∥h(t, τ)∥2.

In order to introduce environment-aware wireless

communications, in the next sections we explore the

potential of selected wireless communications to

estimate the CIR and accessibility of CIR outside

the system.

3 Wireless Cellular Technologies

The concept of wireless cellular networks was

introduced to achieve higher system capacity with

the limited bandwidth allocated to the mobile

networks to cope with the congestion in radio

networks using a single cell. Today, wireless

cellular communication systems provide near global

coverage and are therefore candidates for providing

Fig. 1: The concept of wireless cellular communication and physical radio channels.

WSEAS TRANSACTIONS on COMMUNICATIONS

DOI: 10.37394/23204.2022.21.30

Tomaž Javornik, Andrej Hrovat, Aleš Švigelj

E-ISSN: 2224-2864

252

Volume 21, 2022

wide-area links for CSI estimation on various carrier

frequencies. Since the introduction of cellular

networks in the late 1970s, wireless technology has

evolved significantly, from the first generation (1G)

to the latest (5G). Each generation of wireless

cellular communication introduces some new

concepts, making it significantly different from the

previous one. However, the main concept of

wireless cellular networks has not changed

significantly. The wireless cellular system consists

of a set of base stations (BS) distributed over the

area and a set of terminals often referred to as user

equipment (UE) or mobile stations (MS). The BSs

are interconnected by wired or wireless links in a

network. Radio resources in the frequency, time,

code, or space domains are allocated by a wireless

network to particular communication between BS

and the terminal. To successfully connect to the

network and establish communication with another

terminal device inside or outside the source

communication network, BS and the terminal

communicate through physical wireless channels,

which represent the bearer of one or more logical

channels specified and used in the Medium Access

Control (MAC) layer. The structure of the physical

channel, i.e. modulation and coding scheme, and the

amount of the training sequence depend on the

information transmitted over the physical channel

and the expected channel impairments in the

wireless network. The concept of wireless cellular

communication is illustrated in Fig. 1.

In any wireless cellular communication system

there exist several physical channels. We distinguish

between downlink physical channels, i.e.

transmission from BS to the terminal, and uplink

physical channels, i.e. transmission from the

terminal to BS. In the downlink, some physical

channels are destined for all terminals in the BS

range, while other physical channels are intended

only for a particular terminal. We call the formal

channels broadcast physical channels, while the

latter is referred to as terminal-dedicated physical

channels. Broadcast physical channels include (i)

the synchronization channels, used for carrier

frequency, time, and frame synchronization and

transmission power adjustment, (ii) the control

channels used for broadcasting control information

to all terminals in the BS range, and (iii) the data

channels used for broadcasting information about

the BS functionality as well as data directed to all

terminals in the range.

The set of terminal-dedicated physical channels

includes two types of channels, namely control and

data channel. The dedicated channels in the

downlink are usually paired with the uplink physical

data and control channels.

In the uplink, there exists also random access

channels that are applied to request access to BS,

send data, and control to BS based on the ALOHA

approach. The classes of physical radio channels

between BS and terminal are illustrated in Fig. 1.

The arrows illustrate the direction of

communication. Each generation of the wireless

cellular system uses slightly different terminology

for the BS, the terminal, and the physical channel

naming, mainly to distinguish the technology from

each other.

In this context, we provide a critical overview of

CSI estimation in existing and emerging wireless

communication systems that we believe will be in

operation in the next few years. A particular focus is

on CIR estimation. The overview includes older

digital wireless systems such as GSM and UMTS,

which are currently considered obsolete and

unlikely to be deployed in future wireless systems,

as well as wireless systems that are expected to be in

operation in the next few decades, namely LTE and

5G.

3.1 GSM - Global System for Mobile

Communications

The GSM system is based on the 1G wireless

cellular system, which uses frequency division

multiple access (FDMA) to separate users. In

addition, time division multiple access (TDMA) has

been used to take advantage of digital technology,

so GSM combines TDMA and FDMA multiple

access, [2]. Initially, the bandwidth of 25 MHz is

Fig. 2: The GSM time/frequency multiple access.

WSEAS TRANSACTIONS on COMMUNICATIONS

DOI: 10.37394/23204.2022.21.30

Tomaž Javornik, Andrej Hrovat, Aleš Švigelj

E-ISSN: 2224-2864

253

Volume 21, 2022

divided into 124 carrier frequencies spaced 200 kHz

apart. Each carrier frequency is divided into eight-

time slots, i.e., eight channels, with a duration of

0.577 ms (15/26 ms). Eight of the time slots are

combined to form a TDMA frame, which is

4.615 ms (i.e. 120/26 ms) long. The frames are

organized into so-called multi- and super-frames to

ensure overall synchronization. While the time slot

is the time allocated to a particular user, the

transmission that takes place in a time slot is the

GSM burst and carries physical channels. The GSM

frame structure is illustrated in Fig. 2.

The GSM standard specifies four types of

bursts/physical channels, namely normal bursts (in

the uplink and downlink), synchronization bursts in

the downlink only, frequency correction bursts in

the downlink only, and random access (bursts) in

the uplink only. In the GSM system, the terminal is

called User Equipment (UE), and BS is a base

transceiver station (BTS). Several BS are

interconnected and controlled by the Base Station

Controller (BSC), whose main tasks are to manage

the UE, manage the radio resources, monitor and

control the handover between BS, and establish the

connection between BS and the Mobile Switching

Center (MSC). The assignment of the GSM bursts

to the classes of physical radio channels is plotted in

Table 1.

The normal burst is applied for communication

between BS and UE. The normal burst consists of a

26-bit long training sequence. Eight different

training sequences are specified in the GSM

standard. The same training sequence is used in

each GSM slot transmitted by the same BS while

neighboring BSs using the same radio frequency

channels use different sequences. The GSM training

sequence allows the receiver to estimate the CIR

with five complex taps, [2], separated by a GSM

symbol duration of 3.69 µs, resulting in an excess

delay of almost 18.464 µs.

The purpose of the GSM synchronization burst is

transmitted from BS to help the mobile station

synchronize with the network. Due to its poor auto-

and cross-correlation properties, it is not suitable for

estimating CIR. The frequency correction burst

consists of 3 tail bits at the beginning and end of the

burst, 8.25 bits of guard time, and 142 bits set to

zero. The single-tone signals are not suitable for

estimating CIR due to their narrow band

characteristics. The random-access burst is a short

burst with a longer guard time that allows the

mobile terminal to access the BS. The Random

Access Burst is transmitted only when the mobile

terminal wants to access the wireless cellular

network. This happens from time to time, which

limits its use in estimating CIR.

The GSM system can be used to estimate the CIR,

but the channel is a narrowband channel with only

200 kHz, and the training sequence of a normal

burst can only estimate the multipath excess delay

of 18.464 µs. If possible, another technology should

be applied to estimate the CIR.

Table 1. Types of GSM physical channels, GSM

bursts, and their applicability for CIR estimation.

GSM burst

Usage

Comments

normal

uplink, downlink,

data, control

narrow band 200 kHz,

5 tap CIR,

max. delay spread:

18.5

random access

uplink,

random access

not frequent

transmission

frequency

correction

downlink,

synchronization

narrow band, poor

autocorrelation

synchronization

downlink,

synchronization

narrow band, poor

autocorrelation

3.2 UMTS - Universal Mobile Telecom-

Munications System

The Universal Mobile Telecommunications System

(UMTS) is a third-generation (3G) wireless cellular

communications system. The system architecture of

the radio part consists of User Equipment (UE), the

base station, called Node B, and the Radio Network

Controller. The Node B provides the connection

between the UE and the telephone network. The

Radio Network Controller is responsible for

controlling the Node Bs connected to it. UMTS uses

wideband code division multiple access techniques

in 5 MHz frequency channels. It supports frequency

division and time division duplex transmission. The

Orthogonal Variable Spreading Factor code (OVSF)

Fig. 3: The UMTS frame structure.

WSEAS TRANSACTIONS on COMMUNICATIONS

DOI: 10.37394/23204.2022.21.30

Tomaž Javornik, Andrej Hrovat, Aleš Švigelj

E-ISSN: 2224-2864

254

Volume 21, 2022

is used to spread the signal in the 5 MHz channel.

The OVSF codes have a low correlation with each

other as long as there are no delays or echoes in the

transmission channel. OVSF based spreading also

allows different spreading factors in different

channels. OVSF spreading is also referred to as

channelization codes. The channelization code is

used to separate the physical and control channels in

the uplink, while in the downlink they are primarily

used to separate the different users. The

channelization codes increase the signal bandwidth.

In addition to channelization codes, UMTS also uses

scrambling codes, which are used in the uplink to

separate terminals/users and in the downlink to

separate the transmission of cells. The UMTS frame

structure is depicted in Fig. 3.

Similar to GSM, the physical channels are

classified into broadcast, dedicated, and random

access channels. There are several UMTS physical

broadcast channels, [6]. They are referred to as

downlink common physical channels and include (i)

control channels, namely the primary and secondary

common control physical channel (P-CCPCH and S-

CCPCH), the acquisition indication channel

(AICH), a paging channel (PCH) and physical

channels for the common packet channel access

procedures; (ii) data channels, namely, physical

downlink shared channel (PDSCH); and (iii)

synchronization channels, namely the

synchronization channel (SCH) and the common

pilot channel (CPICH). The common pilot channel

CPICH is the main channel for power and CSI

estimation in the downlink. The CPICH is broadcast

by the Node Bs at constant power. It can be used for

signal quality measurements available as received

signal code power (RSCP) and signal-to-noise ratio

Ec/No. UMTS uplink access channels include the

Physical Random Access Channel (PRACH) and the

Physical Common Packet Channel (PCPCH), which

are used to request access to the Node B or to

transmit data based on the random multiple access

protocol. These channels do not contain sufficient

pilot symbols to be applied for the CIR estimation.

In both the uplink and downlink, there are

dedicated physical channels for control and data

transmission. The Dedicated Physical Control

Channel (DPCCH) is the most important channel for

estimating CIR. In the uplink, the DPCCH is a twin

channel to the Dedicated Physical Data Channel

(DPDCH), and is multiplexed in the I/Q part of the

DPDCH, while in downlink the time multiplexing is

applied. DPCCH is used to carry control

information consisting of known pilot bits to

support channel state estimation for coherent

detection, transmit power control (TPC) commands,

feedback information (FBI), and an optional

transport format combining indicator (TFCI). The

UMTS physical channels with their relation to the

estimation of CIR are listed in Table 2.

It is foreseen that the 3G communication systems

will be replaced in the near future by the 4G and 5G

systems, which provide a higher quality of service

with less complicated, i.e., cheaper, communication

systems for deployment and maintenance, so the

mobile operators are shutting down the 3G

networks. However, the UMTS communication

systems usually use RAKE receivers. The RAKE

receiver estimates the CIR, i.e., the phase shift and

attenuation of each finger of the RAKE receiver so

that the 5 MHz radio CSI in UMTS can be

estimated. The resolution of the time delay of the

channel state is related to the chip duration, i.e., the

frequency of the spreading code, which is

3.84 Mcps or 0.26 µs in the UMTS system, which

Table 2. Types of UMTS physical channels, and their applicability for CIR estimation.

Channel type

Channel name

Comments

Broadcast

primary common control physical channel (P-CCPCH)

secondary common control physical channel (S-CCPCH)

pilot bits for RSCP and Ec/No

acquisition indication channel (AICH)

no pilot bits

paging ch. (PCH)

no pilot bits

physical channel for the common packet ch. access

procedures

not relevant for CIR

physical downlink shared channel (PDSCH)

no pilot bits

synchronization channel (SCH)

relevant for synchronization

common pilot channel (CPICH)

help in estimation of CSI

Dedicated

dedicated physical control channel (DPCCH)

contain pilot bits for CSI estimation

dedicated physical data channel (DPDCH)

no pilot bits

n uplink PDCCH and PDDCH are I/Q

multiplexed

in downlink PDCCH and PDDCH are time

multiplexed

Random access

physical random access channel (PRACH)

only for uplink power adjustment

physical common packet channel (PCPCH)

random access in uplink, not relevant

WSEAS TRANSACTIONS on COMMUNICATIONS

DOI: 10.37394/23204.2022.21.30

Tomaž Javornik, Andrej Hrovat, Aleš Švigelj

E-ISSN: 2224-2864

255

Volume 21, 2022

corresponds to 78 m. The maximum delay spread in

the rural hilly environments for the UMTS channel

is expected to be 20 µs, [7]. The UMTS RAKE

receiver should cope with such excess delay, which

means the CIR is calculated within the receiver, but

unfortunately not available outside the receiver.

3.3 LTE - Long Term Evolution

Long-Term Evolution (LTE) is the evolution of

UMTS towards an all-IP broadband network. The

radio access part of LTE, Evolved Universal

Terrestrial Radio Access (E-UTRA), provides a

framework for increasing data rates and overall

system capacity, reducing latency, and improving

spectral efficiency and performance at cell edges,

[8]. It is designed to support IP connections. The BS

in LTE is called eNode B. E-UTRA applies

orthogonal frequency division multiplexing

(OFDM) in the downlink and both OFDMA and a

precoded version of OFDM called Single Carrier

Frequency-Division Multiple Access (SC-FDMA)

in the uplink. In both OFDM and SC-FDMA

transmission modes, a cyclic prefix is appended to

the transmitted symbols to handle multipath

propagation. LTE supports both frequency and time

division duplex transmission. The basic

communication element in LTE is a time-frequency

block called a resource block (RB). The duration of

the resource block is 0.5 ms. It consists of seven-

time symbols with a bandwidth of 180 kHz. Each

symbol contains 12 subcarriers with a bandwidth of

15 kHz. The resource element (RE) represents a

time-frequency unit with the duration of one symbol

and the bandwidth of one subcarrier. One symbol

occupying all 12 subcarriers is called RB column in

this text and thus consists of 12 REs. The resource

blocks are distributed over several frequency bands

with a bandwidth of 1.4 MHz, 3 MHz, 5 MHz,

10 MHz, 15 MHz and 20 MHz. In the time domain,

LTE transmission is structured in radio frames. Each

of these radio frames is 10 ms long and consists of

10 subframes of 1 ms each, i.e., two RBs. The frame

structure of the LTE system is drawn in Fig. 4.

In LTE and LTE advanced, there are multiple

physical downlink and uplink channels. The

physical channels, which carry data are called

shared channels, while the channels that relay the

control information are named control channels. The

physical shared channels can also carry control

information. In addition, there are channels to

support broadcast and multicast data, namely the

physical broadcast channel and the physical

multicast channel, and special physical channels to

transport important control information, namely the

physical control format indicator channel and the

physical hybrid automatic repeat request indicator

channel in the downlink, and the physical random

access channel in the uplink.

The training signals in LTE are not part of the

physical channels but are specified as reference

signals (RS) that are transmitted within selected

resource elements. In the downlink, there are several

reference signals associated with different downlink

physical channels. The signals broadcast to all users

are synchronization signals, cell-specific reference

signals, mobile broadcast multimedia service single

frequency network reference signals, positioning

reference signals, and CSI reference signals. The

user equipment specific or demodulation reference

signal is dedicated to a single user. In the uplink,

only three reference signals exist, namely, the

sounding reference signal, the demodulation

reference signal for the physical uplink shared

channel (PUSCH), and the demodulation reference

signal for the physical uplink control channel

(PUCCH). The LTE reference signals are listed in

Table 3.

Synchronization signals (SS) are broadcast signals

used to synchronize frames, subframes, slots, and

symbols in the time domain and to determine the

center of the channel bandwidth. SS is transmitted

twice in each frame. There is a primary SS, which is

transmitted on the 62 central subcarriers at the last

symbol in time slots 0 and 10, and a secondary SS,

which is transmitted on the 62 central subcarriers at

Fig. 4: The LTE frame structure.

WSEAS TRANSACTIONS on COMMUNICATIONS

DOI: 10.37394/23204.2022.21.30

Tomaž Javornik, Andrej Hrovat, Aleš Švigelj

E-ISSN: 2224-2864

256

Volume 21, 2022

the second to last symbol in time slots 0 and 10. The

SS can be applied for coarse CIR estimation in the

center of the transmission band, but the training

signals occupy only narrow bandwidth of 930 kHz.

Cell-specific reference signals are broadcast over

the entire cell. They are used to estimate the

Reference Signal Received Power (RSRP),

Reference Signal Received Quality (RSRQ), and

Channel Quality Indicator (CQI).

It corresponds to the Common Pilot (CPICH)

channel in UMTS. The cell-specific reference

signals are transmitted in subframes that support the

transmission of the physical downlink shared

channel. They can be applied for CSI estimation in

bandwidth or a single resource block of 180 kHz.

The multimedia broadcast multicast (MBM)

transmitted in subframes single frequency network

(SFN) (MBMSFN) reference signals are transmitted

in the resource blocks allocated for the MBM

service, which is today rarely used. Similarly, as for

cell-specific reference signals CIR estimation is

obtained for a single RB with a bandwidth of

180 kHz.

LTE provides indoor positioning capabilities. To

support this functionality the positioning reference

signals (PRS) are broadcast in the predefined

positioning subframes. The PRS signals are

transmitted over the entire frequency BS range and

are therefore useful for estimating CIR. In addition,

no data is transmitted in the resource blocks selected

to transmit the PRS signal from any eNode Bs in the

network.

The user equipment (UE) specific reference

signals also referred to as demodulation reference

signals are transmitted in the resource blocks

allocated for the PDSCH. They co-exist with the

cell-specific reference signals. They were

introduced primarily to support signal

demodulation, beamforming, and MIMO

transmission. They are transmitted while the eNode

B is connected to the UE.

Resource elements allocated for data transmission

often experience different interference from

resource elements allocated for cell-specific

reference signals. This difference is in particular

observed for the unloaded networks. In order to

improve this and enable MIMO and multiuser

MIMO transmission, the CSI reference signals are

specified in the LTE standard. They are transmitted

in the RB assigned for PDSCH. In order not to

interfere with data transmission the CSI reference

signals are transmitted in one subframe per 5, 10,

20, 30, or 80 frames. The CSI reference signals are

only transmitted when communication is active,

which limits the use of the CSI signals.

Sounding Reference Signal (SRS) is a reference

signal sent from the UE to the eNode B to estimate

the quality of the uplink channel over a wider

bandwidth. The eNode B can use this information

for uplink scheduling of data bits to specific

subcarriers. The eNode B can also use SRS for

timing estimation in the uplink, as part of timing

alignment, especially in situations where there are

no PUSCH/PUCCH transmissions in the uplink for

an extended period of time.

The demodulation reference signal associated

with the Physical Uplink Shared Channel (PUSCH)

is applied to support the demodulation of the uplink

transmission in the PUSCH. They occupy 12

subcarriers in the middle of the resource block so

that they can be used for channel estimation for

channels with a bandwidth of 180 kHz and when the

UE is connected to the eNode B.

There are several formats of Physical Uplink

Control Channel (PUCCH) reference signals,

namely 1, 1a, 1b, 2, 2a, 2b, and, 3. Three columns in

the time-frequency block are dedicated to the

reference signal in PUCCH formats 1, 1a, and 1b,

while there are only two for formats 2a, 2b, and 3.

Table 3. Types of LTE reference signals and their applicability for CSI estimation.

Signal type

LTE reference signals (RS)

Comments

Broadcast

synchronization signals (SS)

coarse CSI estimation, BW = 930 kHz

cell-specific RS

transmitted in subframes allocated for PDSCH, BW

= 180 kHz

multimedia broadcast multicast (MBM)

transmitted in subframes

single frequency network (SFN) RS

allocated for MBM service, BW = 180 kHz

positioning RS (PRS)

transmitted in the whole LTE band

Dedicated -

downlink

user equipment specific (demodulation) RS

transmitted when UE is connected

channel state information (CSI) RS

transmitted when PDSCH is transmitted

Dedicated - uplink

sounding RS

channel quality in uplink

demodulation RS with PUSCH

when UE is connected to eNode B

demodulation RS with PUCCH

when control information is transmitted

WSEAS TRANSACTIONS on COMMUNICATIONS

DOI: 10.37394/23204.2022.21.30

Tomaž Javornik, Andrej Hrovat, Aleš Švigelj

E-ISSN: 2224-2864

257

Volume 21, 2022

LTE and LTE advanced technologies offer

tremendous potential for monitoring CSI in multiple

frequency bands. However, with current

specifications, the information from CSI cannot be

retrieved by the system through standardized

application programming interfaces. However, the

basic concepts of LTE and LTE Advanced are

adopted by the next generation of mobile radio,

”New Radio”.

3.4 NR - New Radio Generation (5G)

5G wireless cellular technology was developed to

meet the requirements for higher data rates and

lower latency compared to existing 4G technology,

[9]. To achieve this, a scalable and flexible air

interface has been developed under 3GPP. Also, the

new spectrum is allocated in cm and mm frequency

bands in addition to the frequency bands below

6 GHz. The 5G specifications also include support

for massive multi-antenna and multi-beam antenna

systems. The radio access part of 5G systems is

referred to as new radio (NR) and specifies the time

and frequency division duplex transmission. A 5G

base station is referred to as a gNode B (next

generation Node B) to distinguish it from a 4G base

station (evolved Node B, eNode B).

NR uses OFDMA similar to LTE, but supports

many OFDM subcarrier spacings, namely 15 kHz,

30 kHz, 60 kHz, 120 kHz, and 240 kHz. Higher

subcarrier bandwidth results in shorter symbol

duration and consequently lower latency in

information transmission. The following symbol

durations correspond to the specified subcarrier

spacings 66.67 µs, 33.33 µs, 16.67 µs, 8.33 µs, and

4.17 µs. The transmissions from NR are organized

into time frames of 10 ms, consisting of 10

subframes of 1 ms. The subframes contain a

variable number of time slots. The number of time

slots per subframe depends on the cyclic prefix

(normal or extended) and the spacing of the

subcarriers. A slot consists of 14 OFDM symbols

for the normal cyclic prefix and 12 for the extended

cyclic prefix. The resource block in NR is specified

differently than in LTE, namely the resource block

consists of 12 subcarriers and only one symbol,

which means that the RB does not contain a timing

component. The frame structure of NR is shown in

Fig. 5.

The NR specifies a set of physical channels for

transmission in the downlink (Physical Broadcast

Channel (PBCH), Physical Downlink Shared

Channel (PDSCH), Physical Downlink Control

Channel (PDSCH)) and in the uplink (Physical

Uplink Access Channel, Physical Uplink Shared

Channel (PUSCH), Physical Uplink Control

Channel (PUCCH)).

Fig. 5: The NR frame.

Table 4. Types of NR reference signals and their applicability for CSI estimation.

Signal type

NR reference signals (RS)

Comments

Broadcast

synchronization signals (SS)

coarse CSI estimation, BW = 3.6 to 57.6 MHz

demodulation RS for physical broadcast channel (PBCH)

coarse CSI estimation, BW = 3.6 -57.6 MHz

channel state information (CSI) RS

complete NR frequency band

tracking RS

BW equal to RS bandwidth

phase tracking RS

BW equal to RS BW, used only above 6 GHz

Demodulation RS

at the physical uplink shared channel (PUSCH)

UE is connected to BS, BW equal to RS bandwidth

at the physical uplink control channel (PUCCH)

UE is connected to BS, BW equal to RS bandwidth

at physical downlink control channel (PDCCH)

UE is connected to BS, BW equal to RS bandwidth

at physical downlink shared channel (PDSCH)

UE is connected to BS, BW equal to RS bandwidth

Uplink/system

sounding RS

channel quality in uplink

phase tracking (PT) RS

UE is connected to BS

WSEAS TRANSACTIONS on COMMUNICATIONS

DOI: 10.37394/23204.2022.21.30

Tomaž Javornik, Andrej Hrovat, Aleš Švigelj

E-ISSN: 2224-2864

258

Volume 21, 2022

The NR also specifies several reference signals

used by the physical layer for synchronization,

channel quality measurement, and channel

estimation. Some of them are closely related to the

physical channels. The physical signals can be

classified as broadcast signals (synchronization

signals (SS), demodulation reference signals for the

broadcast physical channel, CSI reference signals,

phase tracking reference signals), demodulation

reference signals associated with physical channels,

namely PUSCH, PUCCH, PDSCH and PDCCH,

and uplink system reference signals, namely

sounding reference signal and uplink phase tracking

reference signals.

There are two synchronization signals, a primary

and a secondary. The primary synchronization

signal occupies 127 subcarriers in the first symbol

of the frame, while the secondary synchronization

signal is transmitted in 127 subcarriers in the third

symbol. The 2nd and 4th symbols of the frame and

the unoccupied subcarriers in the 3rd symbol are

assigned to the physical broadcast channel. BPSK

modulation is used for the synchronization symbols.

In addition to synchronizing the UE with the base

station, PSS and SSS are also used to identify the

physical cell identity (PCI), to perform RSRP,

RSRQ, and SNIR measurements, and as an

additional demodulation reference signal for the

PBCH. The bandwidth occupied for the SS depends

on the subcarrier spacing and therefore varies

between 3.6 MHz and 57.6 MHz. The SS can be

applied for an initial CSI estimate in NR.

The demodulation reference signal (DMRS) for

PBCH is inserted into the resource blocks to support

the decoding of the information transmitted in the

PBCH. The PBCH is transmitted in the 48 RBs in

the synchronization signal/PBCH block. 3 DMRS

signals are inserted in each resource block, so there

are 3*48 = 144 reference signals. The DMRS

symbols are generated with a pseudo-noise

generator whose seed depends on PCI. The DMRS

is QPSK modulated and occupies a bandwidth

between 3.6 MHz and 57.6 MHz

The CSI reference signal (CSI-RS) is not a new

concept in NR, it has already been used in LTE but

is more flexible in NR. The transmission of CSI-RS

is limited to the entire bandwidth of the NR, but not

only to the transmissions in the part of the spectrum

where the data is transmitted. The gNode B is now

aware of the CSI for the complete NR bandwidth. In

the frequency domain, the CSI-RS can be

transmitted in all resource blocks or every two

resource blocks. In the time domain, CSI-RS

transmission can be periodic, aperiodic, and semi-

persistent. The periodic CSI-RS can be transmitted

in the Nth time slot, where N is between 4 and 640.

Semi-persistent transmission is similar to periodic

transmission, but transmission can be temporarily

turned off. In aperiodic mode, transmission is on

request from the upper layers. The CSI-RS

transmission can be non-zero power or zero power.

Transmission with zero power is announced by

gNode B to UE, i.e., no signal is transmitted at the

specified RB. Consequently, UE can estimate the

noise and interference power. The nonzero

transmission is mainly used to estimate the channel

characteristics. The NR supports the CSI estimation

in the whole NR band, even if there is no data

transmission.

A tracking reference signal (TRS) is a downlink

synchronization signal used by the UE to track time

and frequency variations of the carrier frequency

and frame timing of BS. The synchronization signal

allows course tracking and synchronization, while

the tracking reference signal allows high-precision

synchronization and tracking. The tracking

reference signal is a CSI-RS with special

parameters. The assignment of the resource block to

TRS RS depends on the configuration of TRS.

The phase tracking reference signal is inserted

into RB to cope with phase noise, which is an

important degradation at higher frequencies. It can

be applied for CSI estimation only at higher

frequency bands.

Demodulation Reference Signal (DMRS) for the

physical downlink control channel is inserted into

the physical downlink control channel to estimate

the propagation channel characteristics of the

PDCCH. The DMRS occupies 25% of the resource

blocks allocated to the PDCCH, i.e., every fourth

subcarrier in the symbol is allocated to the DMRS.

The allocation of the DMRS is fixed and does not

depend on other network planning parameters.

DMRS for the physical downlink shared channel

is always associated with PDSCH. They are used to

assist in the demodulation of PDSCH symbols. The

resource elements assigned to DMRS are flexible

and can be controlled by a set of parameters, which

control the number of DMRS in a frequency and

time domain. The parameters are determined based

on the frequency and time of the channel. The CSI

estimation of DMRS is limited to the RB bandwidth

The NR uses multiple uplink reference signals to

assist the BS in signal demodulation, channel

WSEAS TRANSACTIONS on COMMUNICATIONS

DOI: 10.37394/23204.2022.21.30

Tomaž Javornik, Andrej Hrovat, Aleš Švigelj

E-ISSN: 2224-2864

259

Volume 21, 2022

estimation, and supporting measurements. The

demodulation reference signals for PUCCH and

PUSCH aid in signal demodulation. The sounding

reference signal (SRS) is aimed to assist in channel

estimation and measurements. The phase tracking

reference signals (PT-RS) are used for phase noise

compensation.

Transmission of the DMRS associated with

PUCCH enables coherent detection. Several formats

of PUCCH are specified, and DMRS transmission

depends on the PUCCH format. DMRS can be

transmitted in the whole resource block, but every

second symbol is allocated to PUCCH, or every

symbol but only on every third subcarrier, or only

on the specific symbols assigned to PUCCH. The

CSI can be estimated only during active

transmission.

The PUSCH is always combined with the DMRS.

The DMRS allocation depends on the allocation of

the PUSCH. The DMRS can be transmitted in every

second sub-carrier frequency in each time symbol or

at the beginning or middle of the time slot.

Additional DMRS can be added for fast channel

variation in time or frequency.

The sounding reference signal (SRS) is used to

measure the characteristics of the uplink channel. It

is transmitted on the BS request. The SRS can

occupy 1, 2, or four symbols in the time domain and

can be transmitted in up to 272 resource blocks.

Within the resource block, only a part of the

subcarriers can be allocated for the SRS, namely the

second or the fourth subcarrier.

The phase tracking reference signal (PTRS) is

applied to compensate frequency and phase offset of

the UE. It is transmitted in RB allocated for the

PUSCH. The phase noise has no significant effect at

carrier frequencies below 6 GHz, but at higher

frequencies, the frequency shift must be

compensated. The PTRS is often applied to reduce

the amount of DMRS transmitted with PUSCH. The

PTRS is transmitted during communication.

The 5G standard brings an important innovation

to CSI estimation, namely the ability to estimate

CIR in the time/frequency blocks where no data is

being transmitted. This is enabled by transmitting

CSI non-zero power reference signal in blocks not

reserved for information transmission. This feature

opens up the possibility of using the 5G system to

obtain global knowledge about radio channels and

spectrum occupancy.

4 Wireless Local Area Networks

(WLANs)

Wireless Fidelity (Wi-Fi) is a low-cost wireless

access technology that allows many electronic

devices to connect to the Internet. According to the

Wi-Fi Alliance, Wi-Fi devices are any Wireless

Local Area Network (WLAN) product based on the

Institute of Electrical and Electronics Engineers

(IEEE) 802.11 standards, [10]. The IEEE 802

standard refers to a family of IEEE standards that

address Local Area and Metropolitan Area

Networks. The IEEE 802.11 family of standards is a

subset of IEEE 802 standards that deal with Medium

Access Control (MAC) and Physical Layer (PHY)

specifications. The 802.11 specifies the number of

over-the-air modulation schemes that use the same

basic protocol. The first standard in this series is

IEEE 802.11.-1997. It specified three alternative

physical layer technologies: diffuse infrared, which

supports 1 Mbit/s, frequency hopping spread

spectrum, which supports 1 Mbit/s or 2 Mbit/s and

direct sequence spread spectrum, which supports 1

Mbit/s or 2 Mbit/s in the 2.4 GHz ISM frequency

band. Offering too many technology options

prevented the widespread adoption of this

technology. This was improved in the next version

of the standard, IEEE 802.11b, which offers

11 Mbit/s peak data rates using 22 MHz channels in

the 2.4 GHz frequency band with the same media

access technologies. The standard uses

complementary code keying (CCK) modulation, a

direct spread spectrum modulation that encodes 4 or

8 bits in 8 QPSK symbols. The next version of the

standard, IEEE 802.11a, allows the use of the 5 GHz

frequency band to avoid interference with other

systems in the 2.4 GHz band ISM, and introduces

OFDM modulation, which allows flexible use of the

20 MHz frequency band allocated per channel. The

OFDM uses 64 IFFT. The subcarrier bandwidth is

315.2 kHz. The lowest 6 and highest 6 subcarriers

are not in use but provide a frequency guard

interval. The remaining 52 subcarriers are used for

data transmission (48 subcarriers) and four of them,

namely 21, -7, 7, and 21, are applied for channel

estimation. The symbol duration is 3.2 µs, but 0.8 µs

guard interval or cyclic prefix is added to handle

multipath propagation. This results in a symbol

duration of 4.0 µs. The IEEE 802.11g standard also

extends the use of OFDM to the 2.4 GHz ISM band.

The OFDM burst with guard bands is illustrated in

Fig. 6.

WSEAS TRANSACTIONS on COMMUNICATIONS

DOI: 10.37394/23204.2022.21.30

Tomaž Javornik, Andrej Hrovat, Aleš Švigelj

E-ISSN: 2224-2864

260

Volume 21, 2022

Fig. 6: The Wi-Fi OFDM burst.

The 802.11n amendment includes many

improvements that increase WLAN range,

reliability, and throughput, including new

modulation schemes, multiple antennas at the

transmitter and receiver, and transmission in the

40 MHz bandwidth. The data rate is approaching

600Mbps. The standard supports the 2.4 GHz and

5 GHz frequency bands. Other enhancements to the

IEEE 802.11ac standard include the use of 80 and

160 MHz bands to support more antennas at the

transmitter and receiver, and specification of

modulation schemes with high bandwidth efficiency

such as 256 QAM. Pilots are transmitted in

additional subchannels, e.g., at a bandwidth of

40 MHz, the following subcarriers are occupied by

pilots: -53, -25, -11, 11, 25, 53. In the time domain,

transmissions are organized in physical frames. The

physical frame of the IEEE standard consists of a

preamble and a payload. The preamble consists of a

short training field (STF) with two symbols. The

STF occupies 25% of the subcarriers. It is used for

the initial time and frequency synchronization. The

next part of the preamble is the long training field

(LTF), which consists of 2 symbols. The LTF

occupies all 52 subcarriers and is used for fine time

and frequency synchronization and CIR estimation.

This is followed by the signal part, which consists of

a symbol containing information about the payload,

namely the data rate, i.e. the coding, the length of

the payload, the parity information, and information

about the tail bits. In the standards that support

higher transmission rates, the additional information

is transmitted in the preamble, which increases the

length of the preamble. The duration of the symbol

in the preamble varies, from 192 µs for CCK

transmission to 20 µs for OFDM transmission

according to the IEEE 802.11g standard. The

physical layer frame format for IEEE 802.11g is

shown in Fig. 7.

The latest Wi-Fi standards support broad

bandwidth and allocation of training sig that can be

used for CIR estimation. Compared to the cellular

system, the Wi-Fi chipset produces allows access to

some radio parameters of the technology through

well-specified APIs, making this wireless

technology a good choice for experiments with CSI

estimation at different frequency bands.

5 Wireless Sensor Networks (WSNs)

Recently, wireless sensor networks become very

popular to collect information for a huge number of

sensors. Sensors spread over a wide area make the

technology very convenient for channel estimation

in spite of the wireless sensor technologies usually

do not support broadband communications

occupying a wide frequency band. Among wireless

sensor technologies, those based on IEEE 802.15.4

and IEEE 802.15.4a standards are often in use.

5.1 IEEE 802.15.4 Standard

The IEEE 802.15.4 standard, [11], specifies the

physical layer and media access control for low-rate

wireless Personal Area Networks (PANs). The

MAC and PHY layers are complemented by various

technologies such as ZigBee or RPL (Routing

Protocol for Low-Power and Lossy Networks). The

devices, which comply with the IEEE 801.15.4

standard, operate in several frequency bands: ISM

2.4 GHz band with sixteen 5 MHz channels,

915 MHz frequency band with ten 2 MHz channels,

and 868 MHz frequency band with only one

0.6 MHz channel. The packet data unit (PDU) of the

physical layer contains a synchronization header,

i.e. a preamble and start of packet delimiter, a PHY

header to specify the packet length, and the payload

data. The 32-bit preamble is designed for the

acquisition of symbol and chip timing and may be

used for coarse frequency adjustment. The physical

layer supports two methods for estimating channel

quality, namely energy detection and link quality

indication. Energy detection is an estimate of the

received signal power that is primarily used for

channel assessment. Link quality indication is

measured for each received packet, taking into

account the detected packet energy and the signal-

to-noise ratio. The standard does not provide

sufficient means for CSI estimation.

Fig. 7: IEEE 802.11g time frame.

WSEAS TRANSACTIONS on COMMUNICATIONS

DOI: 10.37394/23204.2022.21.30

Tomaž Javornik, Andrej Hrovat, Aleš Švigelj

E-ISSN: 2224-2864

261

Volume 21, 2022

The IEEE 802.15.4-2011 standard has evolved to

the IEEE 802.15.4e standard to meet the QoS

requirements of industrial communications. The

new MAC behaviors are significantly different from

those considered in IEEE 802.15.4-2011. It

introduces Time Slotted Channel Hopping (TSCH),

which uses fixed-size TDMA slots and multichannel

hopping to provide the ability to monitor CSI across

the 2.4 GHz ISM frequency band by sending a

continuous wave of constant frequency signal as a

payload on different frequency channels.

5.2 IEEE 802.15.4a Standard

The original 802.15.4 standard cannot meet the

requirements of the various applications intended

for wireless PANs, especially in terms of

transmission rate and precise localization.

Therefore, the standard has been extended to include

ultra-wideband (UWB) technology, which operates

in the unlicensed UWB spectrum, [12]. In addition,

the standard specifies a wireless technology to

support longrange and reliable communications, but

this specification has not yet found many

implementations in silicon. Therefore, we will focus

on the UWB part of the standard, which is currently

the only on-chip technology that provides a

documented API to obtain the CSI in the wide

frequency band.

Three ISM UWB frequency bands are available,

namely the sub-gigahertz band (0.250 to

0.750 GHz), the low band (3.244 to 4.742 GHz),

and the high band (5.944 to 10.234 GHz). Within

these bands, 16 channels are specified to support

different data rates, namely 110 kbits/s, 851 kbits/s,

6.81 Mbits/s, and 27.24 Mbits/s. Channel 0 is

allocated in the sub-gigahertz band, channels 1-4 in

the low band, and the remaining channels in the

high band. A compliant device must operate in at

least one of the mandatory channels 0, 3, and 9. The

channel bandwidth is approximately 500 MHz. The

range of UWB devices is between 10 and 100 m and

the channel access method is CSMA/CA or Aloha

The packet format of the IEEE 802.15.4a is

shown in Fig. 8. The Physical Layer Protocol Data

Unit (PPDU) or packet contains a preamble, the

Start of Frame Delimiter (SFD), the Physical Header

(PHR), and the Physical Service Data Unit (PSDU).

The preamble is used for synchronization, frame

timing, and CSI estimation. The preamble is

transmitted at a nominal data rate of 110 kbits/s. The

length of the preamble varies from 16 to 4096. The

PHR transmits information for decoding the

Physical Service Data Unit (PSDU). The length of

the PSDU varies from 0 to 1209 symbols. Several

data rates are supported, namely 0.11, 0.85, 6.81, or

27.24 Mbits/s.

Burst position modulation (BPM) and binary

phase shift keying (BPSK) are used to encode the

transmitted information. The chip or pulse duration

is about 2 ns, which corresponds to a bandwidth of

500 MHz. The CIR is estimated using the

synchronization preamble. The synchronization

preamble consists of 16, 64, 1024, or 4096 symbols.

Each symbol of the preamble contains a ternary

preamble code -1,0,1 with a length of 31 or

optionally 127. The code with a length of 31 is

distributed with 16 chips, which duration is

approximately 2 ns. The code is transmitted only in

the first chip. Similarly, the code with a length of

127 is spread over 4 chips. The IEEE 802.15.4a

preamble is plotted in Fig- 9. This preamble allows

the CIR to be estimated for excess delays below the

duration of a symbol, i.e., about 1 µs. For example,

the Decawave chipset DW1000 calculates the 1016

complex samples CIR with a sampling rate of about

1 ns, which corresponds to the delay caused by a

path difference between radio rays of 30 cm.

Fig. 8. IEEE 802.15.4a packet format.

Fig. 9: IEEE 802.15.4a preamble.

WSEAS TRANSACTIONS on COMMUNICATIONS

DOI: 10.37394/23204.2022.21.30

Tomaž Javornik, Andrej Hrovat, Aleš Švigelj

E-ISSN: 2224-2864

262

Volume 21, 2022

6 Impact of Channel Variation Rate

on the Estimated CSI

In communication systems, the time over which the

wireless channel is expected to be static is called the

coherence time Tc. The coherence time is related to

the Doppler spread, which is the result of the narrow

band signal spectrum widening due to multipath

propagation and the relative motion of the

transmitter and receiver. The Doppler spread is

represented as a Doppler spectrum bounded by the

maximum Doppler frequency shift fm =, where c

is the speed of light, v is the terminal velocity, and fc

is the carrier frequency. A simple approximation for

the coherence time is

This approximation indicates a time interval in

which a Rayleigh fading signal can vary

significantly. In reality, however, the CIR is random,

so the coherence time should be considered a

statistical measure.

In other words, the coherence time can be

considered as a period of time during which the CIR

is essentially invariant, i.e., the symbols transmitted

during this period experience the same channel

impairments. The coherence time can be defined as

an interval in which the channel impulse

autocorrelation function is above 0.5, [13].

This definition is very restrictive so in practice, the

geometric mean of both definitions is usually in use

Coherence time for some selected LTE and NR

bands and some typical user terminal speeds are

shown in Table 5. To ensure connectivity, the

channel coherence time must be longer than the

symbol or subframe length. The symbol time in

LTE systems is 0.06667 ms, while the subframe

length is 1.0 ms. In the LTE bands, the requirements

are met up to the speed of the vehicle on the

highways, while the LTE communication for fast

trains may be faulty in highly scattered

environments. The communications operating in the

mm frequency band may exhibit problems even at

lower user terminal speeds. From Table 5, it can be

seen that the channel characteristic is not constant

for the entire duration of the subframe, which lasts

1 ms when the user terminal is in motion. In other

words: only indoor communication may be based on

the assumption that the radio channel does not

significantly vary with time. The problem is less

pronounced at NR, where symbols are even shorter

at high-frequency bands, namely, 0.03333 ms,

01667 ms, 0.00833 ms, and 0.00417 ms, and the

basic transmission element is the resource block,

whose duration is only a single symbol. For Wi-Fi,

the symbol duration is 0.004 ms, but the packets can

be longer and reach 10 ms, leading to the conclusion

that Wi-Fi can be used to estimate the channel to

walking speed, Table 6 Similar conclusions can be

drawn for the various wireless sensor network

technologies.

Table 5. Coherence time in ms for some LTE and

NR bands and speed of user equipment.

[GHz]

km/h

120km/h

240

km/h

0.7

163.20

10.88

5.44

2.72

0.9

126.93

8.46

4.23

2.12

1.8

63.47

4.23

2.12

1.06

2.1

54.40

3.63

1.81

0.91

2.6

43.94

2.93

1.46

0.73

3.6

31.73

2.12

1.06

0.53

5.9

19.36

1.29

0.65

0.32

28.0

4.08

0.27

0.14

0.07

47.0

2.43

0.16

0.08

0.04

7 Discussion

To benefit from the estimated CSI, the information

must be stored in an intelligent database. The CSI

can be represented as CIR h(t, τ) or channel transfer

functions H(t, f) and CIR-derived parameters such

as the attenuation or gain of the channel, coherence

bandwidth, coherence time, interference from other

systems, i.e. Furthermore, the information about the

carrier frequency, the channel bandwidth, and

measurement system must be stored for estimating

the CSI. We are concerned with propagation

characteristics that change slowly and therefore can

be stored in a database for further use. This category

includes indoor wireless channels, where the

terminal is static or moving at low speed, and

outdoor channels, but only considering shadowing

effects and not the effects of local scattering, which

can change when the terminal slightly changes its

position or orientation.

WSEAS TRANSACTIONS on COMMUNICATIONS

DOI: 10.37394/23204.2022.21.30

Tomaž Javornik, Andrej Hrovat, Aleš Švigelj

E-ISSN: 2224-2864

263

Volume 21, 2022

Table 6. Coherence time in ms for WiFi

technology.

[GHz]

km/h

120

km/h

240

km/h

2.4

47.62

3,17

1.59

0.79

5.0

22,86

1.52

0.76

0.38

6.0

19.05

1.27

0.63

0.32

60.0

1.90

0.13

0.06

0.03

Currently, only chipsets that obey IEEE 802.15.4

standards provide raw information about the CIR,

but we expect this information to be available for

other emerging wireless technologies. The channel

representation of the DW1000 chipset contains 1016

samples of real and imaginary channel coefficients

represented by 16 bits. The sampling time is 1 ns,

which corresponds to a channel bandwidth of

500 MHz. The frequency resolution is thus

MHz.

Similarly, we can calculate the CIR for the LTE

and NR, where the ∆f = 15 kHz. The bandwidth, the

number of resource blocks, the subcarrier

bandwidth, the number of samples to represent the

CIR, the sampling time, and the maximum excess

delay are given in Table 7. The parameters given for

LTE differ significantly from those of the UWB

system due to the large differences in system

bandwidths. The sampling intervals, column 6, are

longer and, as expected, they decrease with

bandwidth.

The LTE parameters are similar to the WiFi

parameters. WiFi exploited 52OFDM symbols to

estimate the CIR. The OFDM symbol occupies a

bandwidth of 313.2 kHz, giving an effective

bandwidth of about 16 MHz and a sampling interval

of 61.5 ns. The maximum excess delay covered

corresponds to the maximum excess delay expected

in LTE and LTE Advanced systems.

Contemporary communication systems are

equipped with multiple antennas.

To consider this aspect as well, the CIR should be

estimated for each pair of antennas between the

transmitter and receiver. NR and LTE support the

estimation of CSI per pair of antennas by

introducing the concept of antenna ports and

defining the training signal in resource elements

such that they do not interfere with each other.

Due to noise and interference, the communication

systems provide the average value of the estimated

CSI. To obtain complete information about CSI, the

CSI statistical parameters should be estimated such

us the probability density function or at least the

standard deviation of the probability density

function. Unfortunately, at the current stage of CSI

estimation, no wireless communication system

provides this data.

The parameters to be stored in the database must

be identified. The wireless channel varies due to

random variations in the environment around the

transmitter and the receiver, the time variation of the

interference, and the random movement and rotation

of the transmitter and receiver. The information

stored in the database consists of parameters that do

not change significantly in time, such as the mean

and standard deviation of the received signal

strength, the CIR, etc. We have identified the

following parameters to be stored in the database:

• The power received from a particular base

station, which is in the 4G and 5G

communication system called Reference Signal

Received Power (RSRP). In systems with

limited channel estimation functionality, the

parameter is rarely available.

Table 7. LTE system: bandwidth, number of resource blocks, subcarrier bandwidth, number of samples for

channel impulse representation, sampling time, and maximum excess delay.

Bandwidth

MHz

Number of

resource blocks

Subcarrier

bandwidth

MHz

Effective bandwidth

MHz

Number of

samples for

h(τ)

Sampling interval

Maximum excess

delay

µs

1.4

1.08

925.93

66.67

3.0

2.70

180

370.37

66.67

5.0

4.50

300

222.22

66.67

10.0

9.00

600

111.11

66.67

15.0

13.50

900

74.07

66.67

20.0

100

18.00

1200

55.56

66.67

50.0

250

45.00

3000

22.22

66.67

WSEAS TRANSACTIONS on COMMUNICATIONS

DOI: 10.37394/23204.2022.21.30

Tomaž Javornik, Andrej Hrovat, Aleš Švigelj

E-ISSN: 2224-2864

264

Volume 21, 2022

• Received signal strength, which is composed of

the received interference power, and noise in

addition to the received power of the respective

base station. In many systems, the parameter is

referred to as the Received Signal Strength

Indicator (RSSI). Due to its simple estimation,

the RSSI is accessible as a parameter in almost

all wireless communication systems.

• Carrier frequency, the parameter is known or

can be estimated from wireless communication

system parameters.

• Channel bandwidth, the parameter is known

from the wireless standard or, if configurable, it

can be estimated from the beacons, broadcast

from the base station.

• Channel impulse response (CIR) which

includes:

 imaginary and real part of the CIR

coefficients,

 sampling interval, i.e. time between

samples.

In early wireless communication systems, CIR is

not estimated. In fact, for narrowband wireless

communication systems, the attenuation of the

wireless channel is sufficient for the operation of the

system. In today’s wireless communication systems,

the CIR is estimated for the bands with an active

connection. However, NR supports the estimation of

CSI even for bands not occupied by an active

connection.

In addition to the above parameters, some

parameters related to the rate of channel variation

would be beneficial, e.g., channel coherence time,

Doppler frequency spread, and channel coherence

bandwidth.

7 Conclusions

The environment-aware wireless communications

exploit the stored information about the CIR in local

or global databases in order to decrease the number

of training symbols in the transmission. However,

this information has to be obtained from previous

transmissions. In this respect, we critically review

the potential of today’s terrestrial wireless

communication systems including wireless cellular

technologies (GSM, UMTS, LTE, NR), wireless

local area networks (WLANs), and wireless sensor

networks (WSNs), for estimating CSI in term of

CIR, wireless channel power delay profile, the ratio

between training and information symbols and the

rate of channel variation, and the potential use of

time invariable CSI in environment aware wireless

communications. While early wireless

communication systems provide means for

estimating narrowband wireless channel parameters

such as channel attenuation, the next generation,

including 2G, 3G, 4G, Wi-Fi, and some wireless

sensor network technologies, is capable of

estimating broadband wireless channel parameters.

The estimated CSI, except for some basic ones such

as RSRP, and RSSI, is not available through defined

and publicly available APIs, and the CSI is

measured only for wireless channels with active

communication. The NR provides the possibility to

measure CSI also in wireless channels without

active communication. Only UWB wireless

technology, based on the IEEE 802.15.4 (2011)

standard, provides a good experimental platform for

observing CSI and CIR and performing some

combined sensing and communication experiments.

The contributions of this paper are (i) the survey

of the CSI estimation for the existing terrestrial

wireless technologies, (ii) the study of the existing

terrestrial wireless cellular technologies for

environment-aware wireless communications, (iii)

the study of the existing wireless sensor networks,

and wireless local area network technologies for

environment-aware wireless communications, and

(iv) the guidelines for future wireless technologies

to be suitable for the environment-aware wireless

communications.

Acknowledgement:

This research was funded by Slovenian Research

Agency under grant no. P2-0016 and grant no. J2-

2507.

References

[1] Osmocom-Analog, Nordic mobile telephone,

[Accessed 17-June-2022]. [Online],

Available:

https://en.wikipedia.org/wiki/Twisted_pair.

[2] S. Redl, S. Redl, M. Weber, and M. Oliphant,

An Introduction to GSM, Artech House, 1995.

[3] M. Pesko, T. Javornik, A. Košir, M. Štular,

and M. Mohorčič, Radio environment maps:

The survey of construction methods, KSII

Transactions on Internet and Information

Systems, vol. 8, no. 11, pp. 3789–3809,

November 2014.

WSEAS TRANSACTIONS on COMMUNICATIONS

DOI: 10.37394/23204.2022.21.30

Tomaž Javornik, Andrej Hrovat, Aleš Švigelj

E-ISSN: 2224-2864

265

Volume 21, 2022

[4] Wikipedia. Channel state information,

[Accessed 25-October-2022]. [Online].

Available:

https://en.wikipedia.org/wiki/Channel_state_i

nformation

[5] Molisch, A., Wireless Communications, 2nd

ed. John Wiley and Sons, 2012.

[6] ETSI European Telecommunications

Standards Institute, “Universal Mobile

Telecommunications System (UMTS);

Physical channels and mapping of transport

channels onto physical channels (FDD),”

European Telecommunications Standards

Institute, Sophia Antipolis, France, Standard

3GPP TS 25.211 version 12.1.0 Release 12,

2015.

[7] ETSI European Telecommunications

Standards Institute, Universal Mobile

Telecommunications System (UMTS),

Deployment aspects (3GPP TR 25.943

version 9.0.0 Release 9, European

Telecommunications Standards Institute,

Sophia Antipolis, France, Standard 3GPP TR

25.943 version 9.0.0 Release 9, 2015.

[8] C. Johnson, Long Term Evolution in Bullets,

2nd ed. CreateSpace Independent Publishing

Platform, Jul. 2012.

[9] C. Johnson, 5G New Radio in Bullets.

Independently Published, 2019.

[10] M. S. Gast and M. Loukides, 802.11 Wireless

Networks: The Definitive Guide, 2nd ed.

USA: O’Reilly & Associates, Inc., 2005.

[11] H. Kurunathan, R. Severino, A. Koubaa, and

E. Tovar, IEEE 802.15.4e in a Nutshell:

Survey and Performance Evaluation, IEEE

Communications Surveys & Tutorials, vol.

20, no. 3, pp. 1989–2010, 2018.

[12] E. Karapistoli, F.-N. Pavlidou, I.

Gragopoulos, and I. Tsetsinas, An overview

of the IEEE 802.15.4a Standard, IEEE

Communications Magazine, vol. 48, no. 1, pp.

47–53, Jan. 2010.

[13] T. S. Rappaport et al., Wireless

Communications: Principles and Practice,

Prentice Hall PTR New Jersey, 1996, vol. 2.

Contribution of Individual Authors to the

Creation of a Scientific Article (Ghostwriting

Policy)

Tomaž Javornik, wrote initial draft, run formal

analyzing and investigation.

Andrej Hrovat, provided the initial idea, formulized

and evaluated of overarching research goals and

aims.

Aleš Švigelj, reviewed the draft, lead the project and

acquired the financial support for the project.

Sources of Funding for Research Presented in a

Scientific Article or Scientific Article Itself

This research was funded by Slovenian Research

Agency under grant no. P2-0016 and grant no. J2-

2507.

Creative Commons Attribution License 4.0

(Attribution 4.0 International, CC BY 4.0)

This article is published under the terms of the

Creative Commons Attribution License 4.0

https://creativecommons.org/licenses/by/4.0/deed.en

_US

WSEAS TRANSACTIONS on COMMUNICATIONS

DOI: 10.37394/23204.2022.21.30

Tomaž Javornik, Andrej Hrovat, Aleš Švigelj

E-ISSN: 2224-2864

266

Volume 21, 2022