Wireless Personal Communications

https://doi.org/10.1007/s11277-020-07629-z

A Novel Adaptive Optimized Fast Blind Channel Estimation

for Cyclic Prefix Assisted Space–Time Block Coded
MIMO‑OFDM Systems

Shovon Nandi1 · Narendra Nath Pathak2 · Arnab Nandi3

© Springer Science+Business Media, LLC, part of Springer Nature 2020

Abstract
In this paper an adaptive optimized fast blind channel estimation using cyclic prefix sup-
ported with Space Time Block Coded Multiple Input Multiple Output Orthogonal Fre-
quency Division Multiplexing (STBC-MIMO-OFDM) system is presented. The main
aspire of our technique is to support multiple users at the same time over same frequency
band based on the Multi-Carrier Code-Division Multiple Access (MC-CDMA) approach.
High complexity and low convergence is the main obstacle in earlier blind channel estima-
tion techniques. Modified flower pollination algorithm is implemented to overcome this
problem. The MC-CDMA approach is utilized to implement the blind channel estimation.
The proposed MC-CDMA is used to reduce the error rate included in the Blind Chan-
nel Estimation. As a part of wireless communications, time block coding technique is uti-
lized to transmit several copies of information across the number of antennas. To develop
the consistency of data transfer different received data is used and then MFPA results in
lower fuel cost compared to FPA. MFPA produces better results compared with previous
methods.

Keywords  MC-CDMA · OFDM · PAPR · PCC · FDCE · PTS

1 Introduction

OFDM is mostly utilized in division of several new emerged broadband wireless and
wired communication systems, for the implementation of high transmission rate and low
complexity over fading channels [1]. In general OFDM consist of various channels, there
are many narrow-band sub channels, and it is transmitted through parallel transmissions
to keep up the high data transmission rate and to expand the duration of symbol to bat-
tle ISI [2]. The main reason behind the OFDM frame is that it has an ability to provide

* Shovon Nandi
[email protected]
1
ECE Department, Bengal Institute of Technology, Kolkata, India
2
ECE Department, Dr. B. C. Roy Engineering College, Durgapur, India
3
ECE Department, National Institute of Technology, Silchar, India

13
Vol.:(0123456789)
 S. Nandi et al.

good performance in multipath channels with significant time delay spreads [3]. When the
OFDM with a long cyclic prefix, it can be convert the frequency selective channel into
multiple independent frequency flat sub channels without a small bit error rate [4]. The
promising strategy is space–time block code (STBC), which is used to give diversity gain
with the help of various transmit antennas, particularly when the diversity is received. It is
too expensive to deploy and uses a simple linear decoder at the end [5].
While transmitting the channels through parallel way the STBC gives the characteristics
to enhance the system performance in wireless communications [6]. The advantage influ-
ences that STBC is combined with OFDM as a space-to time block code-Orthogonal fre-
quency division multiplexing (STBC-OFDM) for different wireless applications [7]. In the
broadband system, the STBC is incorporated with OFDM with the use of STBC at a block
level structure basically known as ST-OFDM [8]. The main drawback in STBC–OFDM
system is large PAPR and long cyclic prefix (CP). The high-control amplifier has no delay
with the help of PAPR hence it will increase the system cost [9]. The PAPR of transmitting
signals on different sorts of antennas are reduced by using efficient technique called clip-
ping, it produces clipping noise and reduces the performance of the system [10].
Another problem is Long CP incorporated with the information of OFDM image to
reduce the ISI caused by various channels [5]. The applications and CP results of trans-
mission efficiency is limited in this technique, which is the major problem of Long CP in
OFDM because it is larger than the CIR [11]. The motivation behind the CP is to distribute
the ISI among dynamic OFDM images. It will change the linear convolution into a cyclic
convolution of the channel reaction with transmitted signal [12]. Two issues which are
introduced by the ISI and inter carrier interference (ICI), efficiency is reduced in the sys-
tem [13]. Because of the low transmission speed produced by CP, an iterative cancelation
strategy called as Remaining ISI Cancelation (RISIC) is used to manage this problem [14].
Conventional techniques consist of tail cancelation, cyclic restoration and removing
the interference because of insufficient CP [15]. The CP acquires in both power and spec-
tral overheads, which can be estimated by extent the length of the CP up to information
block length. Hence to limit this transparency a twofold channel mechanism is required
with high channel memory and short coherence time [16]. To overcome the issue of ISI
and ICI because of channel variation, different techniques was utilized such as, Polyno-
mial Cancelation Coding (PCC), Coordinated Isolates, Time-are Sifting, Taylor Game Plan
Advancement And Minimum Mean-Squared Error (MMSE) [17]. In this research, win-
dowing strategies and some progressive technologies were utilized to dispose the ISI and
ICI in STBC-OFDM system [18]. Also different Frequency-Domain Channel Estimation
(FDCE) techniques are utilized to track and calculate the wireless channels of moderate
blurring with or without channel statistic knowledge [19].
In the paper the recent research works done in blind channel estimation system which is
explained in part 2. The proposed methodology and its explanations were given in Sect. 3.
The simulation results and the performance of the proposed method is contrasted using
other traditional methods are given in Sect. 4 followed by conclusion in Sect. 5.

2 Related Works

Mahmudul et al. [20] exhibited another reduction strategy named as PAPR in light of Lin-
ear Predictive Coding (LPC). The critical issue of OFDM is great nonlinear deformation
in real hardware executions of the amplifier in extreme power due to the highest PAPR.

13
A Novel Adaptive Optimized Fast Blind Channel Estimation for…

Signal whitening property of LPC is used in this method as a pre processing step in OFDM
systems.Predictable content of for PAPR problem in OFDM stationary stochastic processes
is removed by error filtering and the input data cycle interaction is reduced. Based on linear
prediction coding it effectively reduces PAPR and also easy and efficient method which
provides good BER performance. Proper trade-off should be considered between PAPR
and BER performances to determine the optimum order of the error filter.
Prieto et al. [21] presented a novel approach to increase the combined channel and FO
estimation in MIMO-OFDM systems named as IJEP in the existence of ISI–ICI due to
highly disruptive channels in the preamble. Using SS as preamble it is shown that, in the
distorted part estimation can cancel the existing interference and obtaining a primary FO
and channel in non-distorted part. Interference cancellation process in the preamble drives
to new estimators computed with the whole preamble has been proved. Converging to the
theoretical CRB depends only on the amount of iterations carried out whose performances
are similar to the ideal case of a sufficient CP. In this method computational cost is less,
BER performance is enhanced when the IJEP is utilized in the preamble and subsequently
expands the battery life duration. At the same time it also creates some problems such as
less accuracy.
Ma et  al. [22] developed a simple suppression technique linearly integrates ISI free
part of a CP and its related part in an OFDM signal, which demonstrates the effect of ICI
caused by a time shifting channel. Compared with the applicable heuristic combining
weights from the literature for the suppression of ICI, in different optimum merged weights
were derived. Through simulations, it was proposed that the merged weights outperforms
than other heuristic joining weights. It was developed to diminish the calculation difficulty
of joining weights. Moreover, combining the proposed system with ICI improvement tech-
niques for further enhance the system performance.
Singh et al. [23] proposed a hybrid algorithm contains both PTS and SLM algorithms
to reduce the high peak to average power ratio problem in OFDM. Various advantages of
using OFDM such as robustness and large spectral efficiency against ISI, but still some
drawbacks are there such as; high PAPR is the major problem that occurs in OFDM sys-
tems. Various methods available for the minimization of PAPR such as tone reservation
(TR), clipping and filtering, Partial Transmit Sequence (PTS), active constellation scheme,
interleaving and selected mapping (SLM). Clipping and filtering are the simplest tech-
niques, but not suitable when high number of sub-carriers was available. The general meth-
ods utilized for PAPR reduction is PTS and SLM. PTS and SLM algorithms have great
PAPR reduction performance.
Chunjiang et al. [24] designed a hybrid PTS with SLM algorithm to reduce the peak to
average power ratio in OFDM signals. The OFDM technique is utilized for 4G communi-
cations. It has high top to normal power ratio (PAPR), and to reduce the PAPR in OFDM
systems different research has been conducted. In this paper in the reduction of PAPR the
SLM and PTS algorithms have high-quality performance and it reduces the information
overhead than the Golay complement sequence algorithm, less mutilation than the μ-law
compression algorithm, and less BER than the clipping algorithm. To overcome these
problems, another PAPR reduction algorithm was proposed, which is applied to PTS and
SLM algorithms to maintain a strategic distance from the disadvantages. According to the
simulation results, the proposed algorithm are much better in reducing the PAPR than the
SLM and PTS algorithms, with computational complexity, BER and data overhead.
Sarmadi et al. [25] proposed a BCE Orthogonally Coded MIMO-OFDM Systems. Spe-
cific properties of the OSTB codes are used, to assess the limited motivation reaction (FIR)
channel parameters in the time area as opposed to doing this in the frequency area freely

13
 S. Nandi et al.

for each subcarrier. To enhance miserliness on the channel constant model, it is compared
with direct per subcarrier channel estimation methodologies and grants coherent process
across the subcarriers. In the visually damaged MIMO channel, estimation techniques can
manage level blurring channels. The performance of MIMO-OFDM systems may critically
rely on the nature of the accessible channel state data (CSI) at the receiver. Various visually
damaged and partially distracted strategies can gauge frequency-selective MIMO channels
directly. The methods cannot get benefits of the specific structure of the underlying STC.
From the assessment channel parameters in the frequency space is experiencing a frail
miserliness of the channel model and a high computational complexity.

3 Proposed Methodology

3.1 STBC‑MIMO‑OFDM Systems for Fast Blind Channel Estimation

Channel estimation is the important part of mobile wireless channel, which is used to
increase the performance, particularly for 4G and LTE systems. Estimating the time-var-
ying channel frequency answer for the OFDM symbols is known as Channel estimation.
It is required for both symbol detection and ISI cancellation in STBC coded OFDM based
MIMO systems. High complexity and low convergence characteristics in which more num-
ber of OFDM images are expected to determine the convergence values, these are the most
important obstacle of previous blind channel estimation techniques. The vital aim of the
survey is to maximize the convergence speed and minimize the complexity of the blind
channel estimation, for the purpose of ISI suppression in Cyclic Prefix (CP) assisted blind
channel estimation STBC coded OFDM based MIMO systems. In addition the main contri-
butions of this paper are as follows:
The proposed blind channel estimation scheme, including CDMA approach with the
steepest descent optimization first order iterative optimization algorithm used for approxi-
mation of an integral and for finding the minimum of a function developed to enhance the
channel estimation performance of CP assisted STBC coded OFDM based MIMO systems
in the fading environment. Besides, an Adaptive Stein’s unbiased Estimator utilizes Modi-
fied Flower Pollination Algorithm (MFPA) for the purpose of tuning the hyper parameters
at each level of estimation. Finally, detection of the proposed channel estimation scheme
is carried out, based on the analysis the channel coefficients are detected two uncertainty
matrices.
In blind channel estimation the noise subspace is calculated by correlation matrix of
received signals. A vast number of symbols are used to remove the barriers in the circular
characteristics of the channel matrix through the MC-CDMA approaches. In Fig.  1 this
method N times of similar signals for every STBC coded OFDM symbol were generated
with respect to FFT operation. Efficient results can be obtained using the proposed MFPA
based blind channel estimation with the minimal OFDM symbol.
The proposed daze channel estimation approach is implemented in MATLAB and the
performance also compared with the conventional visually impaired channel estimation
approaches.

13
A Novel Adaptive Optimized Fast Blind Channel Estimation for…

Coding

Input Digital STBC OFDM

MCCDMA MFPA
Data Modulation Coding

STBC OFDM
Digital Modulation Received Data
Decoding

Decoding

Fig. 1  Block diagram for proposed model

3.2 OFDM Based MIMO

MIMO is applied in wireless communication to send and receive more than one data, for
antenna technology various antennas are utilized at source and the destination. Each end
of the communication circuits contains antennas combined to limit the errors and improve
information speed. One of the smart antenna technologies is MIMO; the further technolo-
gies are MISO, SISO and SIMO.
OFDM is a digital signal modulation in which particular data stream is dividing into
several narrowband channels to reduce the ISI and successful air-interface. Due to its wired
conditions, OFDM systems are also called as Discrete Multi-Tone transmissions, and used
as a piece of Asymmetric Digital Subscriber Line, High-piece rate and high-speed Digi-
tal Subscriber Line. In OFDM system different modulation schemes are used for different
subcarriers. For instance, the clients close to the Base Station (BS) may have a moderately
decent channel quality; consequently higher order modulation schemes are used to grow
the data rates. By contrast, the clients who are survived in a highly stacked urban zone the
subcarriers quality is poor, hence low order modulation schemes can be invented.
Blind channel estimation technique is essential for receiver design, it is also required
for interference suppression where the multiple receiver antennas are used and improve
the performance parameters of MIMO-OFDM. These results enhanced by the compari-
son of channel estimation methods with channel parametric model and across the subcar-
riers it allows logical processing. MIMO cannot attempt to diminish multipath propagation
and OFDM evades the signal equalization, hence MIMO-OFDM is a generally powerful
combination even the transmitter does not acquire channel state information (CSI), MIMO-
OFDM can reach very high spectral efficiency.

3.3 STBC OFDM Coding

In wireless communications, to transmit several data stream across many antennas

Space–Time Block Coding is used and it also utilized to develop the consistency of data
transfer of the different received data. The OFDM system was utilized as a part of 4G com-
munications, which contains a high Peak to Average Power Ratio (PAPR); then based on

13
 S. Nandi et al.

the different research conducted PAPR in OFDM systems is reduced. While transmitting
the channels through parallel or in a different way the STBC transmit the decent variety to
enhance the system performance in wireless communications.
Error filtering technique mentioned in the proposed method removes the expected con-
tent of stationary stochastic processes. It can diminish the autocorrelation of information
sequences and capable solution for a PAPR issue in OFDM transmissions. New approach
is a powerful reduction scheme in PAPR without reducing the power of spectral level, error
performance or overall computational complexity of the systems. It was also proved that
the present method is a stand-alone technique of all the modulation schemes and can be
applied to several sub carriers under both additive white Gaussian noise and wireless Ray-
leigh fading channel.

3.4 Signal Detection

In this manner, the dissimilarity among the signals depends on separation, noise spread
and signal-in addition to noise curves. To confines both the partition and the spread full
description recorded. The most broadly utilized calculation is called d-prime (d0), and its
recipe is straightforward:
d0 = separations spread;

Based on the difference between the means separations is performed, and based the
standard deviation of the probability densities the spread is matched. A complete descrip-
tion of the detected signal depends on the noise taken after an ordinary (Gaussian) cir-
culation with a fixed variable, autonomous of the signal quality. This supposition of IID
(autonomous and identically distributed) Gaussian noise is normally sensible approxima-
tions. The essential goodness of d0 , it is broadly utilized and its value does not rely on the
criterion of the subject, until now relatively it is a genuine measure of the inside reaction.

3.5 MC‑CDMA

The novel blind channel estimation system for the MC-CDMA systems that solve the chan-
nel estimation problem ensures a high performance in terms of bit-error-rate, minimizes
the transmitted signals and induces a very low computational complexity.
In OFDM the transmission scheme transmits signals in high-speed, high-capacity and also
MC-CDMA transmits spread symbols by a plurality of sub-carriers.
MC-CDMA is an outstanding strategy for fast remote information transmission. In the MC-
CDMA structure, two propelled innovation was incorporated, for example, OFDM and the
CDMA. The main goal is strengthening the OFDM against multipath condition and improves
the capacity of multiuser multiplexing administrations that is achieved via CDMA system. To
obtain high information rate services, a few 4G systems uses MC-CDMA to limit the ISI that
occurs when it is transmitted through multipath wireless channels.MC-CDMA received lots of
interests from the research community during the earlier period of few years, especially for the
data transmission in the wireless communication.

13
A Novel Adaptive Optimized Fast Blind Channel Estimation for…

3.6 Modified Flower Pollination Algorithm

The new proposed algorithm uses a blossom fertilization process of blooming plants. The
principle characteristics of blossom fertilization are characterized into four standards. To build
up Flower pollination Algorithm (FPA), it utilizes an arrangement of definitely understood test
functions and design benchmark. Then investigates the simulations and compares its perfor-
mance with a modified bloom fertilization algorithm, and in fact it is an optimization process
of plant species.
All the above factors and processes of blossom fertilization are related to achieve the ideal
reproduction of the blooming plants. Population initiation process of FPA is modified by tak-
ing the limit each fuel type into consideration to provide larger search space to find the opti-
mal solution. Apart from the modifications on population initiation, LI (Lambda Iteration)
approach also simplifies the generating units cost function and calculates average production
cost of each generating unit.
FPA improvement with DSPS (Dynamic Switching Probability Strategy) is applied
in this particular economic dispatch problem. DSPS set up the probability switch value
depends on number of iterations to control the local and global search. With the probability
of global search is done in the early stages of search and continues to decrease its value
until it reaches certain solutions. DSPS is modified with probability switches as the initial
value.

3.6.1 Modified Flower Pollination Algorithm Steps

FPA was established in 2012 by Yang. Propelled by the stream fertilization procedure of
blooming plants are the accompanying assumptions:

Step 1: Biotic and cross-fertilization is considered as a procedure of worldwide fertiliza-

tion process, and dust conveying pollinators move in Le’vy flights way
Step 2: For nearby fertilization, a biotic and self-fertilization is utilized
Step 3: Pollinators example is creepy crawlies creates blossom steadiness, which is equal
to a proliferation on the likelihood corresponding to the closeness of two blooms
Step 4: The exchanging of neighborhood pollination and worldwide pollination can
be controlled by a switch [0, 1] with a slight inclination toward local probabil-
ity p fertilization. Keep in mind the end goal is detail refreshing of recipes; we
need to change the previously mentioned rules into refreshing conditions. In the
worldwide fertilization step, blossom dust seeds are expressed by pollinators. For
instance, creepy crawlies and dust can go to long separation hence the bugs can
frequently fly and move on longer range

The most important process of FPA is the underlying populations switching from local
to worldwide fertilization.They have a main impact on the computational burden and con-
vergence solution. To develop the algorithm performance, the first modification is check-
ing the opposite guesses simultaneously, from this the fitter can chooses a best solution.
According to the probability theory the likelihood of two different solutions have a chance
of one is 50% better than the other. Starting with the fitter of the two guess have the poten-
tial of starting point closer to the optimal solution.

13
 S. Nandi et al.

3.7 Blind Channel Estimation

3.7.1 Problem Formulation

Let M be a p × q real matrix where q ≤ n . Then, for any m × s real symmetric N. The reso-
lution to the subsequent problem in optimization is,
{ }
U
max uv M NM subject to M U M = Jq (1)
M

Is given by any matrix M∗ whose column space is the same as the subspace planned by
the q principal eigenvectors of N and, for any such M∗,
q
{ } ∑
uv M∗U NM∗ = vi (2)
j=1

where vi (i = 1,…,q) is the q largest Eigen values of N. Replacing N by O and setting

q = 2PN
The resolution to the consequential optimization problem
{ }
U
max uv M OM subject to M U M = J2PN0 (3)
M

Is given by any matrix M∗ that satisfies the following properties:

( )
Range { M∗} = range { B̃ h′ }
uv { Mu∗ OM∗ } = UV { ∧ }.

Replace M in (3) B( ̃ Ẽ −1 , where, hΔ

̃ h) ̃ [h̃ U … h̃ U ]U is the vector optimization variables
0 N0 −1
and
{ }
‖̃ ‖
Ẽ Δ diag ‖ ̃ ‖
‖h0 ‖ … ‖‖
hN0−1 ‖ ⊗ J2P
‖ (4)

With such replacement, the constraint in (3) becomes

M U M = Ẽ −1 (h)
̃ B( ̃ Ẽ −1 = J2PN
̃ h)
0

where orthogonality property has been used. Hence, the constraint M U M = J2PN will be
0
satisfied for any h̃  , it can be dropped. Dropping this constraint obtains the following uncon-
strained problem:
−1
max uv{Ẽ −1 B̃ U (h)S
̃ B(̃ h)
̃ E}.
̃ (5)

Note that the problems in (2) and (4) are not equal to each other because, the matrix M
in (4) has a particular structure while the matrix M in (2) is unstructured. This implies that,
the sets of solutions (2)–(4) can lead to different sets of optimal matrices M for these two
problems. In Particular, the greatest value of the objective function in (4) cannot exceed (2)
| ̃ Ẽ −1 ||
uv|{Ẽ −1 B̃ u (h)S
̃ B(̃ h)
̃ �
= uv{∧}.
| |h=h

13
A Novel Adaptive Optimized Fast Blind Channel Estimation for…

Consequently, as the maxima of the objective functions in both (2) and (4) coincide the
set of all probable solutions to the latter problem in terms of B( ̃ is a subset. In the set of
̃ h)
all promising solutions the former problem in terms of M and the true channel vector h′
belongs to the subspace of all vectors that maximize (4). To simplify the objective function
in (4), this can be rewritten as,
−1
uv{Ẽ −1 B̃ U (h)S
̃ B(̃ h)
̃ E}
̃ = {vec{B( ̃ Ẽ −1 }U (j2PN ⊗ S)vec{B(
̃ h) ̃ Ẽ −1 }.
̃ h) (6)
0

̃ � )} = 𝛹 h� have
Using vec{B(h

vec{B( ̃ Ẽ −1 } = (D
̃ h) ̃ −1 ⊗ J2KTN )vec{A(
̃ h)}.
̃
(7)
0

̃ Ẽ −1 } = (Ẽ −1 ⊗ J2KTN )𝜓 h̃
̃ h)
vec{B( 0

Using (7) in (6) yields

uv{Ẽ −1 B̃ U (h)S
̃ B( ̃ Ẽ −1 } = h̃ u 𝜓 U (Ẽ −1 ⊗ J2KTN )(J2PN ⊗ S) ⋅ (Ẽ −1 ⊗ J2KTN ) 𝜓 h̃
̃ h)
0 0 0

uv{Ẽ −1 B̃ U (h)S
̃ B( ̃ Ẽ −1 } = h̃ U 𝜓 U (Ẽ −2 ⊗ S) 𝜓 h.
̃ h) ̃

Hence, the problem (5) is equivalent to

max h̃ U 𝜓 U (Ẽ −2 ⊗ S) 𝛹 h.
̃
(8)
h̃

Using (4), we obtained the special case of one subcarrier, Ẽ −2 = (1∕‖ ̃ ‖2

‖h‖ )J2P , and the
optimization problem (8)

h̃ u 𝜓 U (J2P ⊗ S) 𝜓 h̃
max .
h̃ ‖h̃ ‖2
‖ ‖
Hence, the channel estimation problem (8) in this particular case reduces the problem of
finding the principal eigenvector of the matrix 𝜓 u (J2p ⊗ S) 𝜓  . Therefore, it is convenient
to reformulate the problem in the time domain, where the subcarriers is independent of the
number of variables, results a small value even for very large subcarrier numbers. The lat-
ter equation rewrite the problem (8) in terms of time-domain variables, obtains
2
max g̃ U N u g̃ Subject to g̃ U Yiu g̃ = ‖ ‖
‖hi ‖ , i = 0, … N0 − 1 (9)
g̃

where g̃ Δ √1 (D ⊗ JMN )U h̃ .
N0
And the superscript “t” refers to the time domain. Equivalently, rewriting (9) as
min− g̃ U N u g̃
g̃

Subject to g̃ U Yiu g̃ = ‖ ‖2 i = 0, … N0 − 1 (10)

‖hi ‖ ,
It is important to stress that, in practice the true covariance matrix S is unavailable
therefore sample estimate is given by,
k
1 ∑ � �U
Ŝ = ZZ (11)
K j=1 j j

13
 S. Nandi et al.

It should be used instead of S, where Zj′ is the j th snapshot of the vector Z ′ and K is the
total number of STBC-OFDM blocks that are used to estimate Ŝ .
Having Ŝ  , the channel norms can be estimated using (11). Also, the noise power can be
estimated by averaging the noise subspace Eigen values of this matrix.

4 Results and Discussions

In this section the experimental results are compared with the existing methods such as
Forward Error Correcting Method (FEC) and Peak Reduction Coding (PRC) is used to
smooth the progress of the estimation process. The periodic transmission of known data,
it limits the NMSE system performance which is the drawback in FEC. Both in terms of
performance bounds as well as in estimation procedures SNR estimation has been consid-
ered. The implementation of the proposed work is done by the MATLAB tool. The results
proved the feasibility and efficiency of the proposed work.

4.1 Performance Metrics

An evaluation metric is a measure, which effectively analyze the existing approaches and
compared with the proposed method. Evaluation metrics normally encompass with set of
measures followed a general method for evaluation. The results of NMSE and MSE are
calculated to compute the efficiency of proposed method. The parameters are illustrated in
the following section.

(i) Normalized Mean Square Error (NMSE)

The NMSE (Normalized Mean Square Error) is an estimator of the general deviations
between calculated and expected values. Considering the fact that the NMSE of channel
estimation will be also influenced by Eb ∕N0 , we assumed that the value of NMSE linearly
decreases with the increase in Eb ∕N0.The NMSE is calculated by,
{ }
| (k) ̂ (k) |2
E |hm − hm |
| |
NMSE = { } = e2
| (k) |2
E |hm |
| |

hm = set of measured concentrations.

(ii) Mean Square Error (MSE)
Mean square error is the average error between the estimated signal and the real input
signal. Obviously, the smaller the MSE is the better efficiency of the proposed work. The
MSE can be computed as follows:
√
1 ∑k
MSE = (Ex − Ey )2
k i=1

where Ex , Ey = Actual and Estimated Signals.

13
A Novel Adaptive Optimized Fast Blind Channel Estimation for…

Fig. 2  Transmitting information as digital signal

Fig. 3  Received information as digital signal and binary FSK demodulation

Here the Input data transmitting information as digital Signal and send it to the demodu-
lation. Then errors can be detected and corrected which is caused by random processes.
Instead of continuous monitoring digital signals can also be sampled. Figures 2 and 3 rep-
resent the input signals which transfers digital information into demodulated form.

13
 S. Nandi et al.

Fig. 4  Wave form for binary FSK modulation corresponding binary information

Figure 4 represents the waveform of signals in FSK modulated signals; hence exten-
sive numerical processing is required to attain a large dynamic range and accurate fre-
quency discrimination. Digital information is spread over discrete changes in the fre-
quency of a carrier wave, which is called Frequency-shift keying (FSK) also known
as frequency modulation (FM). Logic level implementation of BFSK modulator and
demodulator can be obtained by this method.

4.2 Comparative Analysis and Discussion

In order to illustrate the advantages, the present technique is compared with the error
rate performance of STBC with the existing algorithms such as Forward Error Correct-
ing and Peak reduction Coding. Comparison analysis of NMSE for the proposed work
and existing work is shown in figures.
With the carrier frequency offset and channel estimation error in frequency selective
fading channels the symbol error rate of OFDM systems is expanded in Fig. 5. Hence
the utilized SER value is 432.13. It is seen that, in terms of SER the present system
performs better than the existing method.The method has better performance than the
previous techniques.
From this we will expand the symbol error rate of BCE systems with carrier fre-
quency offset and channel estimation error in frequency selective fading channels. In the
blind channel estimation, relationship among the SER and SNR were examined. From
the Fig. 6 it is clear that the performance has been enhanced substantially and the value
of SNR is 5.600.
In Fig.  7 the Average MSE is less compared to all other functions. The results are
contrasted by the parameters like MSE and SNR. The relation between Mean Square

13
A Novel Adaptive Optimized Fast Blind Channel Estimation for…

Fig. 5  STBC OFDM MIMO SER versus SNR

Fig. 6  Blind channel estimation SER versus SNR

Error (MSE) with the SNR at different SNR values in dB is described in this research.
The technique explains the QAM output in dB. It is examined that the BER performance
is calculated according to the estimation. The PAPR of the OFDM MIMO is 9.7523
(db).

13
 S. Nandi et al.

Fig. 7  STBC OFDM MIMO MSE versus SNR

Fig. 8  Blind channel estimation of SNR versus NMSE

The BCE of NMSE and SER performances were tested. From the Fig. 8 it is evalu-
ated that the performance of proposed method is better than other techniques. Blind
Channel Estimation value of NMSE is 0.111.

13
A Novel Adaptive Optimized Fast Blind Channel Estimation for…

5 Conclusion

In this paper an adaptive optimized fast blind channel estimation using cyclic prefix sup-
ported with (STBC-MIMO-OFDM) system is presented. The high complexity and low
convergence is the major problems in multicarrier communications. The main aspire
of our technique is permit the system to support multiple users at the same time over
same frequency band based on the MC-CDMA approach. Modified flower pollination
algorithm is implemented to overcome this problem. In this paper simple and very effi-
cient method is proposed, which effectively reduce the above problem and provide good
SER performance and lower fuel cost compared with FPA. It also shows the modulation
design for the present scheme is appropriate and can be work for multiple number of
subcarrier.

References
1. Armstrong, J. (2009). OFDM for optical communications. Journal of light wave technology, 27(3),
189–204.
2. Hwang, T., Yang, C., Wu, G., Li, S., & Li, G. Y. (2009). OFDM and its wireless applications: A sur-
vey. IEEE Transactions on Vehicular Technology, 58(4), 1673–1694.
3. Wang, X., Ho, P., & Yiyan, W. (2005). Robust channel estimation and ISI cancellation for OFDM sys-
tems with suppressed features. IEEE Journal on Selected Areas in Communications, 23(5), 963–972.
4. Hassib, M. D., Singh, M., Ismail, M., & Nordin, R. (2012). Efficient and low complexity STBC-OFDM
scheme over fading channel. In 2012 18th Asia-Pacific Conference on Communications (APCC) (pp.
402–406). IEEE.
5. Ku, M.-L., & Huang, C.-C. (2008). A refined channel estimation method for STBC/OFDM systems in
high-mobility wireless channels. IEEE Transactions on Wireless Communications, 7(11), 4312–4320.
6. Chen, H.-Y., Meng-Lin, K., Shyh-Jye, J., & Huang, C.-C. (2010). A robust channel estimator for high-
mobility STBC-OFDM systems. IEEE Transactions on Circuits and Systems I: Regular Papers, 57(4),
925–936.
7. Jeon, S., Han, J.-S., Choi, J.-M., & Seo, J.-S. (2015). Cooperative space-time block coded full-duplex
relaying over frequency-selective channel. IET Communications, 9(7), 960–968.
8. Wang, D., Minn, H., & Al-Dhahir, N. (2008). A robust asynchronous multiuser STBC-OFDM trans-
mission scheme for frequency-selective channels. IEEE Transactions on Wireless Communications,
7(10), 3725–3731.
9. Naeiny, M. F., & Marvasti, F. (2011). Selected mapping algorithm for PAPR reduction of space-fre-
quency coded OFDM systems without side information. IEEE Transactions on Vehicular Technology,
60(3), 1211–1216.
10. Kwon, U.-K., Kim, D., Kim, K., & Im, G.-H. (2007). Amplitude clipping and iterative reconstruction
of STBC/SFBC-OFDM signals. IEEE Signal Processing Letters, 14(11), 808–811.
11. Kwon, U.-K., Im, G.-H., & Kim, E.-S. (2007). An iteration technique for recovering insufficient cyclic
prefix and clipped OFDM signals. IEEE Signal Processing Letters, 14(5), 317–320.
12. Alexander, P., Haley, D., & Grant, A. (2011). Cooperative intelligent transport systems: 5.9-GHz field
trials. Proceedings of the IEEE, 99(7), 1213–1235.
13. Shin, C., Heath, R. W., & Powers, E. J. (2008). Non-redundant precoding-based blind and semi-blind
channel estimation for MIMO block transmission with a cyclic prefix. IEEE Transactions on Signal
Processing, 56(6), 2509–2523.
14. Ahn, S.-K., & Yang, K. (2013). A novel subblock-based frequency-domain equalizer over doubly-
selective channels. IEEE Communications Letters, 17(8), 1517–1520.
15. Won, H.-C., & Im, G.-H. (2005). Iterative cyclic prefix reconstruction and channel estimation for a
STBC OFDM system. IEEE Communications Letters, 9(4), 307–309.
16. Guo, Q., Ping, L., & Huang, D. (2009). A low-complexity iterative channel estimation and detec-
tion technique for doubly selective channels. IEEE Transactions on Wireless Communications, 8(8),
4340–4349.
17. Chen, S., & Zhu, C. (2004). ICI and ISI analysis and mitigation for OFDM systems with insufficient
cyclic prefix in time-varying channels. IEEE Transactions on Consumer Electronics, 50(1), 78–83.

13
 S. Nandi et al.

18. Sahin, A., & Arslan, H. (2011). Edge windowing for OFDM based systems. IEEE Communications
Letters, 15(11), 1208–1211.
19. Liu, H., & Schniter, P. (2008). Iterative frequency-domain channel estimation and equalization for
single-carrier transmissions without cyclic-prefix. IEEE Transactions on Wireless Communications,
7(10), 3686–3691.
20. MahmudulHasan, M. (2014). A new PAPR reduction technique in OFDM systems using linear predic-
tive coding. Wireless Personal Communications, 75(1), 707–721.
21. del Amo, C. P., & García, M. J. F.-G. (2013). Iterative joint estimation procedure for channel and fre-
quency offset in multi-antenna OFDM systems with an insufficient cyclic prefix. IEEE Transactions on
Vehicular Technology, 62(8), 3653–3662.
22. Ma, C.-Y., Liu, S.-W., & Huang, C.-C. (2014). Low-complexity ICI suppression methods utilizing
cyclic prefix for OFDM systems in high-mobility fading channels. IEEE Transactions on Vehicular
Technology, 63(2), 718–730.
23. Singh, A., & Singh, H. (2016). Peak to average power ratio reduction in OFDM system using hybrid
technique. Optik-International Journal for Light and Electron Optics, 127(6), 3368–3371.
24. Duanmu, C., & Chen, H. (2014). Reduction of the PAPR in OFDM systems by intelligently applying
both PTS and SLM algorithms. Wireless Personal Communications, 74(2), 849–863.
25. Sarmadi, N., Shahbazpanahi, S., & Gershman, A. B. (2009). Blind channel estimation in orthogonally
coded MIMO-OFDM systems: A semidefinite relaxation approach. IEEE Transactions on Signal Pro-
cessing, 57(6), 2354–2364.

Publisher’s Note  Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional affiliations.

Shovon Nandi  is an Assistant Professor in the Electronics & Commu-

nication Engineering department of Bengal Institute of Technology,
India under Maulana Abul Kalam Azad University of Technology
(MAKAUT), India. He has a total teaching experience of more than a
decade. He completed his M. Tech from University of Calcutta. His
research interests include Wireless Communications, Channel Estima-
tion, Multi-carrier Communications such as MIMO OFDM, MC-
CDMA systems and wireless sensor networks.

Narendra Nath Pathak  received Ph.D. degree from National Institute

of Technology, Durgapur, India, is attached with Dr. B. C. Roy Engi-
neering College as a Professor & Head of Electronics & Communica-
tion Engineering Department. He is the author of several international
journals and international conference proceedingsand reviewer of
many international journals. His research interests include Microwave
Engineering, Antenna Array and Soft Computing, Multi-carrier Com-
munication and Channel Estimation Process.

13
A Novel Adaptive Optimized Fast Blind Channel Estimation for…

Arnab Nandi  received Ph.D. degree from National Institute of Tech-

nology, Durgapur, India, is attached with National Institute of Technol-
ogy, Silchar, India, as an Assistant Professor of Electronics & Commu-
nication Engineering Department. He has published several research
papers ininternational journals of repute and national and international
conferences and reviewer of many international journals. His research
interests include Ad hoc Wireless Networks, Wireless Sensor Net-
works, Cross Layer Issues and Medium Access Control.

13