Beam Forming Techniques in

Smart Antenna
Dhaval N Patel Guided By: Prof. Parita B Parmar
(140430705013)
Contents
• Smart Antenna Analogy
• Types of smart antenna
• DOA Estimation
• Adaptive Beam forming
• Analysis of Beamforming
• Hybrid Algorithm
• Conclusion
• References
 SMART-ANTENNA ANALOGY
• How a smart-antenna system works, let us
imagine two persons carrying on a conversation
inside a dark room.
SMART-ANTENNA ANALOGY
• Electrical smart-antenna systems work the
same way using two antennas instead of the
two ears and a digital signal processor instead
of a brain.
SMART-ANTENNA ANALOGY
• Digital signal processor measures the time
delays from each antenna element, it computes
the direction of arrival (DOA) of the signal-of-
interest (SOI).
• Smart antenna is an antenna array which uses a
number of antenna elements and signal
received at each antenna element is adaptively
combined to improve the overall performance.
Types of Smart Antenna
1. Switched beam smart antenna

2. Adaptive smart antenna

Switched beam smart antenna
• In switched beam smart antenna, antenna
system has several fixed beam patterns.
• As the mobile unit moves throughout the cell,
the switched-beam system detects the signal
strength, chooses the appropriate predefined
beam pattern, and continually switches the
beams as necessary.
Switched beam smart antenna
Adaptive smart antenna
• Systems are really smart because they are able to
dynamically react to the changing RF
environment with its infinite scenario based
patterns.
• They can direct the main beam toward the
direction of arriaval (DOA) while suppressing the
antenna pattern in the direction of the
interferers.
• It can customize an appropriate radiation pattern
for each individual user.
• Dynamically adjust the antenna pattern to
enhance reception while minimizing or fully
rejecting interference .
Adaptive smart antenna
• At 2.4 GHz, The desired signal is taken as a
cosine signal at an angle of 15° and Two
interfering signals are considered at angles of -
15° and 30°.
• When d=0.5λ, the SLL is low and the null
depth is highest resulting in greatest
interference suppression capability.
• Thus, d=0.5λ is the optimum separation
distance between two antenna elements.
• The number of antenna elements increases, the
beamwidth decreases, thus increasing the
directivity of the antenna array.
Direction of Arrival Estimation
• We have also seen that there is a Fourier
relationship between the beam pattern and the
excitation at the array.
• This allows the direction of arrival (DOA)
estimation problem to be treated as equivalent
to spectral estimation.
• There are many techniques : correlation,
Maximum Likelihood, MUSIC, ESPRIT and Matrix
Pencil.
DOA Algorithm
• Comparative performance and geometrical
optimization between

1. MUSIC (Multiple Signal Classification),

2. MVDR ( Minimum Variance Distortion less
Response) and
3. Min Norm

• direction of arrival algorithms used in linear

antenna array.
MVDR algorithm simulations for
different number of array elements.
MUSIC algorithm for different array
element spacing
• DOA for MVDR and MUSIC algorithm for close
signals arriving at -30°, -40° and 50°.
 DOA Algoritham
• The performance of algorithms improves with
more elements in the array and with increasing
the number of samples.
• MUSIC, as the most stable and accurate
algorithm, which provides high resolution.
Adaptive Beamforming
1. LMS (Least Mean Square) Algorithm :
• In this algorithm, the weights are updated at
every iteration by estimating the gradient of the
quadratic Mean Square Error (MSE).
• Then moving the weights in the negative
direction of the gradient by a small amount,
known as the step size.
• The convergence of this algorithm is directly
proportional to the step-size parameter μ.
1. LMS (Least Mean Square) Algorithm :
• The array weights are updated according to the
following equation:

• where the error signal is given by

2. SMI Algorithm:
• This algorithm is suitable for a rapidly changing
environment as it converges much faster than
the LMS algorithm.
• Thereby allowing the tracking of the desired
signal.
• However, computational complexity and matrix
singularities can cause problems.
3. RLS Algorithm :
• The convergence speed of LMS algorithm depends
on the eigen values of the array correlation matrix.
• In a fast changing mobile environment which yields
an array correlation matrix with large eigen value
spread, the LMS algorithm converges with a slow
speed.
• This problem is solved with the RLS algorithm by
replacing the gradient step size μ with a gain
matrix.
4. CGM Algorithm:
• CGM algorithm increases the rate of
convergence by iteratively searching for the
optimum solution by choosing conjugate
(perpendicular) paths for each new iteration.

• µ(n) is the step size and D(n) is the direction

vector.
 Analysis of Beamforming
• The SMI algorithm is found to have a narrower
beamwidth and a higher directivity as
compared to the LMS algorithm.
• The beamwidth and SLL obtained for the RLS
algorithm is slightly higher than that for the SMI
algorithm.
• CGM algorithm gives the maximum null depths
at the two interferences.
 Analysis of Beamforming
• LMS algorithm consumes less power as
compared to RLS algorithm at their respective
optimal number of iterations.
• Optimal number of iterations increases linearly
with path loss exponent and decreases
logarithmically with BER.
Hybrid Algorithm
• The hybridization of LMS and SMI algorithm
improves the beamforming.
• But it is computationally intensive, since SMI
technique requires inversion operation.
• LMS algorithm is directly proportional to the
step-size parameter μ.
• If the step size is too small, the convergence is
slow.
• If the step size is too large, the LMS algorithm
will overshoot the optimum weights of interest.
Hybrid Algorithm
• Now initial weights of the LMS algorithm are
computed by GA with predefined specifications
instead of random selection of weights.
• In this way hybrid algorithm (LMS/GA) is
adapted for the present problem.
Radiation pattern of 20 element linear SA with
direction of desired signal 150º and nulls at 50º
and 100º

standard LMS algorithm (LMS/GA) hybrid algorithm

Conclusion
• It is seen that with increase in antenna
elements M, the beamwidth reduces, making
the array more directional and the SLL reduces,
thus improving beamforming.
• MUSIC, as the most stable and accurate
algorithm for DOA.
• CGM algorithm has the fastest convergence.
• Deep nulls are generated at unwanted direction
using hybrid algorithm.
References
• C. Balanis, Antenna Theory: Analysis and Design. John Wiley, 1997.
• Direction of Arrival Estimation, Introduction, BeamForming,
University of Torranta.
• Luka Lazovic, Ana Jovanovic “Comparative performance study of
DO A algorithm applied on linear antenna array in smart antenna
systems” 2nd Mediterranean Conference on Embedded Computing
,,/" MECD – 2013.
• Anupama Senapati, Kaustabh Ghatak, Jibendu Sekhar Roy “A
Comparative Study of Adaptive Beamforming Techniques in Smart
Antenna Using LMS Algorithm and Its Variants”, DOI
10.1109/CINE.2015.21, 2015 International Conference on
Computational Intelligence & Networks 2375-5822/15 $31.00 ©
2015 IEEE
• Prerna Saxena, A.G. Kothari “Performance Analysis of Adaptive
Beamforming Algorithms for Smart Antennas ” doi:
10.1016/j.ieri.2014.09.101, 2014 International Conference on
Future Information Engineering
References
• Rashi Mehrotra, Ranjan Bose “Green Design for Smart
Antenna System using Iterative Beamforming Algorithms”,
International Conference on Computing, Networking and
Communications, Green Computing, Networking and
Communications Symposium 2015, 978-1-4799-6959-
3/15/$31.00 ©2015 IEEE
• Amara Prakasa Rao and N.V.S.N. Sarma, “Hybrid evolutionary
algorithm based beamforming for smart antenna system”
doi.10.1080/00207217.2013.859304, International Journal of
Electronics Letters-2013, 1:4, 189-197
• Vijendra Mishral, Gaurav Chaitanya “Analysis of LMS, RLS and
SMI Algorithm on the Basis of Physical Parameters for Smart
Antenna” 978-1-4799-3064-7/14/$31. 00©20 14 IEEE
Questions.