Professional Documents
Culture Documents
Nptel QS Bank WSN
Nptel QS Bank WSN
Unit 2 - Week 1:
Course
outline Assignment Week 1
The due date for submitting this assignment has passed. Due on 2018-02-21, 23:59 IST.
How to access As per our records you have not submitted this assignment.
the portal
1) A wireless network without a centralized access point may be: 1 point
Week 1:
An infrastructure network
Lecture 1: An ad hoc network
Introduction:
All of these
Wireless Ad
Hoc Networks- None of these
Part- I
No, the answer is incorrect.
Lecture 2: Score: 0
Introduction:
Wireless Ad Accepted Answers:
Hoc Networks- An ad hoc network
Part- II
2) Which of the following technologies enable wireless ad-hoc networking 1 point
Lecture 3: Self-
organizing ZigBee
Behaviour of
Wireless Ad
4G(LTE)
Hoc Networks All of the above
Week 3 Sensing
Transmitting
Week 4
Processing
None of these
Week 5
Week 6
https://onlinecourses-archive.nptel.ac.in/noc18_cs09/unit?unit=6&assessment=86 1/3
27/07/2020 Wireless Adhoc And Sensor Networks - - Unit 2 - Week 1:
No, the answer is incorrect.
Week 7 Score: 0
Accepted Answers:
Week 8 Transmitting
DOWNLOAD 5) Throughput in wireless ad hoc networks with N nodes decreases at a rate of 1 point
VIDEOS
√N
N^2
N^3
2N
6) In Sprite, a selfish node may save a receipt but does not forward the message. What is the 1 point
countermeasure for this selfish action?
7) The nodes which are unable to perform an operation because of either power failure or 1 point
environmental events are termed as
Selfish node
Failed nodes
Dead node
Malicious node
9) Sprite is a system based on credit, which is used to provide ___________ for mobile nodes 1 point
to cooperate and report actions honestly.
Penalty
Bit coin
Incentives
Acknowledge
https://onlinecourses-archive.nptel.ac.in/noc18_cs09/unit?unit=6&assessment=86 2/3
27/07/2020 Wireless Adhoc And Sensor Networks - - Unit 2 - Week 1:
Accepted Answers:
Incentives
10)Nodes which deliberately disrupt the correct operation of the routing protocol are known as: 1 point
Selfish nodes
Failed nodes
Badly failed nodes
Malicious nodes
Funded by
Powered by
https://onlinecourses-archive.nptel.ac.in/noc18_cs09/unit?unit=6&assessment=86 3/3
27/07/2020 Wireless Adhoc And Sensor Networks - - Unit 3 - Week 2
Unit 3 - Week 2
Course
outline Assignment Week 2
The due date for submitting this assignment has passed. Due on 2018-02-21, 23:59 IST.
How to access As per our records you have not submitted this assignment.
the portal
1) RTS/CTS period is called 1 point
Week 1:
Waiting period
Week 2 Contention period
Running period
Lecture 6: MAC
Protocols in
None of these
MANETs- Part- I
No, the answer is incorrect.
Lecture 7: MAC Score: 0
Protocols in
Accepted Answers:
MANETs- Part-
II Contention period
https://onlinecourses-archive.nptel.ac.in/noc18_cs09/unit?unit=15&assessment=87 1/3
27/07/2020 Wireless Adhoc And Sensor Networks - - Unit 3 - Week 2
No, the answer is incorrect.
DOWNLOAD
VIDEOS Score: 0
Accepted Answers:
30 Bytes
DSR
AODV
ZRP
None of the above
QAMNET
QMR
ODMRP
All the above
7) When a group of nodes are required to contact, which type of the following routing is useful? 1 point
Multicast
Unicast
Broadcast
None of these
8) Join query and join reply are used by which of the following multicast routing protocol? 1 point
ODMRP
QAMNET QMR ZMR
QMR
ZMR
9) The routing scheme which use both the mask and destination address as 0.0.0.0 is: 1 point
Network-specific
Default
Host-specific
None of the above
https://onlinecourses-archive.nptel.ac.in/noc18_cs09/unit?unit=15&assessment=87 2/3
27/07/2020 Wireless Adhoc And Sensor Networks - - Unit 3 - Week 2
10)MARCH exploits the properties of ___________ antennas and overhearing properties of 1 point
MANETs.
Single directional
Bi-directional
Omnidirectional
None of these
Funded by
Powered by
https://onlinecourses-archive.nptel.ac.in/noc18_cs09/unit?unit=15&assessment=87 3/3
27/07/2020 Wireless Adhoc And Sensor Networks - - Unit 4 - Week 3
Unit 4 - Week 3
Course
outline Assignment Week 3
The due date for submitting this assignment has passed. Due on 2018-02-28, 23:59 IST.
How to access As per our records you have not submitted this assignment.
the portal
1) In TCP-ELFN, CWND = 1 causes No significant change, because – 1 point
Week 1:
Optimal window is relatively small
Week 2 It is similar to No change
Default value of RTO is small
Week 3 All of these
https://onlinecourses-archive.nptel.ac.in/noc18_cs09/unit?unit=24&assessment=93 1/3
27/07/2020 Wireless Adhoc And Sensor Networks - - Unit 4 - Week 3
No, the answer is incorrect.
Week 7
Score: 0
6) When ATCP is in the lossy state, reception of an ECN or ICMP message will move ATCP into1 point
which of the following state?
Free
Congested
Forward
None of these
8) In TCP-F, when a routing agent detects the disruption of a route, it explicitly sends a 1 point
_________ packet to the sender.
9) In TCP-BuS, to avoid timeout events during the route re-construction (RRC) phase, the 1 point
timeout values for buffered packets from source to PN will be
Set to one
Doubled
Same
Half
https://onlinecourses-archive.nptel.ac.in/noc18_cs09/unit?unit=24&assessment=93 2/3
27/07/2020 Wireless Adhoc And Sensor Networks - - Unit 4 - Week 3
Accepted Answers:
Doubled
Selecting
Probing
Distributing
None of these
Funded by
Powered by
https://onlinecourses-archive.nptel.ac.in/noc18_cs09/unit?unit=24&assessment=93 3/3
27/07/2020 Wireless Adhoc And Sensor Networks - - Unit 5 - Week 4
Unit 5 - Week 4
Course
outline Assignment Week 4
The due date for submitting this assignment has passed. Due on 2018-03-07, 23:59 IST.
How to access As per our records you have not submitted this assignment.
the portal
1) UAV-SDN aims to achieve – 1 point
Week 1:
Autonomous UAV control
Week 2 Remote UAV control
Remote control and configuration of network of UAVs
Week 3 Swarm of UAVs
Lecture 17: Whether or not a given node was present at a given location at a given time instant
Opportunistic Whether or not it belongs to the cut set
Mobile
Networks- Part- Characterizes a node via which all communication must pass
III Whether or not a given edge was present at a given time instant
Lecture 18: UAV No, the answer is incorrect.
Networks- Part-
Score: 0
I
Accepted Answers:
Lecture 19: UAV
Whether or not a given edge was present at a given time instant
Networks- Part-
II 3) What does a Protocol Translation Unit do? 1 point
Lecture 20: UAV
Networks- Part- Allows two different routing protocols to communicate
III Allows devices with heterogeneous MAC layers to communicate
Week 4: Lecture Provides backward compatibility to older versions of a given protocol
Material Collects messages from highly mobile nodes similar to throwboxes
Quiz : No, the answer is incorrect.
Assignment
Score: 0
Week 4
Accepted Answers:
Assignment
Allows two different routing protocols to communicate
Solution Week 4
4) Which is not a typical requirement for reputation/trust management schemes for OMNs, in 1 point
Week 5
general?
https://onlinecourses-archive.nptel.ac.in/noc18_cs09/unit?unit=33&assessment=92 1/3
27/07/2020 Wireless Adhoc And Sensor Networks - - Unit 5 - Week 4
No, the answer is incorrect.
DOWNLOAD
VIDEOS Score: 0
Accepted Answers:
Huge network bandwidth
5) OMNs are very similar to MANETs in the sense that they both lack in network infrastructure. 1 point
However, their primary difference is the lack of ____
End-to-end communication
Storage
Processing power
All of these
6) Multi-UAV systems have preferably ____________ antennas, whereas single UAV systems 1 point
have omni-directional antennas.
Omni-directional
Bi-directional
Directional
None of these
7) Similar to vehicular networks, which are termed VANETs, UAV networks are popularly termed 1 point
as:
UANETs
FANETs
WINETs
All of these
8) Yaw, Pitch and Roll values are determined from which sensor? 1 point
Accelerometer
Gyroscope
Magnetometer
Barometer
9) An UAV SDN implementation encompasses data and ________ planes of communication. 1 point
Process
Resource
Control
Display
https://onlinecourses-archive.nptel.ac.in/noc18_cs09/unit?unit=33&assessment=92 2/3
27/07/2020 Wireless Adhoc And Sensor Networks - - Unit 5 - Week 4
Accepted Answers:
Control
10)In ProPHET, the use of ___ allows to update the contact probability with a node with whom 1 point
there has been no contact lately
Paging
Aging
Indexing
Casing
Funded by
Powered by
https://onlinecourses-archive.nptel.ac.in/noc18_cs09/unit?unit=33&assessment=92 3/3
27/07/2020 Wireless Adhoc And Sensor Networks - - Unit 6 - Week 5
Unit 6 - Week 5
Course
outline Assignment Week 5
The due date for submitting this assignment has passed. Due on 2018-03-14, 23:59 IST.
How to access As per our records you have not submitted this assignment.
the portal
1) Solution for optimal coverage includes the important parameters as: 1 point
Week 1:
Coverage requirement
Week 2 Detection time
Number of sensors
Week 3 All of the above
https://onlinecourses-archive.nptel.ac.in/noc18_cs09/unit?unit=42&assessment=99 1/3
27/07/2020 Wireless Adhoc And Sensor Networks - - Unit 6 - Week 5
No, the answer is incorrect.
Week 6
Score: 0
In the setup phase, clusters are created and cluster heads are determined
The nodes join the cluster nearest to them with the strongest signal
In the steady state phase, CSMA-based solutions are used for inter-cluster communication
In the steady state phase, TDMA-based solutions are used for inter-cluster communication
At least k sensors
At least 1 sensors
At least 2 sensors
At least 3 sensors
https://onlinecourses-archive.nptel.ac.in/noc18_cs09/unit?unit=42&assessment=99 2/3
27/07/2020 Wireless Adhoc And Sensor Networks - - Unit 6 - Week 5
10)Which is not a variant of S-MAC? 1 point
Funded by
Powered by
https://onlinecourses-archive.nptel.ac.in/noc18_cs09/unit?unit=42&assessment=99 3/3
27/07/2020 Wireless Adhoc And Sensor Networks - - Unit 7 - Week 6
Unit 7 - Week 6
Course
outline Assignment Week 6
The due date for submitting this assignment has passed. Due on 2018-03-21, 23:59 IST.
How to access As per our records you have not submitted this assignment.
the portal
1) 1. Which of the following statements does not describe the S-MAC protocol correctly? 1 point
Week 1:
Energy dissipation is reduced in S-MAC through idle listening
Week 2 Implementation of the algorithm is simple
The sleep and listen vary constantly based on the network traffic conditions
Week 3 All of the above
https://onlinecourses-archive.nptel.ac.in/noc18_cs09/unit?unit=51&assessment=100 1/3
27/07/2020 Wireless Adhoc And Sensor Networks - - Unit 7 - Week 6
Week 6 No, the answer is incorrect.
Assignment
Score: 0
Solution Week 6 Accepted Answers:
All of the above
Week 7
5) Which is not a cause for undesired energy wastage in WSNs? 1 point
Week 8
Sensing of events
DOWNLOAD Idle listening
VIDEOS Overhearing in case of densely deployed sensor networks
Collision of transmitted frames
SPIN
CFFR
All of these
None of these
CDMA
FDMA
TDMA
OFDMA
Directed Diffusion
SPIN
SPEED
TTDD
9) What is the difference between the 3-color and 4-color topdisc mechanisms? a. 4-color 1 point
topdisc mechanism does not have a cluster head node b. Unlike the 3-color topdisc, nodes may have
red or blue color in 4-color topdisc mechanism c. Grey color is not present in 3-color topdisc mechanism
d. Dark grey color is not present in 3-color topdisc mechanism
https://onlinecourses-archive.nptel.ac.in/noc18_cs09/unit?unit=51&assessment=100 2/3
27/07/2020 Wireless Adhoc And Sensor Networks - - Unit 7 - Week 6
Accepted Answers:
Dark grey color is not present in 3-color topdisc mechanism
In the setup phase, clusters are created and cluster heads are determined
The nodes join the cluster nearest to them with the strongest signal
In the steady state phase, CSMA-based solutions are used for inter-cluster communication
In the steady state phase, TDMA-based solutions are used for inter-cluster communication
Funded by
Powered by
https://onlinecourses-archive.nptel.ac.in/noc18_cs09/unit?unit=51&assessment=100 3/3
27/07/2020 Wireless Adhoc And Sensor Networks - - Unit 8 - Week 7
Unit 8 - Week 7
Course
outline Assignment Week 7
The due date for submitting this assignment has passed. Due on 2018-03-28, 23:59 IST.
How to access As per our records you have not submitted this assignment.
the portal
1) In Congestion Control, a bit can be set in a packet moving in the direction opposite to 1 point
Week 1: congestion in
Week 7 2) There is no communication between congested node or nodes and source in the 1 point
https://onlinecourses-archive.nptel.ac.in/noc18_cs09/unit?unit=60&assessment=103 1/3
27/07/2020 Wireless Adhoc And Sensor Networks - - Unit 8 - Week 7
Week 7 No, the answer is incorrect.
Assignment
Score: 0
Solution Week 7 Accepted Answers:
Data Traffic
Week 8
5) ________________ is the best suitable ranging technique among these? (Consider no time- 1 point
DOWNLOAD synchronization among nodes.)
VIDEOS
ToA (one-way message transfer)
Euclidean distance propagation technique
ToA (two-way message transfer)
All of the above
6) For achieving network-wide localization, which are the important parameters for evaluating 1 point
any localization algorithm?
Localization coverage
Average localization accuracy
Average energy consumption
All of these
Topology
Link between nodes
Coverage
All of these
Star topology
Cluster topology
Mesh topology
All of these
https://onlinecourses-archive.nptel.ac.in/noc18_cs09/unit?unit=60&assessment=103 2/3
27/07/2020 Wireless Adhoc And Sensor Networks - - Unit 8 - Week 7
All of these
10)No. of reference points required for localizing a node in 3D coordinate systems is ______. 1 point
4
3
5
None of these
Funded by
Powered by
https://onlinecourses-archive.nptel.ac.in/noc18_cs09/unit?unit=60&assessment=103 3/3
27/07/2020 Wireless Adhoc And Sensor Networks - - Unit 9 - Week 8
Unit 9 - Week 8
Course
outline Assignment Week 8
The due date for submitting this assignment has passed. Due on 2018-04-04, 23:59 IST.
How to access As per our records you have not submitted this assignment.
the portal
1) Which of the following jamming attacks is time specific? 1 point
Week 1:
Constant jamming
Week 2 Intermittent jamming
Deceptive jamming
Week 3 Reactive jamming
https://onlinecourses-archive.nptel.ac.in/noc18_cs09/unit?unit=69&assessment=107 1/3
27/07/2020 Wireless Adhoc And Sensor Networks - - Unit 9 - Week 8
DOWNLOAD 5) A Bluetooth network consists of one primary device(s) and up to ____ secondary devices. 1 point
VIDEOS
Five
Two
Seven
Nine
Passive
Active
Dead
None of these
7) One malicious node behaves like multiple nodes, this type of attack is known as 1 point
Byzantine attack
Sybil attack
DoS attack
Spoofing
Forwarding attack
Sinkhole attack
Selectivity attack
None of these
https://onlinecourses-archive.nptel.ac.in/noc18_cs09/unit?unit=69&assessment=107 2/3
27/07/2020 Wireless Adhoc And Sensor Networks - - Unit 9 - Week 8
10)Traditional cryptographic algorithms cannot be used directly in WSN, because sensor nodes 1 point
are
Resource constrained.
Memory-less
Clock-less
Processor-less
Funded by
Powered by
https://onlinecourses-archive.nptel.ac.in/noc18_cs09/unit?unit=69&assessment=107 3/3
NPTEL MOOC Estimation: Assignment #1
1. In the context of the estimation framework considered, the probability density function
(PDF) of the unknown parameter h is termed as the
a. Likelihood Density
b. Parameter Density
c. Prior Density
d. Posterior Density
Ans c
2. In the context of the estimation framework considered, the probability density function
(PDF) of the unknown parameter h conditioned on the observation vector y is termed as
the
a. Likelihood Density
b. Parameter Density
c. Prior Density
d. Posterior Density
Ans d
3. The estimation framework described in the lectures is,
a. Maximum Cost Estimation
b. Minimum Mean Squared Error Estimation
c. Minimum Error Estimation
d. Maximum Likelihood Estimation
Ans b
4. Given the parameter h and observation vector y, the expression for the MMSE estimate
is,
a.
b.
c.
d.
Ans a
5. Consider jointly Gaussian random variables h, y with mean 0 each and variances , .
Let denote the correlation coefficient between h,y. The marginal probability density
functions of h, y are,
a.
b.
c.
d.
Ans d
6. Consider jointly Gaussian random variables h, y with mean 0 each and variances , .
Let denote the correlation coefficient between h, y. The joint probability density
function of h, y is,
a.
b.
c.
d.
Ans b
7. Consider jointly Gaussian random variables h, y with mean 0 each and variances , .
Let denote the correlation coefficient between h, y. The conditional probability density
function of h given y is,
a.
b.
c.
d.
ans a
8. Consider a scenario with jointly Gaussian random variables h,y, where , ,
= 4, = 9 and correlation coefficient = 0.85. Find the cross-correlation .
a. 5.0
b. 2.55
c. 0.85
d. 5.1
ans d
9. Consider a scenario with jointly Gaussian random variables h,y, where , ,
= 4, = 9 and correlation coefficient = 0.85. Find the MMSE estimate
a. 1.275y-2.825
b. 0.566y+2.43
c. 0.566y-0.7
d. 1.275y+1.725
Ans: 0.566y-0.7
10. Consider a scenario with jointly Gaussian random variables h,y, where , ,
= 4, = 9 and correlation coefficient = 0.85. Find the MMSE
a. 1.11
b. 0.99
c. 1.01
d. 1
Ans:1.11
NPTEL MOOC Estimation: Assignment #2
1. Consider MMSE estimation for the wireless sensor network (WSN) scenario as described
in lectures with each observation y(k) = h + v(k), for 1 k N, i.e. number of
observations N and IID Gaussian noise samples of variance . What is the expression
for the MMSE estimate ĥ of the unknown parameter h
a.
b.
c.
d.
Ans b
2. Consider MMSE estimation for the wireless sensor network (WSN) scenario as described
in lectures with each observation y(k) = h + v(k), for 1 k N, i.e. number of
observations N and IID Gaussian noise samples of dB variance . As the number of
observations N , the MMSE estimate ĥ of the unknown parameter h tends to,
a. 0
b. 1
c. ML Estimate
d. Prior Mean
Ans c
3. Consider MMSE estimation for the wireless sensor network (WSN) scenario as described
in lectures with each observation y(k) = h + v(k), for 1 k 5, i.e. number of
observations N = 5 and IID Gaussian noise samples of dB variance = - 3 dB i.e.
10log10 = -3. Let the observations be y(1) = 1, y(2) = 1, y(3) = 2, y(4) = 3/2, y(5) = 5/2.
Let and . What is the maximum likelihood estimate ĥ of the
unknown parameter h
a. 1.6
b. 1
c. 1.5
d. 2
Ans a
4. Consider MMSE estimation for the wireless sensor network (WSN) scenario as described
in lectures with each observation y(k) = h + v(k), for 1 k 5, i.e. number of
observations N = 5 and IID Gaussian noise samples of dB variance = - 3 dB i.e.
10log10 = -3 dB. Let the observations be y(1) = 1, y(2) = 1, y(3) = 2, y(4) = 3/2, y(5) =
5/2. Let and . What is the MMSE estimate ĥ of the unknown
parameter h
a. 10/7
b. 8/7
c. 5/7
d. 9/7
Ans d
5. Consider MMSE estimation for the wireless sensor network (WSN) scenario as described
in lectures with each observation y(k) = h + v(k), for 1 k 5, i.e. number of
observations N = 5 and IID Gaussian noise samples of dB variance = - 3 dB i.e.
10log10 = -3 dB. Let the observations be y(1) = 1, y(2) = 1, y(3) = 2, y(4) = 3/2, y(5) =
5/2. Let and . What is the posterior probability density function of the
unknown parameter h
a. Gaussian
b. Exponential
c. Rayleigh
d. Uniform
Ans a
6. Given the fading channel estimation problem where the output symbol y(k) is y(k) = hx(k)
+ v(k), with h, x(k), v(k) denoting the real channel coefficient, pilot symbol and noise
sample respectively. Let x = [x(1), x(2),…, x(N)]T denote the pilot vector of transmitted
pilot symbols and y = [y(1), y(2),…, y(N)]T denote the corresponding received symbol
vector. Let v(k) be IID Gaussian noise with zero-mean and variance . Let ,
denote the prior mean, variance of the parameter h. The maximum likelihood estimate of
the channel coefficient h is,
a.
b.
c.
d.
Ans c
7. Given the fading channel estimation problem where the output symbol y(k) is y(k) = hx(k)
+ v(k), with h, x(k), v(k) denoting the real channel coefficient, pilot symbol and noise
sample respectively. Let x = [x(1), x(2),…, x(N)]T denote the pilot vector of transmitted
pilot symbols and y = [y(1), y(2),…, y(N)]T denote the corresponding received symbol
vector. Let v(k) be IID Gaussian noise with zero-mean and variance . Let ,
denote the prior mean, variance of the parameter h. The covariance matrix of the
output vector y is,
a.
b.
c.
d.
Ans b
8. Given the fading channel estimation problem where the output symbol y(k) is y(k) = hx(k)
+ v(k), with h, x(k), v(k) denoting the real channel coefficient, pilot symbol and noise
sample respectively. Let x = [x(1), x(2),…, x(N)]T denote the pilot vector of transmitted
pilot symbols and y = [y(1), y(2),…, y(N)]T denote the corresponding received symbol
vector. Let v(k) be IID Gaussian noise with zero-mean and variance 2. Let , denote
the prior mean, variance of the parameter h. The MMSE estimate of the channel
coefficient h is,
a.
b.
c.
d.
Ans d
9. Consider the fading channel estimation problem where the output symbol y(k) is y(k) =
hx(k) + v(k), with h, x(k), v(k) denoting the real channel coefficient, pilot symbol and
T
1 3
noise sample respectively. Let x 2 1 denote the pilot vector of transmitted
2 2
pilot symbols and y 2 2 1 1
T
denote the corresponding received symbol
vector. Let v(k) be IID Gaussian noise with zero-mean and dB variance = 3 dB. Let
= 1, = 1/2 denote the prior mean, variance of the parameter h. The MMSE estimate
of the channel coefficient h is,
a. 15/23
b. 20/23
c. 21/23
d. 17/23
Ans a
10. Consider the fading channel estimation problem where the output symbol y(k) is y(k) =
hx(k) + v(k), with h, x(k), v(k) denoting the real channel coefficient, pilot symbol and
T
1 3
noise sample respectively. Let x 2 1 denote the pilot vector of transmitted
2 2
pilot symbols and y 2 2 1 1
T
denote the corresponding received symbol
vector. Let v(k) be IID Gaussian noise with zero-mean and variance . Let = 1, =
1/2 denote the prior mean, variance of the parameter h. As the noise variance ,
the MMSE estimate of the channel coefficient h becomes,
a. 15/23
b. 7/4
c. 1
d. 0
Ans c
NPTEL MOOC Estimation: Assignment #3
1. Consider MMSE estimation for the wireless sensor network (WSN) scenario as described
in lectures with each observation y(k) = h + v(k), for 1 k N, i.e. number of
observations N and IID Gaussian noise samples of variance 2. What is the expression for
the MSE (Mean Squared Error) of the ML estimate ĥ of the unknown Gaussian
parameter h
a)
b)
c)
d)
Ans (a)
2. Consider MMSE estimation for the wireless sensor network (WSN) scenario as described
in lectures with each observation y(k) = h + v(k), for 1 k N, i.e. number of
observations N and IID Gaussian noise samples of variance 2. What is the expression for
the MMSE i.e. Minimum Mean Squared Error in the estimate ĥ of the unknown
Gaussian parameter h
a)
b)
c)
d) None of these
e)
Ans (b)
3. Consider MMSE estimation for the wireless sensor network (WSN) scenario as described
in lectures with each observation y(k) = h + v(k), for 1 k 5, i.e. number of observations
N = 5 and IID Gaussian noise samples of dB variance 2 = - 3 dB i.e. 10log102 = -3. Let
the observations be y(1) = 1, y(2) = 1, y(3) = 2, y(4) = 3/2, y(5) = 5/2. Let mu_h = ½ and
\sigma_h^2 = ¼, h be Gaussian. The MSEs of the ML and MMSE estimates are given as,
a) 1/14,1/10
b) 1/10,1/14
c) 1/10,1/4
d) 1/10,1/8
Ans (b)
4. Consider MMSE estimation for the wireless sensor network (WSN) scenario as described
in lectures with each observation y(k) = h + v(k), for 1 k N, i.e. number of observations
is N and IID Gaussian noise samples of dB variance 2 = - 3 dB i.e. 10log102 = -3. Let
\sigma_h^2 = 1/4, h be Gaussian. What is the minimum number of observations N
required such that the true parameter h is within \sigma/4 of the estimate \hat{h} with
probability greater than 99.9%?
a) 172
b) 170
c) 171
d) 174
Ans (a)
5. Given the fading channel estimation problem where the output symbol y(k) is y(k) = hx(k)
+ v(k), with h, x(k), v(k) denoting the real channel coefficient, pilot symbol and noise
sample respectively. Let x = [x(1),x(2),…,x(N)]T denote the pilot vector of transmitted
pilot symbols and y = [y(1), y(2),…, y(N)]T denote the corresponding received symbol
vector. Let v(k) be IID Gaussian noise with zero-mean and variance 2. Let mu_h,
\sigma_h^2 denote the prior mean, variance of the Gaussian parameter h. What is the
expression for the MSE (Mean Squared Error) of the ML estimate ĥ of the unknown
parameter h
a)
b)
c)
d)
Ans (d)
6. Given the fading channel estimation problem where the output symbol y(k) is y(k) = hx(k)
+ v(k), with h, x(k), v(k) denoting the real channel coefficient, pilot symbol and noise
sample respectively. Let x = [x(1),x(2),…,x(N)]T denote the pilot vector of transmitted
pilot symbols and y = [y(1), y(2),…, y(N)]T denote the corresponding received symbol
vector. Let v(k) be IID Gaussian noise with zero-mean and variance 2. Let mu_h,
\sigma_h^2 denote the prior mean, variance of the Gaussian parameter h. What is the
expression for the MMSE i.e. Minimum Mean Squared Error in the estimate ĥ of the
unknown parameter h
a)
b)
c)
d)
Ans (b)
7. Given the fading channel estimation problem where the output symbol y(k) is y(k) = hx(k)
+ v(k), with h, x(k), v(k) denoting the real channel coefficient, pilot symbol and noise
sample respectively. Let x = [x(1),x(2),…,x(N)]T denote the pilot vector of transmitted
pilot symbols and y = [y(1), y(2),…, y(N)]T denote the corresponding received symbol
vector. Let v(k) be IID Gaussian noise with zero-mean and variance 2. Let mu_h,
\sigma_h^2 denote the prior mean, variance of the Gaussian parameter h. As the number
of samples N becomes very large, what is the expression for the MMSE i.e. Minimum
Mean Squared Error in the estimate ĥ of the unknown parameter h
a)
b)
c)
d) None of these
Ans (b)
8. Consider the fading channel estimation problem where the output symbol y(k) is y(k) =
hx(k) + v(k), with h, x(k), v(k) denoting the real channel coefficient, pilot symbol and
T
1 3
noise sample respectively. Let x 2 1 denote the pilot vector of transmitted
2 2
pilot symbols and y 2 2 1 1
T
denote the corresponding received symbol
vector. Let v(k) be IID Gaussian noise with zero-mean and dB variance 2 = 3 dB. Let
mu_h = 1, \sigma_h^2 = 1/2 denote the prior mean, variance of the Gaussian parameter h.
The MSEs of the ML and MMSE estimates are given as,
a) 4/15,4/23
b) 4/11,4/19
c) 4/19,4/11
d) 4/23,4/15
Ans (a)
9. Given the fading channel estimation problem where the output symbol y(k) is y(k) = hx(k)
+ v(k), with h, x(k), v(k) denoting the real channel coefficient, pilot symbol and noise
sample respectively. Let x = [x(1),x(2),…,x(N)]T denote the pilot vector of transmitted
pilot symbols and y = [y(1), y(2),…, y(N)]T denote the corresponding received symbol
vector. Let v(k) be IID Gaussian noise with zero-mean and variance 2. Let mu_h,
\sigma_h^2 denote the prior mean, variance of the Gaussian parameter h. What is the
minimum number of observations N required such that the true parameter h is within
\sigma/8 of the estimate \hat{h} with probability greater than 99.99%? Assume the pilot
symbols are constant modulus with |x(k)|^2 = 4.
a)
b)
c)
d)
Ans (d)
10. Consider the fading channel estimation problem where the output symbol y(k) is y(k) =
hx(k) + v(k), with h, x(k), v(k) denoting the complex channel coefficient, pilot symbol and
noise sample respectively. Let x 2 j 1 j 1 2 j 1 j
T
denote the pilot
vector of transmitted pilot symbols and y 2 2 j j 2 j 1 j denote the
T
corresponding received symbol vector. Let v(k) be IID Gaussian noise with zero-mean
and dB variance 2 = 3 dB. Let mu_h = 1 + j, \sigma_h^2 = 1/2 denote the prior mean,
variance of the complex symmetric Gaussian parameter h. The MMSE estimate of ĥ is,
a) 1/12
b) 1/7
c) 1/9
d) None of these
Ans (c)
NPTEL MOOC Estimation: Assignment #4
1. Consider the fading channel estimation problem where the output symbol y(k) is y(k) =
hx(k) + v(k), with h, x(k), v(k) denoting the complex channel coefficient, pilot symbol and
noise sample respectively. Let x 2 j 1 j 1 2 j 1 j T denote the pilot
vector of transmitted pilot symbols and y 2 2 j j 2 j 1 j T denote the
corresponding received symbol vector. Let v(k) be IID Gaussian noise with zero-mean
and dB variance 2 = 3 dB. Let mu_h = 1 + j, \sigma_h^2 = 1/2 denote the prior mean,
variance of the complex symmetric Gaussian parameter h. The MSE of the MMSE
estimate of ĥ is,
a. 1/9
b. 1/8
c. 1/10
d. 1/7
Ans a
2. Consider the fading channel estimation problem where the output symbol y(k) is y(k) =
hx(k) + v(k), with h, x(k), v(k) denoting the complex channel coefficient, pilot symbol and
noise sample respectively. Let x 2 j 1 j 1 2 j 1 j
T
denote the pilot
vector of transmitted pilot symbols and y 2 2 j j 2 j 1 j denote the
T
corresponding received symbol vector. Let v(k) be IID Gaussian noise with zero-mean
and dB variance 2 = 3 dB. Let mu_h = 1 + j, \sigma_h^2 = 1/2 denote the prior mean,
variance of the complex symmetric Gaussian parameter h. The MSEs of the real,
imaginary parts of ĥ are,
a. 1/2
b. 1/9
c. 1/18
d. 1/4
Ans c
3. Consider the fading channel estimation problem where the output symbol y(k) is y(k) =
hx(k) + v(k), with h, x(k), v(k) denoting the real channel coefficient, pilot symbol and
noise sample respectively. Let x 2 j 1 j 1 2 j 1 j
T
denote the pilot
vector of transmitted pilot symbols and y 2 2 j j 2 j 1 j denote the
T
corresponding received symbol vector. Let v(k) be IID Gaussian noise with zero-mean
and dB variance 2 = 3 dB. Let mu_h = 1 + j, \sigma_h^2 = 1/2 denote the prior mean,
variance of the parameter h. What is the probability that both the real and imaginary parts
of the MMSE estimate ĥ lie within a radius of ½ from the real and imaginary parts of the
unknown complex symmetric Gaussian parameter h respectively
a.
b.
c.
d.
Ans b
4. Consider MMSE estimation for the wireless sensor network (WSN) scenario as described
in lectures with each observation y(k) = h + v(k), for 1 k N, i.e. number of
observations N and independent non-identical Gaussian noise samples v(k) of variance
k2 . What is the expression for the MMSE estimate ĥ of the unknown Gaussian
parameter h
a.
b.
c.
d.
Ans d
5. Consider MMSE estimation for the wireless sensor network (WSN) scenario as described
in lectures with each observation y(k) = h + v(k), for 1 k N, i.e. number of
observations N and independent non-identical Gaussian noise samples v(k) of variance
k2 . What is the expression for the MSE of the MMSE estimate ĥ of the unknown
Gaussian parameter h
a.
b.
c.
d.
Ans a
7. Consider LMMSE estimation for the wireless sensor network (WSN) scenario as
described in lectures with each observation y(k) = h + v(k), for 1 k N, i.e. number of
observations N and IID Gaussian noise samples of variance 2. What is the expression for
the LMMSE estimate ĥ of the unknown parameter h, which is not necessarily Gaussian
a.
b.
c.
d.
Ans d
8. Consider a multi-antenna channel estimation scenario with N = 4 pilot vectors x(1) = [1,
1]T, x(2) = [1, -2]T, x(3) = [2, 1]T, x[4] = [1, -1]T. Let the corresponding received vector
be y = [2, 1, 1, 3]T. Let the channel coefficients be IID Gaussian with dB variance
\sigma_h^2 = -3 dB and dB noise variance \sigma^2 = 3 dB. The number of antennas M
in this scenario is,
a. 1
b. 2
c. 3
d. 4
Ans b
9. Consider a multi-antenna channel estimation scenario with N = 4 pilot vectors x(1) = [1,
1]T, x(2) = [1, -2]T, x(3) = [2, 1]T, x[4] = [1, -1]T. Let the corresponding received vector
be y = [2, 1, 1, 3]T. Let the channel coefficients be IID Gaussian with dB variance
\sigma_h^2 = -3 dB and dB noise variance \sigma^2 = 3 dB. The corresponding pilot
matrix X is
1 1 2 1
a.
1 2 1 1
1
1
2
1
b.
1
2
1
1
c. 1 1 2 1 1 2 1 1
1 1
1 2
d.
2 1
1 1
Ans d
10. Consider a multi-antenna channel estimation scenario with N = 4 pilot vectors x(1) = [1,
1]T, x(2) = [1, -2]T, x(3) = [2, 1]T, x[4] = [1, -1]T. Let the corresponding received vector
be y = [2, 1, 1, 3]T. Let the channel coefficients be IID Gaussian with dB variance
\sigma_h^2 = -3 dB and dB noise variance \sigma^2 = 3 dB. The pilot matrix X for this
scenario satisfies the property that
a. It is invertible
b. It has identical columns
c. It has orthogonal columns
d. None of the above
Ans c
NPTEL MOOC Estimation: Assignment #5
1. Consider the LMMSE (Linear Minimum Mean Squared Error) multi-antenna channel
estimation problem with N transmitted pilot vectors x(k) = [x1(k), x2(k),…,xM(k)]T, 1 k
N and N received symbols y(1), y(2), ..., y(N). Let the channel vector be h = [h1, h2, …,
hM]T. Let the channel coefficients be IID Gaussian with variance and noise variance is
. The optimization problem for MMSE estimation of h is,
a. min y Xh
b. min y Xh
2
c. min E
2
hˆ h , where hˆ c y
T
d. min E y Xh
2
Ans c
2. Consider the LMMSE (Linear Minimum Mean Squared Error) multi-antenna channel
estimation problem with N transmitted pilot vectors x(k) = [x1(k), x2(k),…,xM(k)]T, 1 k
N and N received symbols y(1), y(2), ..., y(N). Let the channel vector be h = [h1, h2, …,
hM]T. Let the channel coefficients be IID Gaussian with variance and noise variance is
. The LMMSE estimate of h is,
a.
b.
c.
d.
Ans a
3. Consider the LMMSE (Linear Minimum Mean Squared Error) multi-antenna channel
estimation problem with N transmitted pilot vectors x(k) = [x1(k), x2(k),…,xM(k)]T, 1 k
N and N received symbols y(1), y(2), ..., y(N). Let the channel vector be h = [h1, h2, …,
hM]T. Let the channel coefficients be IID Gaussian with variance and noise variance is
. The LMMSE estimates of h at high and low SNRs are approximately,
a. ,
b. ,
c. ,
d. ,
Ans b
4. Consider the LMMSE (Linear Minimum Mean Squared Error) multi-antenna channel
estimation problem with N transmitted pilot vectors x(k) = [x1(k), x2(k),…,xM(k)]T, 1 k
N and N received symbols y(1), y(2), ..., y(N). Let the channel vector be h = [h1, h2, …,
hM]T. Let the channel coefficients be IID Gaussian with variance and noise variance is
. The error covariance for the LMMSE estimate of h is,
a.
b.
c.
d.
Ans c
5. Consider a multi-antenna channel estimation scenario with N = 4 pilot vectors x(1) = [1,
1]T, x(2) = [1, -2]T, x(3) = [2, 1]T, x[4] = [1, -1]T. Let the corresponding received vector
be y = [2, 1, 1, 3]T. Let the channel coefficients be IID Gaussian with dB variance =0
dB and dB noise variance = -3 dB. The LMMSE estimate of h is,
a.
b.
c.
d.
Ans d
6. Consider a multi-antenna channel estimation scenario with N = 4 pilot vectors x(1) = [1,
1]T, x(2) = [1, -2]T, x(3) = [2, 1]T, x[4] = [1, -1]T. Let the corresponding received vector
be y = [2, 1, 1, 3]T. Let the channel coefficients be IID Gaussian with dB variance =0
dB and dB noise variance = -3 dB. The ML estimate of h is,
a.
b.
c.
d.
Ans b
7. Consider a multi-antenna channel estimation scenario with N = 4 pilot vectors x(1) = [1,
1]T, x(2) = [1, -2]T, x(3) = [2, 1]T, x[4] = [1, -1]T. Let the corresponding received vector
be y = [2, 1, 1, 3]T. Let the channel coefficients be IID Gaussian with dB variance =0
dB and dB noise variance = -3 dB. The error covariance of the LMMSE estimate of h
is,
a.
b.
c.
d.
Ans a
8. Consider a multi-antenna channel estimation scenario with N = 4 pilot vectors x(1) = [1,
1]T, x(2) = [1, -2]T, x(3) = [2, 1]T, x[4] = [1, -1]T. Let the corresponding received vector
be y = [2, 1, 1, 3]T. Let the channel coefficients be IID Gaussian and dB noise variance
= -3 dB. Let >> I. The error covariance of the LMMSE estimate of h
reduces to,
a.
b.
c.
d.
Ans a
9. Consider the MIMO channel estimation problem with N = 4 pilot vectors x(1) = [1, 1]T,
x(2) = [1, -2]T, x(3) = [2, 1]T, x[4] = [1, -1]T. The received output vectors y are y(1) = [1,
-1, 1]T, y(2) = [2, -2, 1]T, y(3) = [-2, 1, -2]T, y(4) = [1, 2, -1]T . Let the channel coefficients
be IID Gaussian with dB variance = 0 dB and dB noise variance = -3 dB. The size
of the MIMO system is,
a. 4x3
b. 4x2
c. 2x3
d. None of these
Ans c
10. Consider the MIMO channel estimation problem with N = 4 pilot vectors x(1) = [1, 1]T,
x(2) = [1, -2]T, x(3) = [2, 1]T, x[4] = [1, -1]T. The received output vectors y are y(1) = [1,
-1, 1]T, y(2) = [2, -2, 1]T, y(3) = [-2, 1, -2]T, y(4) = [1, 2, -1]T . Let the channel coefficients
be IID Gaussian with dB variance = 0 dB and dB noise variance = -3 dB. The
covariance matrices of the channel matrix and of the noise matrix respectively
are,
a. ,
b.
c. ,
d.
Ans a
NPTEL MOOC Estimation: Assignment #6
1. Consider the LMMSE (Linear Minimum Mean Squared Error) MIMO channel estimation
problem with N transmitted pilot vectors x(k) = [x1(k), x2(k),…,xM(k)]T, 1 k N and N
received symbol vectors y(k) = [y1(k), y2(k),…,yR(k)]T, 1 k N. Let the channel
coefficients be IID Gaussian with variance \sigma_h^2 and noise variance is \sigma^2.
The expression for the LMMSE estimate of the MIMO channel matrix H is,
a.
b.
c.
d.
Ans a
2. Consider the MIMO channel estimation problem with N transmitted pilot vectors x(k) =
[x1(k), x2(k),…,xM(k)]T, 1 k N and N received symbol vectors y(k) = [y1(k),
y2(k),…,yR(k)]T, 1 k N. Let the channel coefficients be IID Gaussian with variance
\sigma_h^2 and noise variance is \sigma^2. The expression for the ML estimate of the
MIMO channel matrix H is, [Hint: This can be obtained by assuming \sigma_h^2 X^TX
>> \sigma^2 I],
a.
b.
c.
d.
Ans b
3. Consider the LMMSE (Linear Minimum Mean Squared Error) MIMO channel estimation
problem with N transmitted pilot vectors x(k) = [x1(k), x2(k),…,xM(k)]T, 1 k N and N
received symbol vectors y(k) = [y1(k), y2(k),…,yR(k)]T, 1 k N. Let the channel
coefficients be IID Gaussian with variance \sigma_h^2 and noise variance is \sigma^2.
The expression for the error covariance of the LMMSE estimate of the MIMO channel
matrix H is,
a.
b.
c.
d.
Ans d
4. Consider the MIMO channel estimation problem with N transmitted pilot vectors x(k) =
[x1(k), x2(k),…,xM(k)]T, 1 k N and N received symbol vectors y(k) = [y1(k),
y2(k),…,yR(k)]T, 1 k N. Let the channel coefficients be IID Gaussian with variance
\sigma_h^2 and noise variance is \sigma^2. The expression for the covariance of the ML
estimate of the MIMO channel matrix H is, [Hint: This can be obtained by assuming
\sigma_h^2 X^TX >> \sigma^2 I]
a.
b.
c.
d.
Ans c
5. Consider the MIMO channel estimation problem with N = 4 pilot vectors x(1) = [1, 1]T,
x(2) = [1, -2]T, x(3) = [2, 1]T, x[4] = [1, -1]T. The received output vectors y are y(1) = [1,
-1, 1]T, y(2) = [2, -2, 1]T, y(3) = [-2, 1, -2]T, y(4) = [1, 2, -1]T . Let the channel coefficients
be IID Gaussian with dB variance \sigma_h^2 = 0 dB and dB noise variance \sigma^2 = -
3 dB. The LMMSE estimate of the MIMO channel matrix H is,
a.
b.
c.
d.
Ans a
6. Consider the MIMO channel estimation problem with N = 4 pilot vectors x(1) = [1, 1]T,
x(2) = [1, -2]T, x(3) = [2, 1]T, x[4] = [1, -1]T. The received output vectors y are y(1) = [1,
-1, 1]T, y(2) = [2, -2, 1]T, y(3) = [-2, 1, -2]T, y(4) = [1, 2, -1]T . Let the channel coefficients
be IID Gaussian with dB variance \sigma_h^2 = 0 dB and dB noise variance \sigma^2 = -
3 dB. The ML estimate of the MIMO channel matrix H is,
a.
b.
c.
d.
Ans b
7. Consider the MIMO channel estimation problem with N = 4 pilot vectors x(1) = [1, 1]T,
x(2) = [1, -2]T, x(3) = [2, 1]T, x[4] = [1, -1]T. The received output vectors y are y(1) = [1,
-1, 1]T, y(2) = [2, -2, 1]T, y(3) = [-2, 1, -2]T, y(4) = [1, 2, -1]T . Let the channel coefficients
be IID Gaussian with dB variance \sigma_h^2 = 0 dB and dB noise variance \sigma^2 = -
3 dB. The error covariance of the LMMSE estimate of the MIMO channel matrix H is,
a.
b.
c.
d.
Ans c
9. Consider an Inter Symbol Interference channel yk xk xk 1 vk . Let an r
3 1
2 2
= 3 tap channel equalizer be designed for this scenario based on symbols y(k+2), y(k+1),
y(k) to detect x(k). What is the effective channel matrix H for this scenario
3 1
2 2
a. H
1 3
2 2
3 1
0
b. H 2 2
3 1
0
2 2
3 1
2 0 0
2
3 1
c. H 0 0
2 2
0 0
3
1
2 2
3 1
0 0
2 2
3 1
d. H 0 0
2 2
3
1
0 0
2 2
Ans c
10. Consider the LMMSE (Linear Minimum Mean Squared Error) multi-antenna channel
estimation problem with N transmitted pilot vectors x(k) = [x1(k), x2(k),…,xM(k)]T, 1 k
N and N received symbols y(1), y(2), ..., y(N). Let the channel vector be h = [h1, h2, …,
hM]T. Let the channel covariance E{hh^T} = R_h and noise covariance E{vv^T} = R_v.
The expressions for the LMMSE estimate, error covariance matrix of h respectively are,
a. ,
b.
c.
d. ,
Ans a
NPTEL MOOC Estimation: Assignment #7
1. Consider an Inter Symbol Interference channel yk xk xk 1 vk . Let an r
3 1
2 2
= 2 tap channel equalizer be designed for this scenario based on symbols y(k+1), y(k) to
detect x(k). Let symbols x(k) be IID zero-mean with dB Power P_d = 10 dB and dB noise
variance \sigma^2 = 3 dB. What is the effective channel matrix H for this scenario
a.
b.
c.
d.
Ans (b)
2. Consider an Inter Symbol Interference channel yk xk xk 1 vk . Let an r
3 1
2 2
= 2 tap channel equalizer be designed for this scenario based on symbols y(k+1), y(k) to
detect x(k). Let symbols x(k) be IID zero-mean with dB Power P_d = 10 dB and dB noise
variance \sigma^2 = 3 dB. What are the covariance matrices of the input, noise vectors
x(k), v(k) respectively for this scenario
a. 10I3X3,2I2X2
b. 10I3X3,3I2X2
c. 10I2X2,2I3X3
d. None of these
Ans (a)
3. Consider an Inter Symbol Interference channel yk xk xk 1 vk . Let an r
3 1
2 2
= 2 tap channel equalizer be designed for this scenario based on symbols y(k+1), y(k) to
detect x(k). Let symbols x(k) be IID zero-mean with dB Power P_d = 10 dB and dB noise
variance \sigma^2 = 3 dB. What is the LMMSE equalizer vector c ?
a.
b.
c.
d.
Ans(c)
4. Consider an Inter Symbol Interference channel yk xk xk 1 vk . Let an r
3 1
2 2
= 2 tap channel equalizer be designed for this scenario based on symbols y(k+1), y(k) to
detect x(k). Let symbols x(k) be IID zero-mean with dB Power P_d = 10 dB and dB noise
variance \sigma^2 = 3 dB. What is the resulting LMMSE equalizer ?
a. +
b. +
c. +
d. +
Ans(d)
5. Consider an Inter Symbol Interference channel yk xk xk 1 vk . Let an r
3 1
2 2
= 2 tap channel equalizer be designed for this scenario based on symbols y(k+1), y(k) to
detect x(k). Let symbols x(k) be IID zero-mean with dB Power P_d = 10 dB and dB noise
variance \sigma^2 = 3 dB. What is the MSE of LMMSE equalization ?
a. 1.638
b. 2
c. 1.75
d. None of these
Ans (a)
6. Consider an Inter Symbol Interference channel yk xk xk 1 vk . Let an r
3 1
2 2
= 2 tap channel equalizer be designed for this scenario based on symbols y(k+1), y(k) to
detect x(k + 1) instead of x(k). Let symbols x(k) be IID zero-mean with dB Power P_d =
10 dB and dB noise variance \sigma^2 = 3 dB. What is the LMMSE equalizer vector c ?
a.
b.
c.
d. None of the above
Ans (a)
7. Consider an Inter Symbol Interference channel yk xk xk 1 vk . Let an r
3 1
2 2
= 2 tap channel equalizer be designed for this scenario based on symbols y(k+1), y(k) to
detect x(k + 1) instead of x(k). Let symbols x(k) be IID zero-mean with dB Power P_d =
10 dB and dB noise variance \sigma^2 = 3 dB. What is the MSE of LMMSE
equalization?
a. 0.875
b. 1.75
c. 0.9699
d. 0.75
Ans (c)
8. OFDM is a technology which is used in
a. 4G LTE
b. 3G HSDPA
c. 2G GSM
d. All of the above
Ans a
9. The acronym OFDM stands for
a. Optimal Frequency Diversity Module
b. Orthogonal Fourier Dispersion Module
c. Optimal Fourier Duplex Multiplexing
d. Orthogonal Frequency Division Multiplexing
Ans d
10. For an L-tap channel, what is the minimum length of cyclic prefix needed to lead to a
circular convolution of the channel and input at the receiver?
a. L
b. L-1
c. L+1
L
d.
2
Ans b
NPTEL MOOC Estimation: Assignment #8
1. Consider an N = 4 subcarrier OFDM system with pilot symbols X(0) = 3+3j, X(1) = -2-2j,
X(2) = -1+j, X(3) = 2-j. Let the corresponding received samples in the time domain be
y(0) = -1-j, y(1) = 2+2j, y(2) = 3-2j, y(3) = 3-2j. Let the noise samples v(k), 0 k 3 be
zero-mean IID Gaussian with variance 2 = -3 dB. There are L = 2 IID channel taps with
zero-mean and dB variance \sigma_h^2 = -6 dB. The length of cyclic prefix is the
minimum required. The transmitted block of samples in the time domain with cyclic
prefix is,
a. , , ,
b. , , , ,
c. , , ,
d. , , ,
Ans c
2. Consider an N = 4 subcarrier OFDM system with pilot symbols X(0) = 3+3j, X(1) = -2-2j,
X(2) = -1+j, X(3) = 2-j. Let the corresponding received samples in the time domain be
y(0) = -1-j, y(1) = 2+2j, y(2) = 3-2j, y(3) = 3-2j. Let the noise samples v(k), 0 k 3 be
zero-mean IID Gaussian with variance 2 = -3 dB. There are L = 2 IID channel taps with
zero-mean and dB variance \sigma_h^2 = -6 dB. The length of cyclic prefix is the
minimum required. The received output symbols across the subcarriers in the frequency
domain are,
a. 7+3j, 2j, -3-3j, -8
b. 7-3j, 2j, -3-3j, -8
c. -7+3j, -2j, 3+3j, 8
d. 7-3j, 2j, -3+3j, -8
Ans b
3. Consider an N = 4 subcarrier OFDM system with pilot symbols X(0) = 3+3j, X(1) = -2-2j,
X(2) = -1+j, X(3) = 2-j. Let the corresponding received samples in the time domain be
y(0) = -1-j, y(1) = 2+2j, y(2) = 3-2j, y(3) = 3-2j. Let the noise samples v(k), 0 k 3 be
zero-mean IID Gaussian with variance 2 = -3 dB. There are L = 2 IID channel taps with
zero-mean and dB variance \sigma_h^2 = -6 dB. The length of cyclic prefix is the
minimum required. The LMMSE estimates of the channel coefficients across subcarriers
0, 1 respectively are,
a. ,
b. ,
c. ,
d. ,
Ans d
4. Consider an N = 4 subcarrier OFDM system with pilot symbols X(0) = 3+3j, X(1) = -2-2j,
X(2) = -1+j, X(3) = 2-j. Let the corresponding received samples in the time domain be
y(0) = -1-j, y(1) = 2+2j, y(2) = 3-2j, y(3) = 3-2j. Let the noise samples v(k), 0 k 3 be
zero-mean IID Gaussian with variance 2 = -3 dB. There are L = 2 IID channel taps with
zero-mean and dB variance \sigma_h^2 = -6 dB. The length of cyclic prefix is the
minimum required. The LMMSE estimates of the channel coefficients across subcarriers
2, 3 respectively are,
a. ,
b. ,
c. ,
d. ,
Ans a
5. Consider an N = 4 subcarrier OFDM system with pilot symbols X(0) = 3+3j, X(1) = -2-2j,
X(2) = -1+j, X(3) = 2-j. Let the corresponding received samples in the time domain be
y(0) = -1-j, y(1) = 2+2j, y(2) = 3-2j, y(3) = 3-2j. Let the noise samples v(k), 0 k 3 be
zero-mean IID Gaussian with variance 2 = -3 dB. There are L = 2 IID channel taps with
zero-mean and dB variance \sigma_h^2 = -6 dB. The length of cyclic prefix is the
minimum required. The ML estimates of the channel coefficients across subcarriers 2, 3
are,
a. ,
b. ,
c. ,
d. ,
Ans b
6. Consider an N = 4 subcarrier OFDM system with pilot symbols X(0) = 3+3j, X(1) = -2-2j,
X(2) = -1+j, X(3) = 2-j. Let the corresponding received samples in the time domain be
y(0) = -1-j, y(1) = 2+2j, y(2) = 3-2j, y(3) = 3-2j. Let the noise samples v(k), 0 k 3 be
zero-mean IID Gaussian with variance 2 = -3 dB. There are L = 2 IID channel taps with
zero-mean and dB variance \sigma_h^2 = -6 dB. The length of cyclic prefix is the
minimum required. The MSEs of LMMSE estimation of the channel coefficients across
subcarriers 0, 1, 2, 3 respectively are,
a.
b.
c.
d.
Ans a
7. Consider an N = 4 subcarrier OFDM system with pilot symbols X(0) = 3+3j, X(1) = -2-2j,
X(2) = -1+j, X(3) = 2-j. Let the corresponding received samples in the time domain be
y(0) = -1-j, y(1) = 2+2j, y(2) = 3-2j, y(3) = 3-2j. Let the noise samples v(k), 0 k 3 be
zero-mean IID Gaussian with variance 2 = -3 dB. There are L = 2 IID channel taps with
zero-mean and dB variance \sigma_h^2 = -6 dB. The length of cyclic prefix is the
minimum required. The noise samples V(l) on the lth subcarrier are,
a. Zero-mean, Gaussian, variance ½
b. Zero-mean, Non-Gaussian, variance ½
1
c. Zero-mean, Gaussian, variance
8
d. Zero-mean, Gaussian, variance 2
Ans d
8. Consider an N = 4 subcarrier OFDM system with pilot symbols X(0) = 3+3j, X(1) = -2-2j,
X(2) = -1+j, X(3) = 2-j. Let the corresponding received samples in the time domain be
y(0) = -1-j, y(1) = 2+2j, y(2) = 3-2j, y(3) = 3-2j. Let the noise samples v(k), 0 k 3 be
zero-mean IID Gaussian with variance 2 = -3 dB. There are L = 2 IID channel taps with
zero-mean and dB variance \sigma_h^2 = -6 dB. The length of cyclic prefix is the
minimum required. The channel coefficients H(l) on the lth subcarrier are,
a. Zero-mean, Gaussian, variance ½
b. Zero-mean, variance ½
1
c. Zero-mean, Gaussian, variance
8
d. Zero-mean, variance 2
Ans b
9. Consider an N = 4 subcarrier OFDM system with pilot symbols X(0) = 3+3j, X(1) = -2-2j,
X(2) = -1+j, X(3) = 2-j. Let the corresponding received samples in the time domain be
y(0) = -1-j, y(1) = 2+2j, y(2) = 3-2j, y(3) = 3-2j. Let the noise samples v(k), 0 k 3 be
zero-mean IID Gaussian with variance 2 = -3 dB. There are L = 2 IID channel taps with
zero-mean and dB variance \sigma_h^2 = -6 dB. The length of cyclic prefix is the
minimum required. The LMMSE estimates of the channel taps h(0), h(1) respectively
are,
a. ,
b. ,
c. -
d.
Ans c
10. Consider an N = 4 subcarrier OFDM system with pilot symbols X(0) = 3+3j, X(1) = -2-2j,
X(2) = -1+j, X(3) = 2-j. Let the corresponding received samples in the time domain be
y(0) = -1-j, y(1) = 2+2j, y(2) = 3-2j, y(3) = 3-2j. Let the noise samples v(k), 0 k 3 be
zero-mean IID Gaussian with variance 2 = -3 dB. There are L = 2 IID channel taps with
zero-mean, Gaussian (only for this part) and dB variance \sigma_h^2 = -6 dB. The
length of cyclic prefix is the minimum required. If the block of pilot symbols above is
repeatedly transmitted, what is the number of OFDM pilot blocks M required such that
1
the real and imaginary parts of the estimate Ĥ 1 lie within a radius of the real and
8
imaginary parts of the true parameter H(1) with probability greater than 99.99%
a.
2Q 1 5 106
2
b. Q 5 10
1 6 2
c. Q 5 10
1 5 2
d. Q 5 10
1 5 2
Ans d
26/07/2020 Estimation for Wireless Communications – MIMO/OFDM Cellular and Sensor Networks - - Unit 2 - Week 1 - Basics o…
Courses » Estimation for Wireless Communications – MIMO/OFDM Cellular and Sensor Networks
Course
outline Assignment-1
The due date for submitting this assignment has passed. Due on 2017-08-06, 23:59 IST.
How to Access As per our records you have not submitted this assignment.
the Portal ?
1) In the context of estimation, the probability density function (PDF) of the observations, 1 point
Week 1 - Basics viewed as a function of the unknown parameter ℎ is termed as the
of Estimation,
Maximum Likelihood Function
Likelihood (ML) Estimation Function
Lecture 01 - Objective Function
Basics – Sensor Cost Function
Network and
Noisy No, the answer is incorrect.
Observation Score: 0
Model
Accepted Answers:
Lecture 02 - Likelihood Function
Likelihood
Function and 2) Consider the wireless sensor network (WSN) estimation scenario described in lectures with 1 point
Maximum each observation 𝑦(𝑘) = ℎ + 𝑣(𝑘), for 1 ≤ 𝑘 ≤ 𝑁 , i.e. number of observations is 𝑁 . The ML
Likelihood (ML)
estimate given by the sample mean has the following property.
Estimate
Lecture 03 - It is biased
Properties of Gaussian distributed
Maximum
Likelihood (ML)
1
Estimate – Variance decreases as , where 𝑁 is number of observations
2
Mean and 𝑁
Number of 1 2
∑ 𝑦 (𝑘)
𝑁
Samples 𝑘=1
Required
1
Quiz : 𝑁 𝑁
Assignment-1 ∏ 𝑦(𝑘)
( )
𝑘=1
Assignment-1
Solution
https://onlinecourses-archive.nptel.ac.in/noc17_ee19/unit?unit=86&assessment=93 1/4
26/07/2020 Estimation for Wireless Communications – MIMO/OFDM Cellular and Sensor Networks - - Unit 2 - Week 1 - Basics o…
𝑚𝑖𝑛 {𝑦(𝑘), 1 ≤ 𝑘 ≤ 𝑁 }
Week 2 - Vector
Estimation None of the above
Pseudo-Inverse,
Properties of LS 2
𝜎
Estimate,
𝑁
Examples –
Mullti-Antenna 𝜎
Downlink and 𝑁
MIMO Channel
Estimation 2
𝑁𝜎
Zero-forcing 𝑁
equalizer,
5) Consider the wireless sensor network (WSN) estimation scenario described in lectures with 1 point
Approximation
error of equalizer each observation 𝑦(𝑘) = ℎ + 𝑣(𝑘), for 1 ≤ 𝑘 ≤ 𝑁 , i.e. number of observations is 𝑁 and noise
samples 𝑣(𝑘) are IID Gaussian noise samples with zero mean and variance 𝜎 2 . For this scenario, what
Week 6 - is the mean of the maximum likelihood estimate ℎ̂ of the unknown parameter ℎ
Introduction to
Orthogonal
Frequency
ℎ
Division
Multiplexing
(OFDM) and Pilot 𝑁ℎ
Based OFDM
Channel ℎ
Estimation, 𝑁
Example
1
Week 7 - OFDM – 𝑁
Week 8 - 1
∑ (𝑦(𝑘) − 𝜃)
2
𝑁
Sequential Least 𝑘=1
Squares (SLS)
Estimation – 𝑁
Scalar/ Vector 1
∑ (𝑦(𝑘) − 𝜃)
Cases, 𝑁
𝑘=1
Applications -
Wireless Fading
Channel 1
𝑁
Estimation, SLS 𝑁
∑ (𝑦(𝑘) + 𝜃)
Example 𝑘=1
https://onlinecourses-archive.nptel.ac.in/noc17_ee19/unit?unit=86&assessment=93 2/4
26/07/2020 Estimation for Wireless Communications – MIMO/OFDM Cellular and Sensor Networks - - Unit 2 - Week 1 - Basics o…
𝑁
1
∑ 𝑦(𝑘)
𝑁
𝑘=1
7) Consider now a slightly modified version of the wireless sensor network (WSN) estimation 1 point
scenario described in class with each observation 𝑦(𝑘) = ℎ + 𝑣(𝑘), for 1 ≤ 𝑘 ≤ 𝑁 . Let the noise
samples be IID Gaussian with mean 𝜃 and variance 𝜎 2 each. What is mean of the maximum likelihood
estimate ℎ̂ ?
𝜃
ℎ +
𝑁
ℎ
+ 𝜃
𝑁
ℎ + 𝜃
8) Consider now a slightly modified version of the wireless sensor network (WSN) estimation 1 point
scenario described in class with each observation 𝑦(𝑘) = ℎ + 𝑣(𝑘), for 1 ≤ 𝑘 ≤ 𝑁 . Let the noise
samples be lID Gaussian with mean 𝜃 and variance 𝜎 2 each. What is variance of the maximum
likelihood estimate ℎ̂ ?
2 2
𝜎 + 𝜃
2
𝜎 2
+ 𝜃
𝑁
2
𝜎
2
𝜎
+ 𝜃
𝑁
9) Consider now a slightly modified version of the wireless sensor network (WSN) estimation 1 point
scenario described in class with each observation 𝑦(𝑘) = ℎ + 𝑣(𝑘), for 1 ≤ 𝑘 ≤ 𝑁 . Let the noise
samples be IID Gaussian with mean 𝜃 and variance 𝜎 2 each. What is the distribution of the maximum
likelihood estimate ℎ̂ ?
Uniform
Exponential
Rayleigh
None of the above
https://onlinecourses-archive.nptel.ac.in/noc17_ee19/unit?unit=86&assessment=93 3/4
26/07/2020 Estimation for Wireless Communications – MIMO/OFDM Cellular and Sensor Networks - - Unit 2 - Week 1 - Basics o…
Accepted Answers:
None of the above
10)Consider now a slightly modified version of the wireless sensor network (WSN) estimation 1 point
scenario described in class with each observation 𝑦(𝑘) = ℎ + 𝑣(𝑘), for 1 ≤ 𝑘 ≤ 𝑁 . Let the noise
samples be IID Gaussian with mean 𝜃 and variance 𝜎 2 = −6𝑑𝐵 i.e. 10 log10 𝜎 2 = −6 . What is the
number of observations 𝑁 required such that the probability that the maximum likelihood estimate ℎ̂ lies
1
within a radius of of the unknown parameter ℎ is greater than 99.9% ? Let Q denote the Gaussian Q
16
−1 −3 2
(2√2
‾𝑄 (𝜃 + 10 ))
−1 −4 2
(8𝑄 (5 × 10 ))
−1 −4
8𝑄 (5 × 10 )
−1 −4
𝜃 + 2𝑄 (5√2
‾ × 10 )
Funded by
Powered by
https://onlinecourses-archive.nptel.ac.in/noc17_ee19/unit?unit=86&assessment=93 4/4
26/07/2020 Estimation for Wireless Communications – MIMO/OFDM Cellular and Sensor Networks - - Unit 3 - Week 2 - Vector E…
Courses » Estimation for Wireless Communications – MIMO/OFDM Cellular and Sensor Networks
Course
outline Assignment-2
The due date for submitting this assignment has passed. Due on 2017-08-06, 23:59 IST.
How to Access As per our records you have not submitted this assignment.
the Portal ?
1) Consider the fading channel estimation problem where the output symbol 𝑦(𝑘) 1 point
Week 1 - Basics is 𝑦(𝑘) = ℎ𝑥(𝑘) + 𝑣(𝑘) with ℎ, 𝑥(𝑘), 𝑣(𝑘) denoting the real channel coefficient, pilot
of Estimation,
Maximum
symbol and noise sample respectively. Let 𝐱 = [𝑥(1), 𝑥(2), … , 𝑥(𝑁 )]𝑇 denote the pilot
Likelihood (ML) vector of transmitted pilot symbols and 𝐲 = [𝑦(1), 𝑦(2), … , 𝑦(𝑁 )]𝑇 denote the
corresponding received symbol vector. Let 𝑣(𝑘) be IID Gaussian noise with zero-mean
Week 2 - Vector
Estimation
and variance 𝜎 2 . The expected value of the maximum likelihood ˆ
ℎ is,
Lecture 06 -
𝑇
Estimation of 𝐱 𝐲
Complex
Parameters – 𝑇
𝐱 𝐲
Symmetric Zero 𝑇
𝐱 𝐱
Mean Complex
Gaussian Noise
𝑇
ℎ𝐱 𝐲
Lecture 07 -
Wireless Fading
Channel ℎ
Estimation –
Pilot Symbols No, the answer is incorrect.
and Likelihood Score: 0
Function
Accepted Answers:
Lecture 08 - ℎ
Wireless Fading
Channel 2) Consider the fading channel estimation problem where the output symbol 𝑦(𝑘) 1 point
Estimation – is 𝑦(𝑘) = ℎ𝑥(𝑘) + 𝑣(𝑘) , with ℎ, 𝑥(𝑘), 𝑣(𝑘) denoting the real channel coefficient, pilot
Pilot Training 𝑇
symbol and noise sample respectively. Let 𝐱 denote the pilot vector of
3 1
based = [ , 1, , −2]
2 2
Maximum 𝑇
Likelihood ML transmitted pilot symbols and 𝐲 = [2, − 12 , −1, 32 ] denote the corresponding received
Estimate symbol vector. Let 𝑣(𝑘) be IID Gaussian noise with zero-mean and dB variance
= −3𝑑𝐵 . The maximum likelihood estimate of the channel coefficient ℎ is,
2
Lecture 09 - 𝜎
Wireless Fading
Channel
Estimation – 1
Mean and 15
Variance of Pilot
Training Based 2
Channel 15
Estimate
Maximum 1
−
Likelihood 15
Lecture 10 - 2
−
Example – 15
Wireless Fading
https://onlinecourses-archive.nptel.ac.in/noc17_ee19/unit?unit=95&assessment=140 1/5
26/07/2020 Estimation for Wireless Communications – MIMO/OFDM Cellular and Sensor Networks - - Unit 3 - Week 2 - Vector E…
Channel No, the answer is incorrect.
Estimation for Score: 0
Downlink Mobile
Communication Accepted Answers:
2
−
Quiz : 15
Assignment-2
3) Consider the fading channel estimation problem where the output symbol𝑦(𝑘) is 1 point
Assignment-2 𝑦(𝑘) = ℎ𝑥(𝑘) + 𝑣(𝑘) , with ℎ, 𝑥(𝑘), 𝑣(𝑘) denoting the real channel coefficient, pilot
Solution 𝑇
symbol and noise sample respectively. Let 𝐱 =
[
3
2
, 1,
1
2
, −2] denote the pilot vector of
Week 3 - Cramer- 𝑇
Rao Bound transmitted pilot symbols and 𝐲 = [2, − , −1, ] denote the corresponding received
1 3
2 2
(CRB), Vector symbol vector. Let 𝑣(𝑘) be IID Gaussian noise with zero-mean and dB variance
Parameter
= −3𝑑𝐵 . The variance of the maximum likelihood estimate of the channel coefficient
2
Estimation, 𝜎
Multi-Antenna ℎ is,
Downlink Mobile
Channel
Estimation 1
15
Week 4 - Least
Squares (LS) 2
15
Principle,
Pseudo-Inverse,
Properties of LS 3
15
Estimate,
Examples –
Mullti-Antenna 4
15
Downlink and
MIMO Channel No, the answer is incorrect.
Estimation Score: 0
Estimation,
Example −
4
−
6
𝑗
17 17
Week 7 - OFDM –
8 4
Comb Type Pilot 17
−
17
𝑗
(CTP)
Transmission, No, the answer is incorrect.
Channel Score: 0
Estimation in
Time/ Frequency Feedback:
Domain, CTP
Example,
Accepted Answers:
Frequency 8 4
Domain 17
+
17
𝑗
Equalization
(FDE), Example- 5) Consider the fading channel estimation problem where the output symbol 𝑦(𝑘) 1 point
FDE is 𝑦(𝑘) = ℎ𝑥(𝑘) + 𝑣(𝑘) , with ℎ, 𝑥(𝑘), 𝑣(𝑘) denoting the complex channel coefficient,
pilot symbol and noise sample respectively. Let
Week 8 -
Sequential Least 𝐱 = [1 − 2𝑗 2 + 𝑗 1 − 𝑗 2 − 𝑗]
𝑇
denote the pilot vector of transmitted pilot
Squares (SLS) symbols and 𝐲 = [1 − 𝑗 − 1 + 2𝑗 1 1 − 2𝑗]𝑇 denote the corresponding
Estimation – received symbol vector. Let 𝑣(𝑘) be IID Gaussian noise with zero-mean and dB variance
Scalar/ Vector
2
https://onlinecourses-archive.nptel.ac.in/noc17_ee19/unit?unit=95&assessment=140 2/5
26/07/2020 Estimation for Wireless Communications – MIMO/OFDM Cellular and Sensor Networks - - Unit 3 - Week 2 - Vector E…
Cases, 𝜎
2
= −9𝑑𝐵 . The variance of the maximum likelihood estimate of the channel coefficient
Applications -
Wireless Fading ℎ is,
Channel
Estimation, SLS
Example
1
120
136
150
196
136
6) Considerthe fading channel estimation problem where the output symbol 𝑦(𝑘) 1 point
is 𝑦(𝑘) = ℎ𝑥(𝑘) + 𝑣(𝑘) , with ℎ, 𝑥(𝑘), 𝑣(𝑘) denoting the complex channel coefficient,
pilot symbol and noise sample respectively. Let
𝐱 = [1 − 2𝑗 2 + 𝑗 1 − 𝑗 2 − 𝑗]
𝑇
denote the pilot vector of transmitted pilot
symbols and 𝐲 = [1 − 𝑗 − 1 + 2𝑗 1 1 − 2𝑗]𝑇 denote the corresponding
received symbol vector. Let 𝑣(𝑘) be IID Gaussian noise with zero-mean and dB variance
𝜎
2
= −9𝑑𝐵 . The variance of the real part of the maximum likelihood ˆ
ℎ is,
240
196
272
75
272
7) Consider the fading channel estimation problem where the output symbol 𝑦(𝑘) 1 point
is 𝑦(𝑘) = ℎ𝑥(𝑘) + 𝑣(𝑘) , with ℎ, 𝑥(𝑘), 𝑣(𝑘) denoting the complex channel coefficient,
pilot symbol and noise sample respectively. Let
𝐱 = [1 − 2𝑗 2 + 𝑗 1 − 𝑗 2 − 𝑗]
𝑇
denote the pilot vector of transmitted pilot
symbols and 𝐲 = [1 − 𝑗 − 1 + 2𝑗 1 1 − 2𝑗]𝑇 denote the corresponding
received symbol vector. Let 𝑣(𝑘) be IID Gaussian noise with zero-mean and dB variance
𝜎
2
= −9𝑑𝐵 . The variance of the imaginary part of the maximum likelihood ˆ
ℎ is,
272
300
75
392
https://onlinecourses-archive.nptel.ac.in/noc17_ee19/unit?unit=95&assessment=140 3/5
26/07/2020 Estimation for Wireless Communications – MIMO/OFDM Cellular and Sensor Networks - - Unit 3 - Week 2 - Vector E…
No, the answer is incorrect.
Score: 0
Accepted Answers:
1
272
8) Consider the fading channel estimation problem where the output symbol 𝑦(𝑘) 1 point
is 𝑦(𝑘) = ℎ𝑥(𝑘) + 𝑣(𝑘) , with ℎ, 𝑥(𝑘), 𝑣(𝑘) denoting the complex channel coefficient,
pilot symbol and noise sample respectively. Let
𝐱 = [1 − 2𝑗 2 + 𝑗 1 − 𝑗 2 − 𝑗]
𝑇
denote the pilot vector of transmitted pilot
symbols and 𝐲 = [1 − 𝑗 − 1 + 2𝑗 1 1 − 2𝑗]𝑇 denote the corresponding
received symbol vector. Let 𝑣(𝑘) be IID Gaussian noise with zero-mean and dB variance
𝜎
2
= −9𝑑𝐵 . The errors in the real and imaginary parts of the maximum likelihood ˆ
ℎ are
Equal
Correlated
Independent
All of the above
9) Consider the fading channel estimation problem where the output symbol 𝑦(𝑘) 1 point
is 𝑦(𝑘) = ℎ𝑥(𝑘) + 𝑣(𝑘) , with ℎ, 𝑥(𝑘), 𝑣(𝑘) denoting the real channel coefficient, pilot
symbol and noise sample respectively. Let 𝐱 = [𝑥(1), 𝑥(2), … , 𝑥(𝑁 )]𝑇 denote the pilot
vector of transmitted pilot symbols and 𝐲 = [𝑦(1), 𝑦(2), … , 𝑦(𝑁 )]𝑇 denote the
corresponding received symbol vector. Let 𝑣(𝑘) be independent Gaussian noise with zero-
mean and variance 𝑘𝜎 2 . And let R denote the covariance matrix of the noise vector
𝐯 = [𝑣(1), 𝑣(2), … , 𝑣(𝑁 )] . The ML estimate of the channel coefficient is,
𝑇
𝑁
1 𝑦(𝑘)
ℎ̂ = ∑
𝑁 𝑘
𝑘=1
𝑇
𝐱 𝐲
𝑇
𝐱 𝐱
𝑇 −1
𝑥 𝐑 𝑦
𝑁
𝑥(𝑘)𝑦(𝑘)
∑
𝑘
𝑘=1
ℎ̂ =
𝑁
2
𝑥 (𝑘)
∑
𝑘
𝑘=1
10)Consider the fading channel estimation problem where the output symbol 𝑦(𝑘) 1 point
is 𝑦(𝑘) = ℎ𝑥(𝑘) + 𝑣(𝑘) , with ℎ, 𝑥(𝑘), 𝑣(𝑘) denoting the real channel coefficient, pilot
symbol and noise sample respectively. Let 𝐱 = [𝑥(1), 𝑥(2), … , 𝑥(𝑁 )]𝑇 denote the pilot
vector of transmitted pilot symbols and 𝐲 = [𝑦(1), 𝑦(2), … , 𝑦(𝑁 )]𝑇 denote the
corresponding received symbol vector. Let 𝑣(𝑘) be independent Gaussian noise with zero-
mean and variance 𝑘𝜎 2 . And let R denote the covariance matrix of the noise vector
https://onlinecourses-archive.nptel.ac.in/noc17_ee19/unit?unit=95&assessment=140 4/5
26/07/2020 Estimation for Wireless Communications – MIMO/OFDM Cellular and Sensor Networks - - Unit 3 - Week 2 - Vector E…
2
𝜎
‖𝑥 2 ‖
𝑇 −1
𝑥 𝐑 𝑥
2
𝜎
𝑁
2
∑ 𝑘|𝑥(𝑘)|
𝑘=1
2
𝜎
𝑇 −1
𝑥 𝐑 𝑥
Funded by
Powered by
https://onlinecourses-archive.nptel.ac.in/noc17_ee19/unit?unit=95&assessment=140 5/5
26/07/2020 Estimation for Wireless Communications – MIMO/OFDM Cellular and Sensor Networks - - Unit 4 - Week 3 - Cramer-…
Courses » Estimation for Wireless Communications – MIMO/OFDM Cellular and Sensor Networks
Course
outline Assignment-3
The due date for submitting this assignment has passed. Due on 2017-08-15, 23:59 IST.
How to Access As per our records you have not submitted this assignment.
the Portal ?
1) Given the log-likelihood ln 𝑝(𝐲; ℎ) of the parameter ℎ and unbiased estimator 1 point
Week 1 - Basics ∞
2
of Estimation, ˆ
ℎ, the quantity (∫ −∞ ˆ
ℎ
∂
∂ℎ
ln 𝑝(𝐲; ℎ)𝑝(𝐲; ℎ)𝑑𝐲
)
equals
Maximum
Likelihood (ML)
2
Week 2 - Vector
Estimation
1
Week 3 - Cramer-
Rao Bound 0
(CRB), Vector
Parameter
Estimation, 𝐼 (ℎ)
Multi-Antenna
Downlink Mobile No, the answer is incorrect.
Channel Score: 0
Estimation
Accepted Answers:
1
Lecture 11 -
Cramer Rao
2) Given the log-likelihood ln 𝑝(𝐲; ℎ) of the parameter ℎ , the quantity 1 point
Bound (CRB)
2
for Parameter
equals,
∂
(𝐸 { ln 𝑝(𝐲; ℎ)})
Estimation ∂ℎ
Lecture 12 -
Cramer Rao
Bound CRB
2
Example –
Wireless Sensor
Network 1
Lecture 13 -
Vector 0
Parameter
Estimation –
𝐼 (ℎ)
System Model
for Multi No, the answer is incorrect.
Antenna
Score: 0
Downlink
Channel Accepted Answers:
Estimation 0
Lecture 14 -
3) The Cramer-Rao Bound for parameter ℎ is derived as, 1 point
Likelihood
Function and
Least Squares
∂
https://onlinecourses-archive.nptel.ac.in/noc17_ee19/unit?unit=103&assessment=141
{ } 1/5
26/07/2020 Estimation for Wireless Communications – MIMO/OFDM Cellular and Sensor Networks - - Unit 4 - Week 3 - Cramer-…
Cost Function 𝐸{
∂
ln 𝑝(𝐲; ℎ)}
for Vector ∂ℎ
Parameter
Estimation ∂
2
𝐸 𝑝(𝐲; ℎ))
{( ∂ℎ }
Lecture 15 -
Least Squares
1
Cost Function 2
∂
for Vector 𝐸
{( ∂ℎ
ln 𝑝(𝐲;ℎ)
) }
Parameter
Estimation
2
Vector 𝐸
∂
ln 𝑝(𝐲; ℎ))
{( }
Derivative ∂ℎ
Gradient
No, the answer is incorrect.
Quiz : Score: 0
Assignment-3
Accepted Answers:
Assignment-3 1
2
Solution 𝐸
{(
∂
ln 𝑝(𝐲;ℎ))
}
∂ℎ
Week 4 - Least 4) Consider now a slightly modified version of the wireless sensor network (WSN) 1 point
Squares (LS) estimation scenario described in class with each observation 𝑦(𝑘) = 𝑘ℎ + 𝑣(𝑘) , for
Principle,
1 ≤ 𝑘 ≤ 𝑁 . Let the noise samples be IID Gaussian with mean zero and variance 𝜎
2
Pseudo-Inverse,
Properties of LS
each. The quantity is
∂
ln 𝑝(𝐲; ℎ)
Estimate, ∂ℎ
Examples –
Mullti-Antenna
Downlink and 1
𝑁
2
MIMO Channel 𝜎
2
∑ (𝑦(𝑘) − 𝑘ℎ)
Estimation 𝑘=1
2
Week 5 - Inter 𝜎
Symbol 𝑁
∑ 𝑘(𝑦(𝑘)−𝑘ℎ)
2
Interference, 𝑘=1
Channel
Equalization, 𝜎
2
Zero-forcing 𝑁
equalizer,
2
∑ (𝑦(𝑘)−𝑘ℎ)
Approximation 𝑘=1
error of equalizer
𝑁
1
∑ 𝑘(𝑦(𝑘) − 𝑘ℎ)
Week 6 - 𝜎
2
𝑘=1
Introduction to
Orthogonal No, the answer is incorrect.
Frequency
Score: 0
Division
Multiplexing Accepted Answers:
(OFDM) and Pilot 𝑁
1
Based OFDM 2
∑ 𝑘(𝑦(𝑘) − 𝑘ℎ)
𝜎
Channel 𝑘=1
Estimation,
Example 5) Consider now a slightly modified version of the wireless sensor network (WSN) 1 point
estimation scenario described in class with each observation 𝑦(𝑘) = 𝑘ℎ + 𝑣(𝑘) , for
Week 7 - OFDM – 1 ≤ 𝑘 ≤ 𝑁 . Let the noise samples be IID Gaussian with mean zero and variance 𝜎
2
(FDE), Example- 𝑁
https://onlinecourses-archive.nptel.ac.in/noc17_ee19/unit?unit=103&assessment=141 2/5
26/07/2020 Estimation for Wireless Communications – MIMO/OFDM Cellular and Sensor Networks - - Unit 4 - Week 3 - Cramer-…
Cases, 6) Consider now a slightly modified version of the wireless sensor network (WSN) 1 point
Applications -
estimation scenario described in class with each observation 𝑦(𝑘) = 𝑘ℎ + 𝑣(𝑘) , for
Wireless Fading
Channel 1 ≤ 𝑘 ≤ 𝑁 . Let the noise samples be IID Gaussian with mean zero and variance 𝜎 2
Estimation, SLS each. The Cramer Rao Bound for the parameter ℎ is,
Example
6 2
𝜎
𝑁 (𝑁 +1)(2𝑁 +1)
2
𝜎
4 2
𝜎
𝑁 (𝑁 +1)
2
𝜎
2
𝑁
7) Consider now a slightly modified version of the wireless sensor network (WSN) 1 point
estimation scenario described in class with each observation 𝑦(𝑘) = 𝑘ℎ + 𝑣(𝑘) , for
1 ≤ 𝑘 ≤ 𝑁 . Let the noise samples be IID Gaussian with mean zero and variance 𝜎
2
the 𝑖𝑡ℎ antenna respectively. Let number of antennas 𝑀 = 2 and consider the
transmission of 𝑁 = 3 pilot vectors. The pilot matrix 𝐗 is of size
3 × 2
2 × 3
2 × 2
3 × 3
9)Consider the multiple antenna downlink channel estimation problem described in 1 point
class with 𝑥𝑖 (𝑘) denoting the pilot symbol transmitted from the 𝑖𝑡ℎ antenna at time
𝑘, 𝑦(𝑘), ℎ 𝑖 denoting the corresponding received symbol and fading channel coefficient for
the 𝑖𝑡ℎ antenna respectively. Let number of antennas 𝑀 = 2 and consider the
transmission of 𝑁 = 3 pilot vectors. Let the vector of received output symbols be
𝑇
https://onlinecourses-archive.nptel.ac.in/noc17_ee19/unit?unit=103&assessment=141 3/5
26/07/2020 Estimation for Wireless Communications – MIMO/OFDM Cellular and Sensor Networks - - Unit 4 - Week 3 - Cramer-…
denoted by 𝐲 = [𝑦(1), 𝑦(2), … , 𝑦(𝑁 )]𝑇 . The equivalent cost function to be minimized
to compute the estimate of the channel vector ℎ is,
𝑁 𝑀
2
∑ ∑ (𝑦(𝑘) − ℎ𝑖 𝑥(𝑘))
𝑘=1 𝑖=1
𝑁
1 𝑇 2
∑ (𝑦(𝑘) − 𝐡 𝑥(𝑘))
2
2𝜎
𝑘=1
∑ |𝐲 − 𝐗𝐡|
𝑘=1
𝑁 𝑀
2
∑ ∑ (𝑦(𝑘) − ℎ𝑖 )
𝑘=1 𝑖=1
Funded by
https://onlinecourses-archive.nptel.ac.in/noc17_ee19/unit?unit=103&assessment=141 4/5
26/07/2020 Estimation for Wireless Communications – MIMO/OFDM Cellular and Sensor Networks - - Unit 4 - Week 3 - Cramer-…
Powered by
https://onlinecourses-archive.nptel.ac.in/noc17_ee19/unit?unit=103&assessment=141 5/5
26/07/2020 Estimation for Wireless Communications – MIMO/OFDM Cellular and Sensor Networks - - Unit 5 - Week 4 - Least Sq…
Courses » Estimation for Wireless Communications – MIMO/OFDM Cellular and Sensor Networks
Course
outline Assignment-4
The due date for submitting this assignment has passed. Due on 2017-08-20, 23:59 IST.
How to Access As per our records you have not submitted this assignment.
the Portal ?
1) Consider the maximum likelihood (ML) multi-antenna channel estimation 1 point
Week 1 - Basics problem with 𝑁 transmitted pilot vectors 𝐱(𝑘) = [𝑥1 (𝑘), 𝑥2 (𝑘), … , 𝑥𝑀 (𝑘)] , 𝑇
of Estimation,
1 ≤ 𝑘 ≤ 𝑁 and 𝑁 received symbols 𝑦(1), 𝑦(2), . . . , 𝑦(𝑁 ) . Let the channel vector be
Maximum
𝐡 = [ℎ 1 , ℎ 2 , … , ℎ 𝑀 ] . Let the pilot matrix be 𝐗. The ML estimate of 𝐡 is,
𝑇
Likelihood (ML)
Week 2 - Vector
𝑇 −1 𝑇
Estimation (𝐗 𝐗) 𝐗 𝐲
Week 3 - Cramer- (𝐗
𝑇
𝐗)
−1
𝐗
𝑇
𝐡
Rao Bound
(CRB), Vector
Parameter 𝐗
−1
𝐲
Estimation,
Multi-Antenna
𝑇 −1 𝑇
Downlink Mobile (𝐗𝐗 ) 𝐗 𝐲
Channel
Estimation No, the answer is incorrect.
Score: 0
Week 4 - Least Accepted Answers:
Squares (LS) 𝑇 −1 𝑇
(𝐗 𝐗) 𝐗 𝐲
Principle,
Pseudo-Inverse, 2) Consider
Properties of LS
a multi-antenna channel estimation scenario with 𝑁
pilot vectors 1 point
= 4
𝐱(1) = [3, 2] , 𝐱(2) = [−2, 1] , 𝐱(3) = [−1, −2] , 𝐱(4) = [−1, 3] . Let the
𝑇 𝑇 𝑇 𝑇
Estimate,
Examples – corresponding received vector be 𝐲 = [3, 2, 2, 1] . Let the dB noise variance be
𝑇
Mullti-Antenna
−4.77𝑑𝐵. The corresponding pilot matrix 𝐗 is,
Downlink and
MIMO Channel
Estimation
⎡ 3 ⎤
Lecture 16 - ⎢ ⎥
Solution ⎢ −2 ⎥
Maximum ⎢ ⎥
Likelihood ML ⎢ 1 ⎥
Estimate ⎢
−1
⎥
Pseudo Inverse ⎢ ⎥
⎢ −2 ⎥
Lecture 17 - ⎢ ⎥
−1
Properties of ⎢ ⎥
Least Squares ⎣ 3 ⎦
Estimate –
Mean
⎡ ⎤
https://onlinecourses-archive.nptel.ac.in/noc17_ee19/unit?unit=109&assessment=150 1/5
26/07/2020 Estimation for Wireless Communications – MIMO/OFDM Cellular and Sensor Networks - - Unit 5 - Week 4 - Least Sq…
Covariance and ⎡ 3 2 ⎤
Distribution ⎢ ⎥
⎢ −2 1 ⎥
Lecture 18 - ⎢ ⎥
−1 −2
Least Squares ⎢ ⎥
Multi Antenna ⎣ −1 3 ⎦
Downlink
Maximum
Likelihood 3 −2 −1 −1
Estimation
corresponding received vector be 𝐲 = [3, 2, 2, 1]𝑇 . Let the dB noise variance be
Quiz :
−4.77𝑑𝐵. The pseudo-inverse of the pilot matrix 𝐗 is,
Assignment-4
Assignment-4
Solution 1
16 −13 −4 −9
87
[ 7 7 −9 16 ]
Week 5 - Inter
Symbol
Interference, 1
−11 17 14 21
Channel 87
[ 5 −27 9 12 ]
Equalization,
Zero-forcing
equalizer, ⎡ 42 32 ⎤
Approximation ⎢ ⎥
error of equalizer 1 ⎢ −29 16 ⎥
87 ⎢ ⎥
−21 −72
⎢ ⎥
Week 6 -
⎣ −16 38 ⎦
Introduction to
Orthogonal
Frequency 33 −23 8 25
1
Division
87
Multiplexing [ 12 −14 34 −45 ]
(OFDM) and Pilot
Based OFDM No, the answer is incorrect.
Channel Score: 0
Estimation,
Accepted Answers:
Example
1
16 −13 −4 −9
87
Week 7 - OFDM – [ 7 7 −9 16 ]
Comb Type Pilot
(CTP) 4) Consider a multi-antenna channel estimation scenario with 𝑁
pilot vectors 1 point
= 4
Transmission, 𝐱(1) = [3, 2] , 𝐱(2) = [−2, 1] , 𝐱(3) = [−1, −2] , 𝐱(4) = [−1, 3] . Let the
𝑇 𝑇 𝑇𝑇
Channel
Estimation in corresponding received vector be 𝐲 = [3, 2, 2, 1]𝑇 . Let the dB noise variance be
Time/ Frequency −4.77𝑑𝐵 . The estimate of the channel vector 𝐡 is,
Domain, CTP
Example,
Frequency 68
1
Domain
87
Equalization [ 32 ]
(FDE), Example-
FDE
1
21
87
[ 17 ]
Week 8 -
Sequential Least
Squares (SLS)
Estimation –
https://onlinecourses-archive.nptel.ac.in/noc17_ee19/unit?unit=109&assessment=150 2/5
26/07/2020 Estimation for Wireless Communications – MIMO/OFDM Cellular and Sensor Networks - - Unit 5 - Week 4 - Least Sq…
Scalar/ Vector −12
1
Cases, 87
[ 45 ]
Applications -
Wireless Fading
Channel 5
Estimation, SLS 1
87
Example [ 33 ]
87
[ 33 ]
Let the corresponding received vector be 𝐲 = [3, 2, 2, 1]𝑇 . Let the dB noise variance be
−4.77𝑑𝐵. The covariance of the ML estimate is,
1
5 0
261
[ 0 5 ]
1
0
3
1
[ 0 ]
3
1
6 0
261
[ 0 5 ]
1
6 −1
261
[ −1 5 ]
261
[ −1 5 ]
6) Considera multi-antenna channel estimation scenario. In general, when are the 1 point
estimation errors of the various channel coefficients uncorrelated for a scenario with IID
noise samples of variance 𝜎 2 each?
When the pilot matrix is square
When the pilot matrix is invertible
When the columns of pilot matrix are orthogonal
None of the above
No, the answer is incorrect.
Score: 0
Accepted Answers:
When the columns of pilot matrix are orthogonal
7) Considera multi-antenna channel estimation scenario with 𝑁 = 3 pilot vectors 1 point
𝐱(1) = [1, 2] , 𝐱(2) = [3, −1] , 𝐱(3) = [1, 1] . Let the corresponding received
𝑇 𝑇 𝑇
1 3 1
11 11 11
1 1 1
[ − ]
3 6 6
https://onlinecourses-archive.nptel.ac.in/noc17_ee19/unit?unit=109&assessment=150 3/5
26/07/2020 Estimation for Wireless Communications – MIMO/OFDM Cellular and Sensor Networks - - Unit 5 - Week 4 - Least Sq…
1 3 1
[ 2 −1 1 ]
2 3 1
− −
11 11 11
1 1 1
[ − ]
3 6 6
2 3 1
− −
6 6 6
1 1 1
[ − ]
11 11 11
1 1 1
[ − ]
3 6 6
7
11
5
[ ]
6
6
11
1
[ ]
6
5
6
1
[ ]
11
5
6
1
[ − ]
11
5
[ ]
6
vector be 𝐲 = [2, 1, 2]𝑇 . Let the dB noise variance be 3𝑑𝐵. The covariance of the ML
estimate is,
1
0
11
1
[ 0 ]
6
2
0
11
2
[ 0 ]
6
2 0
[ 0 2 ]
https://onlinecourses-archive.nptel.ac.in/noc17_ee19/unit?unit=109&assessment=150 4/5
26/07/2020 Estimation for Wireless Communications – MIMO/OFDM Cellular and Sensor Networks - - Unit 5 - Week 4 - Least Sq…
2 2
[ 2 2 ]
2
[ 0 ]
6
11
Funded by
Powered by
https://onlinecourses-archive.nptel.ac.in/noc17_ee19/unit?unit=109&assessment=150 5/5
26/07/2020 Estimation for Wireless Communications – MIMO/OFDM Cellular and Sensor Networks - - Unit 6 - Week 5 - Inter Sy…
Courses » Estimation for Wireless Communications – MIMO/OFDM Cellular and Sensor Networks
Course
outline Assignment - 5
The due date for submitting this assignment has passed. Due on 2017-08-31, 23:59 IST.
How to Access As per our records you have not submitted this assignment.
the Portal ?
1) Consider the maximum likelihood (ML) multi-antenna channel estimation 1 point
Week 1 - Basics problem with 𝑁 transmitted pilot vectors 𝐱(𝑘), pilot matrix 𝐗 and receive vector 𝐲. Let
of Estimation,
the channel vector be 𝐡 = [ℎ 1, ℎ 2, … , ℎ 𝑀 ]𝑇 . Let the noise samples 𝑣(𝑘) be
Maximum
Likelihood (ML) independent Gaussian with zero-mean and variance 𝜎𝑘2 . Let 𝐑 denote the covariance
matrix of the noise vector 𝐯 = [𝑣(1), 𝑣(2), … , 𝑣(𝑁 )]𝑇 . The ML estimate of the
Week 2 - Vector channel vector 𝐡 is,
Estimation
Week 3 - Cramer- −1 𝑇 −1 𝑇 −1
(𝐗𝐑 𝐗 ) 𝐗 𝐑 𝐲
Rao Bound
(CRB), Vector
Parameter 𝑁
−1
𝑁
Estimation, ∑
1
𝐱(𝑘)𝐱
𝑇
(𝑘) ∑
1
𝐱(𝑘)𝑦(𝑘)
2
Multi-Antenna
2
( 𝜎
𝑘
) ( 𝜎
𝑘 )
𝑘=1 𝑘=1
Downlink Mobile
Channel −1
Estimation 𝑁
2 𝑇
𝑁
2
∑ 𝜎 𝐱(𝑘)𝐱 (𝑘) ∑ 𝜎 𝐱(𝑘)𝑦(𝑘)
𝑘 𝑘
( ) ( )
𝑘=1 𝑘=1
Week 4 - Least
Squares (LS)
𝑇 −1 𝑇
Principle, (𝐗𝐑𝐗 ) 𝐗 𝐑𝐲
Pseudo-Inverse,
Properties of LS No, the answer is incorrect.
Estimate, Score: 0
Examples –
Mullti-Antenna Accepted Answers:
−1
Downlink and 𝑁 𝑁
1 𝑇 1
MIMO Channel ∑ 2
𝐱(𝑘)𝐱 (𝑘) ∑ 2
𝐱(𝑘)𝑦(𝑘)
( 𝜎 ) ( 𝜎 )
Estimation 𝑘=1
𝑘
𝑘=1
𝑘
2) Consider the MIMO channel estimation problem with pilot vectors 1 point
Week 5 - Inter
Symbol 𝐱(1) = [3, −2]
𝑇
, 𝐱(2) = [−2, 3]
𝑇
, 𝐱(3) = [4, 2]
𝑇
, 𝐱(4) = [2, 2]
𝑇
. The received
Interference, output vectors y are
Channel
Equalization,
𝐲(1) = [−2, 1, −3]
𝑇
, 𝐲(2) = [−1, 3, 3]
𝑇
, 𝐲(3) = [−1, −2, 2]
𝑇
, 𝐲(4) = [−3, −1, 1]
𝑇
.
Zero-forcing The size of the MIMO system is,
equalizer,
Approximation
error of equalizer 3 × 2
Lecture 21 -
2 × 2
Channel
Equalization
and Inter 2 × 3
Symbol
https://onlinecourses-archive.nptel.ac.in/noc17_ee19/unit?unit=117&assessment=152 1/6
26/07/2020 Estimation for Wireless Communications – MIMO/OFDM Cellular and Sensor Networks - - Unit 6 - Week 5 - Inter Sy…
Interference ISI
Model
3 × 3
Lecture 22 -
Least Squares No, the answer is incorrect.
based Zero Score: 0
Forcing
Accepted Answers:
Channel
3 × 2
Equalizer
Lecture 23 - 3) Consider the MIMO channel estimation problem with pilot vectors 1 point
Example of ISI 𝐱(1) = [3, −2]
𝑇
, 𝐱(2) = [−2, 3]
𝑇
, 𝐱(3) = [4, 2]
𝑇
, 𝐱(4) = [2, 2]
𝑇
. The received
Channel and
Least Squares
output vectors y are
based Zero 𝐲(1) = [−2, 1, −3]
𝑇
, 𝐲(2) = [−1, 3, 3]
𝑇
, 𝐲(3) = [−1, −2, 2]
𝑇
, 𝐲(4) = [−3, −1, 1]
𝑇
.
Forcing As described in the lectures, the pilot matrix 𝐗 for the MIMO channel estimation problem
Lecture 24 - above is,
Equalization
and
Approximation ⎡ 3 −2 ⎤
Error for Zero ⎢ ⎥
Forcing ⎢ −2 3 ⎥
𝐗 =
Channel ⎢
4 2
⎥
Equalizer ⎢ ⎥
⎣ 2 2 ⎦
Lecture 25 -
Example
𝐗 = [3 − 2 − 2 3 4 2 2 2]
Equalization
and
Approximation 𝐗 = [3 − 2 − 2 3 4 2 2 2]
𝑇
Quiz :
Assignment - 5 No, the answer is incorrect.
Score: 0
Assignment-5
Solution Accepted Answers:
3 −2 4 2
𝐗 =
Week 6 - [ −2 3 2 2 ]
Introduction to
Orthogonal 4) Consider the MIMO channel estimation problem with pilot vectors 1 point
Frequency
Division 𝐱(1) = [3, −2]
𝑇
, 𝐱(2) = [−2, 3]
𝑇
, 𝐱(3) = [4, 2]
𝑇
, 𝐱(4) = [2, 2]
𝑇
. The received
Multiplexing output vectors y are
(OFDM) and Pilot
𝐲(1) = [−2, 1, −3]
𝑇
, 𝐲(2) = [−1, 3, 3]
𝑇
, 𝐲(3) = [−1, −2, 2]
𝑇
, 𝐲(4) = [−3, −1, 1]
𝑇
.
Based OFDM
Channel As described in the lectures, the output matrix 𝐘 for the MIMO channel estimation
Estimation, problem above is,
Example
Frequency ⎣ −3 −1 1 ⎦
Domain
Equalization 𝑇
⎡ −2 1 −3 ⎤
(FDE), Example-
⎢ ⎥
FDE
⎢ −1 3 3 ⎥
𝐘 =
⎢ ⎥
−1 −2 2
Week 8 - ⎢ ⎥
Sequential Least ⎣ −3 −1 1 ⎦
Squares (SLS)
Estimation – 𝑇
Scalar/ Vector −2 1 −3 −1 3 3
𝐘 =
Cases, [ −1 −2 2 −3 −1 1 ]
Applications -
Wireless Fading
https://onlinecourses-archive.nptel.ac.in/noc17_ee19/unit?unit=117&assessment=152 2/6
26/07/2020 Estimation for Wireless Communications – MIMO/OFDM Cellular and Sensor Networks - - Unit 6 - Week 5 - Inter Sy…
Channel No, the answer is incorrect.
Estimation, SLS Score: 0
Example
Accepted Answers:
𝑇
⎡ −2 1 −3 ⎤
⎢ ⎥
⎢ −1 3 3 ⎥
𝐘 =
⎢ ⎥
−1 −2 2
⎢ ⎥
⎣ −3 −1 1 ⎦
5) Consider the MIMO channel estimation problem with pilot vectors 1 point
𝐱(1) = [3, −2]
𝑇
, 𝐱(2) = [−2, 3]
𝑇
, 𝐱(3) = [4, 2]
𝑇
, 𝐱(4) = [2, 2]
𝑇
. The received
output vectors y are
𝑇 𝑇 𝑇 𝑇
𝐲(1) = [−2, 1, −3] , 𝐲(2) = [−1, 3, 3] , 𝐲(3) = [−1, −2, 2] , 𝐲(4) = [−3, −1, 1]
𝑇 𝑇 −1
𝐘𝐗 (𝐗 𝐗)
𝑇 −1 𝑇
(𝐗 𝐗) 𝐗 𝐘
𝑇 𝑇 −1
𝐘𝐗 (𝐗𝐗 )
𝑇 −1 𝑇
(𝐗𝐗 ) 𝐗 𝐘
6) Consider the MIMO channel estimation problem with pilot vectors 1 point
𝐱(1) = [3, −2]
𝑇
, 𝐱(2) = [−2, 3]
𝑇
, 𝐱(3) = [4, 2]
𝑇
, 𝐱(4) = [2, 2]
𝑇
. The received
output vectors y are
𝑇 𝑇 𝑇 𝑇
𝐲(1) = [−2, 1, −3] , 𝐲(2) = [−1, 3, 3] , 𝐲(3) = [−1, −2, 2] , 𝐲(4) = [−3, −1, 1]
⎡ 3 −2 ⎤
⎢ ⎥
⎢ −2 3 ⎥
𝐗 =
⎢ ⎥
4 2
⎢ ⎥
⎣ 2 2 ⎦
⎡ 3 −2 ⎤
⎢ ⎥ 1
⎢ −2 3 ⎥ 0
33
𝐗 =
⎢ ⎥ 1
4 2 [ 0 ]
⎢ ⎥ 21
⎣ 2 2 ⎦
1
0 3 −2 4 2
33
𝐗 =
] [ −2 2 ]
1
[ 0 3 2
21
⎡ ⎤
https://onlinecourses-archive.nptel.ac.in/noc17_ee19/unit?unit=117&assessment=152 3/6
26/07/2020 Estimation for Wireless Communications – MIMO/OFDM Cellular and Sensor Networks - - Unit 6 - Week 5 - Inter Sy…
⎡ 3 −2 ⎤
⎢ ⎥ 1
0
⎢ −2 3 ⎥
33
𝐗 =
⎢ ⎥ 1
4 2 [ 0 ]
⎢ ⎥ 21
⎣ 2 2 ⎦
7) Consider the MIMO channel estimation problem with pilot vectors 1 point
𝐱(1) = [3, −2]
𝑇
, 𝐱(2) = [−2, 3]
𝑇
, 𝐱(3) = [4, 2]
𝑇
, 𝐱(4) = [2, 2]
𝑇
. The received
output vectors y are
𝐲(1) = [−2, 1, −3]
𝑇
, 𝐲(2) = [−1, 3, 3]
𝑇
, 𝐲(3) = [−1, −2, 2]
𝑇
, 𝐲(4) = [−3, −1, 1]
𝑇
.
The estimate of the MIMO channel matrix 𝐇 is,
1
21 −17 8
33
[ 13 20 17 ]
⎡ 2
12
⎤
⎢ 17 29 ⎥
⎢ 1 15 ⎥
− −
⎢ 17 29 ⎥
⎢ 25 23
⎥
⎣− ⎦
17 29
1
22 21 11
15
[ 15 −17 27 ]
14 7
⎡− − ⎤
33 21
⎢ ⎥
13 1
⎢− ⎥
33 21
⎢ ⎥
5 21
⎣− ⎦
33 21
8) Consider the MIMO channel estimation problem with pilot vectors 1 point
𝐱(1) = [3, −2]
𝑇
, 𝐱(2) = [−2, 3]
𝑇
, 𝐱(3) = [4, 2]
𝑇
, 𝐱(4) = [2, 2]
𝑇
. The received
output vectors y are
𝐲(1) = [−2, 1, −3]
𝑇
, 𝐲(2) = [−1, 3, 3]
𝑇
, 𝐲(3) = [−1, −2, 2]
𝑇
, 𝐲(4) = [−3, −1, 1]
𝑇
.
Let the noise samples be IID Gaussian zero-mean with variance −6𝑑𝐵. What are the
variances of the estimates of coefficients in any row of the MIMO channel matrix?
1 1
,
66 98
1 1
,
132 84
5 9
,
36 88
https://onlinecourses-archive.nptel.ac.in/noc17_ee19/unit?unit=117&assessment=152 4/6
26/07/2020 Estimation for Wireless Communications – MIMO/OFDM Cellular and Sensor Networks - - Unit 6 - Week 5 - Inter Sy…
this scenario based on symbols 𝑦(𝑘), 𝑦(𝑘 + 1) to detect 𝑥(𝑘) . What is the effective
channel matrix 𝐇 for this scenario?
1
1 0
3
1
[ 0 1 ]
3
1
1
3
1
[ 1 ]
3
1
1 0
3
1
[ 0 1 ]
3
1
1
3
1
[ 1 ]
3
1
[ 0 1 ]
3
Funded by
https://onlinecourses-archive.nptel.ac.in/noc17_ee19/unit?unit=117&assessment=152 5/6
26/07/2020 Estimation for Wireless Communications – MIMO/OFDM Cellular and Sensor Networks - - Unit 6 - Week 5 - Inter Sy…
Powered by
https://onlinecourses-archive.nptel.ac.in/noc17_ee19/unit?unit=117&assessment=152 6/6
26/07/2020 Estimation for Wireless Communications – MIMO/OFDM Cellular and Sensor Networks - - Unit 7 - Week 6 - Introduc…
Courses » Estimation for Wireless Communications – MIMO/OFDM Cellular and Sensor Networks
Unit 7 - Week 6 -
Introduction to Orthogonal Frequency Division
Multiplexing (OFDM) and Pilot Based OFDM Channel
Estimation, Example
Course
outline Assignment - 6
The due date for submitting this assignment has passed. Due on 2017-09-04, 23:59 IST.
How to Access As per our records you have not submitted this assignment.
the Portal ?
1) Consider an Inter Symbol Interference channel 1 point
Week 1 - Basics 𝑦(𝑘) = 𝑥(𝑘) +
1
𝑥(𝑘 − 1) + 𝑣(𝑘) . Let an 𝑟 = 2 tap channel equalizer be designed for
of Estimation, 3
Maximum this scenario based on symbols 𝑦(𝑘), 𝑦(𝑘 + 1) to detect 𝑥(𝑘) . Let the equalizer vector be
Likelihood (ML) denoted by 𝐜. The least squares problem for estimation of 𝐜 is,
Week 2 - Vector
Estimation ‖ ⎡ 1 0 ⎤ ‖
2
⎡0⎤
‖ ⎢ ⎥ ‖
‖ ⎢ ⎥ 1
− ⎢ 3 1 ⎥ 𝐜‖
Week 3 - Cramer- ‖⎢
1
⎥ ‖
⎢ ⎥
Rao Bound ‖⎣
0⎦
1 ‖
⎣ 0 ⎦ ‖
(CRB), Vector ‖ 3
Parameter
Estimation, 2
‖ ⎡ 1 0 ⎤ ‖
Multi-Antenna ‖
⎡0⎤
⎢ ⎥ ‖
Downlink Mobile ‖ ⎢ ⎥ 1
1 ⎥ 𝐜‖
0 − ⎢ 3
Channel ‖⎢ ⎥
⎢ ⎥
‖
‖⎣ ‖
Estimation 1⎦ ⎣ 0
1
⎦ ‖
‖ 3
Week 4 - Least 2
1
Squares (LS) ‖
0 1 0 ‖
‖ 3 ‖
Principle, ‖[
− 𝐜
‖
1
Pseudo-Inverse, 1] [ 0 1 ] ‖
‖ 3
Properties of LS
Estimate, 2
Examples – ‖
0 1
1
‖
‖ 3 ‖
Mullti-Antenna − 𝐜
‖[ 1 ‖
Downlink and 1] [ 1 ]
‖ 3
‖
MIMO Channel
Estimation No, the answer is incorrect.
Score: 0
Week 5 - Inter
Accepted Answers:
Symbol 2
Interference, ‖
⎡0⎤
⎡ 1 0 ⎤ ‖
‖ ⎢ ⎥ ‖
Channel ‖⎢ ⎥ 1
− ⎢ 3 1 ⎥ 𝐜‖
Equalization, ‖⎢
1
⎥ ‖
⎢ ⎥
Zero-forcing ‖⎣
0⎦
1 ‖
⎣ 0 ⎦ ‖
equalizer, ‖ 3
Approximation
error of equalizer 2) Consider an Inter Symbol Interference channel 1 point
𝑥(𝑘 − 1) + 𝑣(𝑘) . Let an 𝑟 = 2 tap channel equalizer be designed for
1
𝑦(𝑘) = 𝑥(𝑘) +
3
Week 6 - this scenario based on symbols 𝑦(𝑘), 𝑦(𝑘 + 1) to detect 𝑥(𝑘) . Let the equalizer vector be
Introduction to
denoted by 𝐜. The zero-forcing (ZF) equalizer vector 𝐜 is,
https://onlinecourses-archive.nptel.ac.in/noc17_ee19/unit?unit=123&assessment=155 1/5
26/07/2020 Estimation for Wireless Communications – MIMO/OFDM Cellular and Sensor Networks - - Unit 7 - Week 6 - Introduc…
Orthogonal
Frequency
1
Division 1
Multiplexing 3
[1]
(OFDM) and Pilot
Based OFDM
Channel 1
23
Estimation, 91
[ 11 ]
Example
Lecture 26 - 1
9
Introduction to 3
[ 13 ]
Orthogonal
Frequency
Division 1
3
Multiplexing
91
[ 27 ]
OFDM – Cyclic
Prefix CP and
Circular
No, the answer is incorrect.
Convolution Score: 0
Lecture 27 -
Accepted Answers:
Introduction to 3
1
Orthogonal 91
[ 27 ]
Frequency
Division 3) Consider an Inter Symbol Interference channel 1 point
Multiplexing
𝑥(𝑘 − 1) + 𝑣(𝑘) . Let an 𝑟 = 2 tap channel equalizer be designed for
1
OFDM – FFT at 𝑦(𝑘) = 𝑥(𝑘) +
3
Receiver and this scenario based on symbols 𝑦(𝑘), 𝑦(𝑘 + 1) to detect 𝑥(𝑘) . Let the equalizer vector be
Flat Fading
denoted by 𝐜. The zero-forcing equalizer is,
Lecture 28 -
Channel
Estimation ˆ
𝑥(𝑘) =
1
𝑦(𝑘 + 1) +
1
𝑦(𝑘)
Across Each 3 3
Subcarrier in
Orthogonal ˆ
𝑥(𝑘) =
11
𝑦(𝑘 + 1) +
23
𝑦(𝑘)
Frequency 91 91
Division
Multiplexing ˆ
𝑥(𝑘) = 3𝑦(𝑘 + 1) +
13
𝑦(𝑘)
OFDM 3
Lecture 29 - 3 81
ˆ
𝑥(𝑘) = 𝑦(𝑘 + 1) + 𝑦(𝑘)
Example 91 91
Orthogonal
Frequency
No, the answer is incorrect.
Division Score: 0
Mulltiplexing Accepted Answers:
OFDM – 3 81
Transmission of ˆ
𝑥(𝑘) = 𝑦(𝑘 + 1) + 𝑦(𝑘)
91 91
Samples with
Cyclic Prefix 4) Consider an Inter Symbol Interference channel 1 point
𝑥(𝑘 − 1) + 𝑣(𝑘) . Let an 𝑟 = 2 tap channel equalizer be designed for
1
Lecture 30 - 𝑦(𝑘) = 𝑥(𝑘) +
3
Example this scenario based on symbols 𝑦(𝑘), 𝑦(𝑘 + 1) to detect 𝑥(𝑘) . Let the equalizer vector be
Orthogonal
denoted by 𝐜. Let the effective channel matrix for this scenario be denoted by 𝐇. The
Frequency
Division approximation error for the zero-forcing equalizer (counting elements starting with 0) is,
Mulltiplexing
OFDM – FFT at
Receiver and 𝑇 𝑇
−1
1 − 𝐇 (𝐇𝐇 ) 𝐇
Channel ⌊ ⌋
1,1
Estimation
Quiz : 𝑇 𝑇
−1
1 − 𝐇 (𝐇𝐇 ) 𝐇
Assignment - 6 ⌊ ⌋
2,2
Assignment-6
Solution 1 − 𝐇𝐇
𝑇
−1
⌊( ) ⌋
1,1
Week 7 - OFDM –
Comb Type Pilot −1
𝑇
(CTP) 1 −
⌊(
𝐇𝐇 ) ⌋
Transmission, 1,1
https://onlinecourses-archive.nptel.ac.in/noc17_ee19/unit?unit=123&assessment=155 2/5
26/07/2020 Estimation for Wireless Communications – MIMO/OFDM Cellular and Sensor Networks - - Unit 7 - Week 6 - Introduc…
Domain, CTP Accepted Answers:
Example, 𝑇 𝑇
−1
Frequency 1 −
⌊
𝐇 (𝐇𝐇 ) 𝐇
⌋
1,1
Domain
Equalization 5) Consider an Inter Symbol Interference channel 1 point
(FDE), Example-
𝑥(𝑘 − 1) + 𝑣(𝑘) . Let an 𝑟 = 2 tap channel equalizer be designed for
1
FDE 𝑦(𝑘) = 𝑥(𝑘) +
3
this scenario based on symbols 𝑦(𝑘), 𝑦(𝑘 + 1) to detect 𝑥(𝑘) . Let the equalizer vector be
Week 8 - denoted by 𝐜. The approximation error for the zero-forcing equalizer is,
Sequential Least
Squares (SLS)
Estimation – 1
Scalar/ Vector 3
Cases,
Applications - 5
Wireless Fading 91
Channel
Estimation, SLS 9
Example 91
27
91
91
for this scenario based on symbols 𝑦(𝑘 + 1), 𝑦(𝑘) to detect 𝑥(𝑘) . What is the effective
3 1
⎡ − ⎤
4 4
⎢ ⎥
𝐇 = ⎢ ⎥
⎢ ⎥
1 3
⎣ − ⎦
4 4
3 1
⎡ − 0 ⎤
4 4
⎢ ⎥
𝐇 = ⎢ ⎥
⎢ ⎥
3 1
⎣ 0 − ⎦
4 4
3 1
⎡ − 0 0 ⎤
4 4
⎢ ⎥
⎢ ⎥
⎢ 3 1 ⎥
𝐇 = 0 − 0
⎢ 4 4 ⎥
⎢ ⎥
⎢ ⎥
3 1
⎣ 0 0 − ⎦
4 4
3 1
⎡ 0 0 − ⎤
4 4
⎢ ⎥
⎢ ⎥
⎢ 3 1 ⎥
𝐇 = 0 − 0
⎢ 4 4 ⎥
⎢ ⎥
⎢ ⎥
3 1
⎣ − 0 0 ⎦
4 4
⎡ ⎤
https://onlinecourses-archive.nptel.ac.in/noc17_ee19/unit?unit=123&assessment=155 3/5
26/07/2020 Estimation for Wireless Communications – MIMO/OFDM Cellular and Sensor Networks - - Unit 7 - Week 6 - Introduc…
⎡ 3
−
1
0 ⎤
⎢ 4 4 ⎥
𝐇 = ⎢ ⎥
⎢ ⎥
3 1
⎣ 0 − ⎦
4 4
for this scenario based on symbols 𝑦(𝑘 + 1), 𝑦(𝑘) to detect 𝑥(𝑘) . Let the effective
channel matrix 𝐇 for this scenario. The equalizer is,
4 108
ˆ
𝑥(𝑘) = − 𝑦(𝑘 + 1) − 𝑦(𝑘)
91 91
4 108
ˆ
𝑥(𝑘) = 𝑦(𝑘 + 1) + 𝑦(𝑘)
91 91
4 108
ˆ
𝑥(𝑘) = 𝑦(𝑘 + 1) − 𝑦(𝑘)
91 91
4 108
ˆ
𝑥(𝑘) = − 𝑦(𝑘 + 1) + 𝑦(𝑘)
91 91
for this scenario based on symbols 𝑦(𝑘 + 1), 𝑦(𝑘) to detect 𝑥(𝑘) . Let the effective
channel matrix 𝐇 for this scenario. The approximation error is,
91
10
91
11
91
12
91
91
for this scenario based on symbols 𝑦(𝑘 + 1), 𝑦(𝑘) to detect 𝑥(𝑘 + 1) . Let the effective
channel matrix 𝐇 for this scenario. The approximation error is,
91
91
91
91
https://onlinecourses-archive.nptel.ac.in/noc17_ee19/unit?unit=123&assessment=155 4/5
26/07/2020 Estimation for Wireless Communications – MIMO/OFDM Cellular and Sensor Networks - - Unit 7 - Week 6 - Introduc…
No, the answer is incorrect.
Score: 0
Accepted Answers:
1
91
4G LTE
4G WiMAX
WiFi
All of the above
No, the answer is incorrect.
Score: 0
Accepted Answers:
All of the above
Funded by
Powered by
https://onlinecourses-archive.nptel.ac.in/noc17_ee19/unit?unit=123&assessment=155 5/5
26/07/2020 Estimation for Wireless Communications – MIMO/OFDM Cellular and Sensor Networks - - Unit 8 - Week 7 - OFDM – …
Courses » Estimation for Wireless Communications – MIMO/OFDM Cellular and Sensor Networks
Course
outline Assignment - 7
The due date for submitting this assignment has passed. Due on 2017-09-11, 23:59 IST.
How to Access As per our records you have not submitted this assignment.
the Portal ?
1) Consider an 𝑁 = 4 subcarrier OFDM system with 𝐿 = 2 channel taps 1 point
Week 1 - Basics denoted by ℎ(0), ℎ(1). Conventional channel estimation is employed with the pilot
of Estimation, symbols transmitted on all the subcarriers given as
Maximum
𝑋(0) = 1 − 𝑗, 𝑋(1) = −1 + 2𝑗, 𝑋(2) = −2 − 𝑗, 𝑋(3) = 1 + 𝑗 . Let the
Likelihood (ML)
corresponding received samples in the time domain be
Week 2 - Vector 𝑦(0) = −1 + 𝑗, 𝑦(1) = 2 − 𝑗, 𝑦(2) = 2 + 2𝑗, 𝑦(3) = −1 − 𝑗 . Let the noise
Estimation samples 𝑣(𝑘), 0 ≤ 𝑘 ≤ 3 be zero-mean IID Gaussian with variance 𝜎 2 . Also, let the
cyclic prefix be of length one symbol. The transmitted block of samples in the time
Week 3 - Cramer-
Rao Bound
domain is,
(CRB), Vector
Parameter
Estimation, 1 − 𝑗, −1 + 2𝑗, −2 − 𝑗, 1 + 𝑗
Multi-Antenna
Downlink Mobile 1 1 1 1 1 5 1
− + 𝑗, − 𝑗, − − 𝑗, 1 + 𝑗
Channel 4 4 2 2 4 4 2
Estimation
1 1 1 1 1 1 5 1
1 + 𝑗, − + 𝑗, − 𝑗, − − 𝑗, 1 + 𝑗
Week 4 - Least 2 4 4 2 2 4 4 2
Squares (LS)
Principle, 1 + 𝑗, 1 − 𝑗, −1 + 2𝑗, −2 − 𝑗, 1 + 𝑗
Pseudo-Inverse,
Properties of LS No, the answer is incorrect.
Estimate, Score: 0
Examples –
Mullti-Antenna Accepted Answers:
1 1 1 1 1 1 5 1
Downlink and 1 + 𝑗, − + 𝑗, − 𝑗, − − 𝑗, 1 + 𝑗
2 4 4 2 2 4 4 2
MIMO Channel
Estimation 2) Consider an 𝑁 = 4 subcarrier OFDM system with 𝐿 = 2 channel taps 1 point
denoted by ℎ(0), ℎ(1). Conventional channel estimation is employed with the pilot
Week 5 - Inter symbols transmitted on all the subcarriers given as
Symbol
𝑋(0) = 1 − 𝑗, 𝑋(1) = −1 + 2𝑗, 𝑋(2) = −2 − 𝑗, 𝑋(3) = 1 + 𝑗 . Let the
Interference,
Channel corresponding received samples in the time domain be
Equalization, 𝑦(0) = −1 + 𝑗, 𝑦(1) = 2 − 𝑗, 𝑦(2) = 2 + 2𝑗, 𝑦(3) = −1 − 𝑗 . Let the noise
Zero-forcing
equalizer, samples 𝑣(𝑘), 0 ≤ 𝑘 ≤ 3 be zero-mean IID Gaussian with variance 𝜎 2 . Also, let the
Approximation cyclic prefix be of length one symbol. The action of the channel on the transmitted block
error of equalizer of OFDM samples can be modeled as,
4
Orthogonal
None of the above
Frequency
Division No, the answer is incorrect.
Multiplexing
Score: 0
OFDM Channel
Estimation Accepted Answers:
Lecture 32 -
None of the above
Comb Type
4) Consider an 𝑁 = 4 subcarrier OFDM system with 𝐿 = 2 channel taps 1 point
Pilot CTP
Based denoted by ℎ(0), ℎ(1). Conventional channel estimation is employed with the pilot
Orthogonal symbols transmitted on all the subcarriers given as
Frequency 𝑋(0) = 1 − 𝑗, 𝑋(1) = −1 + 2𝑗, 𝑋(2) = −2 − 𝑗, 𝑋(3) = 1 + 𝑗 . Let the
Division
Multiplexing
corresponding received samples in the time domain be
OFDM Channel 𝑦(0) = −1 + 𝑗, 𝑦(1) = 2 − 𝑗, 𝑦(2) = 2 + 2𝑗, 𝑦(3) = −1 − 𝑗 . Let the noise
Estimation samples 𝑣(𝑘), 0 ≤ 𝑘 ≤ 3 be zero-mean IID Gaussian with variance 𝜎 2 . Also, let the
Lecture 33 - cyclic prefix be of length one symbol. The received symbols across the subcarriers are
Example Comb
Type Pilot CTP
Based 1 − 𝑗, −2 + 𝑗, 1 − 𝑗, 1 + 2𝑗
Orthogonal
Frequency
Division 2 + 𝑗, −3 − 4𝑗, 5𝑗, −3 + 2𝑗
Multiplexing
OFDM Channel −1 + 𝑗, 2 − 𝑗, 2 + 2𝑗, −1 − 𝑗
Lecture 34 -
Frequency 2 + 𝑗, −3 − 𝑗, 1 + 3𝑗, −1 − 𝑗
Domain
Equalization No, the answer is incorrect.
FDE for Inter Score: 0
Symbol
Interference ISI Accepted Answers:
Removal in 2 + 𝑗, −3 − 4𝑗, 5𝑗, −3 + 2𝑗
Wireless
System 5) Consider an 𝑁 = 4 subcarrier OFDM system with 𝐿 = 2 channel taps 1 point
denoted by ℎ(0), ℎ(1). Conventional channel estimation is employed with the pilot
Quiz :
Assignment - 7 symbols transmitted on all the subcarriers given as
𝑋(0) = 1 − 𝑗, 𝑋(1) = −1 + 2𝑗, 𝑋(2) = −2 − 𝑗, 𝑋(3) = 1 + 𝑗 . Let the
Assignment-7
Solution
corresponding received samples in the time domain be
𝑦(0) = −1 + 𝑗, 𝑦(1) = 2 − 𝑗, 𝑦(2) = 2 + 2𝑗, 𝑦(3) = −1 − 𝑗 . Let the noise
Week 8 - samples 𝑣(𝑘), 0 ≤ 𝑘 ≤ 3 be zero-mean IID Gaussian with variance 𝜎 2 . Also, let the
Sequential Least
Squares (SLS)
https://onlinecourses-archive.nptel.ac.in/noc17_ee19/unit?unit=129&assessment=157 2/5
26/07/2020 Estimation for Wireless Communications – MIMO/OFDM Cellular and Sensor Networks - - Unit 8 - Week 7 - OFDM – …
Estimation – cyclic prefix be of length one symbol. The estimate of the channel coefficient 𝐻 (1)
Scalar/ Vector
across subcarrier 1 is,
Cases,
Applications -
Wireless Fading
Channel −1 + 2𝑗
Estimation, SLS
Example 1 3
+ 𝑗
2 2
4 3
− − 𝑗
5 5
1 − 2𝑗
6) Consider an 𝑁
subcarrier OFDM system with 𝐿 = 2 channel taps
= 4 1 point
denoted by ℎ(0), ℎ(1). Comb type channel estimation is employed with the transmitted
symbols on the subcarriers given as
𝑋(0) = 1 − 𝑗, 𝑋(1) = −1 + 2𝑗, 𝑋(2) = −2 − 𝑗, 𝑋(3) = 1 + 𝑗 . Let the pilot
samples 𝑣(𝑘), 0 ≤ 𝑘 ≤ 3 be zero-mean IID Gaussian with variance 𝜎 2 . Also, let the
cyclic prefix be of length one symbol. The estimate of channel coefficient 𝐻 (1) across
subcarrier 1 is,
1
1 − 𝑗
2
−2 + 𝑗
−1 − 2𝑗
3
− 𝑗
2
samples 𝑣(𝑘), 0 ≤ 𝑘 ≤ 3 be zero-mean IID Gaussian with variance 𝜎 2 . Also, let the
cyclic prefix be of length one symbol. The estimate of channel coefficient 𝐻 (2) across
subcarrier 2 is,
3 1
+ 𝑗
2 2
2 − 𝑗
−1 − 2𝑗
1 3
+ 𝑗
2 2
https://onlinecourses-archive.nptel.ac.in/noc17_ee19/unit?unit=129&assessment=157 3/5
26/07/2020 Estimation for Wireless Communications – MIMO/OFDM Cellular and Sensor Networks - - Unit 8 - Week 7 - OFDM – …
No, the answer is incorrect.
Score: 0
Accepted Answers:
−1 − 2𝑗
has 𝐿 = 2 taps, denoted by ℎ(0), ℎ(1). Let the corresponding received samples in the
time domain be 𝑦(0) = 2 + 𝑗, 𝑦(1) = 3 + 2𝑗, 𝑦(2) = −1 − 𝑗, 𝑦(3) = 2 − 3𝑗 . Let
the noise samples 𝑣(𝑘), 0 ≤ 𝑘 ≤ 3 be zero-mean IID Gaussian with variance 𝜎 2 . Also,
let the cyclic prefix be of length one symbol. The transmitted block of samples in the time
domain is,
3 3 1 1 5 1 7 1 3 3
− 𝑗 − 𝑗 − 𝑗 − − 𝑗 − 𝑗
4 2 4 2 4 2 4 2 4 2
3 3 1 1 1 5 1 7 3 3
− 𝑗 + 𝑗 + 𝑗 − 𝑗 − 𝑗
2 4 2 4 2 4 2 4 2 4
has 𝐿 = 2 taps, denoted by ℎ(0), ℎ(1). Let the corresponding received samples in the
time domain be 𝑦(0) = 2 + 𝑗, 𝑦(1) = 3 + 2𝑗, 𝑦(2) = −1 − 𝑗, 𝑦(3) = 2 − 3𝑗 . Let
the noise samples 𝑣(𝑘), 0 ≤ 𝑘 ≤ 3 be zero-mean IID Gaussian with variance 𝜎 2 . Also,
let the cyclic prefix be of length one symbol. The symbols received across the subcarriers
in the frequency domain are,
has 𝐿 = 2 taps, denoted by ℎ(0), ℎ(1). Let the corresponding received samples in the
time domain be 𝑦(0) = 2 + 𝑗, 𝑦(1) = 3 + 2𝑗, 𝑦(2) = −1 − 𝑗, 𝑦(3) = 2 − 3𝑗 . Let
the noise samples 𝑣(𝑘), 0 ≤ 𝑘 ≤ 3 be zero-mean IID Gaussian with variance 𝜎 2 . Also,
let the cyclic prefix be of length one symbol. The estimate of the channel coefficient 𝐻 (2)
across subcarrier 2 is,
https://onlinecourses-archive.nptel.ac.in/noc17_ee19/unit?unit=129&assessment=157 4/5
26/07/2020 Estimation for Wireless Communications – MIMO/OFDM Cellular and Sensor Networks - - Unit 8 - Week 7 - OFDM – …
7 4
− 𝑗
5 5
19 22
− 𝑗
13 13
19 3
+ 𝑗
10 10
2 9
− 𝑗
5 5
Funded by
Powered by
https://onlinecourses-archive.nptel.ac.in/noc17_ee19/unit?unit=129&assessment=157 5/5
26/07/2020 Estimation for Wireless Communications – MIMO/OFDM Cellular and Sensor Networks - - Unit 9 - Week 8 - Sequent…
Courses » Estimation for Wireless Communications – MIMO/OFDM Cellular and Sensor Networks
Unit 9 - Week 8 -
Sequential Least Squares (SLS) Estimation – Scalar/
Vector Cases, Applications - Wireless Fading Channel
Estimation, SLS Example
Course
outline Assignment-8
The due date for submitting this assignment has passed. Due on 2017-09-18, 23:59 IST.
How to Access As per our records you have not submitted this assignment.
the Portal ?
1) Consider an 𝑁 = 4 subcarrier OFDM system with conventional channel 1 point
Week 1 - Basics estimation i.e. pilot symbols transmitted on all the carriers, given as,
of Estimation, 𝑋(0) = 3 − 𝑗, 𝑋(1) = 2 + 3𝑗, 𝑋(2) = −1 − 2𝑗, 𝑋(3) = −2 + 𝑗 . The ISI channel
Maximum
Likelihood (ML) has 𝐿 = 2 taps, denoted by ℎ(0), ℎ(1). Let the corresponding received samples in the
time domain be 𝑦(0) = 2 + 𝑗, 𝑦(1) = 3 + 2𝑗, 𝑦(2) = −1 − 𝑗, 𝑦(3) = 2 − 3𝑗 . Let
Week 2 - Vector the noise samples 𝑣(𝑘), 0 ≤ 𝑘 ≤ 3 be zero-mean IID Gaussian with variance 𝜎 2 . Also,
Estimation let the cyclic prefix be of length one symbol. The estimate of the channel coefficient 𝐻 (3)
across subcarrier 3 is,
Week 3 - Cramer-
Rao Bound
(CRB), Vector 7 4
− 𝑗
Parameter 5 5
Estimation,
Multi-Antenna 19 22
Downlink Mobile − 𝑗
13 13
Channel
Estimation 19 3
+ 𝑗
10 10
Week 4 - Least
Squares (LS) 2
−
9
𝑗
Principle, 5 5
MIMO Channel
Estimation 2) Consider an 𝑁 = 4 subcarrier OFDM system with comb type channel 1 point
estimation and transmitted symbols
Week 5 - Inter 𝑋(0) = 3 − 𝑗, 𝑋(1) = 2 + 3𝑗, 𝑋(2) = −1 − 2𝑗, 𝑋(3) = −2 + 𝑗 . The ISI channel
Symbol has 𝐿 = 2 taps, denoted by ℎ(0), ℎ(1). Let 𝑙 = 0, 3 denote the pilot subcarriers. Let the
Interference,
Channel
corresponding received samples in the time domain be
Equalization, 𝑦(0) = 2 + 𝑗, 𝑦(1) = 3 + 2𝑗, 𝑦(2) = −1 − 𝑗, 𝑦(3) = 2 − 3𝑗 . Let the noise samples
Zero-forcing 𝑣(𝑘), 0 ≤ 𝑘 ≤ 3 be zero-mean IID Gaussian with variance 𝜎 . Also, let the cyclic prefix
2
equalizer,
Approximation
be of length two symbols. The estimate of the channel tap ℎ(0) is,
error of equalizer
22 5
− − 𝑗
Week 6 - 10 10
Introduction to
https://onlinecourses-archive.nptel.ac.in/noc17_ee19/unit?unit=134&assessment=159 1/5
26/07/2020 Estimation for Wireless Communications – MIMO/OFDM Cellular and Sensor Networks - - Unit 9 - Week 8 - Sequent…
Orthogonal 22
+
5
𝑗
Frequency 10 10
Division
Multiplexing 22
−
5
𝑗
(OFDM) and Pilot 10 10
Based OFDM
Channel −
22
+
5
𝑗
Estimation, 10 10
Channel 3) Consider a Frequency Domain Equalization system with block length 𝑁 . Let 1 point
Estimation in
Time/ Frequency 𝑌 (𝑙)and 𝐻 (𝑙) denote respectively the output and channel coefficient corresponding to
Domain, CTP subcarrier 𝑁 . The estimate of the transmitted symbol 𝑥(𝑘) is given as,
Example,
Frequency
Domain 𝑁 −1
𝑙𝑘
𝑌 (𝑙) −𝑗2𝜋
Equalization ∑ 𝑒 𝑁
𝐻 (𝑙)
(FDE), Example- 𝑙=0
FDE
𝑁 −1
𝑙𝑘
1 𝑌 (𝑙) 𝑗2𝜋
Week 8 - ∑ 𝑒 𝑁
𝑁 𝐻 (𝑙)
Sequential Least 𝑙=0
Squares (SLS)
Estimation – 𝑁 −1
Scalar/ Vector
𝑙𝑘
1 𝐻 (𝑙) 𝑗2𝜋
∑ 𝑒 𝑁
Cases, 𝑁
𝑙=0
𝑌 (𝑙)
Applications -
Wireless Fading
Channel 𝑁 −1
𝐻 (𝑙) −𝑗2𝜋
𝑙𝑘
Estimation, SLS ∑ 𝑒 𝑁
𝑌 (𝑙)
Example 𝑙=0
Equalization 1
∑
𝑌 (𝑙)
𝑒
𝑗2𝜋
𝑁
Symbol
Interference ISI 4) Consider a sequential fading channel estimation problem where the output 1 point
Removal in symbol 𝑦(𝑘) is 𝑦(𝑘) = ℎ𝑥(𝑘) + 𝑣(𝑘) , with ℎ, 𝑥(𝑘), 𝑣(𝑘) denoting the real channel
Wireless
Channels coefficient, pilot symbol and noise sample respectively. Let 𝐱 = [−2 3 2]𝑇 denote the
pilot vector of transmitted pilot symbols at 𝑁 = 3 and 𝐲 = [3 2 − 1]𝑇 denote the
Lecture 36 -
Example corresponding received symbol vector at 𝑁 = 3. Let the transmitted and received
Frequency symbols respectively at 𝑁 + 1 = 4 be 𝑥(4) = −1, 𝑦(4) = −2 respectively. Let 𝑣(𝑘)
Domain be IID Gaussian noise with zero-mean and dB variance 𝜎 2 = −3𝑑𝐵 . Let 𝑝(𝑁 ) denote
Equalization
FDE for Inter
the variance at time 𝑁 . The expression for the gain 𝐾 (𝑁 + 1) at time 𝑁 + 1 is,
Symbol
Interference ISI
𝑝(𝑁 )𝑥(𝑁 +1)
Removal in
2
Wireless 𝜎 +𝑝(𝑁 )𝑥(𝑁 +1)
Channels
2
𝑝(𝑁 )𝑥 (𝑁 +1)
Lecture 37 - 2 2
𝜎 +𝑝(𝑁 )𝑥 (𝑁 +1)
Introduction to
Sequential
2
Application in
2
𝜎 +𝑝(𝑁 )𝑥(𝑁 +1)
Wireless
Channel 𝑝(𝑁 )𝑥(𝑁 +1)
Estimation 𝜎
2
+𝑝(𝑁 )𝑥 (𝑁 +1)
2
Coefficient –
2 2
𝜎 +𝑝(𝑁 )𝑥 (𝑁 +1)
https://onlinecourses-archive.nptel.ac.in/noc17_ee19/unit?unit=134&assessment=159 2/5
26/07/2020 Estimation for Wireless Communications – MIMO/OFDM Cellular and Sensor Networks - - Unit 9 - Week 8 - Sequent…
Estimate and 5) Consider a sequential fading channel estimation problem where the output 1 point
Variance
symbol 𝑦(𝑘) is 𝑦(𝑘) = ℎ𝑥(𝑘) + 𝑣(𝑘) , with ℎ, 𝑥(𝑘), 𝑣(𝑘) denoting the real channel
Update
Equation coefficient, pilot symbol and noise sample respectively. Let 𝐱 = [−2 3 2]𝑇 denote the
Lecture 39 -
pilot vector of transmitted pilot symbols at 𝑁 = 3 and 𝐲 = [3 2 − 1]𝑇 denote the
Example corresponding received symbol vector at 𝑁 = 3. Let the transmitted and received
Sequential symbols respectively at 𝑁 + 1 = 4 be 𝑥(4) = −1, 𝑦(4) = −2 respectively. Let 𝑣(𝑘)
Estimation of
Wireless
be IID Gaussian noise with zero-mean and dB variance 𝜎 2 = −3𝑑𝐵 . Let 𝑝(𝑁 ) denote
Channel the variance at time 𝑁 . The gain 𝐾 (𝑁 + 1) at time 𝑁 + 1 = 4 is,
Coefficient
Quiz : 1
−
Assignment-8 16
Assignment-8 1
−
Solution 17
1
−
18
1
−
19
6) Consider a sequential fading channel estimation problem where the output 1 point
symbol 𝑦(𝑘) is 𝑦(𝑘) = ℎ𝑥(𝑘) + 𝑣(𝑘) , with ℎ, 𝑥(𝑘), 𝑣(𝑘) denoting the real channel
coefficient, pilot symbol and noise sample respectively. Let 𝐱 = [−2 3 2]𝑇 denote the
pilot vector of transmitted pilot symbols at 𝑁 = 3 and 𝐲 = [3 2 − 1]𝑇 denote the
corresponding received symbol vector at 𝑁 = 3. Let the transmitted and received
symbols respectively at 𝑁 + 1 = 4 be 𝑥(4) = −1, 𝑦(4) = −2 respectively. Let 𝑣(𝑘)
be IID Gaussian noise with zero-mean and dB variance 𝜎 2 = −3𝑑𝐵 . Let 𝑝(𝑁 ) denote
the variance at time 𝑁 . The prediction error at time 𝑁 + 1 = 4 is,
38
−
17
36
−
17
34
−
17
32
−
17
7) Consider a sequential fading channel estimation problem where the output 1 point
symbol 𝑦(𝑘) is 𝑦(𝑘) = ℎ𝑥(𝑘) + 𝑣(𝑘) , with ℎ, 𝑥(𝑘), 𝑣(𝑘) denoting the real channel
coefficient, pilot symbol and noise sample respectively. Let 𝐱 = [1 − 1 1]𝑇 denote the
pilot vector of transmitted pilot symbols at 𝑁 = 3 and 𝐲 = [2 12 32 ]𝑇 denote the
corresponding received symbol vector at 𝑁 = 3. Let the transmitted and received
symbols respectively at 𝑁 + 1 = 4 be 𝑥(4) = −1, 𝑦(4) = −1 respectively. Let 𝑣(𝑘)
be IID Gaussian noise with zero-mean and dB variance 𝜎 2 = −3𝑑𝐵 . What is the
prediction error at time 𝑁 + 1 = 4 is,
https://onlinecourses-archive.nptel.ac.in/noc17_ee19/unit?unit=134&assessment=159 3/5
26/07/2020 Estimation for Wireless Communications – MIMO/OFDM Cellular and Sensor Networks - - Unit 9 - Week 8 - Sequent…
0
8) Consider a sequential fading channel estimation problem where the output 1 point
symbol 𝑦(𝑘) is 𝑦(𝑘) = ℎ𝑥(𝑘) + 𝑣(𝑘) , with ℎ, 𝑥(𝑘), 𝑣(𝑘) denoting the real channel
coefficient, pilot symbol and noise sample respectively. Let 𝐱 = [1 − 1 1]𝑇 denote the
pilot vector of transmitted pilot symbols at 𝑁 = 3 and 𝐲 = [2 1 3 ]𝑇 denote the
2 2
1
−
4
9) Consider a sequential fading channel estimation problem where the output 1 point
symbol 𝑦(𝑘) is 𝑦(𝑘) = ℎ𝑥(𝑘) + 𝑣(𝑘) , with ℎ, 𝑥(𝑘), 𝑣(𝑘) denoting the real channel
coefficient, pilot symbol and noise sample respectively. Let 𝐱 = [1 − 1 1]𝑇 denote the
pilot vector of transmitted pilot symbols at 𝑁 = 3 and 𝐲 = [2 12 32 ]𝑇 denote the
corresponding received symbol vector at 𝑁 = 3. Let the transmitted and received
symbols respectively at 𝑁 + 1 = 4 be 𝑥(4) = −1, 𝑦(4) = −1 respectively. Let 𝑣(𝑘)
be IID Gaussian noise with zero-mean and dB variance 𝜎 2 = −3𝑑𝐵 . The update i.e.
quantity to be added to estimate ˆ
ℎ (3) at 𝑁 = 3 to generate the estimate ˆ
ℎ (4) at time
𝑁 + 1 = 4 is,
10) 1 point
https://onlinecourses-archive.nptel.ac.in/noc17_ee19/unit?unit=134&assessment=159 4/5
26/07/2020 Estimation for Wireless Communications – MIMO/OFDM Cellular and Sensor Networks - - Unit 9 - Week 8 - Sequent…
Consider a sequential fading channel estimation problem where the output symbol 𝑦(𝑘) is
𝑦(𝑘) = ℎ𝑥(𝑘) + 𝑣(𝑘) , with ℎ, 𝑥(𝑘), 𝑣(𝑘) denoting the real channel coefficient, pilot
symbol and noise sample respectively. Let 𝐱 = [1 − 1 1]𝑇 denote the pilot vector of
transmitted pilot symbols at 𝑁 = 3 and 𝐲 = [2 1 3 ]𝑇 denote the corresponding
2 2
received symbol vector at 𝑁 = 3. Let the transmitted and received symbols respectively
at 𝑁 + 1 = 4 be 𝑥(4) = −1, 𝑦(4) = −1 respectively. Let 𝑣(𝑘) be IID Gaussian noise
with zero-mean and dB variance 𝜎 2 = −3𝑑𝐵 . The variance of the estimate ˆ
ℎ (4) at time
𝑁 + 1 = 4 is,
10
16
Funded by
Powered by
https://onlinecourses-archive.nptel.ac.in/noc17_ee19/unit?unit=134&assessment=159 5/5