Birla Institute of Technology & Science, Pilani
Work-Integrated Learning Programmes Division
First Semester 2016-2017
Comprehensive Examination (EC-3 Regular)
Course No. : BITS ZG553
Course Title : REAL TIME SYSTEMS
Nature
of Exam : Open Book
Weightage : 50%
Duration : 3 Hours
Date
of Exam : 05/11/2016 (AN)
Note:
1. Please follow all the Instructions
to Candidates given on the cover page of the answer book.
2. All parts of a question
should be answered consecutively. Each answer should start from a fresh
page.
3. Assumptions made if any,
should be stated clearly at the beginning of your answer.
Q.1.
Answer
the following questions. Please provide
justifications wherever required, but limit your answers to few lines only.
(a)
What
is the execution time of a job with feasible interval (3, 10], if its response
time is 7 and it was scheduled at time 5? [2]
(b)
In
a priority-driven system, the density of a periodic task is 0.2. Its phase and
execution time are 2 and 1 respectively. Then what is its relative deadline? [2]
(c)
A
periodic task has a period of 10 milliseconds and worst case response time of 6
milliseconds. In worst case scenario, the amount of time a job of this task
needs to wait because of other high priority tasks in the system, is 5
milliseconds. What is the processor utilization of this task in the worst case
scenario ? [2]
(d)
In
a real-time system with only aperiodic jobs, jobs arrived at a faster rate than
before. What will be the effect of it on the depth of the queue storing jobs
ready to be scheduled (i.e. 'ready' queue), if execution times of the jobs are
constant? [2]
(e)
In
a priority driven system with density of the periodic jobs equal to 0.75, a
sporadic job S (0,8, 1) arrived. Please indicate whether S will be accepted by
the scheduler of the system or not. [1]
(f)
Two
processors are having pipelines of different depths. Which among the two is
likely to produce more deterministic output from timing perspective? [1]
(g)
If
the cyclomatic complexity of a software program increases, will it have any
effect on the number of testcases required to completely test the program? If
so, what will be the effect? [1]
(h)
What
type of memory is usually used to store the compressed image of the application
and operating system binary in an embedded system? [1]
(i)
Between
semaphore and mutex, which one will you prefer to use in case of synchronizing
two tasks? [1]
(j)
In
the context of Structured Analysis and Structured Design(SASD) approach,
between Context Diagram(CD) and Data Flow Diagram(DFD), which one should be
created first? [1]
(k)
In
a priority driven system, a resource is required at different time by three
tasks with priorities 1, 3 and 4. Please note that in this system, priority 1
indicates highest priority, the next highest priority is priority 2 and so on.
What is priority ceiling of this resource when it is free? [1]
BITS
ZG553 (EC-3 Regular) First
Semester 2016-2017 Page 2
Q.2.
Following
table shows the resource requirement of three processes for a resource with 10
units.
Processes
|
Maximum Requirement (in
units)
|
Allocated (in units)
|
Needed (in units)
|
P0
|
6
|
3
|
3
|
P1
|
4
|
3
|
1
|
P2
|
5
|
3
|
2
|
Please indicate whether the system is
in safe or unsafe state, from deadlock perspective. Please provide
justification for your answer. [4]
Q.3.
In
an RM algorithm based priority driven system, there are two periodic tasks T1=(p1,e1)
and T2=(p2,e2), where p1,e1,p2,e2
> 0. Also p2 = m p1 + k, where m and k are natural
numbers, m, k > 0 and k < p1. How many steps will be there in
the time-demand plot of the lowest priority task in this system? [4]
Q.4.
Calculate
3 bit CRC of the binary number 1101010001100. Please note that for a 3-bit CRC,
the generator polynomial is binary 1011. [4]
Q.5.
Following
sets of periodic tasks are to be scheduled using RM Algorithm in a priority
driven system.
T1 =
(9,2), T2 = (12, 2), T3 = (15, 4). [7]
Determine
the worst case response time of each task using recursion (i.e. iterative method)?
Q.6.
An
operating system has one aperiodic queue for handling aperiodic jobs. The
average inter-arrival time of the jobs is 10
milliseconds. The average service time of the jobs is 6 milliseconds. Assuming that the
execution time of the jobs are only dependent on the CPU speed and are inversely proportional to the CPU
speed, how much change (increase or decrease) in CPU speed is required, if
average queue depth should be 1. [8]
Q.7.
There
are 3 periodic tasks T1, T2, T3 in a
priority-driven system, and they use a
serially reusable resource R. These tasks along with their critical
sections are defined as T1 = (1, 5, 2, 5, [R, 1;1]), T2 =
(4, 15, 4, 15, [R, 1;3]), T3 = (0, 8, 2, 8, [R, 1;1]). The tasks are
to be scheduled rate monotonically.
Assume that the jobs of each of these three tasks require the resource R at
time instance equal to one time unit
after they start execution.
In
this system, Priority Inheritance Protocol
is used to take care of Priority
Inversions.
Draw
a GANTT chart (the job schedule)
showing different jobs of the tasks scheduled between time unit 0 and 15.
Please show the critical sections involving Resource R as shaded regions.
[8]
***********
Birla
Institute of Technology & Science, Pilani
Work-Integrated
Learning Programmes Division
First
Semester 2016-2017
Comprehensive
Examination
(EC-3
Regular)
Course No. : EEE ZG512
Course Title : EMBEDDED SYSTEMS DESIGN
Nature
of Exam : Open Book
Weightage : 50%
Duration : 3 Hours
Date
of Exam : 06/11/2016 (AN)
Note:
1.
Please follow all the Instructions
to Candidates given on the cover page of the answer book.
2.
All parts of a question should be
answered consecutively. Each answer should start from a fresh page.
3.
Assumptions made if any, should be
stated clearly at the beginning of your answer.
Q.1.
Three
devices A, B, C supporting 7 bit addressing format on an I2C bus
initiate communication at the same time. All devices are capable of being the
master. Device A (address 32H ) attempts to receive 1 byte of data from a
device C . Device B (address 14H) attempts to transmit 3 bytes of
data to device C. Device C (address 13H ) attempts to send 1 byte of
data to device B. Which device will become bus master for the above transfer?
[3]
Q.2.
One
of the VIC registers of LPC2378 microcontroller was defined as follows
#define
VICSoftIntClr (*(volatile unsigned long
*)(VIC_BASE_ADDR + 0x01C))
What is the
significance of the keyword Volatile ? What can be the possible issues if the
term volatile is not used using the declaration of VICSoftIntClr? [3]
Q.3.
What
is the major difference between Memory Management Unit and Memory Protection
unit? Mention atleast 2 factors which
can lead to prefetch abort in an ARMv4 based embedded system [2 + 1 = 3]
Q.4.
Mention
the four conditions for deadlock to occur in a multi-threaded system. For the
given task allocation graph will a deadlock occur? Justify your answer. [2 + 3 = 5]
Q.5.
Distinguish
between Fully Ascending stack and Empty Ascending Stack. What is the
significance of Stack Sentinel in an Embedded System? [2 + 2 = 4]
EEE ZG512 (EC-3 Regular) First Semester 2016-2017 Page 2
Q.6.
a.
Why are we guaranteed that the set of periodic tasks (with their worst-case
computation time and periods) in the following Table is schedulable using EDF
scheduling? Draw the schedule for 30 time units.
Task
No
|
CPU
Burst
|
Period
|
T1
|
3
|
11
|
T2
|
2
|
12
|
T3
|
3
|
15
|
T4
|
4
|
20
|
Q.6. b. Suppose
we have an additional task with a period of 14 ms. What is the longest worst- case computation time (CPU-burst) for the
additional task so that the task set is schedulable
using EDF? [5 + 2 = 7]
Q.7.
You are assigned to develop a “sensor node”
which will be part of a network intended for weather monitoring in an
agricultural field. The sensor node will be replicated and deployed at several
parts of the field. The sensor node should be battery operated and hence
optimization of power should be done
wherever possible in the design. The technical specifications are as follows.
The sensor node
uses LPC2378 (works at 3.3V) as the microcontroller . The sensors node should house
environmental pressure, solar radiation and soil humidity sensors. The
three type of sensors provide an analog output voltage between a value 0 and
3.3V. The resolution of sensors is 15mV (ie output changes in steps of 15mV).
The system also will have a temperature sensor which can be interfaced to the
microcontroller using SPI protocol. Maximum clock frequency that can be
provided to the module is 5MHz. The temperature sensor provides 12-bit data as
output. The temperature sensor module drives out data on falling edge of the
clock and idle state of the clock is logic ‘0’. The weather conditions are to
be periodically detected at intervals of 2 minutes and wirelessly transmitted
to a remote monitoring unit placed at the entrance of field. The data to be
transmitted has to be provided to the radio module using UART protocol at a
baud rate of 115,200. One UART frame will have the format of 1 start bit, 8
data bits, 1 even parity bit and 1 stop bit. A push button is also available on
the sensor node which can be used to manually inspect the sensor node if the
wireless data is not available from a node for certain duration of time. When
the user presses the button the current value of soil humidity along with the
last sensed value of other sensors will be transmitted to the remote station.
[Soil humidity sensor is sampled again and the corresponding reading is transmitted
where as for the other sensors the last sensed value is displayed as the
reading of other sensors will change very slowly over a duration of time] .You can assume that 12 MHz crystal is
available for the design.
[3 + 12 + 4 + 3 + 3 = 25]
(a)
Draw
a flow chart to indicate the functional design of the system. The flow chart
should clearly indicate the sequence of operations in the system.
(b)
Draw
a block diagram for the system. Indicate the on-chip peripherals and
interfacing with the external circuitry. What are the on-chip peripherals
required for the implementation? Show the global clock configuration.
Initialize all the
EEE ZG512 (EC-3 Regular) First Semester 2016-2017
Page 3
on chip peripherals . Clearly initialize the on chip
peripherals like timer, ADC, SPI, UART,
etc . Also indicate the peripheral clock configurations (only for the
peripherals required in the design)
, interrupt configurations etc clearly . How many bytes of data( from all
sensors) should be transmitted at an interval of 2 minutes via UART to the
remote unit ?
(c
) How many interrupts will be used
in the system? What are the priorities of the interrupts? Clearly indicate the major activates
to be carried out within each of the interrupt service routines.
(d)
Can you guarantee an accurate baud
rate configuration of transmitter to 115,200 using LPC2378 for the given crystal frequency?. Since UART is an asynchronous protocol, how would you estimate the allowable tolerance
in variation of baud rate configuration in LPC2378 so that the LPC2378 can still receive data correctly? What are the factors that will be taken into account while determining the
variation?
(e)
Which low power mode of operation
of LPC2378 can be used for the application? Justify.
Can you think of any other method by which the power consumption can be minimized
for an LPC2378 based system ?
************
Birla Institute of
Technology & Science, Pilani
Work-Integrated
Learning Programmes Division
First Semester 2016-2017
Comprehensive Test
(EC-3 Regular)
Course No. :
ISZC424
Course Title :
Software for Embedded Systems
Nature of Exam :
Open Book
Weightage : 50%
Duration :
3 Hours
Date of Exam : 05/11/2016 (FN)
Note:
1. Please follow all the Instructions
to Candidates given on the cover page of the answer book.
2. All parts of a question
should be answered consecutively. Each answer should start from a fresh
page.
3. Assumptions made if any,
should be stated clearly at the beginning of your answer.
Q.1.
Write
a short note on Android application life cycle [6]
Q.2.
A
typical traffic signal system is being considered for a cross road. Each road has
two lanes and each lane is provided a GREEN light indicating an arrow. All
modes and movement of traffic is possible and indicated by GREEN arrow for
movement. Each section has a RED and YELLOW light. A timer is displayed which
shows the remaining time with a countdown timer. Pedestrians’ indications are
provided for them to cross the roads. For such a system answer the following:
(a)
Derive
at 4 distinct use cases depicting the unique system considered [6]
(b)
Design
and Architect (Logical) the system software. Draw the functional block diagram
and write in brief the function of each [8]
(c)
Comment
on use of the Android Framework to develop the entire controller for such a traffic
signal system [6]
Q.3.
Differentiate
between Bluetooth and WiFi interfaces used in networking of embedded systems [6]
Q.4.
Write
a note on Blending functions (alpha blending) including the cropping/transform
functions with an example [6]
Q.5.
For
the following system provide your design and architecture recommendations to
realise the system that collects data for soil using an embedded device which
in turn sends the data over a wireless interface to a device like mobile phone
where the trends of the data are shown from the data so collected. One mobile
phone may be dedicated to one field which may have numerous sensors and
embedded devices.
(a)
Derive
the system use cases [4]
(b)
Draw
the software design block diagram (clearly depict all functional blocks) and
briefly describe the embedded system and the software on the mobile phone [8]
***********
Birla Institute of Technology & Science, Pilani
Work-Integrated Learning Programmes Division
First Semester 2016-2017
Comprehensive Examination (EC-3 Regular)
Course No.
: SS ZG531
Course
Title : PERVASIVE
COMPUTING
Nature of Exam :
Open Book
Weightage : 50%
Duration : 3 Hours
Date
of Exam : 06/11/2016 (FN)
Note:
1.
Please follow
all the Instructions to Candidates given on the cover page of the answer
book.
2.
All parts of a
question should be answered consecutively. Each answer should start from a
fresh page.
3.
Assumptions made
if any, should be stated clearly at the beginning of your answer.
Q.1.
Why
is it that:
1.1
A
wearable computing device like a Smart Watch or a Smart Glass is not considered
as a Pervasive Computing System by itself?
1.2
Calm
Computing paradigm did envision absence of HCI?
1.3
A
room with multiple physical sensors connected to a single computer may provide
limited level of smart services expected of a typical Smart Room but such room
system would still not qualify as a Pervasive Computing environment?
1.4
Smart
Dust concept, although potentially useful, did not become the kind of success
that was envisioned by Berkeley
researchers?
1.5
Context
modelling process, as applicable to Ubiquitous Computing Environments needs to
capture Active as well as Passive Contexts of the environment and its
participants? [2 x 5=10]
Q.2.
If
a kitchen of a typical Indian house needs to be converted into a Smart Kitchen
that is possible to be monitored remotely from your connected handheld device,
how would you identify the required feature-set,
corresponding services, protocols, application software, system
software as well as required set of computing,
sensing, communication networking, power provisioning, actuation
(optional) and security elements?
Please explain each part
of the process concisely and without providing extraneous information.
Would this process of
identification
(in scope of this question) and
subsequent complete design (not in the scope of the expected answer here) benefit or not benefit from
context-awareness functionality?
(Note: Please feel free to make any set of
cohesive real-life assumptions but briefly justify each assumption! You are
also welcome to assume any layout for the kitchen as well as any commonly used
equipment that a typical urban household kitchen uses.)
[1x5+1x10+1=16]
Q.3.
What
is meant by the term Hybrid Intelligent
Systems? Are they necessarily the
same as Pervasive Computing Systems?
[3+ 1=
4]
Q.4.
Explain
the entire process of context modelling
of any Pervasive Computing Environment along with reasons involved without
or with the help of an example. (Note: If you choose the example route, it
should not be the one discussed in your text books, reference books or
interactive sessions). [2 x 5=10]
Q.5.
What
are the security issues having potential to plague a given Cyber-Physical System assuming that it is connected to the Internet for permitting public access to some
consumer-monitored data but does not
use the Internet as the glue for interconnecting its compute-sense-actuate
elements involved in the actual control operations? Identify any five significant issues and briefly explain their respective significance. [2 x 5 = 10]
********
Comments
Post a Comment