Programming environment

For this assignment you must ensure your work executes correctly on the simulator within Microchip Studio 7 as installed on workstations in ECS 249. If you have           installed Microchip Studio on your own then you are welcome to do much of the       programming work on your machine. (The IDE is available at no charge from             https://bit.ly/3llENk7 but comes only in the Microsoft Windows flavour.) You  must allow enough time to ensure your solutions work on the lab machines. If        your submission fails to work on a lab machine, the fault is very rarely that the lab   workstations. “It worked on my machine!” will be treated as the equivalent of “The  dog ate my homework!”

Individual work

This assignment is to be completed by each individual student (i.e., no group work). Naturally you will want to discuss aspects of the problem with fellow students, and  such discussion is encouraged. However, sharing of code fragments is strictly       forbidden without the express written permission of the course instructor        (Zastre). If you are still unsure regarding what is permitted or have other questions about what constitutes appropriate collaboration, please contact me as soon as          possible. (Code-similarity analysis tools will be used to examine submitted work.)     The URLs of significant code fragments you have found and used in your solution      must be cited in comments just before where such code has been used.

Objectives of this assignment

•   Work with timers, interrupts and the LEDs using the C programming language.

C programming language

In this assignment you will write several functions for a C language program. This   programs uses the LEDs on the lab boards (i.e., to keep this assignment simple, you

will not need to use the LCD panel, and therefore need only work with a single C  source file). The code given to you has already configured two of the mega2560   timers (timers 1 and 3). Interrupt handlers for these timers are already provided (i.e., you are not being asked to implement interrupt handlers). A brief                    demonstration illustrating the behavior for each of the parts can be seen in this   video:

https://youtu.be/rWHe6r2pI_4

The assignment is in four parts, ordered from easier to more difficult, each part building upon your confidence in completing the previous part.

A.   Write the code for lcd_state().

B.   Write the code for SOS() which uses both your completed lcd_state() plus arrays with LED and duration data.

C.   Write the code for glow().

D.   Write the code for pulse_glow().

There is also a bonus available for those who have successfully completed all four  parts. It is called light_show() and is to implement the LED pattern shown on the YouTube video (link above).

Part A: lcd_state()

This part of the assignment does not involve interrupts.

In assignment #2 you wrote an assembly-language function named leds_on. For this part you will write a function that accepts two parameters:

1.   The number of an LCD; and

2.   A number indicating the state to which that LCD must be put (with a zero

value meaning “off”, and all other values meaning “on”).                      The function is therefore to turn an LED on or off and immediately return.

To simply your work, your function need only deal with PORTL LEDs as numbered in the following diagram.

You can also take advantage of notational convenience when writing to I/O registers in AVR C. Assuming the DDR register for the port has been properly configured,          turning on LED #0 can be accomplished via:

PORTL |= 0b10000000;

while turning off LED #0 can be performed by:

PORTL &= 0b01111111;

which is equivalent to:

PORTL &= ~(0b10000000);

(Hint: This function might be a good candidate for including the use of a switch statement.)

Part B: SOS()

This part of the assignment does not directly involve the use interrupts.

By now you are familiar with Morse code, specifically that for SOS (dot dot dot, dash dash dash, dot dot dot). We can implement dots and dashes by calls to led_state() and _delay_ms().

You are to write a function to cause SOS to be displayed, but there is a twist here. In the SOS() function are three variables:

1.   uint8_t light[]: an array of 8-bit values indicate the LED pattern. An array value indicates LEDs on or off by bits set or cleared, with bit 0 the state for      LED #0, bit 1 the state for LED #1, etc.

2.   int duration[]: an array of ints (i.e., 16-bit values) representating the duration in milliseconds for an LED pattern.

3.   int length: the number of elements in light[] and the number of elements in duration[].

For example, at index 6 the bit pattern indicates LED #0 is turned on for 100 milliseconds).

So the twist is that you must cause the LEDs to flash appropriately for SOS by using these arrays: they already contain the patterns and durations required for SOS. You can use the arrays by writing a loop in which defined/declared appropriate counter variables, plus calls to led_state() and _delay_ms(). (A for-loop is perhaps a       better choice here than a while-loop).

Part C: glow()

This part of the assignment must use the program-scope variable named count that is incremented by the timer 1 interrupt handler. When testing your solution, we will evaluate glow() separately from pulse_glow(), i.e., we will call one or the other      function not both in the same test run.

Our current LEDs can be either on or off. There is, however, no easy way to set their relative brightness. Put differently, we control LED brightness by adjusting current through the LED, but we cannot adjust the current on our boards.

There is, however, another way to get something like a brightness control that          exploits something called pulse-width modulation. In essence it means turning an LED on and off, doing so very rapidly, and in a way that gives the appearance of        brightness or dimness.

Suppose we wish to have a somewhat dim LED. Consider the following diagram.

This shows that every 500 microseconds, the LED is turned on for 75 microseconds and then turned off for 425 microseconds. This cycle of 500 microseconds goes on  for as long as dimmer LED intensity is needed.

And suppose we want a brighter LED, but not as bright as the fully-on LED. Consider the next diagram.

Now for 500 microseconds, the LED is turned on for 250 microseconds and then turned off for 250 microseconds.

So when the LED is on, we say the signal is high, and the time for which is the signal is high is called the on time. Given than our period here is 500 microseconds, our       first diagram shows an on time of 75 microseconds, and the percentage of on time to the overall period is called the duty cycle (i.e., 75/500 = 0.15 = 15% duty cycle). For our second diagram, the duty cycle is 50% (i.e., 250/500 = 0.5 = 50% duty cycle).      Going further, a duty cycle of 0% would be an LED that is completely off, while a        duty cycle of 100% would be an LED that is completely on.

You are to implement this notion of a duty cycle. Function glow() takes two parameters:

1.   The number of an LCD; and

2.   The duty cycle for that LCD expressed as a floating-point value between 0.0 and 1.0.

In order to implement this function you must use an infinite loop, and therefore you will only be able to ever set the glow for a single LED. The infinite loop will take        advantage of the fact that the global variable count is incremented once a                    microsecond. The loop’s form will most likely need to be somewhat similar to:

Note that PWM_PERIOD is given in the top-most “DO NO TOUCH” section of a4.c.

Warning: Even though we might expect the LEDs to respond in a visually linear    fashion as we change the duty cycle, unfortunately our own visual systems do not respond in a similar way. That is, the differences in brightness that we see as

increase the duty cycle from 10% to 20% to 30% to 40% etc. will not “appear” as      evenly-spaced changes to our eyes and brains. Don’t worry about this (or if you         insist on worrying about this, please spend some time at https://bit.ly/2pGn61A after you have finished the assignment).

Part D: pulse_glow()

This part of the assignment must use the program-scope variables named count  and slow_count that are incremented by the timer 1 and timer 3 interrupt handlers respectively. When testing your solution, we will evaluate pulse_glow() separately from glow(), i.e., we will call one or the other function but not both in the same test run.

Once you have completed glow(), you might have some intuition of how to                  implement code to mimic those glowing LEDs appearing on some devices. That is, to increase and decrease the brightness over time, we need to increase and decrease the duty cycle over time. As our proxy for duty time in Part C was the value assigned to   the variable threshold, this means changing the value of threshold over time.

As with Part C, in order to implement Part D you must use an infinite loop, and           therefore you will only be able have one LED pulsing. The infinite loop will take          advantage of the fact that the global variables count and slow_count are                      incremented once a microsecond and once ever 10 milliseconds respectively. Your   loop will need to modify threshold by increasing it to PWM_PERIOD or decreasing it  to 0 in some way related to changes in slow_count. While your solution need not      mimic exactly the rate of pulsing that is shown in the YouTube video, it must be very similar.

Bonus: light_show()

(This part of the assignment does not involve interrupts.)

Using the array idea from Part B, implement the light-show pattern using a loop that iterates through you own versions of light[] and duration[] to produce the           pattern in the YouTube video.

What you must submit

•   Your completed a4.c. Do not change the name of this file! Do not submit any other project files.

•   Your work must use the provided skeleton files. Any other kinds of solutions will not be accepted.