COMP3018 Coursework 2

Summary

In this exercise you are required to build an Android running tracking application, and 

document its design and architecture in a report. This is an assessed exercise and will account 

for 40% of your final module mark. This is an individual coursework, and your submission 

must be entirely your own work – please pay particular attention to the section of this 

document regarding plagiarism. This is a sizeable and open-ended coursework compared to

the previous assessed exercises. This document sets out general requirements that your 

application should meet rather than specific instructions.

Your application and report should be submitted no later than:

• 3pm on Monday the 11th of January 2021

Submissions should be made electronically via Moodle. Standard penalties of 5% per working 

day will be applied to late submissions.

Your coursework must be submitted as a .zip or .tar.gz file containing your report and 

application, including all relevant source code, configuration and related files, and a compiled 

.apk file – i.e. the contents of the directory containing your Android Studio project. Do not 

submit RAR files.

Specification

The Quantified Self or life-logging movement has been around for a number of years, but 

advances in mobile and wearable computing have increased the ability of people to collect 

data about their physical activities. The most common of these track activity as it happens for 

fitness, health or gamification purposes, for example displaying comparisons with previous 

activities, keeping track of best time or longest distances etc.

Application

The goal of this coursework is to design and implement a mobile application that functions as 

a basic Running Tracker, in that it should allow the user to track their movement when they 

decide to walk, run or jog, principally by logging the change in physical location using GPS.

The application should allow the user to inspect this data in a useful manner. The user might 

expect to want to be able to ask simple questions of the data such as “how far have I run so 

far today?”, “how much have I improved?” or “have I run faster than my best time today?”.

The application should allow the user to annotate their data. They might expect to be able to 

tag a particular exercise activity as good, or bad, or note something about the weather 

conditions, or they might want to associate a photograph with the exercise activity.

At the minimum, your application should support:

• Logging the movement of a user when they go running or walking

• Saving the movement data in an appropriate manner

• Allowing the user to inspect their data in an appropriate manner

• Allowing the user to annotate their data in a useful manner

• In principle allowing other applications to make use of the data

How you approach building this application is up to you, however how you consider and make

appropriate of the major Android application and architecture components is a significant 

part of the assessment criteria.

For this reason, it is important to consider how the task can be broken down into multiple 

atomic components, how they communicate with one another, and how their various 

lifecycles should interact. Other than making data available, there is no requirement that your 

components will be accessed by components outside of the application.

Some hints and tips regarding getting started with location services / GPS monitoring are 

provided below.

Your application must be written in Java or Kotlin and make use of the Android SDK. There 

are no requirements to target a specific Android API version, however you can assume that 

your application will be tested on an emulated device (1080 x 1920 420dpi) running Android 

API version 29 (Android 10.0).

Your application should have appropriate comments and variable / class names, so that a 

reader can easily understand how it works at the code level.

Adding further additional functionality to the application is encouraged, as are, for example, 

different interpretations of what it means to log running – you could consider walking, or 

other kinds of movement activity as might be measured by sensors on an Android device –

however as always your application should meet the above specification primarily. Indeed, 

an appropriate interpretation of the app’s required functionality is an implicit part of this

assessment.

Report

You must provide a report alongside your application that documents its design and technical 

architecture, in particular providing a rationale for the components that you have 

implemented based on their underpinning capabilities and paradigms, their relationship and 

communication with one another, and the behaviour of the application from the user’s point 

of view.

The report should be at minimum 1000 words long, with a maximum length of 1500 words.

There is no set structure for the report, however you may wish to include a diagram showing 

the components and their relationships, and a short explanation of each one, for example 

how the task is broken down into discrete Activity components, how and when Services are 

started, how data is abstracted from underlying storage etc.

Plagiarism

N.B. Use of third party assets (tutorials, images, example code, libraries etc.) MUST be 

credited and referenced, and you MUST be able to demonstrate that they are available 

under a license that allows their reuse.

Making significant use of tutorial code while referencing it is poor academic practice, and 

will result in a lower mark that reflects the significance of your own original contribution.

Copying code from other students, from previous students, from any other source, or 

soliciting code from online sources and submitting it as your own is plagiarism and will be 

penalized as such. FAILING TO ATTRIBUTE a source will result in a mark of zero – and can 

potentially result in failure of coursework, module or degree. 

All submissions are checked using both plagiarism detection software and manually for 

signs of cheating. If you have any doubts, then please ask.

Assessment Criteria

Marks 

Available

Application Functionality

How you have interpreted the provided Running Tracker specification, 

and the functionality that the application provides, including novelty and 

appropriateness

30

Application Structure and Implementation

How you have implemented and made appropriate use of Android 

components and principles

35

Programming style

How easy the application is to understand, with comments explaining 

each part of the code, correct formatting, and meaningful variable names

as well as 

10

Report

Description and justification of the design and architecture 25

Total 100

Each element of your coursework will be assessed against the standard criteria: 

Exceptional (90-100%) The work should exhibit all the characteristics of an Excellent 

grade. Additionally the work should be essentially without fault 

and of the highest possible quality, exhibiting a substantial 

original component.

Outstanding (80-89%) The work should exhibit all the characteristics of an Excellent 

grade. Additionally the results should exhibit independent 

thought and originality. Any short-comings should be no more 

than incidental.

Excellent (70-79%) The work should display a complete and thorough understanding 

of the conceptual and practical issues surrounding the topic. The 

work should be well structured with a clear line of argument and 

the quality of the analysis should be excellent. The work should 

be comprehensive and rigorous. Any software should be 

complete in all respects and exhibit very high quality. There 

should be evidence of reading beyond the core lecture material.

Good (60-69%) The work should show a good understanding of the conceptual 

and practical issues surrounding the topic. Arguments should be 

clearly structured. The quality of analysis and writing should be 

good. The work should be competently conducted using 

recognised and appropriate methods. Any software should be 

complete and usable and exhibit good levels of quality.

Average (50-59%) The work would be expected to display a fair understanding of 

the key conceptual and practical issues, although weakness may 

be present in some areas. There work should have a basic 

structure, and there should be some argument around the 

information available. The analytical content should be fair. Any 

software should be adequate to illustrate principles; it may 

display weakness in areas not central to the work.

Adequate (40-49%) The work would display an incomplete understanding of the 

central issues relating to the topic. The work would lack a clear 

structure and strong argument and the quality of analysis would 

be below average. Any software would be poorly designed, 

incomplete, poorly commented and difficult to understand; it 

would exhibit poor levels of quality.

Poor (below 40%) The work would display a very poor understanding of the area; 

there would be no clear structure and the analysis may be weak 

or incomplete. Any software would be limited in capability, and 

difficult to use.

The following areas will be taken into account for each part of the assessment:

• Demonstrating knowledge of the area

• Quality of the concept, including appropriateness and novelty

• Quality of the technological design, including appropriate use of software design 

concepts, and appropriate good coding practice (abstraction, commenting, naming)

• Quality of the realization, including how well it works and elaborations over and above 

the basic requirements

• Including all of the above aspects, clarity of structure, quality of argument / evidence, 

and insight / novelty

Hints and tips

Using Location / GPS tracking

There are different mechanisms for obtaining the location of the device, including GPS, Wi-Fi 

or cell-tower signal triangulation, and different mechanisms for how this data can be accessed 

by the device.

Increasingly Android is attempting to push this functionality into Google Play services (giving 

Google more control over parts of the Android stack), and this provides a unified approach 

that fuses multiple location systems into one to provide an abstraction over multiple pieces 

of hardware and to reduce battery usage. This requires making use of an emulator with the 

Google APIs installed – generally this will be a different emulator system image.

https://developer.android.com/training/location/receive-location-updates

There is, however, a simpler approach that is perfectly adequate for this coursework, and that 

is to use the LocationManager system service to provide GPS (global positioning system) 

updates that reveal the user’s location.

https://developer.android.com/reference/android/location/package-summary.html

Accessing location requires permission from the user:

import android.content.Context; 

import android.location.Location;

import android.location.LocationListener;

import android.location.LocationManager;

The LocationManager is a system service, and so needs to be retrieved from the service 

manager via getSystemService. Then it can be passed an instance of a LocationListener that 

will receive updates from the GPS provider. The two other parameters specify the minimum 

frequency of updates (i.e. we can say that we want at most 1 update every 5 seconds), and 

distance between updates (i.e. we can say that we only want to be told when the device has 

moved at least 5 metres). The fastest update frequency for GPS is around 1 second, and 

accuracy varies from a few metres upwards depending on environmental conditions.

LocationManager locationManager = 

(LocationManager)getSystemService(Context.LOCATION_SERVICE);

MyLocationListener locationListener = new MyLocationListener();

try {

locationManager.requestLocationUpdates(LocationManager.GPS_PROVIDER, 5, // minimum time interval between updates

5, // minimum distance between updates, in metres

locationListener);

} catch(SecurityException e) {

Log.d("g53mdp", e.toString());

}

The MyLocationListener class receives these location events by implementing the 

LocationListener interface as follows:

public class MyLocationListener implements LocationListener {

@Override

public void onLocationChanged(Location location) {

Log.d("g53mdp", location.getLatitude() + " " + location.getLongitude());

}

@Override

public void onStatusChanged(String provider, int status, Bundle extras) {

// information about the signal, i.e. number of satellites

Log.d("g53mdp", "onStatusChanged: " + provider + " " + status);

}

@Override

public void onProviderEnabled(String provider) {

// the user enabled (for example) the GPS

Log.d("g53mdp", "onProviderEnabled: " + provider);

}

@Override

public void onProviderDisabled(String provider) {

// the user disabled (for example) the GPS

Log.d("g53mdp", "onProviderDisabled: " + provider);

} }

onProviderEnabled and onProviderDisabled methods are called when the user enables or 

disables the GPS, and onStatusChanged gives information about the status of the GPS signal:

https://developer.android.com/reference/android/location/LocationListener.html

The important method call is onLocationChanged, which reports the current location as it is 

measured, and provides a Location object that can be inspected to obtain WGS 84 latitude, 

longitude, altitude (elevation), reported accuracy of the signal etc.

https://developer.android.com/reference/android/location/Location.html

Note that geodesy and global positioning are incredibly complicated subjects in their own 

right - the Earth is in no way perfectly spherical, and we like to think of linear distances on a 

locally flat surface as opposed to degrees around the world – however the Location class hides 

most of this from us. In particular the distanceTo method will calculate the distance between 

two points given as latitude and longitude:

float distance = myLocation.distanceTo(someOtherLocation);

Emulating GPS

It is possible to complete this coursework entirely using the emulator – there is no advantage 

to or necessity of having a physical Android phone. There is also no expectation that you 

handle the everyday practical details of GPS – losing signal, inaccurate signals etc. You can 

assume that it will be tested on an emulated device with “perfect” GPS.

The emulator provides a mock GPS device that feeds NMEA (latitude and longitude position 

updates) to the phone where they will be handled by the LocationManager as if they were 

real updates, via the extended controls menu. This can be found by clicking “…” on the 

emulator side bar.

Furthermore, the emulator can replay a series of GPS events from a GPX file (a standard log 

format for many GPS devices and applications). It is also possible to export from Google Maps 

to GPX.

Three example GPX files have been uploaded to Moodle for use as “real” latitude and 

longitude positions that can be played out.