Java IoT Authors: Pat Romanski, Elizabeth White, Liz McMillan, Yeshim Deniz, Mehdi Daoudi

Related Topics: Java IoT

Java IoT: Article

Targeting GPS - Integrating J2ME, GPS, and the Wireless Web

Targeting GPS - Integrating J2ME, GPS, and the Wireless Web

For location-based services, the open frameworks of J2ME and J2EE create interesting opportunities in the fields of software development and applied statistics. Traditionally, the software industry in these services has been closed and, as a result, the industry has suffered stagnation, particularly in the area of distributed systems and integration.

Just look at this most recent example – U.S. cell phone carriers didn’t meet the FCC October 2001 mandate for automatic location-based tracking for 911 calls over their networks. The most common reasons the carriers gave for missing the deadline were high costs and an inability to install the network infrastructure. With J2ME, XML, J2EE, and GPS, you can use the existing infrastructure (the Internet and your computer) to build and run such services from your garage, all for a very low cost (free).

With Java’s open architecture, you can build more exciting applications. GPS data contains the altitude and speed of a target, allowing the extraction of 3D information and vector coordinates. This can lead to interesting implementations, such as topographic position tracking of multiple targets or predicting the time two targets will intersect.

The use of wireless GPS and J2ME also allows more accurate tracking of targets because the historical GPS data can be stored in a database, enabling the integration of stochastic models and applications to monitor, predict, and correct movements before sending this information to a mobile device. Because of the limited graphical capability of mobile devices, there’s also a need for applications to convey location-based information through the use of colors, a feature of the more recent mobile devices.

The GPS application covered in this article tracks the user on a graphical plot and gives the distance of another target connected to the network. The target can also track the user. By going through the application, you can build a foundation to develop your own GPS services. We’ll discuss how to set up the test environment and show how to parse and handle GPS data on a mobile device and transfer GPS information back and forth between the device and a J2EE server. Thus, by integrating J2EE and J2ME, the mobile device harnesses the power of the server, allowing the development of intelligent, location-based applications.

Setting Up the Test Environment

Setting Up a GPS Receiver
A GPS receiver must meet two requirements for the application to work. First, it should be able to output NMEA 0183 (version 2.0 or higher) data as text; second, it shouldn’t require the receipt of an initialization string before sending data. eTrex from Garmin meets both requirements. Since the receiver hooks into the serial port on a PC (desktop or laptop), you need to obtain a GPS-to-serial-port adapter. Once connected, GPS data will flow through the communication port. Keep in mind that most GPS receivers don’t work indoors, so place the PC near a window where the receiver has clear access to the open sky. The receiver will need to get signals from at least three satellites to determine its position.

For those who don’t wish to go through the trouble and expense of dealing with a GPS receiver, the source code for this article includes a Java class file (HttpReader) that functions as a GPS data stream. (Listings 1–3 and the source code for this article can be downloaded from the JDJ Web site, www.sys-con.com/java/sourcec.cfm.) This class instance reads an HTML page that contains actual GPS data and returns its input stream. While this method is easier than using a GPS receiver, watching a plot of the author driving to his local grocery store is not nearly as much fun.

HTTP Connection
Since MIDP requires the implementation of a subset of HTTP 1.1, this protocol should work for all implementations of J2ME. Therefore the application uses HTTP to communicate with the server.

The server may reside on the local machine or on a LAN. For the server to be accessible over the Web, you need a wired or wireless Internet connection. However, a wireless connection provides the most usability for testing in a real production environment. This may be set up in a number of ways. A mobile phone with Internet access can hook up through a connector to a port on the laptop. Obtaining a wireless modem or doing a direct dial-up to a server hosting the server-side GPS application are other options.

Software Requirements
The GPS application requires the J2ME Wireless Toolkit from Sun and JDK 1.3 or higher. You also need the kXML package from Enhydra to handle XML on the MIDP device, as well as the xspsoft class files that are included in the download.

For the server-side applications, this article uses JBoss/Tomcat J2EE, which can be freely downloaded from www.jboss.org. Of course, you can also use any commercial J2EE implementation. You also need JAXB and JDK 1.4 from Sun.

The server application allows the user to transmit the GPS coordinates and to retrieve other people’s coordinates. If the you don’t wish to bother with EJBs and don’t care about tracking multiple targets, the server-side application can be eliminated without affecting the core GPS functionality on the MIDP device or emulator. (The application could also be rewritten using JSP and JDBC.)

NMEA Format
In the 1980s the National Marine Electronics Association (NMEA) developed the NMEA 0183 Interface Standard for data exchange between marine electronic devices. Today, most global positioning systems use the NMEA interface for data exchange. Thus the J2ME GPS data parser will work on any GPS receiver that implements this standard. Figure 1 contains a sample of a data stream:

NMEA Sentences

Figure 1: GPS data stream

The $GPGGA sentence contains the target’s topological location information. The relevant fields are as follows:

1 Message Header ($GPGGA)
3 - 4 Latitude, North/South
5 - 6 Longitude, East/West
10 Altitude in meters

The latitude 3753.667 denotes 37 degrees, 53 minutes, and 66.7 seconds. There are a couple of problems with this format. First, the longitude and latitude numbers are not base 10 numbers, so it’s difficult for an application to determine relative distances. Second, the KVM does not support floating or double primitive types nor the respective wrappers, making any decimal value an incorrect numerical format.

The answer is to use fixed-point integer calculations to handle decimal values. For example, the value of 100.234 is 100234, with a fixed point of three. The GpsParserImpl class instance normalizes the longitude and latitude coordinates by using the following code:

int nc = (10000 * k[0]) + (10000 * k[1])/60 +
(1000 * k[2])/3600;

where k[0] denotes the degrees (37), k[1] the minutes (53), and k[2] the seconds (667). The normalized integer value for this latitude is 379019.

To find the speed and direction of the target, we extract information from $GPRMC (fields 8 and 9, respectively). If the altitude field is blank, it means the GPS receiver isn’t able to pick up the signal from a fourth satellite, which is necessary for 3D tracking.

In this scenario the target is not moving, so the speed and direction fields contain 0.0. The J2ME application developer may use this information in interesting ways. For instance, you could display all targets moving over 60 m.p.h. in dark blue, or all targets moving north in green. This type of color-coding and dimensional reduction of information allows an enormous amount of location information to be displayed to the user of a limited mobile device. This will undoubtedly be an important area of future growth, research, and standardization for the mobile industry.

Accessing GPS Data Through the Serial Port
Opening the InputStream
Plugging the GPS receiver into the laptop or mobile device starts a stream of GPS data flowing over the serial port. Reading the input stream from the serial port is no different than reading an input stream from a file or an HTTP connection. As mentioned previously, you can use either HttpReader, which functions as a GPS data stream, or SerialReader to open a connection to an actual GPS receiver. Here’s the only difference in the code:

• SerialReader:

String URI = "comm:1;baudrate=4800;bitsper char=8";
InputConnection inputCon = (InputConnection) Connector.open(URI);
InputStream is = inputCon.openInputStream();


String URI = "http://xspsoft:8080/GPS.html";
HttpConnection inputCon = (HttpConnection) Connector.open(URI);
InputStream is = inputCon.openInputStream();

Note that a SerialReader object opens a connection to the stream on communication port 1; however, your port number may be different. If you choose the wrong port, the KVM will throw an IOException stating that the system can’t find the specified file. If the port is occupied, the IOException states that the handle is invalid.

Each GPS receiver sends data at a certain baud rate. Ensure that the baud rate in the String URI matches the baud rate of your GPS receiver. If any of the bytes in the input stream are above 128, the baud rate is incorrect. The input stream will come through as garbage. Also make sure there are no spaces between the semicolons or the program will throw an exception.

Parsing the InputStream
The abstract class GpsParser contains the base code to read the GPS data stream. A concrete, direct subclass instance instantiates either the SerialReader or HttpReader, depending on the user’s preference. GpsParser returns an NMEA sentence as a character array, one line at a time, through the use of the Template Method and an abstract method denoted as com-mand(char[] c).

The GpsParserImpl class contains an example of a concrete implementation of the command instance method, which is executed for each NMEA sentence flowing through the serial port. In the GpsParserImpl class, the match-
Command class method determines whether the current NMEA sentence is the GPGGA sentence. If it is, the command method scans and tokenizes the data into a vector.

Invoking the GpsParserImpl constructor causes the GPS data stream to continually update the Coordinate class with the respective longitude and latitude information. Now any application can invoke the Coordinate accessor class methods – getLatitude, getLongitude, getAltitude, getSpeed, and getDirection – to get the most current position of the target.

Data Access Objects and the Transporting of Information
In this article various data sources, including a relational database, XML files, MIDP record stores, and GPS data streams, are coming over a serial port. Dealing with so many data sources can be overwhelming without a systematic design and structure of the application services. Using data access objects (DAOs) to access these data sources is a good design practice that will simplify the application. One example we’ve already discussed is the coordinate data access object (actually a class) that returns the most current location information of the target.

This application has three primary data access objects on the MIDP device that can marshal and unmarshal XML: SerialDataObject, EntityDataObject, and RecordDataObjectImpl. The SerialDataObject handles the unidirectional transfer of information from the serial port (through Coordinate DAO) to the server database. The EntityDataObject handles the unidirectional transfer from the database to class methods of CoordinateTarget on the MIDP device. The RecordDataObjectImpl handles the bidirectional transfer of information between the server database and the MIDP record store. While transferring data to the record store may be slow, it is, however, a vital component to building enterprise applications involving an MIDP device.

These data objects can be returned by invoking the getDataObject (String DAO) class method of the factory DataObjectFactory. IGps is a superinterface of all these classes, so these class instances are assured of having accessor instance methods for the GPS data. The SessionMinibean object will take care of the marshaling and unmarshaling as well as choosing the correct data access object. For example, to transfer GPS data from the serial port to the server database merely requires the following lines of code:

SessionMinibean smb = new SessionMinibean("SERIAL");

The database server will now contain the user’s longitude, latitude, altitude, speed, and direction. Another user, with the following lines of code, can retrieve this user’s latitude:

SessionMinibean smb = new
int latitude =

Note that the unmarshal method unmarshals the XML message and populates the class methods of Coord-
inateTarget, so any application has access to the target’s location through CoordinateTarget. Class methods are extremely useful for J2ME applications because of the expense of packing, unpacking, and searching for information within the persistent record store.

If you’d like a local record store copy of the data from the database server, use the following lines of code:

SessionMinibean smb = new

You could also invoke the marshal method and update a database on the server.

In short, we have the basis for wireless hot sync capability between the record store on the mobile device and the database server. Thus the user could set his or her GPS and target tracking preferences, upload the information to the database, and then hot sync from any other mobile device. Hot synchronization could also allow multiple people to synchronize their tracking of a single target or group of targets. Although the record store is not used in this article, the functionality exists within the GPS application download.

A Brief Digression on the MIDP Record Store
The J2ME services on the mobile device access persistent data through the Record Management System (RMS) API. In this example, we avoid the overhead of RMS by accessing location-based information through the use of accessor class methods. However, record store access is critical to any enterprise production application. We’ll briefly cover some of the basics.

In the mainframe era space was at a premium, so the flat file was packed in binary and other formats. As a result, programmers built data-access components that packed and unpacked flat files. The same design principle applies to the somewhat more limited mobile devices that use byte records in a flat-file RMS.

The application needs packing and unpacking functionality to store the GPS byte information in a record store. The packBytes instance method in the RecordDataObjectImpl (see Listing 1) writes the values from the getter methods into a DataOutputStream and then converts the stream to a byte array. The setRecord instance method invokes the packBytes method and adds the bytes to the record store.

One issue with the MIDP record store is which record to pack and unpack. In container-managed persistence, the container automatically generates a findByPrimaryKey instance method for the EJB, which returns the object given by the primary key. The GPS application’s DAO on the mobile device includes a similar instance method denoted as setPrimaryKey. This is an important method because accessing the data by the record ID creates program-data dependence on the storage structure, resulting in a poor design of the record store.

The trick to locating records by the primary key, rather than key index, is to implement the data access object (RecordDataObjectImpl) as a RecordFilter. This allows the use of a RecordEnumeration on the current object to unpack the record bytes for the record that contains the primary key. By invoking the setPrimaryKey method, a RecordEnumeration returns those records (only one) that are true for the match method (see Listing 2). It then unpacks the correct record, giving access to the record information through the accessor methods. This process is invisible to the business object that instantiates RecordDataObjectImpl.

It’s a lot of work to look up and access data from a record store. However, notice that the use of the data access object is almost identical to the CMP entity bean in this GPS application. In a sense, the application uses a MiniEJB entity bean on the mobile device. For example, consider the code fragments in Listing 3. For both the MIDP and EJB code, we create a RecordDataObject and set the primary key to an integer that contains the value of 3. In both cases we invoke the accessor methods. The only difference is that in MIDP we must invoke setRecord, which is primarily needed for efficiency.

Note that in both cases the business object doesn’t need to know the record ID on which it is operating. By using a similar design for the MIDP and EJB process, the code is considerably reduced in the business object layer.


com.xspsoft.j2me.db.GpsDataObject gdo = new GpsDataObject();
gdo.setPrimaryKey(new Integer(3));
gdo.setId(new Integer(10));
gdo.setName("Mr. X");


com.xspsoft.gps.bean.GpsDataObject gdo = new GpsDataObject();
gdo.getEntityBeanByKey(new Integer(3));
gdo.setId(new Integer(10));
gdo.setName("Mr. X");

XML and Data Binding in J2ME
It’s worth looking into how the SessionMinibean marshals and unmarshals XML data for the transfer of tabular information. Note that the following method for dealing with XML will also be useful for the serialization of Java objects. The lack of MIDP support for RMI will make XML (and SOAP) a critical component for enterprise applications involving J2ME. Therefore, it’s a good idea for the J2ME developer to become proficient in using and manipulating XML.

The SessionMinibean class instance handles the transfer of GPS information between one of the data objects returned by DataObjectFactory and the RootXml and ClientXml objects, which are discussed later. First, we’ll demonstrate how the kXML package from Enhydra can read and write XML from the mobile device to a servlet. To write the <client myName="Mr. X" myLatitude="320854"/> tag to the SOAP servlet requires the following lines of code:

String URI = "http://xspsoft:
HttpConnection ic =
(HttpConnection) Connector.open(URI);
OutputStream os =
writer = new
org.kxml.io.XmlWriter w =
new XmlWriter(
(Writer) writer);
"Mr. X");

The XmlWriter instance requires a Writer parameter, which can be obtained by wrapping the OutputStream with an OutputStreamWriter. Now, writing to the XmlWriter instance will write directly to the servlet.

Reading from the SOAP servlet is very similar. Consider the following code:

InputStream is = ic.openInputStream();
InputStreamReader reader = new InputStreamReader(is);
XmlReader xr = new XmlReader(reader);
String myName = xr.getValue("myName");
String myLatitude = xr.getValue("myLatitude");

Two classes, ClientXml and RootXml, handle the primary work of marshaling and unmarshaling the data. ClientXml has accessor instance methods to set and get GPS information for the class instance. Consider the marshal method from the ClientXml class:

public void marshal(XmlWriter writer) throws Exception {
XmlWriter w = writer;
w.attribute("id", this.getId().toString());
w.attribute("name", this.name);
w.attribute("latitude", this.getLatitude().toString());
w.attribute("longitude", this.getLongitude().toString());
w.attribute("altitude", this.getAltitude().toString());
w.attribute("speed", this.getSpeed().toString());
w.attribute("direction", this.getDirection().toString());

To write an XML document to the servlet, use the following code:

ClientXml cx = new ClientXml();
cx.setName("Mr. X");
cx.setLatitude(new Integer(327890));
<<more set methods>>
cx.setDirection(new Integer(11));

The output looks like:

<client name="Mr. X" latitude="327890" ….direction="11"/>

The application can marshal any data access object that implements the IGps interface, which is a two-step process. First, invoke the set methods of a ClientXml instance, passing the return values from the DAO get methods as parameters. Second, invoke the ClientXml marshaling method. The unmarshal instance method is similar and can be viewed in the downloaded code. The purpose of the RootXml class instance is to enumerate through a vector of ClientXml objects, invoking its marshal or unmarshal method. In this specific case the application automatically writes (or reads) a complete XML document that contains multiple rows of clients and their respective locations.

The general case is more interesting. We now have a powerful technique: any object with accessor methods can marshal and unmarshal XML data without ever directly parsing XML data.

Plotting Target Location
MainDriver is the core class that runs the MIDlet application. This section gives a brief overview of the process and shows how all the previous programs fit together. The application starts with the following lines of code:

public void startApp() throws MIDletStateChangeException {
new GpsParserImpl();
GpsPlot gp = new GpsPlot();

Invoking the GpsParserImpl constructor begins a thread that starts updating the Coordinate class accessor methods from the GPS receiver. Next we begin another thread by instantiating the GpsPlot inner class. In its constructor, this GpsPlot object starts a TimerTask thread called MiniServlet:

public class MiniServlet extends TimerTask {
public MiniServlet() {
Timer t = new Timer();
t.scheduleAtFixedRate(this, 1000, 10000);
public void run() {
SessionMinibean serial = new SessionMinibean("SERIAL");
SessionMinibean entity = new SessionMinibean("ENTITY");

MiniServlet schedules its run method to invoke every 10,000 milliseconds. The run method marshals XML data from the Coordinate class to the servlet by invoking serial.marshal(). The servlet then updates the database on the server through the use of JAXB and container-managed persistence.

The run method from MiniServlet then invokes entity.marshal(). This unmarshals XML data from the server database (through the same servlet), updating the CoordinateTarget class accessor methods. In short, we’ve passed our coordinates to the server database and retrieved the target’s coordinates. The target is going through the same process, passing its information to the server and retrieving our GPS information.

The GpsPlot object run method stays alive indefinitely with the following code:

while (true) {
try { Thread.sleep
(Exception e){}

The repaint method invokes the drawMan method, given below:
1. private void drawMan(Graphics g) {
2. int lt = Coordinate.getLatitude() - CENTER_Y + CENTER_SCREEN;
3. int ln = Coordinate.getLongitude() - CENTER_X + CENTER_SCREEN;
4. int speed = Coordinate.getSpeed();
6. g.fillRect(ln - 2, lt - 4, 4, 4);
7. g.fillRect(ln - 3, lt - 8, 6, 4);
8. g.fillRect(ln - 2, lt - 9, 4, 1);
9. g.fillRect(ln - 1, lt - 10, 2, 1);
10. g.fillRect(ln - 2, lt - 11, 4, 1);
11. g.fillRect(ln - 1, lt - 12, 2, 1);
12. g.fillRect(ln, lt, 2, 2);
14. int deltaLat=Coordinate.getLatitude()- CoordinateTarget.getLatitude();
15. int deltaLong=Coordinate.getLongitude()- CordinateTarget.getLongitude();
16. int distance=com.xspsoft.j2me.util.Math.dist(deltaLat,deltaLong);
18. g.drawString(new String("Speed:"+speed),5,76,Graphics.TOP|Graphics.LEFT);
19. g.drawString(new String("Target Dist: " + distance), 5, 86, 16|4);
20. }

Before the plot is started, the program finds the first latitude and assigns the value to CENTER_Y and the value of the first longitude to CENTER_X. The center of the screen is coordinate pair (50, 50), although this will change depending on the device. On line 2, we calculate the latitude position by invoking the getLatitude method to find the current latitude. Next we subtract the initial latitude. This gives us the latitude movement, centered at zero. We add 50 to center the plot to the middle of the screen. A similar calculation is used for longitude. The fillRect methods plot the pixels that represent a man-shaped figure. As we move, the figure will move from the center coordinate.

We’d also like to display our current speed (lines 4 and 18) and the distance to another target (lines 14–16, 19). Since distance measurements involve square roots, a function not directly supported by the KVM, the download contains a Math class file that handles the distance calculations.

J2ME, J2EE, and XML are helping to end the age of mundane location-based services in the commercial area. The primary advantage is that these technologies open up the location-based services to a larger, more talented pool of developers. Gone are the days when we used our GPS system to ask, “Where is the BurgerBoy exit?” Instead, parents will be tracking their child online or you’ll be locating that elusive friend. As a consequence, your elusive friend will be putting a security perimeter around himself to detect other targets so he can remain elusive.

This article demonstrates a simple application using everyday technologies that allows a user to determine his or her distance from a moving object. Its potential is limitless.

More Stories By Shane Isbell

Shane Isbell works as a software architect at a wireless carrier.

Comments (18)

Share your thoughts on this story.

Add your comment
You must be signed in to add a comment. Sign-in | Register

In accordance with our Comment Policy, we encourage comments that are on topic, relevant and to-the-point. We will remove comments that include profanity, personal attacks, racial slurs, threats of violence, or other inappropriate material that violates our Terms and Conditions, and will block users who make repeated violations. We ask all readers to expect diversity of opinion and to treat one another with dignity and respect.

IoT & Smart Cities Stories
The platform combines the strengths of Singtel's extensive, intelligent network capabilities with Microsoft's cloud expertise to create a unique solution that sets new standards for IoT applications," said Mr Diomedes Kastanis, Head of IoT at Singtel. "Our solution provides speed, transparency and flexibility, paving the way for a more pervasive use of IoT to accelerate enterprises' digitalisation efforts. AI-powered intelligent connectivity over Microsoft Azure will be the fastest connected pat...
There are many examples of disruption in consumer space – Uber disrupting the cab industry, Airbnb disrupting the hospitality industry and so on; but have you wondered who is disrupting support and operations? AISERA helps make businesses and customers successful by offering consumer-like user experience for support and operations. We have built the world’s first AI-driven IT / HR / Cloud / Customer Support and Operations solution.
Codete accelerates their clients growth through technological expertise and experience. Codite team works with organizations to meet the challenges that digitalization presents. Their clients include digital start-ups as well as established enterprises in the IT industry. To stay competitive in a highly innovative IT industry, strong R&D departments and bold spin-off initiatives is a must. Codete Data Science and Software Architects teams help corporate clients to stay up to date with the mod...
At CloudEXPO Silicon Valley, June 24-26, 2019, Digital Transformation (DX) is a major focus with expanded DevOpsSUMMIT and FinTechEXPO programs within the DXWorldEXPO agenda. Successful transformation requires a laser focus on being data-driven and on using all the tools available that enable transformation if they plan to survive over the long term. A total of 88% of Fortune 500 companies from a generation ago are now out of business. Only 12% still survive. Similar percentages are found throug...
Druva is the global leader in Cloud Data Protection and Management, delivering the industry's first data management-as-a-service solution that aggregates data from endpoints, servers and cloud applications and leverages the public cloud to offer a single pane of glass to enable data protection, governance and intelligence-dramatically increasing the availability and visibility of business critical information, while reducing the risk, cost and complexity of managing and protecting it. Druva's...
BMC has unmatched experience in IT management, supporting 92 of the Forbes Global 100, and earning recognition as an ITSM Gartner Magic Quadrant Leader for five years running. Our solutions offer speed, agility, and efficiency to tackle business challenges in the areas of service management, automation, operations, and the mainframe.
The Jevons Paradox suggests that when technological advances increase efficiency of a resource, it results in an overall increase in consumption. Writing on the increased use of coal as a result of technological improvements, 19th-century economist William Stanley Jevons found that these improvements led to the development of new ways to utilize coal. In his session at 19th Cloud Expo, Mark Thiele, Chief Strategy Officer for Apcera, compared the Jevons Paradox to modern-day enterprise IT, examin...
With 10 simultaneous tracks, keynotes, general sessions and targeted breakout classes, @CloudEXPO and DXWorldEXPO are two of the most important technology events of the year. Since its launch over eight years ago, @CloudEXPO and DXWorldEXPO have presented a rock star faculty as well as showcased hundreds of sponsors and exhibitors! In this blog post, we provide 7 tips on how, as part of our world-class faculty, you can deliver one of the most popular sessions at our events. But before reading...
DSR is a supplier of project management, consultancy services and IT solutions that increase effectiveness of a company's operations in the production sector. The company combines in-depth knowledge of international companies with expert knowledge utilising IT tools that support manufacturing and distribution processes. DSR ensures optimization and integration of internal processes which is necessary for companies to grow rapidly. The rapid growth is possible thanks, to specialized services an...
At CloudEXPO Silicon Valley, June 24-26, 2019, Digital Transformation (DX) is a major focus with expanded DevOpsSUMMIT and FinTechEXPO programs within the DXWorldEXPO agenda. Successful transformation requires a laser focus on being data-driven and on using all the tools available that enable transformation if they plan to survive over the long term. A total of 88% of Fortune 500 companies from a generation ago are now out of business. Only 12% still survive. Similar percentages are found throug...