In this page:

General Packet Radio Service (GPRS) is an important stage in GSM evolution. It offers its users new data services and enables operators to offer more attractive pricing options. Moreover, utilizing the existing GSM infrastructure reduces investments required by the operators.

What is GPRS?

GPRS extends the capabilities of existing GSM networks and enables higher data rates and longer messages. It is a packet based data service for TDMA and GSM networks which enables the transition towards 3G.

GPRS was designed to make mobile data faster, cheaper, and user-friendlier than ever before. By introducing Internet Protocol (IP) and packet switching to mobile networks, it gives mobile users faster data rate and particularly suits bursts as in Internet and intranet traffic. For subscribers, GPRS enables voice and data calls simultaneously. Setting up a Connection is almost instantaneous, and users can be always connected to the mobile Internet, enjoying Web surfing, high-speed delivery of e-mails with large file attachments, and access to corporate LANs.

GPRS was defined by the European Telecommunications Standards Institute (ETSI) as a mean of providing packet radio service on GSM radio networks. At the same time, carriers whose networks were based on North-American TDMA have decided to deploy GPRS technologies in their networks.

Today, GPRS is seen as a preliminary step that will someday lead to the convergence of GSM and TDMA networks.

The main benefits of GPRS are that it reserves radio resources by using them only when data is available to send, and reduces the reliance on traditional circuit-switched networks. GPRS will improve the quality of data services measured in terms of reliability, response time, and available features. For instance, GPRS improves the capacity capabilities of the network suppliers by sharing the same radio resources among all mobile stations in a cell, thus providing effective use of scarce resources.

Back to page menu

Why do we need packet data, and what is packet data really?

As the Wireless Application Protocol (WAP) began to spread throughout the world during the year 2000, some users complained that it was too expensive, slow and cumbersome to use. Actually, most of the characteristics of WAP over GSM that users complained about were not due to bad WAP performance but typical problems of circuit-switched networks. A circuit-switched connection to a site works just like a regular telephone call. You dial to your Internet Service Provider (ISP), connect at 9.6Kbps and do not share this capacity with anyone. For certain streaming sound applications, this might be a good solution but most streaming applications are not of a constant bit rate. For burst like sessions such as WAP browsing, however, it is totally inefficient.  

In a radio channel which is assigned to a user, the user is paying the same regardless of the actual capacity utilized. A single time slot equals the capacity which is required for a voice call. Circuit-switched data requires at least one time slot to be allocated during a data session, regardless of how much data is actually transmitted.

Similarly, the operator is not fully utilizing the existing capacity, because no one else can access the unused channel when using circuit-switched network. As described above, the user has to establish a new connection whenever he or she wants to get some information (if not already connected). Say that someone checks the weather using a WAP phone and disconnects afterwards. If that someone then wants to check the weather in a neighboring area five minutes later, he or she will have to establish a new connection. The connection establishment sometimes takes as much as 20-40 seconds. To summarize, circuit-switched networks are less suitable for data sessions in which one does not need a guaranteed bit rate, the amount of information that is sent and received constantly varies and The cost is high for both the user and the operator. 

Introducing packet data not enables users to share the radio resources (just like traffic is handled over the fixed Internet, where several users share the same connection) and solves the problems mentioned above. This way, one doesn't load the network when not sending or receiving packets. Furthermore, users don't only share the capacity with each other, but also share it with circuit-switched voice and other data users.

This feature is also beneficial for the operator, who now can accommodate more users within the same network. In addition, this feature makes it possible to utilize parts of the network that weren't used before. When a circuit-switched network is fully loaded and some users get blocked calls due to lack of resources, it still has as much as 40 percent of unused capacity. This situation derives from the gaps between the disconnection of a channel and someone else who is connecting. A good analogy is to think of a circuit-switched network as a large box of rocks that can never be 100 percent full because of the irregular shapes of the rocks. Filling the same box with sand makes it possible to fill up the box more efficiently than before. With GPRS, however, the number of users rises and added interference is the result. This situation makes it hard to get 100 percent utilization, but it is still a significant improvement.

All rights reserved www.rad.co.il

Back to page menu

Packet switching in GPRS

Rather than dedicating a radio channel to one mobile user for a fixed period of time, the available radio resources can be concurrently shared by several users who consume resources only when they are actually sending or receiving data. This enables a larger number of GPRS users to share the same bandwidth and be served from a single cell. GPRS therefore allows the operator to maximize system usage and efficiency in a dynamic and flexible way.

Back to page menu

Why do we need GPRS?

GPRS was developed to supply GSM operators with means to meet the growing demand for wireless data services which is a result of the extensive growth of the Internet and corporate intranets. Applications using these networks require relatively high throughput and are characterized by burst like traffic and asymmetrical throughput needs. GPRS is better suited to meet the demand of this burst like data need rather than the traditional circuit-switched systems. GPRS also allocates the bandwidth independently between the uplink and downlink.

Another goal of GPRS is to enable GSM operators to enter the wireless packet data market with minimum expense. GPRS standards make use of standard GSM radio systems which include TDMA framing structures and GSM standard modulation schemes. By doing this, the cost implications regarding the cell equipment are minimized. Besides, GPRS provides the dynamic allocation and assignment of radio channels to packet services. By that, GSM operators can flexibly deploy GPRS without committing their entire network to it.

GPRS overlays packet-based air interfaces on the existing circuit-switched GSM network and allows the user to use a packet-based data service. This is quite a major upgrade to the a circuit-switched networks.

The GPRS standard is delivered in a way that network operators will need only to add few new infrastructure nodes and make software upgrades to some of the existing network elements. With GPRS, information is split into packets before it is being transmitted and at the receiving end the packets are reassembled.

Back to page menu