Standardised by 3GPP, HSPA is the set of technologies that defines the migration path for 3G/WCDMA operators worldwide.
HSPA, which uses the FDD transmission scheme, includes HSDPA (High Speed Downlink Packet Access), HSUPA (High Speed Uplink Packet Access) and HSPA Evolved. These are also known as 3GPP Releases 5 through to 8.
Unlike many other mobile broadband technologies, HSPA provides very efficient voice services in combination with mobile broadband data.
In most HSPA networks, the end-user can expect to enjoy speeds of at least 1Mbps upwards, depending upon the peak speed of the network (anywhere from 1.8Mbps to 14.4 Mbps) with peak uplink speeds of up to 5.7Mbps.
HSPA Evolved introduces MIMO (Multiple-Input, Multiple-Output) capabilities and higher order modulation (64QAM), enabling greater throughput speeds and higher performance. The first HSPA Evolved services are expected to be deployed at the end of 2008.
“HSDPA is the natural next step in the evolution of GSM. It will bring broadband speeds to mobile networks.”
HSPA is a common term to embrace all acronyms for HSDPA and HSUPA as well as HSPA Evolved. The mobile broadband technology is now deployed worldwide with a huge number of devices showing not only volume but the variety of devices depicting the different ways people wish to access information on the move.
The networks are deployed mainly at 1900MHz and 2100MHz with a few operating at the more advantageous spectrum of 850MHz – and with potential reframing and UHF spectrum reallocation, a number of other operators will adopt this spectrum to help increase coverage as HSPA takes off.
There are many technical white papers written in great detail on HSPA and so this section of our website will only deal with the main features of each variant and the benefits they bring to user and to mobile network operator.
HSDPA – High Speed Downlink Packet Access – is predominately a software upgrade on Rel 99 of the UMTS standard. This upgrade is to increase the efficiency and reduce the latency of the link and is achieved by utilising a number of techniques in parallel :
Adaptive Modulation and Coding – Software within the node B (base station) analyses each user on the cell for signal quality and using this information and the cell capacity at the time determines what modulation scheme to ‘talk’ to each device over. So, for good signal quality and a lightly loaded cell, the node B will assign 16QAM modulation enabling peak rates up to 3.6Mb/s and up, dropping back to QPSK with associated lower data rates if conditions become less favourable.
Fast Packet Scheduling – again relies on the devices within the cell reporting their signal strength. The node B can then determine which device to send data to in the next 2ms time frame thus making the most efficient use of the bandwidth available. The node B can also determine how much data to send to individual devices based on their link budget. The HSPA system uses 16 codes to which 15 are assigned to HSPA. The node B then determines how many codes to assign to individual devices within the cell at any given 2ms time slot which in turn determines the overall speed data is sent out at. The node B can assign all time slots and all 15 codes to a single device in the cell and if that device reports good signal conditions, the maximum data rate can be achieved.
Hybrid Automatic Repeat reQuest (HARQ) – this technique is employed to correct errors in the transmission of the packets between the node B and the user’s device. The device requests a retransmission of any packets that are in error whilst storing all the old erroneous packets. The device then soft recombines all the packets to correct the errors. As it stores all the packets in error and uses them to correct the transmission, a more reliable / efficient method is achieved.
HSUPA – High Speed Uplink Packet Access – utilises the same techniques as HSDPA in terms of link adaptation on the modulation deployed and HARQ to improve the uplink and therefore create synchronous data transmissions of up to 5.7Mb/s. A few differences are in the way the scheduling works in order to ‘service’ all the devices uploading and the reduced modulation schemes :
Scheduling – this is a request-grant mechanism similar to the Fast Packet Scheduling above but initiated by the device. The device requests permission to send data and the node B determines, based on cell loading, requests and power levels within the cell, when and how many devices to be granted permission and at what speeds etc.
Non-Scheduled – for certain applications where delay based on the Scheduled request and the node B overhead would be too great such as VoIP, there is another method where the device initiates the transmission. In these cases the power level is set by the device and typically stays constant. With Scheduled request-grant activity, the node B determines the power level of the device transmission and is controlled dynamically to ensure maximum efficiency for all devices on that cell.
HSPA Evolved – Sometimes referred to HSPA+ or I-HSPA (slightly different but amounts to the same end goal for the user). This system will enhance the downlink to provide 42Mb/s by utilising 64QAM modulation and the uplink to 11.5Mb/s through 16QAM. A further enhancement to help in achieving the increase data rates is the addition of MIMO antennas (multiple in multiple out) – usually deployed to enhance the system performance by a factor of 4. Other features include reducing latency by keeping the devices in a different state when inactive and the full backward compatibility to Rel99 / R5 & R6 of UMTS/HSPA.