What is High Speed Packet Access?
High-Speed Packet Access (HSPA) is a mobile data communication protocol used in 3G mobile telecommunications networks. HSPA was developed by the 3rd Generation Partnership Project (3GPP), a collaboration between groups of telecommunications standards organizations, and has been introduced as the evolution of Universal Mobile Telecommunications System (UMTS) and Wideband Code Division Multiple Access (WCDMA).
Why is it called High-Speed Packet Access?
The main characteristics that differentiates HSPA from its predecessor WCDMA are higher speed data transmission; real-time, variable bit rate services; improved system capacity; fast packet access (although it’s not related to the 802.11p AP-based MAC layer fast access feature); and reduced latency.
How does it achieve higher speed data transmission?
HSPA employs packet switching which utilizes the channel more efficiently than circuit switching, delivering data faster. The basic idea of packet switching is to break up messages into smaller parts that are then sent separately through the network. Each part travels along its own path by taking different routes, so multiple paths can be employed simultaneously for a single message. Once all parts of the message arrive at their destination, they are reassembled by the device receiving them. This makes it possible to address multiple destinations in one message while delivering each part of it on an individual basis. Moreover HSPA supports adaptive modulation i.e., it changes between higher numbers (256 QAM) and lower numbers (64 QAM) of quadrature amplitude modulation, depending on whether sufficient radio resources are available.
How does it achieve real-time, variable bit rate services?
Real-time means that the time delay between receiving or sending data is brief enough to be acceptable for the application. Variable bit rate means that the average amount of data transmitted changes dynamically according to transmission conditions, user behaviour etc., whereas fixed bit rate is when all users receive the same amount of data per unit time. Since packet switching allows channels to be shared dynamically, HSPA can offer both higher peak rates (up to 28 Mbit/s under good conditions) and higher average rates than UMTS. To increase its capabilities, HSPA uses multiple-input (and multiple-output) for diversity combining of radio signals sent from the same or different base stations.
How does it achieve improved system capacity?
It can be achieved with cell splitting, which is when one cell site is divided into two smaller areas with separate power and backhaul resources. This increases the number of users served in a given area without increasing interference between users. Adjacent sites are used to create larger cells by allocating dedicated channels to neighboring sectors in an overlay scheme called “sector splitting”. With more spectrally efficient modulations like 256 QAM, data rates per sector can go up to 84 Mbit/s. To gain further efficiencies, carriers can implement features such as dynamic channel allocation, which means each cell site dynamically changes its operating frequencies according to traffic conditions. Another way of increasing capacity is by using the 3GPP Long Term Evolution (LTE) standard.
How does it achieve fast packet access?
The WCDMA R99 HSDPA introduced an early form of this feature that was limited to situations where a single mobile station or user equipment (UE) was transferring data continuously for a period of time to/from the network. That method required either dedicated physical channels, or more likely in cases where the UE only sends occasional small packets, an increase in uplink power levels was necessary to compensate for larger amount of transmission errors due to retransmissions. The latter severely reduced battery life of mobile devices.
Future, more advanced versions of HSDPA announced in release 7 (WCDMA R7) and its follow-up (HSPA+) extend the original early form of fast packet access to situations where a mobile station is transferring more than one data stream simultaneously to/from the network on different transport channels; this requires no increase in uplink power levels due to retransmissions and therefore significantly increases battery life of user equipment while making use of otherwise wasted UE power capacity. This feature can further be improved by also applying it for downlink packets, improving latency and allowing the usage of large TCP window sizes.
How does it achieve reduced latency?
Latency involves waiting time between receiving or sending data. It includes the time taken in the relaying network nodes. To reduce this, HSPA Plus has introduced many features such as adaptive modulation and coding (AMC) for faster adaptation to channel conditions, fast scheduling of packets via dedicated packet schedulers located in the radio-base station nodes, reordering of out-of-sequence packets which are often observed when TCP/IP is used over wireless links, support for small UDP data packets that result in improved efficiency in applications like VoIP where retransmissions are not practical even if they increase latency slightly, and hybrid ARQ (HARQ) i.e., negative acknowledgments using soft or hard combining that reduces latency by allowing fewer retransmissions.
How does it achieve higher data rates?
It does so by applying more spectrally efficient wideband or higher order modulation schemes, e.g., 16-QAM for existing 3G-systems and 64-QAM plus support for carrier aggregation in LTE Advanced. Carrier aggregation allows the bundling of up to 5 component carriers that are located on different frequencies into a single transport channel, which can then be processed using wider signal bandwidths. This supports combined bitrates of up to 1 Gbit/s via dual-channel MIMO techniques.
How does it achieve lower cost per bit?
By adding multi-antenna transmission modes to HSPA, this improves spectral efficiency per cell all the way from 384 kbit/s.
