Three main findings
Motorola has said a multi-user, outdoor performance trial has produced three main findings operators will need to take note of when introducing HSDPA. Video needs prioritisation, as just four active users in a site can degrade performance. Handsets need equalisation technology included. And base stations need lots of processing power.
The first is that sufficient processing power at the base station is needed to reduce latency in certain applications, such as web-browsing. Effectively, Motorola is saying, operators need enough power at the base station to effect complex scheduling, and avoid state-switching users – the cause of latency.
A Motorola statement explained,
” These delays can be attributed to a mechanism known as state switching . This effectively moves a user from a high to a low speed state based on user activity without the user being aware. When the user requests data, such as clicking a hyperlink for a web page, there can be delays of seconds as the radio network transitions back from a slow to a fast state. Users will not expect such delays for services marketed as “Mobile Broadband”.
“There is a solution – avoid switching users down from a high to a low speed state regardless of user activity. As HSDPA enables radio resources to be dynamically shared between users (known as scheduling), the necessity to state switch should be reduced. However, scheduling requires intensive computer processing capabilities at the base station. Network operators should ensure that they have sufficient processing power at the base stations to schedule the highest possible number of calls at launch of the HSDPA service.”
The second lesson learnt, the vendor said, is to adopt handset funcionality for improved performance. Specifically, handsets need equalisation technology to enhance performance when on the move.
“The performance of HSDPA is heavily dependent on device or handset capability. A signal processing function known as an equaliser enhances performance when the user is moving. Initial trial results demonstrate an increase in data rates of as much as 40% for devices that support equalisation. Today only a few device manufacturers claim to have equaliser functionality.”
The third issue Motorola raised is that of prioritisation.
“The trials have shown that video streaming performance degrades when a relatively modest number of users are active. As little as four active users are sufficient to cause video streaming to freeze if scheduling priorities are not set properly. To compensate for this, operators must actively prioritise video over other services or provide more capacity. Operators could defer video services on HSDPA to a later stage, but as video services consume a large amount of UMTS capacity they should be moved to HSDPA for improved efficiency. This will reduce the cost to deliver video services.
To ensure a mobile broadband user experience Motorola recommends early introduction of Quality of Services features for video.”
Motorola’s HSDPA experience is industry leading and through the identification of these three guidelines, a result of the real-life trials, we can help operators significantly enhance the consumer experience of HSDPA at launch, Raghu Rau, senior vice president, global marketing and strategy, Motorola Networks. The trials are specifically designed to help build optimal performance HSDPA-enabled networks and to effectively deploy services on those networks.
Mobile Europe has found Motorola to be a very realistic vendor in terms of released results of HSDPA trials. These findings, slanted as they may be towards Motorola’s own capabilities, should give operators decent pause for thought.