oFono - Open Source Telephony ***************************** Copyright (C) 2008-2009 Intel Corporation. All rights reserved. Mission statement ================= The overall goal of the oFono project is to create a telephony host stack for embedded/mobile and desktop systems based on Linux. It currently targets GSM/UMTS User Equipment (UE) based on 3GPP standards, but is of course not limited to that. Extensions for other telephony systems like CDMA/EVDO are more than welcome. Within oFono there is clear abstraction between the application interfaces based on D-Bus, the hardware level (via drivers) and the integration with other system components (plugins). The whole architecture is modular and flexible. Telephony features ================== The oFono stack will support the majority of the features from the 3GPP specification, but not all of them. The whole standard is pretty complex and some features and not used in any mobile network so far, some of them are outdated and have no relevance anymore. This paragraph tries to give some insights on what has been implemented so far, what is coming in the future and especially what will not be part of oFono. Current implemented features: Modem abstraction Currently there are two modem drivers available. The "atmodem" driver handles hardware based on the 3GPP TS 27.007 standard. The "isimodem" driver handles Nokia based PhoNet modems. Network registration The network registration interface contains support for network detection. It also handles the automatic or manual registration to a mobile network. Additional information about the current network can be easily retrieved and displayed to the user. Voice call handling The voice call interface handles simple call creation and termination. It also supports 3way-calling and multi-party features. This is only for voice control. It doesn't contain support for the voice data path/routing. Advanced voice call control Features like COLR/CLIR/CLIP/COLP are supported by the voice call handling interface. The support for call forwarding and call waiting is also present. Interfaces for call barring and advice of charge do exist, but highly depend on if the operator supports them. Call history The call history is realized via a plugin interface and not handled directly. This allows an easy integration with storage systems like Evolution-Data-Server. Phonebook support The storage of the SIM card is not used. The only supported feature is to export the contacts stored on the SIM card to some third-part entity. SIM card storage is limited, slow and not flexible enough for modem telephony applications. Short message service The SMS support is available, but is currently still missing some integration with the driver framework. Work in progress features: SIM PIN handling At this moment only SIM cards with deactivated PIN request are supported. The full support for SIM PIN is planned, but not yet implemented. Cell broadcast and USSD Both features are work in progress right now. GPRS data connection No work on this has been done so far, but it is planned feature. Not implemented features: SIM card storage The SIM card storage will not be used by oFono. It is slow, limited in size and flexibility. It is an outdated interface that makes no sense in a modern telephony system. Export of stored contacts from the SIM card is supported for legacy reasons and to allow a smooth transition. WAP support The WAP feature is outdated and uses full Internet access with an embedded browser is the future. MMS support The MMS support should not be part of oFono itself. The best idea is to implement a MMS service that runs in the user session and uses oFono's SMS interface to listen for notifications. Similar to the split between BlueZ and OBEX daemon. Especially when it comes to image conversion and other tasks that MMS support requires it is important to NOT do this as a system daemon. EMS support This is an Ericsson specific standard and not widely spread across the handset manufactures. Video telephony Currently there are no plans to support this. The support from the networks and available handsets are still limited. This needs re-evaluation once such a service becomes more prominent. A similar split like with MMS might be a good idea.