This diploma thesis is deals with Voice over IP and is focussed on answering the question, if traditional telephony is long gone. We discovered, that the signalisation in VoIP still hasn't been standardized. This means, with SIP and the H.323-Framework two different and competing protocols exist. From our point of view, SIP will be able to establish itself on the market. Beside these two main actors, many manufacturers provide proprietary signalisation solutions. Due to the fact, that in the Zurcher Kantonalbank a homogen infrastructure is implemented, we propose to consider the possibility of choosing a proprietary solution.

VoIP is exposed to exactly the same security risks as every other internet application. But there are several other risks, which are specially concerning VoIP. Specially to consider is SPIT (Spam over Internet Telephony), whose vulnerability can't be rated at the moment. The theory is able to ward all these dangers of, but in practice some implementation problems emerge. This diploma thesis is able to disprove well known prejudices about VoIP, for example that the quality of voice transmission can't be achieved or that unconquerable security risks exist. The image of VoIP being able to save immense costs has to be handled with care. A new acquisition or a migration to VoIP causes substantially infrastructural and personal costs. The operational costs however can in fact be lowered.

Regarding the Zurcher Kantonalbank, we see VoIP as a great opportunity, because the Zurcher Kantonalbank already has their own, branch-spanning IP-network. This network is able to carry the additional VoIP-traffc without any extensions. Furthermore the telephony infrastructure only needs few adaptations. The only decision which is left and which the Zurcher Kantonalbank has to make, is to decide whether Softphones in combination with Headsets shall be introduced or if the classic telephones on the desk shall be maintained.