Dsdm is a two-channel digital signaling method which is not compatible with the traditional ptt. What are the advantages and disadvantages of dsdm?
As for advantages, to talk about using dsdm technology, there are three main points: first, easy to use; second, high security level; thirdly, shorter call latency.
After all, it’s acknowledged that the continuous innovation in technology provides us with better communication service levels. However, technological development brings along both good and bad stuff. here i’d like to point out some negative aspects of dsdm compared with ptt system: firstly, voice quality will be affected because low bit rate encoding is adopted in transmission process by dsdm technology, secondly, dsdm relies on server, so if the server is unavailable or busy, the service will be affected as well.
To sum up, with regard to its advantages and disadvantages, there’s no perfect product in this world. What matters most is how to optimize the technology for the utmost benefit of us users. DSDM must be qualified as a safe and low-cost technology after all these years’ development and implementation experience. we should attach more attention to aspects like voice quality and emergency cases that require quick response instead of negative points like security risks. (by joff)
How does it work?
The main idea behind DSDM is “in complex environment” requirements; for military application cases which involves various platforms, various users and various networks.
DSDM is an evolutionary service driven architecture (SA) system that has evolved from the United States armed forces’ information technology (IT) system for military use. DSDM aims to deploy IT equipment quickly to adapt to changes in military operations; it also facilitates re-inspection of the systems after deployment, providing flexibility for future modifications.
How does voice quality improve?
The keypoint of DSDM’s voice quality improvement is “delay”. under certain conditions, you might be surprised at how good your call becomes after adding just 10ms delay into the path! It is inevitable that a packet network performs real-time transmission function, but what can we do for this real-time requirement?
there are several options:
1) Let the packet network transmit & handle data loss via retransmission in case of data lost, not much different from traditional telephony networks.
2) Let end-users bear with real-time requirements. When real-time is required, voice samples are not transmitted through packet networks, nor are they converted into Internet Protocol (IP) packets; instead, they are transported in their native format to the destination point directly over a reliable circuit switched connection.
3) take advantage of delay compensation techniques applied in existing public switched telephone networks (PSTN).
This final approach aims at predicting future speech frames by using past information which is cached in memory, and replaying it afterwards. the delay compensation techniques are used to avoid any additional latency that would be introduced if such an approach were applied.
This latter technique is particularly suited for highly interactive applications such as video games with speech communication or dictation computer programs with voice input. one can see that these required characteristics do not contradict each other, but rather complete each other: high-level QoS requirements cannot be met without taking care of low layer QoS constraints and vice versa .
Advantages of dsdm are:
- a) easy to use and cross-platform,
- b) a good match for interactive applications with a tight latency budget,
Disadvantages of dsdm are:
- a) absence of timing information,
- b) absence of trends,
- c) language dependency.
ALSO READ: d2l mnsu
DSDM is an API that makes it easy to design interactive real-time programs with low latency constraints. Without having any knowledge about how the underlying layers are working one is able to develop software systems which meet their deadlines easily and reliably. This is achieved by specifying desired QoS levels (bandwidth, latency…) together with some strategies for meeting them (e.g., dropping packets when the network cannot handle the load). The dsdm core system will try its best to provide these QoS levels but cannot tell if the application meets its requirement or not since it is the application’s responsibility to provide the system with this information.
Our research investigates how we can build a profiling infrastructure that enables us to detect which parts of an application fail to meet their deadlines and why, so as to enable users (or tools) to improve it.
DSDM is not a tool or a framework: there are no dsdm compilation units, no libraries, no bindings… It only exists as source code and documentation. This helps us maintain quality since we do not depend on anything else but our own work and we choose the technologies we use ourselves. As such we have complete control over everything and can adapt it without too much trouble should our needs change. The downside of this approach is that we cannot brag about our tool or framework being state of the art nor can we offer comprehensive documentation. We have to rely on what people themselves make out of it and on some tutorials.
