Computer-implemented method of optimizing bandwidth in a PSAP environment and communication network

1. A computer-implemented method of optimizing bandwidth in a public safety answering point (PSAP) environment comprising a PSAP server having a processor connected to a non-transitory computer readable medium, wherein the method comprises:

receiving a Real Time Processing (RTP), stream relating to an emergency call comprising video data and audio data;

retrieving content related to emergency incidents from a Social Media (SM) server;

feeding the content to a Stream Processing Module (SPM) comprising a Machine Learning (ML) engine for determining a similarity level between the emergency call and any of the emergency incidents retrieved from the SM server; and

in response to the similarity level exceeding a predetermined threshold value, providing an option for downgrading the emergency call to a call comprising only audio data, and

in response to the option being selected, downgrading the RTP stream to an audio only call.

2. The method of claim 1 , comprising:

in response to the similarity level being below the predetermined threshold value, then the emergency call comprising video data and audio data being maintained.

3. A computer-implemented method of optimizing bandwidth in a public safety answering point (PSAP) environment comprising a PSAP server having a processor connected to a non-transitory computer readable medium, wherein the method comprises:

receiving a Real Time Processing (RTP), stream relating to an emergency call comprising video data and audio data;

retrieving content related to emergency incidents from a Social Media (SM) server;

feeding the content to a Stream Processing Module (SPM) comprising a Machine Learning (ML) engine for determining a similarity level between the emergency call and any of the emergency incidents retrieved from the SM server;

in response to the similarity level being below the predetermined threshold value, then the emergency call comprising video data and audio data being maintained;

in response to the similarity level exceeding a predetermined threshold value, providing an option for downgrading the emergency call to a call comprising only audio data, and

in response to the option being selected, downgrading the RTP stream to an audio only call;

after the downgrading to the audio only call, negotiating to adjust the audio only call to a call comprising video data and audio data.

4. The method of claim 3 , comprising:

initially filtering network traffic for identifying whether the received RTP stream related to the emergency call requires transcoding.

5. The method of claim 4 comprising:

transcoding emergency calls that cannot be directly connected to the PSAP server.

6. The method of claim 2 , comprising: A computer-implemented method of optimizing bandwidth in a public safety answering point (PSAP) environment comprising a PSAP server having a processor connected to a non-transitory computer readable medium, wherein the method comprises:

receiving a Real Time Processing (RTP), stream relating to an emergency call comprising video data and audio data;

retrieving content related to emergency incidents from a Social Media (SM) server;

feeding the content to a Stream Processing Module (SPM) comprising a Machine Learning (ML) engine for determining a similarity level between the emergency call and any of the emergency incidents retrieved from the SM server;

in response to the similarity level being below the predetermined threshold value, then the emergency call comprising video data and audio data being maintained;

in response to the similarity level exceeding a predetermined threshold value, providing an option for downgrading the emergency call to a call comprising only audio data, and

in response to the option being selected, downgrading the RTP stream to an audio only call;

performing a training phase in which the SPM retrieves content related to emergency incidents from the SM server and/or from other sources of information and/or from a database of the PSAP.

7. The method of claim 6 , wherein in the training phase includes:

annotating a video and/or an image posted on social media and retrieved from the SM server to define several visual features per frame or shot which train a Hidden Markov Model (HMM) algorithm so as to determine whether a post refers to an emergency incident or not.

8. The method of claim 7 , wherein the visual features comprise color, shots, and object-based features.

9. The method of claim 6 , wherein the content related to emergency incidents comprises images and/or videos and/or location data and/or dates of the corresponding posts.

10. The method of claim 9 , comprising:

in response to the similarity level for an emergency incident retrieved from the SM server and/or from other sources of information exceeding the predetermined threshold value, forwarding the data related to the emergency incident from the PSAP server to a call taker handling the emergency call.

11. The method of claim 10 , wherein the data related to the emergency incident comprises an image and/or the video and/or the location data and/or the date of the corresponding post.

12. A computer-implemented method of optimizing bandwidth in a public safety answering point (PSAP) environment comprising a PSAP server having a processor connected to a non-transitory computer readable medium, wherein the method comprises:

receiving a Real Time Processing (RTP), stream relating to an emergency call comprising video data and audio data;

retrieving content related to emergency incidents from a Social Media (SM) server;

feeding the content to a Stream Processing Module (SPM) comprising a Machine Learning (ML) engine for determining a similarity level between the emergency call and any of the emergency incidents retrieved from the SM server;

in response to the similarity level exceeding a predetermined threshold value, providing an option for downgrading the emergency call to a call comprising only audio data,

in response to the option being selected, downgrading the RTP stream to an audio only call; and

retrieving data related to emergency incidents posted on social media and stored on the SM server and/or from other sources of information periodically, and storing the retrieved data on a database of the PSAP.

13. A communication apparatus comprising a telecommunication device adapted to carry out a computer-implemented method of optimizing bandwidth in a public safety answering point, (PSAP), wherein the communication apparatus comprises:

a telecommunication device having a processor connected to a non-transitory memory, the non-transitory memory having code stored thereon that defines a method performed by the telecommunication device when the processor runs the code, the method comprising:

receiving a Real Time Processing (RTP), stream relating to an emergency call comprising video data and audio data;

retrieving content related to emergency incidents from a Social Media (SM) server;

feeding the content to a Stream Processing Module (SPM) comprising a Machine Learning (ML) engine for determining a similarity level between the emergency call and any of the emergency incidents retrieved from the SM server;

in response to the similarity level being below the predetermined threshold value, then the emergency call comprising video data and audio data being maintained;

in response to the similarity level exceeding a predetermined threshold value, providing an option for downgrading the emergency call to a call comprising only audio data; and

in response to the option being selected, downgrading the RTP stream to an audio only call; and

after the downgrading to the audio only call, negotiating to adjust the audio only call to a call comprising video data and audio data.

14. The communication apparatus of claim 13 , wherein the apparatus is configured as a communication network comprising at least one of: a Session Border Controller (SBC), an Emergency Service Routing Proxy (ESRP), a PSAP server, at least one Social Media server, and a Media Server.