Method for controlling traffic to a computing node in a communication network

Abstract

Method performed by a network node (201) for controlling traffic to a computing node (202) in a communication network, the method comprising receiving (301) incoming content (203) to be forwarded to the computing node (202), dividing (302) the incoming content (203) into prioritized data (203a) and non-prioritized data (203b), routing (303) the prioritized data (203a) over a first communication channel (207a) to the computing node (202), compressing (304) the non-prioritized data (203b) to form compressed data (204) and routing (305) the compressed data (204) over a second communication channel (207b) to the computing node (202).

Description

[0001] METHOD FOR CONTROLLING TRAFFIC TO A COMPUTING NODE IN A COMMUNICATION NETWORK

[0002] TECHNICAL FIELD

[0003] The present disclosure relates generally to communications, and more particularly to methods for controlling traffic to a computing node in a communication network, as well as to methods for receiving incoming content in a communication network.

[0004] BACKGROUND

[0005] Network traffic in communication networks continues to increase. As an example, the Mobility Report from Ericsson dated June 2024 predicts an increase in mobile network traffic of more than 450 exabyte (EB) per month globally by 2024. The network traffic increase is compounded by the growing volumes of generated content and real-time Artificial Intelligence (Al) interactions, highlighting the urgent need for scalable and efficient network solutions to support these demands.

[0006] Even though there is currently no major issue with network bandwidth needed for generative video, it is expected to become a problem soon. Consequently, ensuring scalable and efficient network infrastructure becomes a paramount concern for service providers worldwide to address the need of the new services relying on generative Al.

[0007] Classical methods dealing with excessive use of network bandwidth include data compression combined with traffic prioritization, leading to decreased quality of experience on the receiving side (e.g. incoming video stream with lower resolution or frame rate). A disadvantage with these methods is that the receiver of the content normally gets the result in lower resolution because of the compression. A parallax on the classical data compression methods is to use so called “semantic communication”, wherein a description of the content, instead of the content itself is sent. The idea is that instead of transmitting the actual information, as traditional communication systems do, a semantic communication (which will herein also be referred to as semantic compression) is performed. Thus, semantic compression systems translate the actual content to a semantic representation of the content on the transmitter side. The semantic representation of the content is transmitted to the receiver side and used to generate the content on the receiver side. As is the case with classical data compression methods, in semantic compression, the content generated on the receiver side can have larger or smaller differences to the original content, depending on the accuracy of the semanticinformation. Thus, the operation of both prior art compression methods (i.e., data compression and semantic compression) may have a negative effect on the quality of experience at the receiver side.

[0008] It is therefore necessary to develop new methods for controlling traffic in communication networks, especially in wireless networks, which reduce required network bandwidth without adversely affecting the end-user experience.

[0009] SUMMARY

[0010] It is an aim of the present disclosure to provide methods, a network node and a computing node which at least partially address one or more of the challenges mentioned above. It is a further aim of the present disclosure to provide methods, a network node and a computing node for compressing content in such a way that network bandwidth can be saved while minimizing the effect on the perceived quality of the content at the receiver side.

[0011] According to some embodiments, a computer implemented method performed by a network node for controlling traffic to a computing node in a communication network is provided. The method includes obtaining incoming content to be forwarded to the computing node. The method further includes dividing the incoming content, into prioritized data and non-prioritized data. The method further includes routing the prioritized data over a first communication channel to the computing node. The method further includes compressing the non-prioritized data to form compressed data and routing the compressed data over a second communication channel to the computing node.

[0012] According to other embodiments, a computer implemented method performed by a computing node for receiving incoming content in a communication network is also provided. The incoming content comprises prioritized data and compressed data corresponding to nonprioritized data. The method includes receiving the prioritized data through a first communication channel. The method further includes receiving the compressed data, corresponding to the non-prioritized data, through a second communication channel. The method further includes decompressing the compressed data to obtain the non-prioritized data and combining the prioritized data and the non-prioritized data to obtain the incoming content.

[0013] According to other embodiments, a network node, a computing node, computer programs, computer program products, and non-transitory computer readable mediums are provided to perform the computer implemented methods above.Various embodiments described herein are capable of saving bandwidth while guaranteeing perceived quality of the content forwarded over the communication network to the computing node. This is achieved thanks to the fact that the methods according to the present invention reduce the size of the data to be transmitted by dividing the content thereof into two different categories, prioritized data and non-prioritized data, so that the prioritized data can be transmitted unmodified, while the non-prioritized data undergoes compression. In this way, a reduction in the size / volume of data to be transmitted can be achieved, with minimum effect on the perceived quality of the content.

[0014] BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The accompanying drawings, which are included to provide a further understanding of the disclosure and are incorporated in and constitute a part of this application, illustrate certain non-limiting embodiments of the inventive concepts in the drawings:

[0016] Fig. 1 is a schematic diagram illustrating data flow in a communication network subject to prior art traffic control methods.

[0017] Fig. 2 is a schematic diagram illustrating data flow in a communication network subject to a method for controlling traffic according to an embodiment of the present invention described in Fig. 3.

[0018] Fig. 3 is a flow diagram illustrating a method for controlling traffic in a communication network according to an embodiment of the present invention.

[0019] Fig. 4 is a flow diagram illustrating a method for receiving incoming content in a communication network according to an embodiment of the present invention.

[0020] DETAILED DESCRIPTION

[0021] Some of the embodiments contemplated herein will now be described more fully with reference to the accompanying drawings. Embodiments are provided by way of example to convey the scope of the subject matter to those skilled in the art. Inventive concepts may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of present inventive concepts to those skilled in the art. It should also be noted that these embodiments are not mutually exclusive.Components from one embodiment may be tacitly assumed to be present / used in another embodiment.

[0022] Fig. 1 is a schematic diagram illustrating the data flow in a communication network subject to two different prior art traffic control methods, aimed at saving bandwidth. The method shown in the upper part of Fig. 1 schematically illustrates the data flow corresponding to a traffic control method based on conventional data compression, while the method shown in the lower part thereof illustrates the data flow corresponding to a traffic control method based on semantic compression.

[0023] The traffic control method based on data compression begins when an incoming content 103 is received at a network node 101. In this particular case, the incoming content 103 is an image of a flower (schematically shown in Fig. 1). Then the network node 101 acts as a data compressor and generates a new image, the compressed data 104, which is based on the incoming content 103 but has a reduced size and lower resolution, so it can therefore be transmitted using a smaller bandwidth. The compressed data 104 is subsequently routed over a communication channel 107 to a computing node 102. Finally, the computing node 102 decompresses the compressed data 104 to obtain the generated content 106, which is based on the incoming content 103 but may have some loss of information due to the data compression.

[0024] Similarly, the traffic control method based on semantic compression also begins when an incoming content 103 is received at a network node 101. To facilitate the understanding of the differences existing between the two methods, in this case the incoming content 103 is also an image of a flower. Then the network node 101 acts as a semantic compressor and replaces the incoming content with a description of its original content. In this particular case, said description (“an image of a flower in black and white with eight pedals and an anther”) corresponds to the compressed data 104 and has even smaller size than in the previous traffic control method based on data compression. This description is subsequently routed over a semantic channel 105 to a computing node 102. Finally, the computing node 102 employs the description in the compressed data 104 to generate the generated content 106, which in this case is a flower based on the incoming content 103 but which has some different features, such as the shape of the pedals.

[0025] Fig. 2 is a schematic diagram illustrating the hardware components present in a communication network subject to a traffic control method according to an embodiment of thepresent invention, as well as the corresponding data flow in the communication network. This traffic control method will be described in more detail below, in relation to Fig. 3.

[0026] The embodiment of the invention shown in Fig. 2 comprises a network node 201 provided with processing circuitry 201a and memory 201b coupled to the processing circuitry 201a. The memory 201b has instructions stored therein that are executable by the processing circuitry 201a to cause the network node 201 to perform operations perform comprising any of the operations of the traffic control method described herein, e.g., perform one or more of the steps described below with reference to Fig. 3.

[0027] The network node 201 may be, for example, but not limited to: an access point (APs) (e.g., radio access point), a base station (BS) (e.g., radio base station, Node B, evolved Node B (eNB) and New Radio (NR) NodeB (gNB)) and a user equipment (UE).

[0028] User equipment (UE) include, but are not limited to, a smart phone, mobile phone, cell phone, voice over IP (VoIP) phone, wireless local loop phone, desktop computer, personal digital assistant (PDA), wireless cameras, gaming console or device, music storage device, playback appliance, wearable terminal device, wireless endpoint, mobile station, tablet, laptop, laptop-embedded equipment (LEE), laptop-mounted equipment (LME), smart device, wireless customer-premise equipment (CPE), vehicle-mounted or vehicle embedded / integrated wireless device, etc. Other examples include any UE identified by the 3rd Generation Partnership Project (3GPP), including a narrow band internet of things (NB-loT) UE, a machine type communication (MTC) UE, and / or an enhanced MTC (eMTC) UE.

[0029] The network node 201 is arranged for obtaining incoming content 203. The present invention contemplates the possibility that the incoming content 203 is received from a content provider, but also may be obtained internally (for example by being created / generated internally by the network node 201 or retrieved from an internal memory). The incoming content is to be routed to a receiver which in the shown embodiment is a computing node 202.

[0030] Optionally, the network node 201 may also receive information 208 describing the status of the communication network, as will be further described below.

[0031] The processing circuitry 201a of the network node 201 may act as an element prioritizer 209, dividing the incoming content 203 into prioritized data 203a and non-prioritized data 203b. Theprioritized data 203a is a part or segment of the incoming content 203 with high importance, e.g., more important than one or more other parts of the incoming content 203. Therefore, it is desirable that the computing node 202 is able to receive the prioritized data 203a with high quality, i.e., without the prioritized data 203a being distorted. Thus, if the content is a video sequence, the prioritized data 203a could be, for example and without being limitative, the portion of the images corresponding to the main characters that participate in said video sequence. Conversely, non-prioritized data 203b is a part or segment of the incoming content 203 of lower importance and therefore for which the quality of the reception is less important. In the case of the incoming content being a video sequence, the non-prioritized data 203b could be, for example and without being limitative, the portion of the images corresponding to the background of the scene.

[0032] In the embodiments of the invention shown in Figs. 2 and 3, this division between prioritized and non-prioritized data, may be made depending on the information 208, so the most appropriate content division can be chosen depending on the status of the communication network, e.g., available bandwidth and / or channel quality of the communication channels between the network node 201 and the computing node 202, as well as load and / or energy status of the network node 201 and the computing node 202.

[0033] Moreover, the processing circuitry 201a of the network node 201 may route, e.g., send or transmit, with the help of a first content synthesizer 210 and a channel encoder 211, the prioritized data 203a over a first communication channel 207a to the computing node 202. The prioritized data 203a is thereby sent directly in raw form to the computing node 202, that is without any compression. The first communication channel 207a may in embodiments be a wireless communication channel.

[0034] Conversely, the non-prioritized data 203b may be compressed by the first content synthesizer 210 and converted into compressed data 204 before being actually sent to the computing node 202. The present invention contemplates the following types of compression of the nonprioritized data 203b: data compression (by which data size of the non-prioritized data 203b is reduced, for example and without being limitative by means of a compression algorithm), semantic compression (by which the non-prioritized data 203b is replaced by a description of its content), or both to data compression and to semantic compression.

[0035] More in particular, at least one of the following compression methods could be used to compress the non-prioritized data 203b: Huffman coding, arithmetic coding, run-lengthencoding, delta encoding, Fourier transform compression, wavelet transform, layer 3 audio compression, LZ77 or LZ78 compression, prediction and entropy coding, semantic matching compression, knowledge graph compression, concept based encoding, neural semantic compression, transformed based sparse encoding; and contextual compression.

[0036] In the embodiment shown in Figs. 2 and 3, the channel encoder 211 routes the compressed data 204 over a second communication channel 207b to the computing node 202. In addition, at least one prompt 205 including an explanation on the content of the compressed data 204 may be generated and transmitted to the computing node 202 over a third communication channel 207c. The second communication channel 207b and the third communication channel 207c may in embodiments be wireless communication channels.

[0037] As already anticipated, the embodiment of the invention shown in Fig. 2 also comprises the computing node 202 provided with processing circuitry 202a and memory 202b coupled to the processing circuitry 202a. Said memory 202b has instructions stored therein that are executable by the processing circuitry 202a to cause the computing node 202 to perform operations comprising any of the operations of the method for receiving incoming content 203 described herein, e.g., perform one or more of the steps described below with reference to Fig.

[0038] 4.

[0039] The computing node 202, which may also be referred to as a computing device, a client device or UE, may be, for example, but not limited to: a smart phone, mobile phone, cell phone, voice over IP (VoIP) phone, wireless local loop phone, desktop computer, personal digital assistant (PDA), wireless cameras, gaming console or device, music storage device, playback appliance, wearable terminal device, wireless endpoint, mobile station, tablet, laptop, laptop-embedded equipment (LEE), laptop-mounted equipment (LME), smart device, wireless customer-premise equipment (CPE), vehicle-mounted or vehicle embedded / integrated wireless device, etc. Other examples include any UE identified by the 3rd Generation Partnership Project (3GPP), including a narrow band internet of things (NB-loT) UE, a machine type communication (MTC) UE, and / or an enhanced MTC (eMTC) UE.

[0040] The computing node 202 may be on the same communication network as the network node 201 or on another communication network. The computing node 202 may further be a network endpoint, e.g., located outside the demarcation zone of a mobile operator, such as e.g., a public cloud service or an enterprise server on a private network.The processing circuitry 202a of the computing node 202 may act as a channel decoder 212 and receives prioritized data 203a over the first communication channel 207a and compressed data 204 over the second communication channel 207b. In the particular embodiment of the invention shown in Fig. 2 and 3, the at least one prompt 205 including explanations on the content of the compressed data 204 may further be received over the third communication channel 207c.

[0041] Moreover, the processing circuitry 202a of the computing node 202 may also act as a second channel synthesizer 213 which decompresses the compressed data 204 to restore the nonprioritized data 203b. In this particular embodiment of the invention, decompression may be carried out taking into account the explanation on the content of the compressed data 204 included in the at least one prompt 205. Finally, the second channel synthesizer 213 combines the prioritized data 203a and the non-prioritized 203b to retrieve the incoming content 203.

[0042] The computer implemented method for controlling traffic of Fig. 3 will be now described. This method begins when incoming content 203 is obtained, at step 301, at a network node, such as the network node 201 shown in Fig. 2.

[0043] Optionally, the network node 201 also receives information 208 describing the status of the communication network including one or more of the status of the network node 201, the computing node 202, the first communication channel 207a and / or the second communication channel 207b.

[0044] Then, in step 302 the processing circuitry 201a of the network node 201, e.g., acting as an element prioritizer 209, divides the incoming content 203 into prioritized data 203a and nonprioritized data 203b. The division may comprise the network node 201 identifying parts / elements of the incoming content 203 with high importance, based on the incoming content 203 itself and other available information, and categorizing it as prioritized data 203a. Parts of the incoming content 203 with lower importance may then be categorized as nonprioritized data 203b.

[0045] Different techniques can be used to extract importance of different parts of the incoming content 203, depending on the type of incoming content 203, i.e., if the incoming content 203 comprises text, images, video and / or audio. For example, for text, semantic analysis can determine the significance of elements based on their meaning and context, while dependency parsing can identify key components by analysing grammatical dependencies. Named EntityRecognition (NER) prioritizes named entities like people, locations, and organizations, which often hold significant importance. The Term Frequency-Inverse Document Frequency (TF-IDF) method highlights unique or critical terms by evaluating word importance based on its frequency in a specific document relative to a larger corpus. Contextual embeddings can assess word and phrase importance based on their usage in context. Keyphrase extraction algorithms, such as Rapid Automatic Keyword Extraction (RAKE) or TextRank, are designed to extract key phrases from text. Machine learning models can assign salience scores to different parts of the text, indicating their relative importance. Position of elements in the text with most important elements can appear in the introduction or conclusion.

[0046] For incoming content 203 comprising an image, the extraction of important parts from the image can be done using a visual object detector. Different methods can be used for assessing the most important parts of the image, for example it could be the cardinality of an object in the scene, the relative size of the object, the position of the object, the colouring of the object, the semantics behind the object (i.e., its nature), etc.

[0047] For incoming content 203 comprising audio, the extraction of important parts from audio stream may be a product of frequency analysis (e.g., using fast Fourier transformations), to identify recurring patterns that may indicate relative importance. Other techniques that can be used are detection of significant emotional change in the audio stream, change or rhythm and tempo, spotting of keywords, etc.

[0048] The division may further be made depending on the information 208, so the most appropriate content division can be chosen depending on the status the communication network. In this way, the network node 201 can adapt the division of the incoming content 203 to the current network conditions and e.g., categorize a larger part of the incoming content 203 as prioritized data 203a during good network conditions than during poor conditions.

[0049] Information 208 on the status of communication network may include status of the communication channels between the network node 201 and the computing node 202 i.e., the first communication channel 207a and / or the second communication channel 207b in the embodiments shown in Figs. 2 and 3. The status of the first communication channel 207a and / or the second communication channel 207b, can include, for example and without being limitative, at least one of the following: current bandwidth utilization of the first communication channel 207a and / or of the second communication channel 207b, predicted bandwidth utilization of the first communication channel 207a and / or of the second communicationchannel 207b; and an indication of a channel quality (such as a measure of the signal to noise ratio, latency time, rate of packet loss, error rate and / or a quality of service class identifier) of the first communication channel 207a and / or to the second communication channel 207b.

[0050] Similarly, information 208 describing the status of the network node 201 and / or the computing node 202 can optionally include at least the following: load of the network node 201 and / or the computing node 202, and / or available energy reserves of the network node 201 and / or the computing node 202.

[0051] Throughout the present description, “load” should be understood as the available volatile memory (e.g., RAM) at the network node 201 and / or at the computing node 202, also use of the processing unit of the network node 201 and / or the computing node 202 (e.g., central processing unit (CPU), Neural processing unit (NPU) and also available non-volatile storage (e.g., in a hard disk drive)). Similarly, “available energy reserves” should be understood as the battery capacity, current level and / or rate of discharge of the network node 201 and / or the computing node 202.

[0052] In addition, in some embodiments of the present invention which are compatible with this embodiment thereof, the network node 201 may further receive a description 206 of the incoming content 203 and the division of the incoming content 203 may be made depending on said received description 206. This allows the network node 201 to choose a particular data division which best suits the type of content to be forwarded to the computing node 202.

[0053] The description 206 of the incoming content 203 can include, for example and without being limitative, at least one of the following: current bitrate of the incoming content 203, predicted bitrate of the incoming content 203, an indication of whether the incoming content 203 has been generated by an artificial intelligence model or not and a semantic description of the incoming content 203. The description 206 may e.g., be received from the content provider.

[0054] Alternatively or in addition, in embodiments of the invention, the network node 201 may receive dividing instructions from the content provider and may divide the incoming content 203 according to said received dividing instructions.

[0055] In step 303, the processing circuitry 201a of the network node 201 routes, e.g., with the help of the first content synthesizer 210 and the channel encoder 211 , the prioritized data 203a over the first communication channel 207a to the computing node 202. Prioritized data 203a istherefore transmitted directly in raw form to the computing node 202, that is without any compression and hence without any loss of information. The network node 201 may e.g., transmit the prioritized data 203a over the first communication channel 207a to the computing node 202.

[0056] Conversely, the non-prioritized data 203b is compressed, e.g., by the first content synthesizer 210, in step 304, to form compressed data 204 before being sent to the computing node 202. The non-prioritized data 203b may be compressed using any of the previously described data and / or semantic compression methods.

[0057] In some embodiments of the invention the non-prioritized data 203b is further divided in at least two different sets of non-prioritized data depending on its relevance, each set of nonprioritized data undergoing a different type of compression to form a set of compressed data.

[0058] Finally, in step 305 the processing circuitry 201a of the network node 201 routes, e.g., with the help of the first content synthesizer 210 and the channel encoder 211, the compressed data 204 over the second communication channel 207b to the computing node 202. The network node 201 may e.g., transmit the compressed data 204 over the second communication channel 207b to the computing node 202.

[0059] In some embodiments of the invention such as the embodiment shown in Fig. 2, the method for controlling according to the present invention also comprises generating at least one prompt 205 including an explanation on the content of the compressed data 204 and transmitting the prompt 205 to the computing node 202.

[0060] Alternatively or in addition, the present invention also contemplates the possibility of the network node 201 receiving at least one prompt 205 from a content provider including an explanation on the content of the compressed data 204 and transmitting this prompt 205 to the computing node 202.

[0061] Thus, the network node 202 may in addition to the prioritized data 203a and the compressed data 204 further send one or more prompts 205 including an explanation on the content of the compressed data 204 to the computing node 202.The prompt 205 including an explanation on the content of the compressed data 204 may include for example and without being limitative: an explanation in natural language on the content of the compressed data 204; an explanation in natural language processing embedded in a representation; an audio signal representing the content of the compressed data 204; and / or an image representing the content of the compressed data 204.

[0062] Fig. 4 shows a computer implemented method performed by a computing node, such as the computing node 202 shown in Fig. 2, for receiving incoming content 203 in a communication network according to an embodiment of the present invention. According to the methods of the present invention, the incoming content 203 comprises both prioritized data 203a and compressed data 204 (corresponding to non-prioritized data 203b).

[0063] In step 401 , the prioritized data 203a is received through the first communication channel 207a. Similarly, in step 402, the compressed data 204 is received through the second communication channel 207b.

[0064] The method shown in Fig. 4, further comprises additional step 405, wherein at least one prompt 205 including explanations on the content of the compressed data 204 is received through the third communication channel 207c. However, this step is optional and is only present in some embodiment of the present invention.

[0065] According to the present invention the at least one prompt 205 can comprise, for example a without limitation: an explanation in natural language on the content of the compressed data 204; an explanation in natural language processing embedded in a representation; an audio signal representing the content of the compressed data 204 and an image representing the content of the compressed data 204.

[0066] Next, in step 403, the compressed data 204 is decompressed to retrieve the non-prioritized data 203b. The decompression may be carried out taking into account the explanation on the content of the compressed data 204 included in any prompts 205 received.

[0067] The present invention contemplates several different alternatives for the computing node 202 to know which decompression method to use. Thus, in one embodiment, the compressed data 204 is provided with metadata, e.g., on packet headers, indicating the decompression method to use.Alternatively, the computing node 202 may receive an indication regarding which decompression method to use, during session establishment with the network node 201 before the prioritized data 203a and the compressed data 204 are received.

[0068] The method according to the present invention contemplates subjecting the compressed data 204 to the following possible forms of decompression: data decompression, semantic decompression, or both to data decompression and to semantic decompression.

[0069] More particularly, at least one of the following decompression methods could be used to decompress the compressed data 204 in order to obtain the non-prioritized data 203b: Huffman decoding, arithmetic decoding, run-length decoding, delta decoding, Fourier transform decompression, wavelet transform, layer 3 audio decompression, LZ77 or LZ78 decompression, prediction and entropy decoding, semantic matching decompression, knowledge graph decompression, concept based decoding, neural semantic decompression, transformed based sparse decoding, and contextual decompression.

[0070] Finally in step 404, the prioritized data 203a and the non-prioritized data 203b are combined to retrieve the incoming content 203. Thus, the computing node 202 combines / merges the prioritized data 203a and the non-prioritized data 203b to obtain the incoming content 203.

[0071] Although the computing devices described herein (e.g., the network node 201, the computing node 202) may include the illustrated combination of hardware components, other embodiments may comprise computing devices with different combinations of components. It is to be understood that these computing devices may comprise any suitable combination of hardware and / or software needed to perform the tasks, features, functions and methods disclosed herein. Determining, calculating, obtaining or similar operations described herein may be performed by processing circuitry, which may process information by, for example, converting the obtained information into other information, comparing the obtained information or converted information to information stored in the network node, and / or performing one or more operations based on the obtained information or converted information, and as a result of said processing making a determination. Moreover, while components are depicted as single boxes located within a larger box, or nested within multiple boxes, in practice, computing devices may comprise multiple different physical components that make up a single illustrated component, and functionality may be partitioned between separate components. For example, a communication interface may be configured to include any of the components described herein, and / or the functionality of the components may be partitioned between the processingcircuitry and the communication interface. In another example, non-computationally intensive functions of any of such components may be implemented in software or firmware and computationally intensive functions may be implemented in hardware.

[0072] In certain embodiments, some or all of the functionality described herein may be provided by processing circuitry executing instructions stored in memory, which in certain embodiments may be a computer program product in the form of a non-transitory computer-readable storage medium. In alternative embodiments, some or all of the functionality may be provided by the processing circuitry without executing instructions stored on a separate or discrete device-readable storage medium, such as in a hard-wired manner. In any of those particular embodiments, whether executing instructions stored on a non-transitory computer-readable storage medium or not, the processing circuitry can be configured to perform the described functionality. The benefits provided by such functionality are not limited to the processing circuitry alone or to other components of the computing device, but are enjoyed by the computing device as a whole, and / or by end users and a wireless network generally.

Claims

CLAIMS1. A computer implemented method performed by a network node (201) for controlling traffic to a computing node (202) in a communication network the method comprising: - obtaining (301) incoming content (203) to be forwarded to the computing node (202); - dividing (302) the incoming content (203), into prioritized data (203a) and nonprioritized data (203b);routing (303) the prioritized data (203a) over a first communication channel (207a) to the computing node (202);- compressing (304) the non- prioritized data (203b) to form compressed data (204);routing (305) the compressed data (204) over a second communication channel (207b) to the computing node (202).

2. The method according to claim 1, further comprising receiving information (208) describing the status of one or more of the network node (201), the computing node (202), the first communication channel (207a) and the second communication channel (207b), and wherein the division of the incoming content (203) is made depending on said received information (208).

3. The method according to claim 2, wherein the information (208) describing the status of the first communication channel (207a) and / or the second communication channel (207b), includes at least one of the following:- current bandwidth utilization of the first communication channel (207a) and / or the second communication channel (207b);predicted bandwidth utilization of the first communication channel (207a) and / or the second communication channel (207b); and- an indication of a channel quality of the first communication channel (207a) and / orto the second communication channel (207b).

4. The method according to any of claims 2 to 3, wherein the information (208) describing the status of the network node (201) and / or the computing node (202), includes at least one of the following:- load of the network node (201) and / or the computing node (202), and / or- available energy reserves of the network node (201) and / orthe computing node (202),5. The method according to any of claims 1 to 4, further comprising receiving a description (206) of the incoming content (203), and wherein the division of the incoming content (203) is made depending on said received description (206).

6. The method according to claim 5, wherein the description (206) of the incoming content (203) includes at least one of the following:- current bitrate of the incoming content (203);- predicted bitrate of the incoming content (203);- an indication of whether the incoming content (203) has been generated by an artificial intelligence model or not; and- a semantic description of the incoming content (203).

7. The method according to any of claims 1 to 6, wherein to form the compressed data (204), the non-prioritized data (203b) is subject to:- data compression, by which data size is reduced,- semantic compression, by which data is replaced by a description of its content, or - both to data compression and to semantic compression.

8. The method according to any of claims 1 to 7, further comprising receiving dividing instructions from a content provider, and wherein the division of the incoming content (203) is made depending on said received dividing instructions.

9. The method according to any of claims 1 to 8, further comprising generating at least one prompt (205) including an explanation on the content of the compressed data (204) and transmitting the at least one prompt (205) to the computing node (202).

10. The method according to any of claims 1 to 9, further comprising receiving at least one prompt (205) from a content provider including an explanation on the content of the compressed data (204) and transmitting the at least one prompt (205) to the computing node (202).

11. The method according to any of claims 9 or 10, further comprising routing the at least one prompt (205) over a third communication channel (207c) to the computing node (202).

12. The method according to any of claims 9 to 11, wherein the at least one prompt (205) comprises:- an explanation in natural language on the content of the compressed data (204); - an explanation in natural language processing embedded in a representation;- an audio signal representing the content of the compressed data (204); and - an image representing the content of the compressed data (204).

13. The method according to any of claims 1 to 12, wherein the non-prioritized data (203b) is further divided in at least two different sets of non-prioritized data depending on its relevance, each set of non-prioritized data undergoing a different type of compression to form a set of compressed data.

14. A network node (201) in a communication network, the network node (201) comprising:- processing circuitry (201a); and- memory (201b) coupled to the processing circuitry (201a) and having instructions stored therein that are executable by the processing circuitry (201a) to cause the network node (201) to perform operations comprising any of the operations of claims 1 to 13.

15. A computer program, comprising program code to be executed by processing circuitry (201a) of a network node (201) in a communication network, whereby execution of the program code causes the network node (201) to carry out operations comprising any of the operations of claims 1 to 13.

16. A computer implemented method performed by a computing node (202) for receiving incoming content (203) in a communication network, the incoming content (203) comprising prioritized data (203a) and compressed data (204) corresponding to nonprioritized data (203b), the method comprising:receiving (401) the prioritized data (203a) through a first communication channel (207a);receiving (402) the compressed data (204), corresponding to the non-prioritized data (203b), through a second communication channel (207b);17- decompressing (403) the compressed data (204) to obtain the non-prioritized data (203b); and- combining (404) the prioritized data (203a) and the non-prioritized (203b) data to obtain the incoming content (203).

17. The method according to claim 16, wherein to obtain the non-prioritized data (203b), the compressed data (204) is subject to:- data decompression,- semantic decompression, or- both to data decompression and to semantic decompression.

18. The method according to any of claims 16 or 17, further comprising receiving (405) at least one prompt (205) including an explanation on the content of the compressed data (204), and wherein the explanation on the content of the compressed data (204) included in the at least one prompt (205) is used as input for decompressing (403) the compressed data (204) to obtain the non-prioritized data (203b).

19. The method according to claim 18, wherein the at least one prompt (205) is received through a third communication channel (207c).

20. The method according to any of claims 18 to 19, wherein the at least one prompt (205) comprises:- an explanation in natural language on the content of the compressed data (204); - an explanation in natural language processing embedded in a representation; - an audio signal representing the content of the compressed data (204); and - an image representing the content of the compressed data (204).

21. A computing node (202) in a communication network, the computing node (202) comprising:- processing circuitry (202a); and- memory (202b) coupled to the processing circuitry (202a) and having instructions stored therein that are executable by the processing circuitry (202a) to cause the computing node (202) to perform operations comprising any of the operations of claims 16 to 20.

182. A computer program, comprising program code to be executed by processing circuitry (202a) of a computing node (202) in a communication network, whereby execution of the program code causes the computing node (202) to carry out operations comprising any of the operations of claims 16 to 20.19

Images