EP4689722A1

EP4689722A1 - Joint sensing and communications with deep neural networks

Info

Publication number: EP4689722A1
Application number: EP24739910.8A
Authority: EP
Inventors: Jibing Wang; Erik Richard STAUFFER
Original assignee: Google LLC
Current assignee: Google LLC
Priority date: 2023-06-13
Filing date: 2024-06-13
Publication date: 2026-02-11
Also published as: CN121359048A; KR20260003271A; WO2024259042A1

Abstract

End-to-end joint radar sensing and communication (JSC) capabilities are provided between a first device (101) and second device (112) using deep neural networks (DNNs). A DNN structure (103) at the first device generates a JSC signal waveform (107, 107a-107n) representative of input communication data (102) and a radar signal. The second device (1112) processes the JSC signal waveform at a DNN structure (114) to generate reconstructed communication data (115). The DNN structure of the second device generates a radar sensing feedback (RFB) signal waveform (117) from the JSC signal waveform and reflections (111a-111n) from one or more objects (109a-109n) for transmission to the first device. The first device uses the DNN structure to process the received RFB signal waveform and/or any reflections (110a-110n) of the transmitted JSC signal waveform from the one or more objects (109a-109n) to generate radar sensing information (105) for the one or more objects.

Description

JOINT SENSING AND COMMUNICATIONS WITH DEEP NEURAL NETWORKS Background [0001] The concept of wireless communication systems coexisting with positioning/sensing or radar systems is known as, for example, Joint Communication Radar (JCR), Joint Sensing and Communications (JSC), Joint Communications and Sensing (JCS), Integrated Communication and Sensing (ICS), Joint Radar and Communication Systems (JRCS), Joint Radar Sensing and Communication (JRSC) and others. The acronym JSC is used hereinafter to collectively refer to these types of dual communication/radar systems. Current JSC systems aim to share common radar sensing and communications hardware and air interface resources allowing concurrent communications and radar functions for particular JSC system types. The JSC system types include, without limitation, for example bi-static JSC, multi-static JSC, and mono-static JSC and the like. However, wireless communication system designers typically design and optimise wireless communication systems for providing high data rates with a defined quality of service, while radar system designers typically design and optimise radar systems for target object detection and range/velocity estimation. These different design objectives have historically resulted in different waveforms for wireless communication systems than radar systems. For example, 5G advanced wireless communication systems typically use orthogonal frequency division multiplexing (OFDM) waveforms, and 6G wireless communication systems may use Orthogonal time frequency space (OTFS) waveforms. For radar systems, the waveforms used include, for example, frequency modulated continuous wave (FMCW), linear frequency modulation (LFM), or non-linear frequency modulation (NLFM) waveforms, which can be different for different JSC system types. These waveform differences make it a challenging exercise to efficiently perform JSC for one or more JSC system types within 5G or 6G and beyond wireless communication systems. [0002] Many Fifth Generation (5G) or Sixth Generation (6G) and beyond wireless communications systems have complicated transmitter and receiver processing chains and a standardised waveform (e.g., OFDM waveforms for 5G) designed to meet the performance requirements of the corresponding communications standard. A 5G or 6G transmitter processing chain of a first device may include an arrangement of encoding, interleaving, scrambling, pre-coding, modulation, and radio frequency (RF) analog transmission components for processing input communication data for transmission using OFDM waveforms. A 5G or 6G receiver processing chain of a second device may include an arrangement of RF analog receiving, channel estimation, demodulation, descrambling, de-interleaving, and decoding components for receiving the OFDM waveforms transmitted by the first device and recovering the input communication data transmitted. Combining such hardware and the corresponding waveforms (e.g., OFDM, and OTFS, etc.) with a radar sensing transmitter and receiver chain and its corresponding waveforms (e.g., FMCW, LFM, and NLFM, etc.) for one or more JSC system types while simultaneously meeting the requirements of the corresponding communications standard and accurately performing target object detection and range/velocity estimation is a complex task. Summary of Implementations [0003] This specification relates to methods, apparatus, and systems that augment wireless communication systems (e.g., current 5G advanced or 6G and beyond wireless communication systems) with transmitting/receiving deep neural network (DNN) structures capable of efficiently performing joint sensing and communication (JSC) under various communication and/or radar conditions/environments/ performance requirements (e.g., non-line of sight (NLOS) communications, line of sight (LOS) communications, multi-path interference, multiple access interference, narrowband interference, changes in weather or atmospheric conditions, changes in radar channel conditions, target characteristic, different target sizes and types, channel throughput, channel frequency or frequency band, channel bandwidth, channel delay spread, channel angular spread, or any other type of communication signal interference or radar condition/performance requirement and the like) whilst simultaneously performing accurate target object detection and/or range/velocity estimation and meeting the communication performance requirements of the wireless communication system (e.g., corresponding 5G advanced or 6G and beyond communication standards). [0004] In a first aspect, the present disclosure provides a method performed by a first device, the method comprising: establishing JSC DNN operations for one or more time slots of a communication session with a second device; for each time slot of the one or more time slots, the method further comprising: retrieving input communication data for transmission to the second device in said each time slot; processing the input communication data with a JSC DNN structure of the first device to generate an output JSC signal representing the input communication data and a radar signal; transmitting, to the second device, the output JSC signal as a JSC signal waveform in said each time slot over a communication channel; receiving one or more radar sensing signal(s) for said each time slot based on reflections of the transmitted JSC signal waveform from one or more objects; processing the received radar sensing signal(s) by the JSC DNN structure of the first device to generate radar sensing information / measurements of the one or more objects in the communication channel; and sending the radar sensing information / measurements of one or more objects for said each time slot to one or more upper layer protocols of a protocol stack of the first device. [0005] In a second aspect, the present disclosure provides a method performed by a second device, the method comprising: establishing JSC DNN operations for one or more time slots of a communication session with a first device according to the first aspect; for each time slot of the one or more time slots, the method performing: receiving a JSC signal waveform transmitted from the first device in said each time slot over a communication channel, the JSC signal waveform representing input communication data and a radar signal; receiving one or more further JSC signal waveforms for said each time slot based on reflections of the JSC signal waveform transmission from one or more objects in the communication channel; processing the received JSC signal waveform and further JSC signal waveforms with a JSC DNN structure at the second device to generate reconstructed communication data corresponding to the input communication data and a radar sensing feedback signal associated with the one or more objects; transmitting the radar sensing feedback signal to the first device for use in generating radar sensing information /measurements for the one or more objects; and sending the reconstructed communication data to a data sink of the second device or sending the reconstructed communication data to one or more upper protocol layers of a protocol stack of the second device. [0006] In a third aspect, the present disclosure provides a method performed by a second device, the method comprising: establishing JSC DNN operations for one or more time slots of a communication session with a first device according to the first aspect; for each time slot of the one or more time slots, the method performing: receiving a JSC signal waveform transmitted from the first device in said each time slot over a communication channel, the JSC signal waveform comprising data representative of input communication data and a radar signal; receiving one or more further JSC signal waveforms for said each time slot based on reflections of the JSC signal waveform transmission from one or more objects in the communication channel; processing the received JSC signal waveform and further JSC signal waveforms with a communication DNN structure at the second device, the communication DNN structure configured for generating reconstructed communication data corresponding to the input communication data transmitted from the first device; and sending the reconstructed communication data to a data sink of the second device or sending the reconstructed communication data to one or more upper protocol layers of a protocol stack of the second device. [0007] Further aspects provide computer-readable medium, apparatus, and systems for implementing the methods of the first, second, third and fourth aspects. [0008] Aspects of the methods, apparatus and systems provide numerous advantages including, for example, performing efficient JSC in a communication session between the first and second device of, for example, a 5G or 6G and beyond wireless communication system for various communication channel conditions whilst simultaneously performing accurate radar sensing (e.g., target object detection and/or range/velocity estimation) and meeting the performance requirements of the corresponding communication standard for one or more JSC system types. Additional advantages provide efficient, rapid, and dynamic reconfiguration of JSC DNN operations (e.g., bi-static, multi-static, or mono-static JSC) during a communication session between the first and second device whilst maintaining accurate radar sensing and communications performance. Further advantages include efficient design of JSC waveforms for use in, for example, a 5G or 6G and beyond wireless communication systems and for various radar sensing and communication channel conditions whilst simultaneously performing accurate radar sensing and meeting the requirements of the corresponding communication standard. Brief Description of the Drawings [0009] Embodiments will be described, by way of example, with reference to the following drawings, in which: [0010] FIG.1 is a schematic diagram illustrating an example wireless communication system for performing joint sensing and communication (JSC) using deep neural networks (DNNs); [0011] FIG.2 is a schematic diagram illustrating another example wireless communication system for performing JSC using DNNs that highlights hardware components of the devices; [0012] FIG.3a is a schematic diagram illustrating an example bi-static JSC for the wireless communication system of FIGs.1 or 2; [0013] FIG.3b is a schematic diagram illustrating an example multi-static JSC for the wireless communication system of FIGs.1 or 2; [0014] FIG.3c is a schematic diagram illustrating an example mono-static JSC for the wireless communication system of FIGs.1 or 2; [0015] FIG.3d is a schematic diagram illustrating an example bi-static or multi- static JSC using first, second and third devices in the wireless communication system of FIGs.1 or 2; [0016] FIG.4 is a signal flow diagram illustrating bi-static JSC DNN operations for one or more time slots of a communications session between a first device and a second device; [0017] FIG.5 is a signal flow diagram illustrating multi-static JSC DNN operations for one or more time slots of a communications session between a first device and a second device; [0018] FIG.6 is a signal flow diagram illustrating mono-static JSC DNN operations for one or more time slots of a communications session between a first device and a second device; [0019] FIG.7a is a signal flow diagram illustrating bi-static JSC DNN operations for one or more time slots of a downlink (DL) communications session between a base station (BS) and a user equipment (UE); [0020] FIG.7b is a signal flow diagram illustrating multi-static JSC DNN operations for one or more time slots of a DL communications session between a BS and a UE; [0021] FIG.7c is a signal flow diagram illustrating mono-static JSC DNN operations for one or more time slots of a DL communications session between a BS and a UE; [0022] FIG.8a is a flow diagram illustrating an example JSC DNN process for a first device performing JSC DNN operations in one or more time slots of a communication session with a second device; [0023] FIG.8b is a flow diagram illustrating an example establishment of JSC DNN operations by a first device for one or more time slots of the communication session with the second device; [0024] FIG.9a is a flow diagram illustrating an example JSC DNN process for a second device performing bi-static or multi-static JSC DNN operations in one or more time slots of a communication session with a first device; [0025] FIG.9b is a flow diagram illustrating another example JSC DNN process for a second device performing mono-static JSC DNN operations in one or more time slots of a communication session with a first device; [0026] FIG.9c is a flow diagram illustrating an example establishment of JSC DNN operations by a second device for one or more time slots of the communication session with the first device; [0027] FIG.10a is a schematic diagram illustrating an example master neural network table for selecting or configuring corresponding JSC DNN structures for use in JSC DNN operations for one or more time slots of a communication session between the first device and second device and optionally a third device; [0028] FIG.10b is a schematic diagram illustrating an example first neural network table used by a first device for selecting and configuring corresponding JSC DNN structures for use in JSC DNN operations for one or more time slots of a communication session between the first device and second device and optionally the third device; [0029] FIG.10c is a schematic diagram illustrating an example second neural network table used by a second device for selecting and configuring corresponding JSC DNN structures for use in JSC DNN operations for one or more time slots of a communication session between the first device and second device; [0030] FIG.10d is a schematic diagram illustrating an example third neural network table used by a third device for selecting and configuring corresponding JSC DNN structures for use in JSC DNN operations for one or more time slots of a communication session between the first device and second device, where the third device assists the first device; [0031] FIG.10e is a flow diagram illustrating an example JSC DNN structure selection process for a first device during establishment of JSC DNN operations with a second device; [0032] FIG.11 is a schematic diagram illustrating an example joint training process for training first and second JSC DNN structures for use by first and second devices, respectively; [0033] FIG.12 is a schematic diagram of an example computer-readable medium. [0034] Common reference numerals are used throughout the figures to indicate similar features. Detailed Description [0035] FIG.1 illustrates an example of a wireless communication system 100 including a first device 101 and a second device 112 configured for performing joint sensing and communication (JSC) using deep neural networks (DNNs). In the wireless communication system 100, the first and second devices 101 and 112 establish a communication session. During the communication session, the first and second devices 101 and 112 perform JSC DNN operations in one or more time slots of the communication session. The first and second devices 101 and 112 include corresponding configurable JSC DNN structures 103 and 114 performing JSC DNN operations. The JSC DNN operations include the first and second devices 101 and 112 configuring their corresponding JSC DNN structures 103 and 114, respectively, according to a selected JSC system type (e.g., bi-static, multi-static or mono-static JSC), communication performance requirements, radar sensing requirements, and channel parameters. After configuration, further JSC DNN operations include the first and second devices 101 and 112 performing JSC according to the selected JSC system type in the one or more time slots of the communication session using the respective JSC DNN structures 103 and 114. [0036] The first device 101 includes the JSC DNN structure 103 coupled to a radio frequency (RF) front-end transmitter (Tx) / receiver (Rx) subsystem (RF front- end Tx/Rx subsystem) 104. The first device 101 retrieves input communication data 102 from a data source for transmission in one or more time slots to the second device 112. The input communication data 102 for a time slot is input to the JSC DNN structure 103. The JSC DNN structure 103 processes the input communication data 102 for the time slot to generate an output JSC signal 106a for transmission in the time slot. The output JSC signal 106a integrates the input communication data 102 and a radar signal and/or radar signal characteristics, which is generated by the JSC DNN structure 103. That is, the output JSC signal 106a represents a digital joint communication and radar signal with the input communication data 102 incorporated therein along with radar signal characteristics. For example, the JSC DNN structure 103 generates the output JSC signal 106a from the input communication data 102, where the output JSC signal 106a represents the input communication data 102 for transmission in that time slot to the second device 112 but digitally transformed (or conformed) to operate also as the radar signal. For example, the JSC DNN structure 103 processes the input communication data 102 to generate the output JSC signal 106a representing the input communication data 102 transformed to operate as a radar signal when transmitted. The RF front-end Tx/Rx subsystem 104 processes (e.g., digital-to-analog (DAC) conversion and frequency up conversion to radio frequency) and transmits the output JSC signal 106a in the corresponding time slot as a JSC signal waveform 107 to the second device 112 over wireless communication / radar channel 108 (also referred to as a wireless communication channel or communication channel). The JSC signal waveform 107 for the corresponding time slot represents the input communication data 102 transmitted in that time slot to the second device 112 conformed to operate also as the radar signal. For example, the JSC signal waveform 107 for the corresponding time slot is a waveform that represents the input communication data 102 transmitted in that time slot to the second device 112 and jointly operates as a radar signal in that time slot. The JSC signal waveform 107 for the time slot is a joint radar sensing and communication signal waveform for the time slot with the input communication data 102 for the time slot integrated/incorporated therein. The JSC waveform 107 has communication signal characteristics and radar signal characteristics enabling it to be used for JSC applications. There are one or more objects 109a, 109b, and 109c to 109n (e.g., people, animals, obstacles, trees, buildings, vehicles and/or any other object capable of reflecting the transmitted JSC signal waveform 107) within the environment of the wireless communication channel 108. [0037] Depending on the JSC system type, the first device 101 receives, for the time slot, one or more reflections 110a, 110b, and 110c to 110n of the transmitted JSC signal waveform 107 (e.g., reflected radar signals) from the corresponding one or more objects 109a, 109b, and 109c to 109n and/or a radar sensing feedback (RFB) signal waveform 117 from the second device 112. The RF front-end Tx/Rx subsystem 104 receives and baseband processes (e.g., analog/digital conversion and frequency down conversion to baseband) the one or more reflections 110a to 110n of the JSC signal waveform 107 from the one or more objects 109a to 109n and/or a RFB signal waveform 117 from the second device 112 into one or more radar sensing signal(s) 106b. Since each of the radar sensing signal(s) 106b are derived from the reflections 110a- 110n or 1111a-111n of the JSC signal waveform 107 from the one or more objects 109a to 109n, each of the radar sensing signal(s) 106b includes one or more components of the radar signal and/or the radar signal characteristics incorporated within the transmitted JSC signal waveform 107 resulting from the transmission of the output JSC signal 106a for the time slot. The radar signal components/radar signal characteristics of the reflections 110a-110n or 111a-111n being associated with the corresponding objects 109a-109n. That is, the radar sensing signal(s) 106b for each time slot include data representative of one or more component radar signal(s) or radar signal characteristic(s) associated with the reflections 110a-110n or 111a-111n of the transmitted JSC signal waveform 107 from the one or more objects 109a-109n for the time slot. The radar sensing signal(s) 106b are input and processed by the JSC DNN structure 103 for generating radar sensing information 105 (e.g., doppler, velocity, positioning etc.) of the one or more objects 109a-109n. Radar sensing information 105 can also be referred to as radar sensing measurements or estimates, but will hereinafter be referred to as radar sensing information 105. The first device 101 detects these objects 109a to 109n using the JSC DNN structure 103 to process the one or more radar sensing signal(s) 106b and generate radar sensing information 105 for these objects 109a to 109n and/or those objects detected. [0038] The second device 112 includes an RF front-end Tx/Rx subsystem 113 coupled to another JSC DNN structure 114. The RF front-end Tx/Rx subsystem 113 of the second device 112 receives, for the corresponding time slot, at least a component 107a of the transmitted JSC signal waveform 107 and one or more further reflected JSC signal waveforms 111a, 111b, and 111c to 111n based on reflections of the transmitted JSC signal waveform 107 from the corresponding one or more objects 109a, 109b, 109c to 109n in the environment of the wireless communication channel 108. The JSC signal waveform 107 for the corresponding time slot represents the input communication data 102 transmitted in that time slot to the second device 112 transformed/conformed to operate also as a radar signal. The RF front-end Tx/Rx subsystem 113 receives and processes (e.g., analog-to-digital (ADC) conversion and frequency down conversion to baseband) at least the component 107a of the transmitted JSC signal waveform 107 and the one or more further reflected JSC signal waveforms 111a to 111n into a received JSC signal 116a for the corresponding time slot. The received JSC signal 116a for the corresponding time slot is input to the JSC DNN structure 114. The JSC DNN structure 114 processes the received JSC signal 116a to generate reconstructed communication data 115 corresponding to the input communication data 102 transmitted in the corresponding time slot. [0039] Depending on the JSC system type (e.g., bi-static, multi-static, or mono- static JSC), the JSC DNN structure 114 of the second device 112 can be configured for either: a) reconstructing communications data from the received JSC signal 116a (e.g., mono-static JSC system type), or b) reconstructing communications data from the received JSC signal 116a and generating RFB signal 116b associated with the one or more objects 109a to 109n for transmission as RFB signal waveform 117 to the first device 101 (e.g., bi-static or multi-static JSC system type). If the second device 112 is configured for b), then the RF front-end Tx/Rx subsystem 113 transmits the RFB signal 116b as a RFB signal waveform 117 for the corresponding time slot to the first device 101 for processing by the JSC DNN structure 103 of the first device 101. [0040] The first device 101 subsequently receives one or more radar sensing signals 106b for the corresponding time slot based on a) reflections 110a to 110n of the transmitted JSC signal waveform (e.g., mono-static JSC system type) for the corresponding time slot; b) or RFB signal waveform 117 for the corresponding time slot (e.g., bi-static JSC system type); or c) reflections 110a-110n of the transmitted JSC signal waveform and RFB signal waveform 117 for the corresponding time slot (e.g., multi-static JSC system type). The first device 101 inputs the one or more radar sensing signals 106b for the corresponding time slot to the JSC DNN structure 103. The JSC DNN structure 103 processes the one or more radar sensing signals 106b and generates radar sensing information 105 for the one or more objects 109a-109n (e.g., range, doppler, velocity, position, or other radar measurement and the like) in the environment of the wireless communication channel 108. The radar sensing requirements may establish the types of radar sensing information that the JSC DNN structure 103 generates. For example, the JSC DNN structure 103 generates radar sensing information 105 including one or more items from the group of: generating of the radar sensing information includes generating the radar sensing information (or measurements/estimates) for each of the objects 109a-109n including one or more items from the group of: a range estimate of each of the associated objects 109a-109n; a doppler estimate for each of the associated objects 109a-109n; a location or position estimate for each of the associated objects 109a-109n; a delay spread for each of the associated objects 109a-109n; a doppler spread for each of the associated objects 109a- 109n; an average delay for each of the associated objects 109a-109n; an angular estimate for each of the associated objects 109a-109n; an azimuth and/or elevation estimate of each of the associated objects 109a-109n; and/or any other suitable radar sensing information or parameter estimated for each of the associated objects 109a- 109n. [0041] The first and second devices 101 and 112 have JSC DNN structures 103 and 114 for use in JSC DNN operations during the communication session therebetween. When establishing JSC DNN operations during a communication session the first and second devices 101 and 112 configure their JSC DNN structures 103 and 114 with corresponding trained DNN models depending, at least in part, on the JSC system type the first device 101 selects for generating specific radar sensing information 105 of one or more objects 109a-109n in the environment. As described with reference to FIG.11, machine learning (ML) algorithms are used to jointly train (e.g., using supervised learning or unsupervised learning) a combination of DNN models, or DNNs, for each particular JSC system type and a combination of communication and/or radar sensing channel characteristics (e.g., various channel conditions and radar sensing requirements) to form a trained pair of JSC DNN structures for configuring the JSC DNN structures 103 and 114 by the first and second devices 101 and 112 when performing JSC, respectively. Each trained pair of JSC DNN structures 103 and 114 includes one or more trained transmitting and/or receiving DNN model arrangements based on the JSC system type (e.g., bi-static, multi-static, or mono-static JSC), channel conditions/characteristics, communication performance requirements, and radar sensing requirements used when jointly training these DNN models. Each trained pair of JSC DNN structures is capable of providing end-to-end JSC without the complexity of conventional transmitter and receiver JSC processing chains. The JSC DNN structures 103 and 114 augment and/or replace conventional transmitter and receiver JSC processing chains. [0042] For example, the configuration of a trained pair of JSC DNN structures is used to configure the JSC DNN structure 103 of the first device 101 and the JSC DNN structure 114 of the second device 112. After configuration, the JSC DNN structure 103 processes input communication data 102 to generate transmission JSC signal waveforms that represent the input communications data 102 and radar signals suited to efficiently overcome various channel conditions/environments used during training (e.g., non-line of sight (NLOS) communications, line of sight (LOS) communications, multi-path interference, multiple access interference, and narrowband interference, changes in weather or atmospheric conditions, different target sizes and types, other radar and/or communication signal interference) whilst simultaneously, depending on the JSC system type used during training: a) performing accurate target object detection and/or range/velocity estimation; and b) meeting the communication performance requirements of the wireless communication system 100 (e.g., corresponding 5G or 6G and beyond communication standards). The corresponding JSC DNN structure 114 of the second device 112 overcomes the above-mentioned various channel conditions/environments when processing the received JSC signal waveforms for generating reconstructed communication data 115 meeting the communication performance requirements of the wireless communication system 100 (e.g., corresponding 5G or 6G and beyond communication standards) and/or, depending on JSC system type, generating RFB signal waveforms 117 for transmission to the first device 101 for input to the JSC DNN structure 103 of the first device 101 for performing accurate target object detection and/or range/velocity estimation. [0043] The JSC DNN operations performed during a communication session between the first device 101 and second device 112 are established when the first device 101 identifies the channel condition of the wireless communication/radar channel 108 between the first and second devices 101 and 112, and also based on JSC system type according to JSC performance requirements to be met during the communication session. The JSC performance requirements include the JSC system type, radar sensing requirements, and also communication performance requirements. The radar sensing requirements include, without limitation, for example one or more of range resolution and doppler resolution, and the like, which may affect the frequency range selection for the transmitted JSC signal. The communication performance requirements include, without limitation, for example one or more of throughput, latency, block error rate, and/or other performance metrics/parameters and the like as defined by the communication standard (e.g., 5G or 6G and beyond standard) implemented by the wireless communication system 100. For example, the channel condition of the wireless communication (and/or radar) channel 108 affecting communication and/or radar performance include, without limitation, for example one or more from the group of: a non-line-of-sight communication channel condition; a line-of-sight communication channel condition; weather or atmospheric conditions; radar channel conditions; target characteristic; channel throughput; channel frequency or frequency band; channel bandwidth; channel delay spread; channel doppler spread; channel angular spread; or any other type of condition affecting the communication channel between the first device and second device for JSC. [0044] For example, the first device 101 identifies the JSC system type based on the radar sensing requirements (e.g., range resolution, doppler resolution, frequency band, and the like) for generating radar sensing information 105 for the potential one or more objects 109a-109n (e.g., range, doppler, velocity, position, or other radar measurement) and communication performance requirements (e.g., block error rate, latency, throughput and the like) to be met during the communication session with the second device 112. The first device 101 also identifies the channel condition of the wireless communications channel 108 and using the selected JSC system type, searches, selects and retrieves a suitable trained pair of JSC DNN structures from a plurality of trained pairs of JSC DNN structures, each trained pair of JSC DNN structures associated with a JSC DNN identifier. The selected pair of trained JSC DNN structures meet the JSC performance requirements. After selection, the first device 101 transmits a control message to the second device 112, the control message including the JSC DNN identifier associated with the selected trained pair of JSC DNN structures. The JSC DNN identifier specifies which JSC DNN structure the first device 101 and the second device 112 retrieves and uses for the one or more time slots when performing the JSC DNN operations. [0045] Depending on the type of wireless communication system 100, the control message also specifies one or more communication resource parameters (e.g., downlink (DL) and/or uplink (UL) resource block (RB) configurations, DL and/or UL frequencies, etc.) for use by the first device 101 and second device 112 when receiving the transmitted JSC signal waveform 107 from the first device 101. Depending on the JSC system type, the control message also specifies the communication resource parameters for use by the second device 112 when transmitting RFB signal waveforms 117 corresponding to the one or more time slots to the first device 101. [0046] In bi-static JSC, the first device 101 transmits a JSC signal waveform 107 representing input communication data 102 and a radar signal in a time slot, while the second device 112 provides a RFB signal waveform 117 corresponding to the time slot that assists the first device 101 in radar sensing for the time slot in relation to the transmitted JSC signal waveform 107 in the time slot. In this case, the JSC system type is a bi-static JSC system type where the trained JSC DNN structures 103 and 114 of the first and second devices 101 and 112, respectively, have bi-static JSC processing capabilities and the second device 112 assists the first device 101 in radar sensing for a time slot by providing the first device 101 with a RFB signal waveform 117 corresponding to the time slot. The JSC DNN structure 103 of the first device 101 processes input communication data 102 for a time slot and generates an output JSC signal 106a for transmission in the time slot by the RF front-end Tx/Rx subsystem 104 as a JSC signal waveform 107. The JSC signal waveform 107 represents the input communication data 102 and a radar signal for the time slot. [0047] The JSC DNN structure 103 of the first device 101 also processes one or more radar sensing signal(s) 106b derived from a RFB signal waveform 117 transmitted from the second device 112a to the first device 101 after the JSC DNN structure 114 of the second device 112 generates RFB signal 116b from at least a received component 107a of the transmitted JSC signal waveform 107 and reflected JSC signal waveforms 111a, 111b, and 111c to 111n caused by reflections of the transmitted JSC signal waveform 107 from objects 109a-109n. The JSC DNN structure 114 of the second device 112 processes the received component 107a of the transmitted JSC signal waveform 107 and the reflected JSC signal waveforms 111a, 111b, and 111c to 111n to generate both reconstructed communication data 115 and RFB signal 116b (e.g., doppler, velocity, and range etc.) corresponding to the one or more objects 109a-109n. The second device 112 transmits the RFB signal 116b as RFB signal waveform 117 to the first device 101. The generated reconstructed communication data 115 corresponds to the input communication data 102 that was transmitted within the JSC signal waveform 107 in the corresponding time slot. [0048] The RF front-end Tx/Rx subsystem 104 receives and processes (e.g., ADC and frequency down conversion to baseband) the RFB signal waveform 117 for the corresponding time slot from the second device 112 into a radar sensing signal 106b. The JSC DNN structure 103 of the first device 101 processes the radar sensing signal 106b derived from the radar sensing feedback signal waveform 117 for the corresponding time slot to generate the radar sensing information 105 for the one or more objects 109a to 109n for the corresponding time slot. The bi-static JSC system type provides an advantage that the first device 101 does not need to be in full duplex mode to receive and process reflections 110a to 110n of the transmitted JSC signal waveform 107 from the one or more objects 109a to 109n when generating radar sensing information 105. [0049] In multi-static JSC, the first device 101 and second device 112 perform bi-static JSC but the first device 101 also receives (e.g., in full duplex mode) reflections 110a-110n of the JSC signal waveform 107 transmitted in the time slot from the one or more objects 109a-109n. In this case, the JSC system type is a multi-static JSC system type in which both the JSC DNN structures 103 and 114 are trained to have bi-static JSC processing capabilities, where in addition the JSC DNN structure 103 is also trained to have mono-static JSC processing capabilities. The RF front-end Tx/Rx subsystem 104 of the first device 101 receives and processes (e.g., ADC and frequency down conversion to baseband) the reflections 110a-110n into one or more radar sensing signals 106b. After the RF front-end Tx/Rx subsystem 104 receives and processes the RFB signal waveform 117 from the second device 112 that is associated with the time slot into one of the one or more radar sensing signals 106b, the radar sensing signals 106b are processed by the JSC DNN structure 103 of the first device 101 to generate radar sensing information 105 for the one or more objects 109a-109n in relation to the time slot. The multi-static scenario provides an advantage of additional radar sensing accuracy or tracking accuracy with the first device 101 operating in full duplex mode. [0050] Other bi-static / multi-static scenarios are possible, for example, a third device (e.g., see third device 341 of FIG.3d) provides another RFB signal to the first device 101. For example, the first device 101 may establish JSC DNN operations for the one or more time slots in a communication session with the second device 112, while at the same time the first device 101 recruits the third device to provide radar feedback support in response to the JSC signal waveforms 107 transmitted between the first and second devices 101 and 112. The third device includes a radar sensing DNN structure for only processing, for each of the time slots, the received JSC signal waveform 107 transmitted from the first device 101, and corresponding reflections of the transmitted JSC signal waveform 107 and sending a generated RFB signal for the corresponding time slots to the first device 101. In this scenario, the second device 112 performs JSC DNN operations where the JSC DNN structure 114 of the second device 112 processes the received JSC signal waveforms 107 and reflections thereof 111a-111n from the one or more objects 109a-109n to generate reconstructed communication data 115 corresponding to the input communication data 102 that was transmitted to the second device 112 in the transmitted JSC signal waveform 107 and/or RFB signal waveform 117. As an option, the second device 112 only generates reconstructed communication data 115 for the time slot whilst the third device provides a RFB signal to the first device 101 for the time slot. This configuration provides an advantage of further enhancing the resolution of radar sensing in the wireless communication system 100 and/or providing flexibility by enabling JSC operations between first and second devices 101 and 112 when a second device 112 is only capable of processing the transmitted JSC signal waveform 107 for reconstructing the input communication data 102. [0051] In mono-static JSC, the first device 101 transmits a JSC signal waveform 107 in a time slot, which represents input communication data 102 and a radar signal for the time slot, and simultaneously receives (e.g., in full duplex mode) reflections 110a-110n of the transmitted JSC signal waveform 107 associated with the time slot from the one or more objects 109a-109n. The second device 112 only generates reconstructed communication data 115 from the received JSC signal waveform 107 and/or reflections thereof 111a-111n. In this case, the JSC system type is a mono-static JSC system type in which the first device 101 configures the JSC DNN structure 103 for mono-static JSC processing capabilities and the second device 112 configures the JSC DNN structure 114 for only reconstructing communications data 115. The RF front-end Tx/Rx subsystem 104 of the first device 101 receives and processes (e.g., ADC and frequency down conversion to baseband) the reflections 110a-110n of the transmitted JSC signal waveform 107 into one or more radar sensing signals 106b, which the JSC DNN structure 103 processes to generate radar sensing information 105 associated with the one or more objects 109a-109n. The first device 101 operates in full duplex mode so it can simultaneously transmit the JSC signal waveform 107 and receive the JSC signal waveform reflections 110a-110n from the one or more objects 109a-109n for generating radar sensing information 105 for the one or more objects 109a-109n. This provides the advantage of faster generation of radar sensing information 105 for the one or more objects 109a-109n for each of the time slots because there is minimal latency in the JSC DNN structure 103 receiving the one or more radar sensing signals 106b. [0052] As an option, the first device 101 also has a protocol stack with a plurality of protocol layers. After generating radar sensing information 105 for a particular time slot or for one or more time slots, the first device 101 sends the radar sensing information 105 for each one or more time slots to one or more upper layer protocols of the protocol stack of the first device 101. In the protocol stack of the first device 101, the lower layers are responsible for providing services to the upper layers, and the upper layers use those services to provide their own functions. For example, the radar sensing information 105 is generated at the physical layer of a protocol stack and passed up and processed by each of the upper layers until the application layer of the protocol stack, where the corresponding radar sensing information 105 is used for, without limitation, for example display to a user, further processing, and/or sending to one or more applications of the first device 101 or second device 112 for further processing and/or consumption of the radar sensing information 105. [0053] Similarly, the second device 112 also has a protocol stack with a plurality of protocol layers. After the JSC DNN structure 114 generates reconstructed communication data 115 for a particular time slot or for one or more time slots, the second device 112 sends the reconstructed communication data 115 for each one or more time slots to one or more upper layer protocols of the protocol stack of the second device 112. In the protocol stack of the second device 112, the lower layers are responsible for providing services to the upper layers, and the upper layers use those services to provide their own functions. For example, the reconstructed communication data 115 is generated at the physical layer of a protocol stack and passed up and processed by each of the upper layers until the application layer of the protocol stack, where the corresponding reconstructed communication data 115 is used for, without limitation, for example display to a user, further processing, and/or sending to one or more applications of the second device 112 for further processing and/or consumption of the reconstructed communication data 115. [0054] The first and second devices 101 and 112 can be any type of communication device for use in wireless communication system 100 such as, but not limited to, for example any combination of radio access network elements including terrestrial network (TN) base stations (BSs), non-terrestrial network (NTN) BSs (or network devices including satellites, drones, or high-altitude platform stations (HAPS)), user equipment (UE), or other RAN elements within wireless communication system 100. For example, the first device 101 and second device 112 may be two BSs, or two UEs, or a BS and a UE, or a UE and a BS, or any other combination of communication devices as the application demands. FIG.2 illustrates a wireless communication system 200 in which the first device 201 is a TN BS and the second device 212 is a UE. [0055] FIG.2 illustrates another example of a wireless communication system 200 in which a first device 201 performs JSC DNN communications with a second device 212. Reference numerals of FIG.1 are reused in FIG.2 for similar or the same components or features. In this example, the first device 201 is a TN BS, referred to as BS 201, and the second device 212 is a UE, referred to as UE 212. The BS 201 connects via one or more interfaces to a core network of the wireless communication system 200. For example, the wireless communication system 200 may be a 5G or 6G wireless communication system. [0056] In this example, the BS 201 and UE 212 communicate via downlink (DL) transmissions and uplink (UL) transmissions over wireless communications channel 108. The wireless communications channel 108 may include a DL communication channel (e.g., Physical Downlink Shared Channel (PDSCH)) for transmitting a DL transmission signals from the BS 201 to UE 212 and an UL communication channel (e.g., Physical Uplink Shared Channel (PUSCH)) for transmitting an UL transmission signals from the UE 212 to the BS 201. The DL communication channel may also include a DL control channel (e.g., Physical Downlink Control Channel (PDCCH)) and the UL communication channel may also include an UL control channel (e.g., Physical Uplink Control Channel (PUCCH)). In an example, UE 212 transmits the UL transmission signals via the UL control channel. [0057] The BS 201 is implemented as a computing system/apparatus for performing any of the corresponding methods, JSC DNN operations or processes described herein and/or for implementing any of the corresponding systems, units and/or apparatus as described herein. The BS 201 includes a RF front-end Tx/Rx subsystem 204, one or more transceivers 220, one or more processors 221, and a memory unit 223 connected together. The skilled person would appreciate that other types of computing devices/systems/platforms may alternatively be used to implement the BS 201 and the methods described herein, such as a distributed computing system as the application demands. The BS 201 includes one or more processors 221 (e.g., central processing units (CPUs)). The one or more processors 221 control operation of other components of the BS 201 such as RF front-end Tx/Rx subsystem 204, one or more transceivers 220, the memory unit 223 and the like. The one or more processors 221 may be a single core device or a multiple core device. The one or more processors 221 may include a CPU or a graphical processing unit (GPU). Alternatively, the one or more processors 221 may include specialized processing hardware, for instance a reduced instruction set computer (RISC) processor or programmable hardware with embedded firmware. The BS 201 may include multiple processors. In some embodiments, the one or more processors 221 may be part of a distributed computing system such as a cloud computing system and/or cloud computing platform. [0058] The one or more processors 221 of the BS 201 may be connected to a network interface such as, for example, transceivers 220 including a transmitter (Tx) and a receiver (Rx) for communicating via RF front-end Tx/Rx subsystem 204 over wireless communication channel 108 of the network with other apparatus and systems such as UE 212, other communication devices, network equipment, RAN entities or devices, operators and/or any other apparatus, service, system and/or device as the application demands. The one or more processors 221 may, optionally, be connected with a user interface (UI) for user or operator input for instructing or using the BS 201 and/or underlying computing system and/or for outputting data therefrom. The one or more processors 221 may, optionally, be connected with a display for displaying output to a user or operator. [0059] The BS 201 includes memory system or memory unit 223 including a working or volatile memory. The one or more processors 221 may access the volatile memory in order to process data and may control the storage of data in memory. The volatile memory may include random access memory (RAM) of any type, for example, Static RAM (SRAM), Dynamic RAM (DRAM), or it may include Flash memory, such as a Secure Digital Card. In some embodiments, the memory unit 223 and/or one or more volatile memories may include a multiple of a plurality of memory forming part of the distributed computing system such as the cloud computing system and/or cloud computing platform and the like. The BS 201 also includes a non-volatile memory. The non-volatile memory may store a set of operation or operating system instructions for controlling the operation of the processors 221 in the form of computer readable instructions and/or software instructions in the form of computer readable instructions, which when executed on the one or more processors 221 cause the processors 221 to implement the methods, processes, operations and/or functionality of the JSC DNN operations and/or methods as described herein. The non-volatile memory may be a memory of any kind such as a Read Only Memory (ROM), a Flash memory, Secure Digital drive, a magnetic drive memory or magnetic disc drive memory and the like as the application demands. In some embodiments, the non-volatile memory may include a multiple of a plurality of non-volatile memory forming part of the distributed computing system such as the cloud computing system and/or cloud computing platform and the like. [0060] The non-volatile memory of the memory unit 223 of the BS 201 includes computer program code and/or instructions for implementing a BS DNN Controller (BS DNNC) 224 and/or a BS JSC DNN structure 203. The BS DNNC 224, when executed on the one or more processors 221, controls the JSC DNN operations at the BS 201 using BS JSC DNN structure 203 and BS DNN configuration store 225 stored in memory unit 223. Although the BS DNNC 224 is part of the memory unit 223, this is by way of example only and the BS DNNC is not so limited, the skilled person would appreciate that the BS DNNC 224 may be implemented in hardware and/or software of the BS 201 as the application demands. In this example, the BS DNNC 224 configures the BS JSC DNN structure 203 to perform DL JSC applications and includes a BS DL Transmitting Communication and Radar DNN model (BS DL Tx CR-DNN) 203a coupled to a BS DL Receiving Radar DNN model (BS DL Rx R-DNN) 203b. For example, the BS DL Tx CR-DNN 203a processes input communication data to generate an output JSC signal, which the BS 201 transmits as a JSC signal waveform representing the input communication data transformed to be operable as a radar signal as described with reference to FIG.1. Depending on the selected JSC system type (e.g., bi-static, multi-static, or mono-static) that the BS DL Rx R-DNN 203b is configured for, the BS DL Rx R-DNN 203b receives as input one or more radar sensing signals generated from either: a) the BS 201 receiving reflections 110a-110n of the transmitted JSC signal waveform from one or more objects 109a-109n (e.g. mono-static or multi-static JSC); b) the BS 201 receiving a RFB signal waveform 117 transmitted from UE 212 corresponding to reflections 111a-111n of the transmitted JSC signal waveform received and processed at the UE 212 (e.g., bi-static or multi-static); or both a) and b) (e.g. multi-static JSC). The BS DL Rx R-DNN 203b processes the one or more radar sensing signals to generate radar sensing information (e.g., doppler, tracking, position, etc.) in relation to the one or more objects 109a-109n. The BS DNNC 224 uses the selected JSC system type (e.g., bi-static, multi-static, or mono-static JSC) to retrieve suitable model parameters or weights for BS DL Tx CR-DNN 203a and BS DL Rx R- DNN 203b from the BS DNN configuration store 225. In this example, the BS DNNC 224 configures the BS JSC DNN structure 203 to perform, when executed on the one or more processors 221, any of bi-static, multi-static, or mono-static JSC. [0061] Similarly, the UE 212 is implemented as a computing system/apparatus for performing any of the corresponding methods, JSC DNN operations or processes described herein and/or for implementing any of the corresponding systems, units and/or apparatus as described herein. The UE 212 includes a RF front-end Tx/Rx subsystem 213, one or more transceivers 230, one or more processors 231, and a memory unit 233 connected together. The skilled person would appreciate that other types of computing devices/systems/platforms may alternatively be used to implement the UE 212 and the methods described herein. The UE 212 includes one or more processors 221 (e.g., CPUs). The one or more processors 221 control operation of other components of the UE 212 such as RF front-end Tx/Rx subsystem 213, one or more transceivers 230, the memory unit 233 and the like. The one or more processors 231 may be a single core device or a multiple core device. The one or more processors 231 may include a CPU and/or a GPU. Alternatively, the one or more processors 231 may include specialized processing hardware, for instance a RISC processor or programmable hardware with embedded firmware. The UE 212 may include multiple processors. [0062] The one or more processors 231 of the UE 212 may be connected to a network interface such as, for example, transceivers 230 including a Tx and an Rx for communicating via RF front-end Tx/Rx subsystem 213 over wireless communication channel 108 of the network with other apparatus and systems such as BS 201, other communication devices, network equipment, RAN entities or devices, users or operators and/or any other apparatus, service, system and/or device as the application demands. The one or more processors 231 may, optionally, be connected with a UI for user input for instructing or using the UE 212 and/or underlying computing system and/or for outputting data therefrom. The one or more processors 231 may, optionally, be connected with a display for displaying output to a user. [0063] The UE 212 includes memory system or memory unit 233 including a working or volatile memory. The one or more processors 231 may access the volatile memory in order to process data and may control the storage of data in memory. The volatile memory may include RAM of any type, for example, SRAM, DRAM, or it may include Flash memory, such as a Secure Digital-Card. The UE 212 also includes a non- volatile memory. The non-volatile memory may store a set of operations or operating system instructions for controlling the operation of the processors 231 in the form of computer readable instructions and/or software instructions in the form of computer readable instructions, which when executed on the one or more processors 231 cause the processors 231 to implement the corresponding methods, processes, operations and/or functionality of the JSC DNN operations and/or methods at the UE 212 as described herein. The non-volatile memory may be a memory of any kind such as a ROM, a Flash memory, Secure Digital drive, a magnetic drive memory or magnetic disc drive memory and the like as the application demands. [0064] The non-volatile memory of the memory unit 233 of the UE 212 includes computer program code and/or instructions for implementing a UE DNN Controller (UE DNNC) 234 and/or a UE JSC DNN structure 214. The UE DNNC 234, when executed on the one or more processors 231, controls the JSC DNN operations at the UE 212 using a UE JSC DNN structure 214 and UE DNN configuration store 235 stored in memory unit 233. Although the UE DNNC 234 is part of the memory unit 233, this is by way of example only and the UE DNNC 234 is not so limited, where the skilled person would appreciate that the UE DNNC 234 may be implemented in any combination of hardware and/or software of the UE 212 and/or as the application demands. In this example, when the selected JSC system type is either bi-static or multi-static JSC, the UE DNNC 234 configures the UE JSC DNN structure 214 for DL JSC to include a UE DL Receiving Communication and Radar DNN model (UE DL Rx CR-DNN) 214a coupled to a UE DL Transmitting Radar Feedback DNN model (UE DL Tx RFB-DNN) 214b. The UE DL Rx CR-DNN 214a processes a received JSC signal waveform corresponding to the JSC signal waveform transmission from the BS 201 to generate reconstructed communication data corresponding to the input communication data incorporated in the received JSC signal waveform. The UE DL Rx CR-DNN 214a processes radar sensing signals generated from the UE 212 receiving one or more reflections 111a-111n of the JSC signal waveform transmission from the BS 201 from the one or more objects 109a-109n and generates radar sensing feedback (RFB) information (e.g., doppler, tracking and/or position etc.) for the one or more objects 109a-109n. The UE DL Tx RFB-DNN 214b processes the RFB information for the one or more objects 109a-109n and generates a RFB signal for transmission as an RFB signal waveform 117 to the BS 201, for use by the BS 201 (e.g., for bi-static or multi- static JSC) in generating radar sensing information for the one or more objects 109a- 109n. The UE DNNC 234 uses the selected JSC system type (e.g., bi-static or multi- static JSC) to retrieve the model parameters and/or weights of the UE DL Rx CR-DNN 214a and UE DL Tx RFB-DNN 214b from UE DNN configuration store 235 to configure the UE JSC DNN structure 214 for bistatic or multi-static DL JSC. In this example, the UE JSC DNN structure 214 performs, when executed on the one or more processors 231, bi-static or multi-static radar sensing applications. [0065] Additionally or alternatively, when mono-static JSC system type is selected, the UE DNNC 234 configures the UE JSC DNN structure 214 to perform, when executed on the one or more processors 231, mono-static JSC applications by replacing the UE DL Rx CR-DNN 214a and UE DL Tx RFB-DNN 214b with a UE DL Receiving Communications DNN model (UE DL Rx C-DNN) 214c. The UE DNNC 234 uses the selected JSC system type (e.g., mono-static JSC) to retrieve the model parameters and/or weights of the UE DL Rx C-DNN 214c from UE DNN configuration store 235 to configure the UE JSC DNN structure 214 for mono-static DL JSC. In mono-static JSC applications the BS 201 does not use a RFB signal waveform 117 from the UE 212. Rather, the UE DL Rx C-DNN 214c receives and processes the received JSC signal waveform corresponding to the JSC signal waveform transmission from the BS 201 and/or one or more reflections 111a-111n of the JSC signal waveform to generate reconstructed communication data corresponding to the input communication data incorporated in the received JSC signal waveform. [0066] As an option, similar to the BS 201 performing DL JSC, the UE 212 can perform UL JSC in which the UE DNNC 234 configures the UE JSC DNN structure 214 to perform, when executed on the one or more processors 231, UL JSC applications, where the UE 212 performs JSC transmission and radar sensing. For example, for UL JSC, the UE JSC DNN structure 214 includes a UE UL Transmitting Communication and Radar DNN model (UE UL Tx CR-DNN) 214d and a UE UL Receiving Radar DNN model (UE UL Rx R-DNN) 214e. These UE UL DNN models 214d and 214e operate in a similar manner, depending on the selected JSC system type (e.g., bi-static, multi-static, or mono-static), as the corresponding BS DL DNN models 203a and 203b when the BS 201 performs DL JSC with UE 212. For UL JSC at the BS 201, the BS DNNC 224 configures the BS JSC DNN structure 203 to perform, when executed on the one or more processors 221, UL JSC applications. For example, for UL JSC at the BS 201, the BS JSC DNN structure 203 includes one or more of, depending on the selected JSC system type (e.g., bi-static, multi-static, or mono-static), a BS UL Receiving (Rx) Communication and Radar (CR) DNN model (BS UL Rx CR-DNN) 203c, a BS UL Transmitting (Tx) Radar Feedback (RFB) DNN model (BS UL Tx RFB-DNN) 203d, and/or a BS UL Receiving (Rx) Communication (C) DNN model (BS UL Rx C-DNN) 203e. These BS UL DNN models 203c, 203d, and 203e operate in a similar manner, depending on the selected JSC system type (e.g., bi-static, multi-static, or mono-static), as the corresponding UE DL DNN models 214a, 214b, and 214c when the UE 212 performs DL JSC with BS 201. [0067] A plurality of trained pairs of JSC DNN structures including different combinations of the above DNN models for use by the BS 201 and UE 212 in JSC applications are stored in a first neural network table (e.g., first neural network table 1010 of FIG.10b) of the BS DNN configuration store 225 along with the corresponding channel characteristics/ conditions and JSC system types (e.g., bi-static, multi-static, or mono-static JSC) used when training each trained pair of JSC DNN structures. Each trained pair of JSC DNN structures within the first neural network table is assigned a JSC DNN identifier/index. Similarly, the UE 212 stores a second neural network table (e.g., second neural network table 1020 of FIG.10c) in the UE DNN configuration store 235, which includes the trained JSC DNN structure used by the UE 212 from each pair of trained JSC DNN structures and associated corresponding JSC DNN identifiers. In another example, the first neural network table of the BS 201 stores only the corresponding JSC DNN structure used by the BS 201 of each trained pair of JSC DNN structures for each JSC DNN identifier. The second neural network table of the UE 212 also stores the corresponding JSC DNN structure used by the UE 212 that matches the other half of the trained pair of JSC DNN structures for each JSC DNN identifier. [0068] In a further example, in addition to the BS 201 performing JSC DNN operations with the UE 212 using a bi-static or multi-static JSC system type, the BS 201 also recruits a third device as described with reference to FIG.1 or third device 341 of FIG.3d for assisting in bi-static or multi-static JSC. [0069] In an example, the BS 201 sends the radar sensing information of one or more objects 109a-109n for said each time slot to one or more upper layer protocols of a protocol stack of the BS 201 (e.g., to an application protocol layer of a protocol stack for use by one or more applications executing on the BS 201). Similarly, the UE 212 sends the reconstructed communication data to a data sink of the UE 212 or sends the reconstructed communication data to one or more upper protocol layers of a protocol stack of the UE 212 (e.g., to an application protocol layer of a protocol stack for use by one or more applications executing on the UE 212). [0070] In operation, DL JSC DNN operations can be established when the BS 201 identifies the channel condition of the DL communication channel (PDSCH) between the BS 201 and the UE 212 and selects a JSC system type according to JSC performance requirements. A suitable trained pair of JSC DNN structures is selected from the first neural network table using the identified channel condition and selected JSC system type, where the BS 201 transmits a control message to the UE 212 with a field associated with the corresponding JSC DNN identifier of the selected trained pair of JSC DNN structures. The BS 201 uses the JSC DNN identifier to retrieve the corresponding BS JSC DNN structure of the selected trained pair of JSC DNN structures to configure the BS JSC DNN structure 203 accordingly for DL JSC DNN operations. The UE 212 also uses the JSC DNN identifier to retrieve the corresponding UE JSC DNN structure of the selected trained pair of JSC DNN structures and configure the UE JSC DNN structure 214 accordingly for DL JSC DNN operations. In this example, the JSC system type is either a bi-static or multi-static JSC, where the BS JSC DNN structure 203 includes a BS DL Tx CR-DNN 203a coupled with a BS DL Rx R-DNN 203b. The UE JSC DNN structure 214 includes a UE DL Rx CR-DNN 214a coupled with a UE DL Tx RFB-DNN 214b. [0071] In this example, the control message uses DL control plane signalling (e.g., PDCCH) for specifying the UE JSC DNN structure 214 and the one or more time slots and associated communication resources. For example, the control message is either a radio resource control (RRC) message or a downlink control indicator (DCI) signal. The control message indicates particular resource blocks (RBs)/frequencies used for transmitting the JSC signal waveform 107 over the DL communication channel (e.g., PDSCH) in each of the time slots when performing JSC DNN operations. The UE 212 transmits RFB signal waveform 117 corresponding to each time slot over the wireless communication channel 108. The UE 212 can use, without limitation, for example conventional UL communication control channels (e.g., PDCCH) for transmitting the RFB signal waveform 117. Alternatively or additionally, the control message further includes a set of UL RBs / frequencies and subsequent UL time slots that the UE 212 can use when transmitting the RFB signal waveform 117 over the UL in a subsequent UL time slots to the BS 201. Each RFB signal waveform 117 corresponding to a particular DL time slot that the BS 201 used to transmit the JSC signal waveform 107. [0072] Alternatively, if the pair of JSC DNN structures 203 and 214 of the BS 201 or UE 212 outputs a JSC signal waveform 107 or a RFB signal waveform 117 that does not use RBs or particular carrier frequencies, then the control message only needs to specify the JSC DNN identifier and the one or more DL / UL time slots and/or subsequent DL / UL time slots for when the BS 201 transmits each JSC signal waveform 107 over the DL and for when the UE 212 transmits corresponding RFB signal waveforms 117 over the UL. After configuring the BS 201 and UE 212 for performing DL JSC DNN operations during a communication session, the BS 201 and UE 212 proceed to perform DL JSC DNN operations during a communication session in a similar manner to the JSC DNN operations described with reference to any of FIGs.1 and 3a to 11. [0073] In another example, for UL JSC DNN operations, the BS 201 indicates UL RBs / frequency configurations and UL time slots that the UE 212 may use in transmitting one or more JSC signal waveforms over the UL. In operation, establishment of UL JSC DNN operations occurs when the UE 212 or the BS 201 identifies the channel condition of the UL communication channel (e.g., PUSCH) between the BS 201 and the UE 212, and also identifies a JSC system type for UL JSC DNN operations according to JSC performance requirements. In a similar manner as DL JSC DNN operations, a suitable trained pair of JSC DNN structures is selected (e.g., see first and second neural network tables 1010 and 1020 of FIGs.10b or 10c) using the identified channel condition, the selected JSC system type, and JSC performance requirements. The BS DNNC 224 retrieves the BS JSC DNN structure of the trained pair of JSC DNN structures and configures the BS JSC DNN structure 203 with the retrieved trained JSC DNN structure for performing UL JSC DNN operations in one or more time slots. The UE DNNC 234 retrieves the UE JSC DNN structure of the trained pair of JSC DNN structures and configures the UE JSC DNN structure 214 with the retrieved trained JSC DNN structure for performing UL JSC DNN operations in one or more time slots. [0074] In another example, if the UE 212 selects the trained pair of JSC DNN structures that it intends to use given the channel conditions, JSC system type, and JSC performance requirements, the UE 212 transmits a control message specifying the JSC DNN identifier for the selected trained pair of JSC DNN structures to the BS 201. The BS 201 uses the received JSC DNN identifier to retrieve the corresponding JSC DNN structure of the selected trained pair of JSC DNN structures for use by BS 201 from the first neural network table. The UE DNNC 234 retrieves the UE JSC DNN structure of the trained pair of JSC DNN structures and configures the UE JSC DNN structure 214 with the retrieved trained JSC DNN structure for performing UL JSC DNN operations in one or more time slots. The BS DNNC 224 configures the BS JSC DNN structure 203 with the retrieved trained JSC DNN structure for performing UL JSC DNN operations in one or more time slots. In response to the control message, the BS 201 transmits a control message to the UE 212 specifying one or more communication resource parameters (e.g., DL / UL time slots and/or RB configurations, DL / UL frequencies etc.) for use by the UE 212 in transmitting the JSC signal waveform over the UL and for use by the UE 212 in receiving the RFB signal waveforms 117 from the BS 201 over the DL for use by the UE 212 in radar sensing associated with the one or more objects 109 to 109n. [0075] When the JSC system type is either bi-static or multi-static JSC for UL JSC DNN operations, the BS JSC DNN structure 203 includes a BS UL Rx CR-DNN 203c coupled with a BS UL Tx RFB-DNN 203d. The UE JSC DNN structure 214 includes a UE UL Tx CR-DNN 214d coupled with a UE UL Rx R-DNN 214e. After configuring the BS 201 and UE 212 for performing UL JSC DNN operations during a communication session, the BS 201 and UE 212 proceed to perform UL JSC DNN operations during a communication session in a similar manner in relation to the JSC DNN operations as described with reference to FIGs.1 and 3a to 11. [0076] For the BS 201, the at least one processor 221, with the at least one memory unit 223 and computer program code or instructions stored thereon are arranged to cause the computing system of the BS 201 to at least perform at least the corresponding operations, methods, and/or processes, for example as disclosed in relation to the schematic diagrams, flow diagrams or operations as described with any of FIGs.1 to 12 and related features thereof. For the UE 212, the at least one processor 231, with the at least one memory unit 233 and computer program code or instructions stored thereon are arranged to cause the computing system of the UE 212 to at least perform at least the corresponding operations, methods, and/or processes, for example as disclosed in relation to the schematic diagrams, flow diagrams or operations as described with any of FIGs.1 to 12 and related features thereof. [0077] The JSC communication system 200 including the BS 201 and UE 212 provides an advantage of performing efficient JSC in a communication session between the BS 201 and UE 212 of a 5G / 6G or beyond communication system for various communication channel conditions whilst simultaneously performing accurate sensing (e.g., target object detection and/or range/velocity estimation) and meeting the performance requirements of the corresponding communication standard. Further advantages include efficient design of JSC waveforms for use in a 5G / 6G or beyond communication systems and for various sensing and communication channel conditions whilst simultaneously performing accurate sensing and meeting the requirements of the corresponding communication standard. Additional advantages of the JSC communication system 200 and/or as described herein provide efficient, rapid, and dynamic reconfiguration of JSC DNN operations within a communications session between BS 201 and UE 212 (e.g., bi-static, multi-static, mono-static JSC radar sensing) whilst maintaining accurate sensing and communications performance. [0078] Although a wireless communication system 100 or 200 is described with reference FIGs.1 or 2 and/or as herein described, this is by way of example only and it is not so limited, it is to be appreciated by the skilled person that any type of communication system or network is applicable such as, for example, any telecommunication system or network; any wired communication network; any wireless communication network; a satellite network; a peer-2-peer communication network; a communication system or network using third generation (3G), fourth generation (4G), 5G, and/or 6G and beyond standards technologies; a Wi-Fi communication network; optical communication network; a fibre optic communication network; and/or any other network for communications between the first device and second device; combinations thereof, modifications thereto, and/or as the application demands. Although the first device 101 is described as a BS 201 with reference to FIG. 2 and/or as herein described, this is by way of example only and it is not so limited, it is to be appreciated by the skilled person that the first device 101 may be any type of communication device that is capable of communicating with the second device 112 such as, without limitation, for example a UE 212, a BS, a satellite, a mobile phone or smart phone, a laptop, a computing device, a device using 3G, 4G, 5G, and/or 6G and beyond standards technologies, and/or any other device used for communications with the second device 112; combinations thereof, modifications thereto, and/or as the application demands. Although the second device 112 is described as a UE 212 with reference to FIG.2 and/or as herein described, this is by way of example only and it is not so limited, it is to be appreciated by the skilled person that the second device 112 may be any type of communication device that is capable of communicating with the first device 101 such as, without limitation, for example a UE 212, a BS, a satellite, a mobile phone or smart phone, a laptop, a computing device, a device using 3G, 4G, 5G, and/or 6G and beyond standards technologies, and/or any other device used for communications with the first device 101; combinations thereof, modifications thereto, and/or as the application demands. [0079] FIG.3a illustrates an example of a wireless communication system 300a with a first device 301 and a second device 312 for performing bi-static JSC. FIG.3a modifies the wireless communication systems 100 and 200 of FIGs.1 and 2 by further defining the JSC DNN structures of the first and second devices 101 and 112 or BS 201 and UE 212, respectively. The first and second devices 301 and 312 have already established a communication session over wireless communication/radar channel 308 as described herein with reference to FIGs.1 or 2 and/or FIGs.4 to 12. The first and second devices 301 and 312 establish and perform bi-static JSC DNN operations in one or more particular time slots of the communication session over wireless communication/radar channel 308. [0080] Bi-static JSC DNN operations are established when the first device 301 selects a trained pair of JSC DNN structures that correspond to a bi-static JSC system type. The selected trained pair of JSC DNN structures includes a first trained JSC DNN structure and a second trained JSC DNN structure with transmitting/receiving DNN models arranged for bi-static JSC. The first trained JSC DNN structure including a trained Transmitting Communication and Radar DNN model (Tx CR-DNN) 303a coupled to a trained Receiving Radar DNN model (Rx R-DNN) 303b for configuring a JSC DNN structure 303 of the first device 301 for bi-static JSC DNN operations. The second trained JSC DNN structure including a trained Receiving Communication and Radar DNN model (Rx CR-DNN) 314a coupled to a trained Transmitting Radar Feedback DNN model (Tx RFB-DNN) 314b for configuring a JSC DNN structure 314 of the second device 312 for bi-static JSC DNN operations. The Tx CR-DNN 303a and Rx R-DNN 303b of the first JSC DNN structure and the Rx CR-DNN 314a and Tx RFB- DNN 314b of the second JSC DNN structure have been jointly trained and stored for selection by first device 301 as described, for example, with reference to FIGs.10a to 11. [0081] In this example, the first device 301 configures the JSC DNN structure 303 with the trained Tx CR-DNN 303a coupled to the trained Rx R-DNN 303b. The Tx CR-DNN 303a generates an optional feed forward communication and radar (FFCR) signal 303c (or a feed forward signal) that is passed to the Rx R-DNN 303b when performing radar sensing for a particular time slot of the one or more time slots. The FFCR signal 303c includes, without limitation, for example any one or more from the group of: input communication data 302 for the particular time slot; output JSC signal 306a for the particular time slot; inputs to one or more neural network layers of the Tx CR-DNN 303a; outputs of one or more neural network layers of the Tx CR-DNN 303a; and/or a combination thereof. The trained Tx CR-DNN 303a receives as input the input communication data 102 for transmission in the particular time slot and generates an output JSC signal 306a that represents the input communication data 102 and a radar signal for the particular time slot. [0082] The first device 301 transmits the output JSC signal 306a in the particular time slot, via DAC / RF Tx antenna component 304a of RF front-end Tx/Rx subsystem 304, as JSC signal waveform 307. The first device 301 transmits the JSC signal waveform 307 in the particular time slot to the second device 312 over a wireless communication/radar channel 308. The transmission of the JSC signal waveform 307 in the particular time slot represents the input communication data 302 for transmission to the second device 312 and the radar signal. There are one or more objects 309a-309n (e.g., people, animals, obstacles, trees, buildings, vehicles and/or any other object capable of reflecting the transmitted JSC signal waveform 307) within the environment of the wireless communication/radar channel 308. [0083] The second device 312 includes an RF front-end Tx/Rx subsystem 313 with an RF Rx antenna / ADC (RF Rx/ADC) component 313b coupled to the JSC DNN structure 314. The RF Rx/ADC component 313b receives, in the particular time slot, the transmitted JSC signal waveform 307 or the at least one component 307a of the JSC signal waveform 307 and one or more further reflected JSC signal waveforms 311a, 311b, and 311c to 311n based on reflections of the JSC signal waveform 307 from the corresponding one or more objects 309a, 309b, and 309c, to 309n in the environment of the wireless communication/radar channel 308. The RF Rx/ADC component 313b (e.g., frequency down conversion / ADC to baseband) processes the received JSC signal waveform 307 and/or component 307a thereof and one or more further reflected JSC signal waveforms 311a, 311b, and 311c to 311n into a JSC signal 316a for the particular time slot. [0084] The JSC DNN structure 314 is coupled to the RF front-end Tx/Rx subsystem 313. The second device 312 configures the JSC DNN structure 314 to include the trained Rx CR-DNN 314a coupled to the trained Tx RFB-DNN 314b. The JSC signal 316a for the particular time slot is input to the Rx CR-DNN 314a of the JSC DNN structure 314. The Rx CR-DNN 314a processes the JSC signal 316a and generates reconstructed communication data 315 for the particular time slot and also radar sensing feedback (RFB) information 314d for the particular time slot. The reconstructed communication data 315 for the particular time slot corresponds to the input communication data 102 transmitted in the JSC signal waveform 307 in the particular time slot. The RFB information 314d represents, without limitation, for example one or more of range, doppler, velocity, position, azimuth, elevation or other radar measurements / parameters and the like for the one or more objects 309a, 309b, and 309c to 309n in the environment of the wireless communication/radar channel 308. The generated RFB information 314d is input to the Tx RFB-DNN 314b. Tx RFB- DNN 314b processes the RFB information 314d (e.g., range, doppler, velocity, position, or other radar measurement and the like) and generates an RFB signal 316b corresponding to the particular time slot for transmission to the first device 301. [0085] The RF front-end Tx/Rx subsystem 313 of the second device 312 also includes a DAC / RF Tx antenna (DAC/RF Tx) component 313a coupled to the Tx RFB- DNN 314b of the JSC DNN structure 314. The DAC/RF Tx component 313a receives and processes the RFB signal 316b for transmission as a RFB signal waveform 317 corresponding to the particular time slot to the first device 301. This will assist the trained JSC DNN structure 303 of the first device 301 to generate radar sensing information (e.g., range, doppler, velocity, position, or other radar measurement and the like) corresponding to the one or more objects 309a-309n in the environment of the wireless communication / radar channel 308 for the particular time slot. [0086] The first device 301 receives and processes the RFB signal waveform 317 corresponding to the particular time slot, via RF Rx / ADC component 304b of the RF front-end Tx/Rx subsystem 304, as radar sensing signal 306b corresponding to the particular time slot. The radar sensing signal 306b and the FFCR signal 303c corresponding to the particular time slot are input to the Rx R-DNN 303b, which generates radar sensing information 305 of the one or more objects 309a, 309b, and 309c to 309n (e.g., range, doppler, velocity, position, or other radar measurements / parameters and the like) for the particular time slot. [0087] FIG.3b illustrates an example of a wireless communication system 300b with a first device 301 and a second device 312 of FIG.3a arranged for multi-static JSC. FIG.3b modifies the wireless communication systems 100, 200, and 300a of FIGs.1, 2, and 3a by further configuring the JSC DNN structures of the first and second devices 101 and 112, BS 201 and UE 212, or first and second devices 301 and 312 of FIG.3a, respectively, into a multi-static JSC system type. In this example, the first device 301 and a second device 312 perform JSC using DNNs jointly trained for when the JSC system type is multi-static JSC. [0088] In a similar manner as for FIG.3a, the selected trained pair of JSC DNN structures for multi-static JSC system type includes a first trained JSC DNN structure and a second trained JSC DNN structure with transmitting/receiving DNN models arranged for multi-static JSC. The first trained JSC DNN structure including a trained Tx CR-DNN 303a coupled to a trained Rx R-DNN 303b for configuring a JSC DNN structure 303 of the first device 301 and trained for multi-static JSC DNN operations. The second trained JSC DNN structure including a trained Rx CR-DNN 314a coupled to a trained Tx RFB-DNN 314b for configuring a JSC DNN structure 314 of the second device 312 and trained for multi-static JSC DNN operations. The Tx CR-DNN 303a and Rx R-DNN 303b of the first JSC DNN structure and the Rx CR-DNN 314a and Tx RFB- DNN 314b of the second JSC DNN structure are jointly trained for multi-static JSC and stored for selection by first device 301 as described, for example, with reference to FIGs.10a to 11. [0089] The multi-static JSC system type is similar to the bi-static JSC system type described with respect to FIG.3a with the additional modification that the first device 301 operates in full duplex mode for receiving, for each particular time slot, a plurality of reflected JSC signal waveforms 310a, 310b, and 310c to 310n corresponding to the JSC signal waveform 307 transmission in the particular time slot and reflected from the corresponding one or more objects 309a, 309b, and 309c to 309n. The first device 301 receives and processes, via RF Rx/ADC component 304b of the RF front-end Tx/Rx subsystem 304, the reflected JSC signal waveforms 310a, 310b, and 310c to 310n and also the RFB signal waveform 317 corresponding to the particular time slot to generate one or more radar sensing signals 306b corresponding to the particular time slot. The trained Rx R-DNN 303b receives, as input, multiple radar sensing signals 306b. The radar sensing signals 306b and the FFCR signal 303c corresponding to the particular time slot are input to the trained Rx R-DNN 303b, which generates radar sensing information 305 of the one or more objects 309a, 309b, and 309c to 309n (e.g., range, doppler, velocity, position, or other radar measurements/parameters and the like) for the particular time slot. [0090] It is noted, that at least the trained Rx R-DNN 303b will have different trained DNN layers, nodes, weights and biases compared with the trained Rx R-DNN 303b of FIG.3a due to the multiple inputs of the radar sensing signals 306b derived from the reflected JSC signal waveforms 310a, 310b, and 310c to 310n and also the RFB signal waveform 317. The second device 312 is configured the same as for the bi-static radar sensing configuration of FIG.3a and operates in the same manner as described with reference to FIG.3a. [0091] FIG.3c illustrates an example wireless communication system 300c with a first device 301 and a second device 312 of FIG.3a arranged for mono-static JSC. FIG.3c modifies the wireless communication systems 100, 200, 300a, and 300b of FIGs.1, 2, 3a, and 3b by further configuring the JSC DNN structures of the first and second devices 101 and 112, BS 201 and UE 212, or first and second devices 301 and 312, respectively, into a mono-static JSC system type. In this example, the first device 301 and a second device 312 performs JSC using DNNs jointly trained in for when the JSC system type is mono-static JSC. [0092] In a similar manner as for FIG.3b, the selected trained pair of JSC DNN structures for mono-static JSC system type includes a first trained JSC DNN structure and a second trained JSC DNN structure with transmitting/receiving DNN models arranged for mono-static JSC. The first trained JSC DNN structure including a trained Tx CR-DNN 303a coupled to a trained Rx R-DNN 303b for configuring a JSC DNN structure 303 of the first device 301 and trained for mono-static JSC DNN operations. However, for mono-static JSC, the second trained JSC DNN structure only includes a trained Receiving Communications DNN (Rx C-DNN) 314c for configuring the JSC DNN structure 314 of the second device 312 and trained for mono-static JSC DNN operations. The Tx CR-DNN 303a and Rx R-DNN 303b of the first JSC DNN structure and the Rx C-DNN 314c of the second JSC DNN structure are jointly trained for mono- static JSC and stored for selection by first device 301 as described, for example, with reference to FIGs.10a to 11. [0093] The mono-static JSC system type is a subset of the multi-static JSC system type as described with reference to FIG.3b, but with the additional modification that the first device 301 operates in full duplex mode for only receiving, for each particular time slot, the plurality of reflected JSC signal waveforms 310a, 310b, and 310c to 310n corresponding to the JSC signal waveform 307 transmission in the particular time slot and reflected from the corresponding one or more objects 309a, 309b, and 309c to 309n. The second device 312 does not provide a RFB signal waveform 317 as described with reference to FIGs.3a and 3b. Instead, the first device 301 receives and processes, via RF Rx/ADC component 304b of the RF front-end Tx/Rx subsystem 304, the reflected JSC signal waveforms 310a, 310b, and 310c to 310n corresponding to the particular time slot to generate one or more radar sensing signals 306b corresponding to the particular time slot. The trained Rx R-DNN 303b is configured for receiving as input the multiple radar sensing signals 306b. The radar sensing signals 306b and the FFCR signal 303c corresponding to the particular time slot are input to the trained Rx R-DNN 303b, which generates radar sensing information 305 of the one or more objects 309a, 309b, and 309c to 309n (e.g., range, doppler, velocity, position, or other radar measurements / parameters and the like) for the particular time slot. [0094] As well, the RF Rx/ADC component 313b of the second device 312 receives, in the particular time slot, the transmitted JSC signal waveform 307. The RF Rx/ADC component 313b (e.g., frequency down conversion / ADC to baseband) processes the received transmitted JSC signal waveform 307 into a JSC signal 316a for the particular time slot. The received JSC signal 316a for the particular time slot is input to the trained Rx C-DNN 314c, which only generates reconstructed communications data 315 for the particular time slot. As an option, the second device 312 can also additionally receive and process, via RF Rx/ADC component 313b, a plurality of reflected JSC signal waveforms 311a, 311b, and 311c to 311n corresponding to the JSC signal waveform 307 transmission in the particular time slot and reflected from the corresponding one or more objects 309a, 309b, and 309c to 309n for input as a plurality of JSC signals 316a (or combined into a JSC signal 316a) to Rx C-DNN 314c for generating reconstructed communication data 315. [0095] FIG.3d illustrates an example of a wireless communication system 300d with the first device 301 and the second device 312 of FIGs.3a or 3b arranged for bi- static or multi-static JSC plus a third device 341. FIG.3d modifies the wireless communication systems 100, 200, 300a, and 300b of FIGs.1, 2, 3a, and 3b in which the third device 341 uses a modified JSC DNN structure for assisting the first device 301 in bi-static or multi-static JSC. In this example, the first device 301 and a second device 312 perform JSC using DNNs in which the JSC system type is a bi-static or multi-static JSC as described with reference to FIGs.3a and 3b. However, the third device 341 operates in a bi-static radar sensing configuration only and does not reconstruct communication data that is destined for the second device 312. Instead, the RF front-end Tx/Rx 342 of the third device 341 receives, in the particular time slot, the transmitted JSC signal waveform 307 or component 307a thereof and one or more further reflected JSC signal waveforms 340a, 340b, and 340c to 340n based on reflections of the transmitted JSC signal waveform 307 from the corresponding one or more objects 309a, 309b, and 309c to 309n. The RF front-end Tx/Rx 342 processes the received transmitted JSC signal waveform 307 and any one or more further received reflected JSC signal waveforms 340a-340n into a received JSC signal 346a for the particular time slot. The JSC DNN structure of the third device 341 is a Radar Sensing JSC DNN structure 343 (Radar-DNN structure) that only includes a trained Receiving Radar DNN model (Rx R-DNN) coupled to a trained Tx RFB-DNN model. The Rx R-DNN of the Radar DNN structure 343 processes the received JSC signal 346a and generates, for the particular time slot, RFB information for the particular time slot. The generated RFB information is input to the Tx RFB-DNN of the Radar DNN structure 343. Tx RFB-DNN of the Radar DNN structure 343 processes the RFB information (e.g., range, doppler, velocity, position, or other radar measurement and the like) and generates a RFB signal 346b for the particular time slot. The RF front-end Tx/Rx 342 processes the RFB signal 346b for transmitting as a RFB signal waveform 345 for the particular time slot to the first device 301. This additional RFB signal waveform 345 is received by the first device 301 in a similar manner that the RFB signal waveform 317 is received from the second device 312 for the particular time slot as described with reference to FIGs.3a and 3b. The additional RFB signal waveform 345 will assist the trained JSC DNN structure 303 of the first device 301 to generate radar sensing information (e.g., range, doppler, velocity, position, or other radar measurement and the like) in a similar manner as the RFB signal waveform 317 from the second device 312 does in FIGs.3a and 3b (e.g., range, doppler, velocity, position, or other radar measurement and the like). [0096] FIGs.3a to 3c illustrate and describe three JSC system types, namely, bi- static JSC, multi-static JSC, and mono-static JSC whilst FIG.3d illustrates a combination of two JSC system types (e.g., bi-static JSC or multi-static JSC) using the first device 301, the second device 312, and the third device 341. In the following disclosure, FIGs.4 to 6 illustrate and describe signalling diagrams for the three JSC types of bi-static JSC, multi-static JSC and mono-static JSC, by way of example only and the system is not so limited, it is to be appreciated by the skilled person in the art that the processes and/or systems described by the signalling diagrams can be further modified and/or combined based on combination of two or more JSC system types as described and/or as the application demands. [0097] FIG.4 illustrates a signal flow diagram of example bi-static JSC DNN operations 400 for one or more predefined time slots during a communications session between a first device 101 and a second device 112. The first and second devices 101 and 112 of FIG.1 perform bi-static JSC DNN operations 400 using any of the bi-static JSC aspects as described with reference to FIGs.1, 2, 3a, 3b and/or 3d. The signal flow of the bi-static JSC DNN operations 400 for the communication session between the first and second devices 101 and 112 include the following signal flow operations. [0098] In operations 402/422, the first and second devices 101 and 112 establish bi-static JSC DNN communications with each other. That is, the first device 101 establishes JSC DNN operations for one or more time slots of the communication session with the second device 112. During establishment of the bi-static DNN connection, the first and second devices 101 and 112 communicate with each other for defining, agreeing, and/or configuring the type of JSC DNN structures each of the first device 101 and second device 112 will use including the transmitting/receiving DNN models used for performing end-to-end JSC therebetween. As described with reference to FIGs.2, 3a, 3b and 3d, the JSC DNN structure of the first device 101 include a Tx CR- DNN model and Rx R-DNN model. As described with reference to FIGs.2, 3a, 3b and 3d, the JSC DNN structure of the second device 112 will include a Rx CR-DNN model and Tx RFB-DNN model. In this example, the first device 101 identifies the channel conditions/characteristics of the communication / radar channel between the first device 101 and the second device 112, which the first device 101 uses to select a pair of JSC DNN structures of a bi-static JSC system type from first or second neural network tables (provided for the first device 101 or second device 112, respectively) as described with reference to FIGs.10a to 10c. For example, after the first device 101 initiates a conventional or standard communication session with the second device 112, the first device 101 selects the pair of JSC DNN structures for use in the bi-static JSC DNN connection with the second device 112 depending on the communication / radar channel conditions/ environment, the radar and/or communication performance requirements for the JSC DNN connection (e.g., JSC performance requirements), and the data communications characteristics (e.g., voice communication, data communications, multimedia streaming, QoS parameters, network slicing parameters, and the like). In an example, the first device 101 requests the machine learning processing capabilities of the second device 112, which assists the first device 101 in selecting the pair of JSC DNN structures for the first device 101 and the second device 112 for bi-static JSC DNN communications/operations. [0099] After selecting the pair of JSC DNN structures for the bi-static DNN connection, the first device 101 initiates the bi-static DNN connection by sending a bi- static JSC DNN connection request message to the second device 112 with fields specifying a bi-static JSC DNN connection and a JSC DNN indicator corresponding to the selected pair of JSC DNN structures. The JSC DNN indicator enables the second device 112 to select the corresponding bi-static JSC DNN structure to assist the first device 101 perform radar sensing of one or more objects 109a-109n. For example, the first device 101 sends one or more DL control messages to the second device 112 indicating a selected type of JSC DNN structure (e.g., the selected specific JSC DNN identifier) for use by the second device 112 in processing the transmitted JSC signal waveforms. [00100] In operation 403, the first device 101 configures its JSC DNN structure based on the selected pair of JSC DNN structures or the selected specific JSC DNN identifier. The selected JSC DNN structure of the first device 101 performs JSC transmission and radar sensing operations. In this example, the first device 101 configures its JSC DNN structure to include Tx CR-DNN and Rx-R-DNN models as described with reference to FIGs.1, 2, 3a, 3b and 3d. In operation 423, the second device configures its JSC DNN structure based on the selected pair of JSC DNN structures or the selected specific JSC DNN identifier. The selected JSC DNN structure of the second device performs communication data reception and radar feedback operations. For example, in operation 423, upon receiving the one or more control messages from the first device 101 indicating a selected type of JSC DNN structure for use by the second device 112 in processing the received JSC signal waveforms, the second device 112 configures the JSC DNN structure of the second device 112 based on the selected type of JSC DNN structure for use in the one or more time slots. In this example, the second device 112 configures its JSC DNN structure to include a Rx CR- DNN and Tx RFB-DNN models as described with reference to FIGs.1, 2, 3a, 3b and 3d. [00101] For each time slot of the one or more predefined time slots, the first device 101 and second device 112 perform the following operations: In operation 404, the first device 101 retrieves input communication data for transmission in the time slot to the second device 112. In operation 406, the first device 101 processes the input communication data with a Tx CR-DNN of the JSC DNN structure of the first device to generate an output JSC signal representing the input communication data and a radar signal. In operation 408, the first device 101 transmits to the second device 112, the output JSC signal as a Tx JSC signal waveform in the time slot over the communication channel. [00102] In operation 424, the second device 112 receives the Tx JSC signal waveform transmitted from the first device 101 in the time slot over the communication channel. The received Tx JSC signal waveform represents input communication data and a radar signal transmitted in the time slot. In operation 426, the second device (SD) 112 also receives one or more further JSC signal waveforms (e.g., O₁ SD JSC-signal waveform, O₂ SD JSC-signal waveform, …, O_N SD JSC-signal waveform) for the time slot based on reflections of the Tx JSC signal waveform transmission from one or more objects 109a-109n (e.g., O1, O2, …, ON). In operation 428, the second device 112 processes the received Tx JSC-signal waveform and one or more further received JSC signal waveforms for the time slot using the Rx CR-DNN model of the JSC DNN structure at the second device 112 to generate reconstructed communication data corresponding to the input communication data transmitted in the time slot plus radar sensing feedback (RFB) information (e.g., range, doppler, velocity, etc.) associated with the one or more objects 109a-109n. In operation 430, the Tx RFB-DNN processes the RFB information for the time slot to generate a RFB signal for transmission to the first device 101. In operation 432, the second device 112 transmits the RFB signal as a Tx RFB-signal waveform (e.g., RFB signal waveform) to the first device 101 for use by the first device 101 in generating radar sensing information at the first device 101 for the one or more objects 109a-109n. In operation 434, the second device 112 sends the reconstructed communication data for the time slot to a data sink. For example, the second device 112 sends the reconstructed communication data for the time slot to one or more upper protocol layers of a protocol stack of the second device 112. For example, second device 112 sends the reconstructed communication data to an application protocol layer of the protocol stack of the second device 112 for use by one or more applications executing on the second device 112. In operation 436, the second device 112 proceeds to operations 424/426 for receiving another Tx JSC signal waveform transmitted from the first device 101 in a next time slot (or further subsequent time slots) over the communication channel. For example, the next time slot is a subsequent time slot the first device 101 has designated for performing bi-static JSC DNN operations within. [00103] The first device 101, in operation 410, receives the Tx RFB-signal waveform transmitted by the second device 112 for the time slot as a received radar sensing signal, which includes information representative of the RFB information associated with the one or more objects 109a-109n generated at the second device 112 in operations 428 to 430. The Rx R-DNN of the JSC DNN structure of the first device 101 processes the received radar sensing signal to generate radar sensing information of the one or more objects 109a-109n in the communication channel. In operation 414, the first device 101 sends the radar sensing information of one or more objects 109a- 109n for the time slot to one or more upper layer application protocols of a protocol stack of the first device 101. For example, the first device 101 sends the radar sensing information of one or more objects 109a-109n for said each time slot to one or more upper layer protocols of a protocol stack of the first device 101 (e.g., to an application protocol layer of a protocol stack for use by one or more applications executing on the first device 101). The one or more applications may execute on the first device and/or one or more other devices and access the radar sensing information of the one or more objects 109a-109n accordingly. For example, the second device 112 sends the reconstructed communication data for the time slot to one or more upper protocol layers of a protocol stack of the second device 112. For example, the second device 112 sends the reconstructed communication data to an application protocol layer of the protocol stack of the second device 112 for use by one or more applications executing on the second device 112. [00104] In operation 416, the first device 101 determines whether any more time slots of the predefined time slots are available for transmission of input communication data using bi-static JSC operations. If so, then the signal flow of bi-static JSC DNN operations 400 proceeds to operation 404 for transmitting further input communication data in the next time slot. If there are no further predefined time slots using bi-static JSC DNN operations 400, then the signal flow of the bi-static JSC DNN operations 400 proceeds to operation 418. In operation 418, the first device 101 transmits a control message to the second device disabling bi-static JSC communications. The first device 101 disables bi-static JSC communications. In operation 438, upon receiving the control message disabling bi-static JSC communications, the second device 112 also disables bi-static JSC communications. [00105] In operation 439, the first and second devices 101 and 112 continue with the communication session using conventional communications and/or terminate the communication session as the application demands. [00106] FIG.5 illustrates a signal flow diagram of example multi-static JSC DNN operations 500 for one or more predefined time slots during a communications session between a first device 101 and a second device 112. The bi-static JSC DNN operations 400 of FIG.4 are further modified in which the first and second devices 101 and 112 are configured to perform multi-static JSC aspects as described with reference to FIGs.1, 2, 3a, 3b and 3d. The signal flow of the multi-static JSC DNN operations 500 for the communication session between the first and second devices 101 and 112 include the following signal flow operations of: [00107] In operations 502/522, the first and second devices 101 and 112 establish JSC DNN communications with each other in a similar manner as for operations 402/422 of FIG.4. In this case, the first device 101 establishes multi-static JSC DNN operations for one or more predefined time slots of the communication session with the second device 112. During establishment, the first device 101 selects a pair of JSC DNN structures from first or second neural network tables (e.g., as described with reference to FIGs.10a to 10c) associated with the multi-static JSC system type and one or more communication channel conditions/characteristics identified by the first device 101 in relation to the communication channel. As described with reference to FIG.4, the selected pair of JSC DNN structures include a similar JSC DNN structure for the first device 101 and a similar JSC DNN structure for the second device 112 in which the corresponding models are jointly trained for the multi-static JSC system type. [00108] Operations 503 and 523 substantially correspond to operations 403 and 523 of FIG.4, except that the trained Tx CR-DNN model and trained Rx R-DNN model performs JSC transmission and multi-static radar sensing (rather than bi-static JSC). [00109] For each time slot of the one or more predefined time slots, the first device 101 and second device 112 perform the multi-static JSC operations 504 to 508, which substantially corresponds to operations 404 to 408 of FIG.4. Operations 524 to 536 substantially correspond to operations 424 to 436 of FIG.4. In operation 532, the second device 112 transmits the radar sensing feedback (RFB) signal as a Tx RFB-signal waveform to the first device 101 for use by the first device 101 in generating multi-static radar sensing information for the one or more objects 109a-109n. [00110] The first device 101, in operation 510, receives the Tx RFB-waveform transmitted by the second device 112 for the time slot. In contrast to operation 410 of FIG.4, the first device (FD) 101 receives a plurality of reflected Tx JSC signal waveforms (e.g., O₁ FD JSC-signal waveform, O₂ FD JSC-signal waveform, … O_N FD JSC-signal waveform) due to the Tx JSC-signal waveform reflected from the one or more objects 109a-109n (e.g., O₁, O₂, …, O_N). The RF front-end Tx/Rx of the first device 101 processes the received Tx RFB-waveform transmitted from the second device 112 for the time slot and the received plurality of reflected Tx JSC signal waveforms (e.g., O₁ FD JSC-signal waveform, O₂ FD JSC-signal waveform, … O_N FD JSC-signal waveform) for the time slot and generates the received radar sensing signal. In operation 512, the Rx R-DNN of the JSC DNN structure of the first device 101 processes the received radar sensing signal (e.g., reflected FD JSC-signal waveforms and Tx RFB signal waveform) to generate radar sensing information of the one or more objects 109a-109n in the communication channel. In operation 514, the first device 101 sends the radar sensing information of one or more objects 109a-109n for the time slot to one or more upper layers of a protocol stack of the first device 101. For example, the first device 101 sends the radar sensing information of one or more objects 109a-109n for said each time slot to one or more upper layer protocols of a protocol stack of the first device 101 (e.g., to an application protocol layer of a protocol stack for use by one or more applications executing on the first device 101). The one or more applications may execute on the first device and/or one or more other devices and access the radar sensing information of the one or more objects 109a-109n accordingly. [00111] In operation 516, the first device 101 determines whether there are any more time slots of the predefined time slots for transmission of input communication data using multi-static JSC operations. If so, then the signal flow of multi-static JSC DNN operations 500 proceeds to operation 504 for transmitting further input communication data in the next time slot. If there are no further predefined time slots, then the signal flow of the multi-static JSC DNN operations 500 proceeds to operations 518, 538 and 539, which correspond to operations 418, 438 and 439 of FIG.4. [00112] FIG.6 illustrates a signal flow diagram of example mono-static JSC DNN operations 600 for one or more predefined time slots during a communications session between a first device 101 and a second device 112. The mono-static JSC is a subset of multi-static JSC as described with reference to FIG.5. In mono-static JSC, the first device 101 receives reflections of the JSC signal waveform transmitted from the first device 101 for generating radar sensing information, the second device 112 only generates reconstructed data from a received JSC signal waveform transmitted by the first device 101 and/or reflections thereof from the one or more objects 109a-109n. The multi-static JSC DNN operations 500 of FIG.5 is further modified in which the first and second devices 101 and 112 are configured to perform the mono-static JSC aspects as described with reference to FIGs.1, 2 and 3c. The signal flow of the mono-static JSC DNN operations 600 for the communication session between the first and second devices 101 and 112 include the following signal flow operations of: [00113] In operations 602/622, the first and second devices 101 and 112 establish JSC DNN communications with each other in a similar manner as for operations 502/522 of FIG.5. In this case, the first device 101 establishes mono-static JSC DNN operations for one or more predefined time slots of the communication session with the second device 112. As described with reference to FIGs.1, 2, and 3c, the selected pair of JSC DNN structures include a similar JSC DNN structure for the first device 101 and a similar JSC DNN structure for the second device 112 in which the corresponding DNN models have been jointly trained for a mono-static JSC system type. [00114] Operations 603 and 623 are substantially correspond to operations 503 and 523 of FIG.5 except that the trained Tx CR-DNN model and trained Rx R-DNN model of the selected pair of JSC DNN structures of the first device 101 perform JSC transmission and mono-static JSC (rather than multi-static JSC). The trained Rx C- DNN model of the selected pair of JSC DNN structures of the second device 112 performs reconstruction of any input communication data transmitted in a Tx JSC signal waveform for each time slot from the first device 101 and/or reflected versions thereof from the one or more objects 109a-109n. [00115] For each time slot of the one or more predefined time slots, the first device 101 and second device 112 perform mono-static JSC operations 604 to 608, which substantially correspond to operations 504 to 508 of FIG.5, in which the Tx JSC signal waveform representing the input communication data and a radar signal is transmitted in the time slot over the communication channel [00116] In operation 624, the second device 112 receives the Tx JSC signal waveform for the time slot. In operation 632, the second device 112 processes the received Tx JSC signal waveform for the time slot using the Rx C-DNN model of the JSC DNN structure of the second device 112 to generate reconstructed communication data corresponding to the input communication data transmitted in the Tx JSC signal waveform for the time slot. As an option, the Rx-C-DNN model can also process any received reflections of the transmitted Tx JSC signal waveform from the one or more objects 109a-109n when generating reconstructed communications data for the time slot. In operation 634, the second device 112 sends the reconstructed communications data for the time slot to a data sink or to one or more upper protocol layers of a protocol stack of the second device 112. Operation 636 substantially corresponds to operations 436 or 536 of FIGs.4 or 5, respectively. [00117] Operations 610 to 616 of FIG.6 substantially correspond to operations 510 to 516 of FIG.5, but where the Rx R-DNN of the JSC DNN structure of the first device 101 only processes a plurality of received reflected Tx JSC signal waveforms as a result of the transmitted Tx JSC signal waveform e.g., O1 FD JSC-signal waveform, O2 FD JSC-signal waveform, … O_N FD JSC-signal waveform)) into the received radar sensing signal for input to the Rx R-DNN to generate mono-static radar sensing information of the one or more objects 109a-109n in the communication channel for the time slot. In operation 614, the first device 101 sends the mono-static radar sensing information of one or more objects 109a-109n for the time slot to one or more upper layer protocols of the protocol stack of the first device 101. In operation 616, the first device 101 determines whether there are any more time slots of the predefined time slots for transmission of input communication data using mono-static JSC DNN operations. If so, then the signal flow of mono-static JSC DNN operations 600 proceeds to operation 604 for transmitting further input communication data in the next time slot. If there are no further predefined time slots for mono-static JSC DNN operations 600, then the signal flow of the mono-static JSC DNN operations 600 proceeds to operations 618, 638 and 639, which correspond to operations 418, 438 and 439 of FIG.4. [00118] FIG.7a illustrates a signal flow diagram of example bi-static JSC DNN operations 700a for one or more predefined time slots (TSs) (e.g., TS{a, b, c, d}) of a communications session between a first device 201 and a second device 212 in wireless communication system 200. In this example, the first device 201 is a BS, referred to as BS 201, and the second device 212 is a UE, referred to as UE 212. Although this example describes that the first device 201 as a BS 201 and the second device 212 as a UE 212, this is by way of example only and this example is not so limited. It is to be appreciated by the skilled person that the first and second devices 201 and 212 can be any type of communication device for use in wireless communication system 200 such as, but not limited to, for example any combination of radio access network elements including BSs, network devices, UEs, or other RAN elements within wireless communication system 200. For example, the first device 201 and second device 212 may be two BSs, or two UEs, or a BS and a UE, or a UE and a BS, or any other combination of communication devices as the application demands. As an option, the wireless communication system 200 can include one or more other communications devices (e.g., a third device as described in FIG.3d) for assisting BS 201 in bi-static radar sensing for each time slot during bi-static JSC DNN operations 700a of FIG.7a. [00119] In this example, the data communication channel includes a DL data channel, for example, a PDSCH for transmissions of JSC signal waveforms from BS 201 to the UE 212, and a DL control channel, for example, a PDCCH for establishing bi- static JSC DNN operations in the communication session between BS 201 and UE 212. The data communication channel also includes an UL data channel, for example, a PUSCH for transmissions of radar sensing feedback (RFB) signal waveforms 117 from the UE 212 to the BS 201. Alternatively, rather than using an UL data channel, the UE 212 may use an UL control channel, for example, a PUCCH for transmissions of RFB signal waveforms 117 from the UE 212 to the BS 201. [00120] The BS 201 includes a BS DNNC 224 connected to a BS JSC DNN structure 203 that perform bi-static JSC DNN operations as described with reference to FIGs.1, 2, 3a, 3b, 4 and 5. Similarly, the UE 212 includes a UE DNNC 234 connected to a UE JSC DNN structure 214 that also performs bi-static JSC DNN operations as described with reference to FIGs.1, 2, 3a, 3b, 4 and 5. [00121] Operations 702 and 722 are similar to operations 402 and 422 of FIG.4, where the BS DNNC 224 establishes a DL bi-static JSC DNN communications session between BS 201 and UE 212. In operation 702, the BS DNNC 224 determines the PDSCH channel characteristics/condition and retrieves a specific JSC DNN identifier (e.g., DNN ID1) that identifies a pair of JSC DNN structures from the neural network table provided for the BS 201 for a bi-static radar system type that addresses the known channel conditions, communication performance requirements and/or radar performance requirements. The selected pair of JSC DNN structures includes a BS JSC DNN structure and a UE JSC DNN structure for bi-static JSC system type. As an option, the BS 201 receives UE capability information of UE 212 (e.g., processing power such as central processing unit or graphics processing unit capabilities, memory, DNN capabilities, operating system and software version, and the like etc.) and/or UE assistance information (e.g., power saving (low battery), battery levels, processor usage, thermal status, and/or temperature thresholds/overheating). The BS DNNC 224 uses UE capability and/or assistance information to select and retrieve the pair of JSC DNN structures for the bi-static radar system type in which the UE is capable of configuring the UE JSC DNN structure of the pair of JSC DNN structures. The BS JSC DNN structure and US JSC DNN structure are based on those described with reference to, for example, FIG.4. The BS 201 transmits an RRC Establishment Request message including a DNN type or specific JSC DNN identifier (e.g., DNN ID1) associated with the selected BS JSC DNN / UE JSC DNN pair, a set of predefined DL time slots (TSs) (e.g., DL TSs {a, b, c, d}), a set of DL frequencies and/or resource blocks (RBs) for use in each of the set of predefined DL time slots, a set of predefined UL TSs (e.g., UL TS{e, f, g, h}) and a set of UL frequencies and/or resource blocks (RBs) for use in each of the set of predefined UL time slots (e.g., RRC JSC DNN Establishment Req (DNN ID1, DL TS {a, b, c, d}, DL freq. / RBs, UL TS {e, f, g, h}, UL freq. / RBs)). [00122] Operation 703 is similar to operation 403 of FIG.4 where the BS DNNC 224 issues a configuration instruction (e.g., Cfg(DNN ID1)) associated with the selected BS JSC DNN structure, and where the BS 201 configures the BS JSC DNN structure 203 according to the specific JSC DNN identifier (e.g., DNN ID1) using the neural network table provided for the BS 201 (e.g., master neural network table 1000 or first neural network table 1010 as described with reference to FIGs.10a to 10b). For example, the BS 201 retrieves the DNN parameters/weights and configuration parameter of the trained BS Tx CR-DNN and trained BS Rx R-DNN corresponding to the selected BS JSC DNN Structure from the neural network table using the JSC DNN identifier (e.g., DNN ID1). The BS 201 configures the BS JSC DNN structure 203 with the DNN parameters/weights and/or configuration parameters of the trained BS Tx CR-DNN and trained BS Rx R-DNN of the selected BS JSC DNN Structure. After configuration of the BS JSC DNN structure 203, the BS JSC DNN structure 203 includes the trained BS Tx CR-DNN and trained BS Rx R-DNN of the selected BS JSC DNN Structure for bi-static JSC operations. [00123] Operation 723 is similar to operation 423 of FIG.4 where the UE DNNC 234 receives the RRC Establishment Request message (e.g., RRC JSC DNN Establishment Req (DNN ID1, DL TS {a, b, c, d}, DL freq./RBs, UL TS {e, f, g, h}, UL freq./RBs)) and the UE DNNC 234 issues a configuration instruction (e.g., Cfg(DNN ID1)) associated with the selected UE JSC DNN structure, where the UE 212 configures the UE JSC DNN structure 214 according to the specific JSC DNN identifier (e.g., DNN ID1) using the neural network table provided for the UE 212 (e.g., second neural network table 1020 as described with reference to FIGs.10a to 10c). For example, the UE 212 retrieves the DNN parameters/weights and/or configuration parameters of the trained UE Rx CR-DNN and trained UE Tx RFB-DNN corresponding to the selected UE JSC DNN Structure from the neural network table using the JSC DNN identifier (e.g., DNN ID1). The UE 212 configures the UE JSC DNN structure 214 with the DNN parameters/weights and/or configuration parameters of the trained UE Rx CR-DNN and trained UE Tx RFB-DNN of the selected UE JSC DNN Structure. After configuration of the UE JSC DNN structure 214, the UE JSC DNN structure 214 includes the trained UE Rx CR-DNN and trained UE Tx RFB-DNN of the selected UE JSC DNN Structure for bi-static JSC operations. [00124] In operation 722, after configuration of the UE JSC DNN structure 214, the UE DNNC 234 of the UE 212 sends an RRC response indicating acknowledgement (ACK) of configuring the UE JSC DNN structure 214 to the BS 201 (e.g., RRC JSC DNN Establishment Resp (ACK)). The operations 702-703 and 722-723 complete the establishment of the DL bi-static JSC DNN communications session between BS 201 and UE 212. [00125] On receipt of the acknowledgement, in operation 704a, input communication data for transmission in a specific time slot, TS a (e.g., input communication data X for TS a (IDXa)) is applied to the Tx CR-DNN model of the BS JSC DNN structure 203, which generates an output JSC signal (e.g., DNN_IDXa_Sens) representing a combination of IDXa with a radar signal (e.g., Sens). That is, the output JSC signal (e.g., DNN_IDXa_Sens) represents a digital joint communication and radar signal with the input communication data for TS a (e.g., IDXa) incorporated therein and/or as described with reference to FIG.1. For example, the BS JSC DNN structure 203 generates the output JSC signal (e.g., DNN_IDXa_Sens) by processing the input communication data for TS a, where the output JSC signal for TS a (e.g., DNN_IDXa_Sens) represents the input communication data for transmission in TS a to the UE 212 but digitally transformed (or conformed) by the BS JSC DNN structure 203 to operate also as a radar signal for radar sensing. In operation 706a, the BS JSC DNN structure 203 provides the output JSC signal (e.g., DNN_IDXa_Sens) to the BS DNNC 224 for transmission as a transmit JSC signal waveform over PDSCH to the UE 210 in TS a. On receiving DNN_IDXa_Sens, the BS DNNC 224 buffers the DNN_IDXa_Sens data until transmission in TS a. In the specific TS a, the BS 201 processes and transmits the output JSC signal (e.g., DNN_IDXa_Sens) as a JSC signal waveform signal over PDSCH. In operation 708a, the BS DNNC 224 transmits the JSC signal waveform for TS a over PDSCH in TS a (e.g., PDSCH RF Tx JSC WAVEFORM (DNN_IDXa_Sens_TS_a))). The JSC signal waveform for TS a represents a JSC waveform incorporating the input communication data for TS a but which is optimised for the channel conditions /communication and/or radar performance requirements / JSC system type to operate as a radar signal for radar sensing and/or as described with reference to FIG.1. [00126] Operations 724a to 734a substantially correspond to operations 424 to 434 of FIG.4. In operation 724a, the UE 212 receives the transmission JSC signal waveform over PDSCH in TS a and processes (e.g., down converts) into a first received JSC signal for TS a (e.g., RxIDXa_Sens_TS_a). The transmitted JSC signal waveform is also reflected from objects 109a-109n (e.g., O₁, O₂, …, O_N), which generate a plurality of reflected transmitted JSC signal waveforms for TS a (e.g., O₁ UE JSC-signal waveform_TS_a, O₂ UE JSC-signal waveform_TS_a, …, O_N UE JSC-signal waveform_TS_a). In each of operations 726a-1, 726a-2, to 726a-N, the UE 212 receives corresponding reflected transmitted JSC signal waveforms for TS a, where UE DNNC 234 processes the received reflected waveforms from the one or more objects 109a- 109n for TS a, which include, for example, reflected waveforms O₁ UE JSC-signal O₂ UE JSC-signal waveform_TS_a, …, O_N UE JSC-signal into a corresponding plurality of received reflected JSC signals for TS O₂UEJSC_TS_a, …, O_NUEJSC_TS_a). As described in operation 428 of FIG.4, the first received JSC signal for TS a (e.g., RxIDXa_Sens_TS_a) and the plurality of received reflected JSC signals for TS a (e.g., O₁UEJSC_TS_a, O₂UEJSC_TS_a, …, O_NUEJSC_TS_a) are input to and processed by the UE Rx CR-DNN of the UE JSC DNN structure 214 to generate reconstructed communications data for TS a (e.g., RCIDXa) and to generate radar sensing feedback (RFB) information (e.g., range, doppler, velocity) for the one or more objects. The RFB information is input to the UE Tx RFB-DNN of the UE JSC DNN structure 214 to generate a RFB signal for TS a (e.g., Tx RFB_TS_a)). The RFB signal for TS a (e.g., Tx RFB_TS_a) represents range, doppler, velocity, position, or other radar measurement and the like of the one or more objects 109a-109n as described with reference to FIG.4. [00128] In operation 730a, the RFB signal (e.g., Tx RFB_TS_a) is sent to the UE DNNC 234 and transmitted as a RFB signal waveform for TS a in one of the available UL time slots (e.g., TS e), where in operation 732a, the RFB signal for TS a (e.g., Tx RFB_TS_a) is transmitted as a transmission RFB signal waveform for TS a in TS e over PUSCH to the BS 201 (e.g., PUSCH Tx JSC DATA (Tx RFB_TS_a, TS e). Alternatively, the RFB signal for TS a can be sent over PUCCH. In operation 734a, the UE 212 sends the reconstructed communications data for TS a (e.g., RCIDXa) to a data sink such as, for example, one or more upper layer protocols of the protocol stack of the UE 212 and/or as herein described. [00129] The BS 201 receives, in operation 710a, the transmitted RFB signal waveform for TS a in TS e over PUSCH and processes (e.g., down conversion to base band) into a received radar sensing signal for TS a (e.g., Rx RS_RFB_TS_a). The BS DNNC 224 buffers the received radar sensing signal for TS a (e.g., Rx RS_RFB_TS_a) received in TS e until the Rx R-DNN model of the BS JSC DNN structure 203 is ready for processing the received radar sensing signal for TS a. Operations 712a to 714a substantially correspond to operations 412 to 414 of FIG.4. In operation 712a, the BS DNNC 224 inputs the received radar sensing signal Rx RS_RFB_TS_a to the Rx R- DNN model for generating radar sensing information for TS a (e.g., RSTS_a) in relation to the one or more objects 109a-109n. The RSTS_a represents one or more from the group of: range, doppler, velocity, position, or other radar measurement and the like in relation to the one or more objects 109a-109n for TS a. For example, the RSTS_a data represents one or more from the group of: range estimate associated with each of the objects 109a-109n, doppler estimate associated with each of the objects 109a-109n, velocity estimate associated with each of the objects 109a-109n, a location or position estimate associated with each of the objects 109a-109n, a delay spread associated with each of the objects 109a-109n, an average delay associated with each of the objects 109a-109n, an angular estimate associated with each of the objects 109a- 109n, an azimuth and/or elevation estimate associated with each of the objects 109a- 109n, and/or any other suitable radar measurement associated with each of the objects 109a-109n, radar sensing information or radar sensing parameter estimate and the like in relation to the one or more objects 109a-109n for TS a. [00130] For the BS 201 and UE 212 operations 704a to 714a and operations 724a to 734a, respectively, repeat for each set of input communication data for transmission in subsequent time slots (e.g., IDXb for TS b, IDYc for TS c, IDZd for TS d,… etc.) to generate reconstructed communication data for each of the subsequent time slots (e.g., RCIDXb for TS b, RCIDYc for TS c, RCIDZd for TS d,… etc.) at the UE 212 and generate radar sensing information for each of the subsequent time slots (e.g., TS{b, c, d}) for objects 109a-109n (e.g., RSTS_b for TS b, RSTS_c for TS c, RSTS_d for TS d…, etc). For example, operations 704d to 714d and 724d to 734d are performed in a similar manner as for operations 704a to 714a and 724a to 734a where the BS 201 and UE 212 generate radar sensing information for TS d (e.g., RSTS_d) in relation to the one or more objects 109a-109n and reconstructed communications data for TS d (e.g., RCIDZd), which corresponds to the input communication data transmitted in TS d (e.g., IDZd). [00131] As for operations 418 and 438 of FIG.4, the bi-static JSC DNN operations repeat until the BS DNNC 224 determines the bi-static JSC DNN operations should be disabled. In operation 718, the BS DNNC 224 transmits a RRC control message to the UE 212 to disable bi-static JSC communications (e.g., RRC JSC DNN Disable Req (.)). The BS DNNC 224 disables bi-static JSC communications. In operation 738, on receiving the RRC control message disabling bi-static JSC communications, the UE 212 also disables bi-static JSC communications and sends an acknowledgement accordingly (e.g., RRC JSC DNN Disable Resp (ACK)). [00132] FIGs.7b and 7c illustrate signal flow diagrams of example multi-static JSC DNN operations 700b and example mono-static JSC DNN operations 700c, respectively. The multi-static DNN operations 700b of FIG.7b is essentially the same as the bi-static DNN operations 700a of FIG.7a but differs in that the BS 201 also processes further received reflected signals from objects 109a-109n at the BS 201 for further enhancing the accuracy of the radar sensing information. The mono-static DNN operations 700c of FIG.7c differs from the bi-static DNN operations 700a and multi-static DNN operations 700c of FIGs.7a and 7c, respectively, in that the BS 201 only processes further received reflected signals from objects 109a-109n at the BS 201, where the UE 212 does not transmit an RFB signal waveform (e.g., radar sensing feedback signal waveform) to the BS 201 as described in FIGs.7a and 7b. The mono- static DNN operations 700c further simplifies processing of the radar sensing information at the BS 201. The BS 201 and UE 212 use mono-static DNN operations 700c when the UE 212 is incapable of more complex JSC processing or JSC DNN operations. In addition, mono-static DNN operations 700c decrease the latency in generating radar sensing information at the BS 201 compared with bi-static DNN operations 700a and multi-static DNN operations 700b, where the BS 202 waits for the RFB signal waveform corresponding to a TS from the UE 212 before generating radar sensing information for that TS. [00133] FIG.7b illustrates a signal flow diagram of the example multi-static JSC DNN operations 700b for one or more predefined time slots (e.g., TS{a, b, c, d}) of a communications session between BS 201 and UE 212. The multi-static JSC DNN operations 700b modifies the bi-static JSC DNN operations 700a of FIG.7a by including processing of further received reflected signals from objects 109a-109n at the BS 201 for enhancing the radar sensing information. [00134] Operations 702, 722, 703 and 723 of FIG.7b substantially correspond to operations 702, 712, 703 and 723 of FIG.7a apart from the BS DNNC 224 selecting the pair of JSC DNN structures for multi-static JSC system type rather than bi-static JSC system type as in FIG.7a. In this example, the BS JSC DNN structure includes a trained BS Tx CR-DNN model and a trained BS Rx R-DNN model of a multi-static JSC system type. The UE JSC DNN structure includes a trained UE Rx CR-DNN model and a trained UE Tx RFB-DNN model of the multi-static JSC system type. After configuration of the BS JSC DNN structure 203 in a similar manner as described with reference to FIG.7a, in operation 703, the BS JSC DNN structure 203 includes the trained BS Tx CR-DNN and trained BS Rx R-DNN of the selected BS JSC DNN structure for multi-static JSC operations. After configuration of the UE JSC DNN structure 214 in a similar manner as described with reference to FIG.7a, in operation 723, the UE JSC DNN structure 214 includes the trained UE Rx CR-DNN and trained UE Tx RFB-DNN of the selected UE JSC DNN Structure for multi-static JSC operations. In operation 723, on configuration of the UE JSC DNN structure 214, the UE DNNC 234 of the UE 212 sends an RRC response in operation 722, indicating acknowledgement of configuring the UE JSC DNN structure 214 to the BS 201 (e.g., RRC JSC DNN Establishment Resp (ACK)). [00135] The BS 201 and UE 212 perform multi-static DNN operations 700b after the BS 201 receives, in operation 722, an acknowledgement (ACK) of UE 212 successfully configuring the UE JSC DNN structure 214 (e.g., RRC JSC DNN Establishment Resp (ACK)). Multi-static DNN operations 700b are performed in which operations 704a, 706a, 708a, 724a, 726a-1, 726a-2, 726a-N, 730a, 732a/710a, 712a, and 734a substantially correspond to bi-static DNN operations 704a, 706a, 708a, 724a, 726a-1, 726a-2, 726a-N, 730a, 732a/710a, 712a, and 734a of FIG.7a apart from the BS JSC DNN and UE JSC DNN being of a multi-static JSC type. In addition, on transmission of the JSC signal waveform for TS a over PDSCH in operation 708a, the transmitted JSC signal waveform also reflects from objects 109a-109n (e.g., O₁, O₂, …, O_N) to generate a plurality of BS received reflected transmitted JSC signal waveform for TS a (e.g., O₁ BS JSC-signal waveform_TS_a, O₂ BS JSC-signal waveform_TS_a, …, O_N BS JSC-signal waveform_TS_a). For each of operations 710a-1, 710a-2, to 710a-N, the BS 201 receives the corresponding BS reflected transmitted JSC signal waveform for TS a (e.g., O₁ BS JSC-signal waveform_TS_a, O₂ BS JSC-signal waveform_TS_a, …, O_N BS JSC-signal waveform_TS_a). The BS DNNC 224 processes the BS received reflected waveforms from objects 109a-109n for TS a, which include, for example, the reflected waveforms of O₁ BS JSC-signal waveform_TS_a, O₂ BS JSC-signal waveform_TS_a, …, ON BS JSC-signal waveform_TS_a, into a corresponding plurality of radar sensing signals for TS a (e.g., O₁BSJSC_RS_TS_a, O₂BSJSC_RS_TS_a, …, O_NBSJSC_RS_TS_a). The BS DNNC 224 buffers the received radar sensing signals for TS a (e.g., O₁BSJSC_RS_TS_a, O₂BSJSC_RS_TS_a, …, O_NBSJSC_RS_TS_a) until the BS JSC DNN has the multi-static inputs for generating radar sensing information for TS a. As described in operation 528 of FIG.5 and also in operations 724a, 726a-1, 726a-2, 726a-N, 730a, 732a/710a of FIG.7a, the UE 212 generates an RFB signal for TS a (e.g., Tx RFB_TS_a) which is transmitted to the BS 201 as an RFB signal waveform over PUSCH in TS e in operation 732a (e.g., PUSCH Tx JSC DATA (Tx RFB_TS_a, TS e). Alternatively, the RFB signal for TS a can be sent over PUCCH. As well, in operation 710a, the BS 201 receives the transmitted RFB signal waveform for TS a in TS e over PUSCH from the UE 212 and processes the received RFB signal waveform for TS a (e.g., down conversion to base band) into a further received radar sensing signal for TS a (e.g., Rx RS_RFB_TS_a). [00136] Operations 712a to 714a substantially correspond to operations 512 to 514 of FIG.5 and operations 712a and 714a as described with respect to FIG.7a, except that in operation 712a of FIG.7b, the BS DNNC 224 inputs the plurality of received radar sensing signals for TS a (e.g., O₁BSJSC_RS_TS_a, O₂BSJSC_RS_TS_a, …, O_NBSJSC_RS_TS_a) and the further received radar sensing signal for TS a (e.g., Rx RS_RFB_TS_a) from the UE 212 to the Rx R-DNN model of the BS JSC DNN structure 203 for generating radar sensing information for TS a (e.g., RSTS_a) in relation to the one or more objects 109a-109n. [00137] The RSTS_a data for TS a represents one or more from the group of: range, doppler, velocity, position, or other radar measurement and the like in relation to the one or more objects 109a-109n for TS a. For example, the RSTS_a includes data representative of one or more from the group of: range estimate associated with each of the objects 109a-109n, doppler estimate associated with each of the objects 109a-109n, velocity estimate associated with each of the objects 109a-109n, a location or position estimate associated with each of the objects 109a-109n, a delay spread associated with each of the objects 109a-109n, an average delay associated with each of the objects 109a-109n, an angular estimate associated with each of the objects 109a-109n, an azimuth and/or elevation estimate associated with each of the objects 109a-109n, and/or any other suitable radar measurement associated with each of the objects 109a- 109n, radar sensing information or radar sensing parameter estimate and the like in relation to the one or more objects 109a-109n for TS a. [00138] For the BS 201 and UE 212 operations 704a to 714a and operations 724a-734a, respectively, repeat for each set of input communication data for transmission in subsequent time slots TS b, TS c and TS d (e.g., IDXb for TS b, IDYc for TS c, IDZd for TS d,… etc.) to generate reconstructed communication data for each of the subsequent time slots (e.g., RCIDXb for TS b, RCIDYc for TS c, RCIDZd for TS d,… etc.) at the UE 212 and generate radar sensing information for each of the subsequent time slots for objects 109a-109n (e.g., RSTS_b for TS b, RSTS_c for TS c, RSTS_d for TS d…, etc). As for operations 518 and 538 of FIG.5, the multi-static JSC DNN operations repeat until the BS DNNC 224 determines the multi-static JSC DNN operations are to be disabled, where, in FIG.7b, operations 718 and 738 corresponding to operations 718 and 738 of FIG.7a are performed. [00139] FIG.7c illustrates a signal flow diagram of example mono-static JSC DNN operations 700c for one or more predefined time slots (e.g., TS{a, b, c, d}) of a communications session between BS 201 and UE 212. The mono-static JSC DNN operations 700c modifies the multi-static JSC DNN operations 700b of FIG.7b by removing the UE processing of the further received reflected signal from objects 109a- 109n so the RFB signal waveform is not generated at the UE 212, whilst retaining the processing of the BS received reflected signals from objects 109a-109n at the BS 201. This provides an advantage of simplifying processing of the radar sensing information for when the UE 212 is incapable of more complex JSC processing. Another advantage includes throughput in generating radar sensing information for each time slot compared with bi-static or multi-static JSC system types because the BS 201 does not have to wait for reception of a corresponding RFB signal waveform from UE 212 in relation to each time slot. [00140] Operations 702, 722, 703 and 723 of FIG.7b substantially correspond to operations previously described with FIG.7a or 7b apart from the BS DNNC 224 selecting the pair of JSC DNN structures for mono-static JSC system type. In this example, the BS JSC DNN structure includes a trained BS Tx CR-DNN model and a trained BS Rx R-DNN model of a mono-static JSC system type. The UE JSC DNN structure includes a trained UE Rx C-DNN model only. After configuration of the BS JSC DNN structure 203, in operation 703, the BS JSC DNN structure 203 includes the trained BS Tx CR-DNN and trained BS Rx R-DNN of the selected BS JSC DNN Structure for mono-static JSC DNN operations 700c. After configuration of the UE JSC DNN structure 214, in operation 723, the UE JSC DNN structure 214 includes the trained UE Rx C-DNN for mono-static JSC DNN operations. In operation 723, after configuration of the UE JSC DNN structure 214, the UE DNNC 234 of the UE 212 sends an RRC response, in operation 722, indicating acknowledgement of configuring the UE JSC DNN structure 214 to the BS 201 (e.g., RRC JSC DNN Establishment Resp (ACK)). [00141] The BS 201 and UE 212 perform mono-static JSC DNN operations 700c after the BS 201 receives, in operation 722, the acknowledgement of the UE 212 successfully configuring the UE JSC DNN structure 214 (e.g., RRC JSC DNN Establishment Resp (ACK)). Mono-static JSC DNN operations 700c are performed in which operations 704a, 706a, 708a, and 710a-1, 710a-2, 710a-N substantially correspond to multi-static DNN operations 704a, 706a, 708a, and 710a-1, 710a-2, 710a- N of FIG.7b apart from the BS JSC DNN and UE JSC DNN being configured for a mono-static JSC system type. [00142] In operation 708a, the BS DNNC 224 transmits the DNN_IDXa_Sens data to the UE 212 as a transmission JSC signal waveform over PDSCH in TS a (e.g., PDSCH RF Tx JSC WAVEFORM (DNN_IDXa_Sens, TS a))). The UE 212 receives the transmission JSC signal waveform over PDSCH in TS a and processes (e.g., down converts) into a received JSC signal for TS a (e.g., RxIDXa_Sens_TS_a). In operation 724a, the UE DNNC 234 inputs the received JSC signal for TS a (e.g., RxIDXa_Sens_TS_a) to the Rx C-DNN model of the UE JSC DNN structure 214 to generate reconstructed communications data for TS a (e.g., RCIDXa). The UE DNNC 234 of the UE 212 sends the reconstructed communications data for TS a (e.g., RCIDXa) that corresponds to the input communication data IDXa transmitted in TS a to a data sink or to an upper layer protocol of a protocol stack of the UE 212 and/or as described herein. [00143] In operation 708a, on transmission over PDSCH of the JSC signal waveform for TS a, the transmitted JSC signal waveform also reflects from objects 109a-109n (e.g., O₁, O₂, …, O_N) to generate a plurality of BS received reflected transmitted JSC signal waveforms for TS a (e.g., O₁ BS JSC-signal waveform_TS_a, O₂ BS JSC-signal waveform_TS_a, …, O_N BS JSC-signal waveform_TS_a). As for operations 710a-1, 710a-2, to 710a-N of FIG.7b, the BS 201 processes the corresponding BS received reflected transmitted JSC signal waveforms for TS a into one or more received radar sensing signals for TS a (e.g., O₁BSJSC_RS_TS_a, O₂BSJSC_RS_TS_a, …, O_NBSJSC_RS_TS_a). In response to operations 710a-1, 710a- 2, to 710a-N, the BS DNNC 224 inputs the received radar sensing signals for TS a into the Rx R-DNN model of the BS JSC DNN structure 203 for generating radar sensing information for TS a (e.g., RSTS_a) in relation to the one or more objects 109a-109n. [00144] The radar sensing information for TS a (e.g., RSTS_a) in relation to the one or more objects 109a-109n represents one or more from the group of: a range estimate of each associated object, doppler estimate of each associated object, a velocity estimate of each associated object, a position or location estimate of each associated object, a delay spread, a doppler spread, an average delay, an angular estimate of each associated object, an azimuth and/or elevation estimate of each associated object, or other radar measurement and the like in relation to the one or more objects 109a-109n for TS a. For example, the RSTS_a includes data representative of one or more from the group of: range estimate, doppler estimate, velocity estimate, a location or position estimate, a delay spread, an average delay, an angular estimate, an azimuth and/or elevation estimate, and/or any other suitable radar measurement, radar sensing information or radar sensing parameter estimate and the like in relation to the one or more objects 109a-109n for TS a. [00145] For the BS 201 and UE 212 operations 704a to 714a and operations 724a-744a, respectively, repeat for each set of input communication data for transmission in subsequent time slots (e.g., IDXb for TS b, IDYc for TS c, IDZd for TS d,… etc.) to generate reconstructed communication data for each of the subsequent time slots (e.g., RCIDXb for TS b, RCIDYc for TS c, RCIDZd for TS d,… etc.) at the UE 212 and generate radar sensing information for each of the subsequent time slots for objects 109a-109n (e.g., RSTS_b for TS b, RSTS_c for TS c, RSTS_d for TS d…, etc). The mono-static JSC DNN operations repeat until the BS DNNC 224 determines the mono-static JSC DNN operations should be disabled, where, in FIG.7c, the operations 718 and 748 corresponding to operations 718 and 738 of FIG.7a are performed. [00146] Although FIGs.7a to 7c described downlink JSC DNN operations between a BS 201 and a UE 212, this is by way of example only and it is not so limited. It is to be appreciated by the skilled person that the first device 201 and the second device 212 can be a UE and a BS, respectively, where the first device 201 (i.e., UE) and second device 212 (i.e., BS) perform uplink JSC DNN operations over PUSCH. [00147] FIG.8a illustrates a flow diagram of an example JSC DNN process 800 for a first device performing JSC DNN operations in one or more time slots of a communication session with a second device. The JSC DNN process 800 includes the following steps. [00148] In step 802, establishing JSC DNN operations for one or more time slots of a communication session with a second device. In an example, step 802 includes establishing JSC DNN communications and configuring the JSC DNN structure of the first device as described with reference to operations 402 and 403 of FIG.4, and operations 702, 703, 722, and 723 of FIG.7a with respect to a bi-static JSC system type. In another example, step 802 includes establishing JSC DNN communications and configuring the JSC DNN structure of the first device as described with reference to operations 502 and 503 of FIG.5, and operations 702, 703, 722, and 723 of FIG.7b with respect to a multi-static JSC system type. In a further example, step 802 includes establishing JSC DNN communications and configuring the JSC DNN structure of the first device as described with reference to operations 602 and 603 of FIG.6, and operations 702, 703, 722, and 723 of FIG.7c with respect to a mono-static JSC system type. [00149] For each time slot of the one or more time slots, the JSC DNN process 800 performs steps 804 to 816 based on: [00150] In step 804, retrieving input communication data for transmission to the second device in said each time slot. For example, step 804 includes retrieving input communications data as described with reference to any of operations 404, 504, 604, 704a-704d with reference to FIGs.4 to 7c. [00151] In step 806, processing the input communication data with a JSC DNN structure of the first device to generate an output JSC signal, where the output JSC signal represents the input communication data and a radar signal for transmission in said each time slot and/or as described with reference to FIG.1. For example, the JSC DNN structure includes a transmitting communications and radar DNN (Tx CR-DNN) model for outputting JSC signal for transmission in said each time slot when given the input communication data as input. In an example, step 806 includes processing the input communications data using a Tx CR-DNN to generate an output JSC signal as described with reference to operations 406 of FIG.4, and operations 706a-706d of FIG. 7a with respect to a bi-static JSC system type. In another example, step 806 includes processing the input communications data using a Tx CR-DNN to generate an output JSC signal as described with reference to operations 506 of FIG.5, and operations 706a-706d of FIG.7b with respect to a multi-static JSC system type. In a further example, step 806 includes processing the input communications data using a Tx CR- DNN to generate an output JSC signal as described with reference to operations 606 of FIG.4, and operations 706a-706d of FIG.7c with respect to a mono-static JSC system type. [00152] In step 808, transmitting, to the second device, the output JSC signal as a JSC signal waveform in said each time slot over a communication channel. For example, step 808 includes transmitting the output JSC signal as a JSC signal waveform to the second device as described with reference to any of operations 408, 508, 608, 708a-708d with reference to FIGs.4 to 7c. [00153] In step 810, receiving one or more radar sensing signals for said each time slot based on reflections of the transmitted JSC signal waveform from one or more objects. For example, for a bi-static JSC system type, in step 810 the one or more radar sensing signals include receiving a RFB signal waveform transmitted from the second device as described with reference to any of operations 410 and 732a/710a to 732d/710d with reference to any of FIGs.4 and 7a, respectively. For example, the one or more radar sensing signals are generated from the first device receiving, from the second device, a RFB signal waveform generated by the second device using reflections of the transmitted JSC signal waveform from the one or more objects for the time slot. The first device performs RF to baseband processing of the received RFB signal waveform for the time slot to generate into the radar sensing signal(s) for the time slot. In another example, for a multi-static JSC system type, in step 810 the one or more radar sensing signals include receiving a RFB signal waveform transmitted from the second device and also receiving reflections of the transmitted JSC signal waveform from the one or more objects as described with reference to any of operations 510, and operations 732a/710a to 732d/710d and 710a-1 to 710a-N to 710d-1 to 710d-N with reference to FIGs.5 and 7b, respectively. In another example, for a mono-static JSC system type, in step 810 the one or more radar sensing signals include receiving reflections of the transmitted JSC signal waveform from the one or more objects as described with reference to any of operations 610, and operations 732a/710a to 732d/710d and 710a-1 to 710a-N to 710d-1 to 710d-N with reference to FIGs.6 and 7c, respectively. For example, the one or more radar sensing signals are generated from the first device receiving one or more reflected JSC signal waveforms for the time slot due to reflections of the transmitted JSC signal waveform from the one or more objects. The first device performs RF to baseband processing of the received one or more reflected JSC signal waveforms associated with the one or more objects for the time slot into one or more radar sensing signals for the time slot that correspond to the one or more objects. [00154] In step 812, processing the received radar sensing signals by the JSC DNN structure (e.g., BS JSC DNN structure 203 of FIG.2) of the first device to generate radar sensing information of the one or more objects. For example, the JSC DNN structure includes a receiving radar DNN (Rx R-DNN) model for generating radar sensing information when given the received radar sensing signals as input. For example, for a bi-static JSC system type, in step 812 the radar sensing signals representing the received RFB signal waveform are input to a Rx R-DNN model for generating radar sensing information with reference to any of operations 412 and 712a- 712d with reference to any of FIGs.4 and 7a, respectively. In another example, for a multi-static JSC system type, in step 812 the radar sensing signals representig the received RFB signal waveform and also reflections of the transmitted JSC signal waveform from the one or more objects are input to a Rx R-DNN model for generating radar sensing information as described with reference to any of operations 512 and operations 712a-712d with reference to FIGs.5 and 7b, respectively. In another example, for a mono-static JSC system type, in step 812 the radar sensing signals representing the reflections of the transmitted JSC signal waveform from the one or more objects are input to a Rx R-DNN model for generating radar sensing information as described with reference to any of operations 612 and operations 712a-712d with reference to FIGs.6 and 7c, respectively. [00155] In step 814, sending the radar sensing information of the one or more objects for each time slot to one or more upper protocol layers of a protocol stack of the first device. In an example, the first device sends the radar sensing information of one or more objects for said each time slot to one or more upper layer protocols of the protocol stack of the first device. For example, the radar sensing information is sent up to an application protocol layer of a protocol stack for use by one or more applications executing on the first device. [00156] In step 816, determining whether to continue JSC DNN operations. If there are further time slots in the one or more time slots (e.g., ‘Y’), then proceed to step 804 for retrieving input communication data for the next time slot of the one or more time slots. If there are no further time slots for performing JSC DNN operations (e.g., ‘N’), then proceed to step 818. [00157] In step 818, disabling JSC DNN operations for the communication session with the second device. For example, see operations 418 and 438 of FIG.4, operations 518 and 538 of FIG.5, operations 618 and 638 of FIG.6, and operations 718 and 738 of any of FIGs.7a to 7c. [00158] FIG.8b illustrates a flow diagram of an example JSC DNN establishment process of step 802 of FIG.8a performed by a first device when establishing JSC DNN operations for one or more time slots of the communication session with the second device. The JSC DNN establishment process of step 802 includes the following steps of: [00159] In step 802a, sending one or more control messages to the second device indicating a selected JSC DNN structure for use in each time slot. For example, the one or more control messages may be an RRC establishment request message or equivalents thereof as described in operation 702 with reference to any of FIGs.7a to 7c. [00160] In step 802b, configuring the JSC DNN structure of the first device with the selected JSC DNN structure for use in processing the input communication data and receiving the one or more radar sensing signals. For example, the first device configures the JSC DNN structure of the first device as described in operation 703 with reference to any of FIGs.7a to 7c. [00161] The JSC DNN establishment process of step 802 proceeds to step 804 of JSC DNN process 800 of FIG.8. [00162] FIG.9a is a flow diagram illustrating an example JSC DNN process 920 for a second device performing bi-static or multi-static JSC operations in one or more time slots of a communication session with a first device. The bi-static or multi-static JSC DNN operations are based on the bi-static or multi-static JSC DNN operations as described with reference to FIGs.3a, 3b, 4, 5, 7a and 7b. The JSC DNN process 920 for the second device includes the following steps of: [00163] In step 922, establishing JSC DNN operations for one or more time slots of a communication session with a first device. In an example, step 922 includes establishing JSC DNN communications with the first device and configuring the JSC DNN structure of the second device as described with reference to operations 422 and 423 of FIG.4, and operations 722 and 723 of FIG.7a with respect to a bi-static JSC system type. In another example, step 922 includes establishing JSC DNN communications with the first device and configuring the JSC DNN structure of the second device as described with reference to operations 522 and 523 of FIG.5, and operations 722, and 723 of FIG.7b with respect to a multi-static JSC system type. [00164] For each time slot of the one or more time slots, the JSC DNN process 920 performs steps 924 to 936 based on: [00165] In step 924, receiving a JSC signal waveform transmitted from the first device in said each time slot over a communication channel, the JSC signal waveform representing input communication data and a radar signal. For example, step 924 includes receiving the JSC signal waveform at the second device as described with reference to any of operations 424, 524, 724a-724d with reference to FIGs.4, 5, 7a and 7b. [00166] In step 926, receiving one or more further JSC signal waveforms for said each time slot based on reflections of the JSC signal waveform transmission from one or more objects in the communication channel. For example, for a bi-static or multi- static JSC system types, in step 926 the one or more further JSC signal waveforms include reflections of the JSC signal waveform from the one or more objects as described with reference to any of operations 426, 526, and 726a-1 to 726a-N and 726d- 1 to 726d-N with reference to any of FIGs.4, 5, and FIGs.7a and 7b, respectively. [00167] In step 928, processing the received JSC signal waveform and, if any, further JSC signal waveforms (e.g., reflections) with a JSC DNN structure at the second device to generate reconstructed communication data for said each time slot corresponding to the input communication data transmitted in said each time slot and to generate radar sensing feedback (RFB) information associated with the one or more objects for said each time slot. For example, the JSC DNN structure at the second device includes a receiving communications and radar DNN (Rx CR-DNN) model for generating the reconstructed communication data and the RFB information (e.g., range, doppler, velocity etc.) for said each time slot. For example, for a bi-static or multi-static JSC system type, in step 928 the received JSC signal waveform and, if any, further JSC signal waveforms are input to a Rx CR-DNN model for generating reconstructed communication data and RFB information as described in any of operations 428 or 528 with reference to FIGs.4 or 5, respectively. [00168] In step 930, the JSC DNN structure further includes a transmitting (Tx) radar feedback (RFB) DNN for processing the RFB information for said each time slot using the Tx RFB DNN, the RFB DNN generates an RFB signal for transmission as a RFB signal waveform to the first device. For example, for a bi-static or multi-static JSC system type, in step 930 the RFB information is input to an Tx RFB DNN model for generating a RFB signal as described in any of operations 430, 530 or 730a-730d with reference to FIGs.4, 5 or 7a and 7b, respectively. [00169] In step 932, transmitting the RFB signal to the first device as a RFB signal waveform for use by first device in generating radar sensing information for the one or more objects. For example, for a bi-static or multi-static JSC system type, in step 932 the second device transmits the RFB signal as described in any of operations 432, 532 or 732a-732d with reference to FIGs.4, 5 or 7a and 7b, respectively. For example, the RFB signal waveform for each time slot is received in step 810 of JSC DNN process 800 as a received radar sensing signal and processed in step 812 for assisting in generating radar sensing information corresponding to the one or more objects. In some examples, conventionally transmitting the generated RFB signal data to the first device in one or more further time slots over a communications data channel or a communications control channel. [00170] In step 934, the JSC DNN structure (e.g., JSC DNN structure 214 of FIG. 2) of the second device sends the reconstructed communication data for said each time slot to a data sink or to one or more upper protocol layers of a protocol stack of the second device. The reconstructed communication data for each time slot corresponding to the input communication data transmitted in each corresponding time slot. In an example, the second device sends the reconstructed communication data to one or more upper protocol layers of a protocol stack of the second device. For example, to an application protocol layer of the protocol stack for use by one or more applications executing on the second device. [00171] In step 936, determining whether to continue JSC DNN operations. If the second device is to receive further JSC signal waveforms in corresponding further time slots of the one or more time slots (e.g., ‘Y’), then proceed to step 924 for receiving said further JSC signal waveforms transmitted by the first device in the further time slot of the one or more time slots. If there are no further time slots for performing JSC DNN operations (e.g., ‘N’), then proceed to step 938. Additionally or alternatively, determining whether to continue JSC DNN operations includes, for example, receiving a control message from the first device indicating disabling of JSC DNN operations, if such a control message is received (e.g., ‘N’), then proceed to step 938, otherwise proceed to step 924. [00172] In step 938, disabling JSC DNN operations for the communication session with the first device. For example, see operations 418 and 438 of FIG.4, operations 518 and 538 of FIG.5, and operations 718 and 738 of any of FIGs.7a and 7b. [00173] FIG.9b is a flow diagram illustrating another example JSC DNN process 940 for a second device performing mono-static JSC DNN operations in one or more time slots of a communication session with a first device. The mono-static JSC DNN operations are based on the mono-static JSC DNN operations as described with reference to FIGs.3c, 6, and 7c. The example JSC DNN process 940 includes the following steps of: [00174] In step 922, establishing JSC DNN operations for one or more time slots of a communication session with a first device. In an example, step 922 includes establishing JSC DNN communications with the first device and configuring the JSC DNN structure of the second device as described with reference to operations 622 and 623 of FIG.6, and operations 722, and 723 of FIG.7c with respect to a mono-static JSC system type. [00175] For each time slot of the one or more time slots, the JSC DNN process 940 performs steps 924/926, 942 to 948 based on: [00176] In step 924/926, receiving a JSC signal waveform transmitted from the first device in said each time slot over a communication channel, the JSC signal waveform representing input communication data and a radar signal and transmitted in said each time slot. In some examples, step 924/926 includes receiving one or more further JSC signal waveforms for said each time slot based on reflections of the JSC signal waveform transmission from one or more objects. In another example, step 924/926 includes receiving the JSC signal waveform at the second device as described with reference to any of operations 624 and operations 724a-724d with reference to FIGs.6 and 7c. [00177] In step 942, processing the received JSC signal waveform and/or any further JSC signal waveforms associated with reflections from one or more objects with a communication DNN structure at the second device. The communication DNN structure configured for generating reconstructed communication data for each time slot corresponding to the input communication data transmitted in each corresponding time slot from the first device. For example, for a mono-static JSC system type, in step 942 the received JSC signal waveform and/or any further JSC signal waveforms associated with reflections from one or more objects are input to an Rx C-DNN model for processing and generating reconstructed communication data as described in operation 642 with reference to FIG.6. [00178] In step 944, sending the reconstructed communication data for each time slot to a data sink. In an example, the second device sends the reconstructed communication data to one or more upper protocol layers of a protocol stack of the second device. For example, to an application protocol layer of the protocol stack for use by one or more applications executing on the second device. [00179] In step 946, the second device determines whether to continue JSC DNN operations. If there are further time slots assigned for performing JSC DNN operations (e.g., ‘Y’), then proceed to step 924/926 for receiving said further JSC signal waveforms in the further time slots. If there are no further time slots for performing JSC DNN operations (e.g., ‘N’), then proceed to step 948. In another example, if the second device receives a control message from the first device indicating disabling of JSC DNN operations (e.g., ‘N’), then proceed to step 948, otherwise proceed to step 924/926. [00180] In step 948, disabling JSC DNN operations for the communication session with the first device. For example, see operations 618 and 638 of FIG.6 and operations 718 and 738 of FIG.7c. [00181] FIG.9c is a flow diagram illustrating an example JSC DNN establishment process of step 922 of FIG.9a or 9b performed by a second device when establishing JSC DNN operations for one or more time slots of the communication session with the first device. For example, step 922 of JSC DNN processes 920 or 940 performs the JSC DNN establishment process of step 922. The JSC DNN establishment process of step 922 includes the following steps of: [00182] In step 922a, receiving one or more control messages from the first device indicating a selected type of JSC DNN structure for use by the second device in processing the received JSC signal waveforms transmitted from the first device. For example, the one or more control messages may be an RRC establishment request message or equivalents thereof as described in operation 702 with reference to any of FIGs.7a to 7c. [00183] In step 923, configuring the JSC DNN structure of the second device based on the selected type of JSC DNN structure for use in the one or more time slots. The JSC DNN structure configured for generating communication data and/or radar feedback data from a received JSC signal waveform transmitted by the first device. For example, second device configures the JSC DNN structure as described in operations 403, 503, or 703 with reference to any of FIGs.4 to 6 and 7a to 7c. [00184] The JSC DNN establishment process of step 922 proceeds to step 924 of JSC DNN process 920 or step 924/926 of JSC DNN process 940. [00185] FIGs.10a illustrates an example master neural network table 1000 and FIGs 10b, 10c, and 10d illustrate example first, second and third neural network tables 1010, 1020, and 1030 provided for and used by a first device, a second device, and, optionally, a third device, respectively. The first, second and third neural network tables 1010, 1020 and 1030 are sub-tables of the master neural network table 1000. The first device, second device and third device use the neural network tables 1010, 1020 and 1030, respectively, to, for example, select corresponding JSC DNN structures for JSC DNN operations in one or more time slots of a communication session between the first device, the second device and, optionally, the third device as described with reference to FIGs.1 to 9c. [00186] Referring to FIG.10a, the master neural network table 1000 includes a JSC DNN identifier column, First Device JSC DNN structure column, a Second/Third Device JSC DNN Structure column, a Communications Channel Condition/ Characteristic column, and a JSC system type column (e.g., bi-static JSC, multi-static JSC, mono-static JSC etc.). The JSC DNN identifier column holds in each row a specific JSC DNN identifier (e.g., JSC DNN ID=1, 2, …, 10, …, 15a, 15b, … etc.). The First Device JSC DNN Structure column holds in each row the configuration data or trained model weights for the trained DNN models of the First Device JSC DNN structure. The Second/Third Device JSC DNN Structure column holds in each row the configuration data or trained model weights for the trained DNN models of the Second/Third Device JSC DNN structure. The DNN models held in each row of the First Device JSC DNN Structure column and the Second/Third Device JSC DNN Structure column form a pair of JSC DNN structures that have been jointly trained based on the communications condition/characteristic and radar configuration as indicated in the same row of the Communications Channel Condition/Characteristic and JSC system type columns. The Communications Channel Condition/Characteristic holds in each row an indicator of the communication channel condition/characteristic (e.g., LOS, NLOS, WEATHER CONDITION 1/NLOS, etc.) that the trained DNN models in the First Device JSC DNN Structure and the Second/Third Device JSC DNN Structure columns of the same row were jointly trained for. The JSC system type column holds in each row the type of JSC system that the corresponding trained DNN models in the First Device JSC DNN Structure and the Second/Third Device JSC DNN Structure columns of the same row were jointly trained for such as, for example, bi- static JSC, multi-static JSC, mono-static JSC, and/or any combination, modification and/or variation thereof. [00187] Each row of the master neural network table 1000 is populated with a specific JSC DNN identifier for a pair of JSC DNN structures for the first device and the second/third device, the pair of JSC DNN structures for the first device and second/third device having been trained for a particular communication channel condition/characteristic and a particular radar or JSC system type (e.g., bi-static JSC, multi-static JSC, or mono-static JSC). For example, the first row of the master neural network table 1000 includes a JSC DNN ID value of 1, in which the first device JSC DNN structure includes a first device JSC DNN structure includes Tx CR-DNN / Rx R- DNN models, and the second device/third device JSC DNN structure includes Rx CR- DNN/Tx RFB-DNN models, where the Tx CR-DNN / Rx R-DNN models and Rx CR- DNN/Tx RFB-DNN models have been jointly trained for the line of sight (LOS) communications condition and the Tx CR-DNN / Rx R-DNN models and Rx CR- DNN/Tx RFB-DNN models have inputs and outputs corresponding to a bi-static JSC system type. [00188] Referring to FIG.10b, the first neural network table 1010 includes a JSC DNN identifier column, First Device JSC DNN structure column, a Communications Channel Condition/ Characteristic column, and a JSC system type column (e.g., bi- static JSC, multi-static JSC, mono-static JSC etc.). The JSC DNN identifier column holds in each row a specific JSC DNN identifier (e.g., JSC DNN ID=1, 2, …, 10, …, 15a, 15b, … etc.). The First Device JSC DNN Structure column holds in each row (e.g., for each JSC DNN Identifier) the configuration data or trained model weights for the trained DNN models of the First Device JSC DNN structure, where corresponding rows (e.g., same JSC DNN Identifiers) of the Second/Third Device JSC DNN Structure column of the second and third neural network tables 1020 and 1030 holds the configuration data or trained model weights for the trained DNN models of the Second/Third Device JSC DNN structure. The DNN models/weights etc. held in each row of the First Device JSC DNN Structure column and the Second/Third Device JSC DNN Structure column for the same JSC DNN Identifier form a pair of JSC DNN structures that have been jointly trained based on the communications condition/characteristic and radar configuration as indicated in the same row (same JSC DNN Identifier) of the Communications Channel Condition/Characteristic and JSC system type columns. The Communications Channel Condition/Characteristic holds in each row an indicator of the communication channel condition/characteristic (e.g., LOS, NLOS, WEATHER CONDITION 1 / NLOS, etc.) that the trained DNN models in the First Device JSC DNN Structure of the first neural network table 1010 and the Second/Third Device JSC DNN Structure columns of the second/third neural network tables 1020/1030 of the same row were jointly trained for. The JSC system type column holds in each row the type of JSC system that the corresponding trained DNN models in the First Device JSC DNN Structure of the first neural network table 1010 and the Second/Third Device JSC DNN Structure columns of the second/third neural network tables 1020/1030of the same row were jointly trained for such as, for example, bi-static JSC, multi-static JSC, mono-static JSC, and/or any combination, modification and/or variation thereof. [00189] As an option, the first neural network table 1010 may include Second/Third Device JSC DNN Structure columns of the master neural network table 1000, which can enable the first device, when accessing the first neural network table 1010 to determine a suitable pair of JSC DNN structures based on the capabilities of both the first and second devices. For example, the first device can determine the complexity of the Second Device JSC DNN structure and determine whether the capabilities of the second device mean the second device can configure and operate the Second Device JSC DNN structure of a pair of JSC DNN structures. If not, then the first device can identify another pair of JSC DNN structures in which the second device is capable of configuring and operating the corresponding Second Device JSC DNN structure. The first device then notifies, using a control message, the second device the JSC DNN Identifier of the selected pair of JSC DNN structures. [00190] As an option and in some examples, the master neural network table 1000 is provided for and/or accessible by the first device or the second device for use in selecting a pair of JSC DNN structures based on communications channel conditions and JSC system type. This provides an advantage of the first device selecting a suitable pair of JSC DNN structures in which the second device is capable of executing and/or operating the Second Device JSC DNN Structure of the selected pair of JSC DNN structures. For example, the first device receives capability information or processing capacity information associated with the second device and uses this information to select a pair of JSC DNN structures in which the second device is more likely capable of generating reconstructed data and/or RFB signal waveforms using the Second Device JSC DNN structure of the selected pair of JSC DNN structures. Alternatively, the first device is provided with a first neural network table 1010 that is a subset of the master neural network table 1000 as illustrated in FIG.10b. Alternatively, the second device is provided with a neural network table 1020 that is a subset of the master neural network table 1000 as illustrated in FIG.10c. [00191] Referring to FIG.10b, the first neural network table 1010 is provided for and used by the first device is a subset of the master neural network table 1000 and only includes the JSC DNN ID column, the corresponding First Device JSC DNN structure column, and the Communications Channel Condition/Characteristic and JSC system type columns. [00192] Referring to FIG.10c, the second neural network table 1020 is provided for and used by the second device is a subset of the master neural network table 1000 and only includes the JSC DNN ID column and the corresponding Second Device JSC DNN structure from the Second/Third Device JSC DNN structure column. [00193] Referring to FIG.10d, the third neural network table 1030 is provided for and used by the third device is also a subset of the master neural network table 1000 and only includes the JSC DNN ID column and the corresponding Third Device JSC DNN structure from the Second/Third Device JSC DNN Structure column. [00194] Referring to FIGs.10c and 10d, in this example, the second neural network table 1020 of the second device includes JSC DNN IDs 1, 2, 10 and 15a apart from those JSC DNN IDs allocated to the third device such as, for example, JSC DNN ID 15b, which is allocated to the third device JSC DNN structure as seen in the third neural network table 1030. Similarly, the third neural network table 1030 of the third device includes JSC DNN IDs apart from those assigned to the second device. [00195] In some examples, the second neural network table 1020 includes all JSC DNN IDs assigned to the second device and also the corresponding columns of the master neural network table 1000. This allows the second device to identify an JSC DNN ID and select a first JSC DNN structure for a particular channel condition/characteristic and radar configuration for establishing a JSC DNN operations in a communication session with the first device. The first device is assigned the second JSC DNN structure and so is supporting the radar sensing the second device is performing. The first, second and/or third devices use the first, second and third neural network tables 1010, 1020 and/or 1030 for selecting and configuring their corresponding JSC DNN structures in relation to the JSC system type and/or radar/channel conditions or performance requirements (e.g., JSC performance requirements) as described with reference to any of FIGs.1 to 9b. [00196] In operation, the first device selects a pair of JSC DNN structures from the first neural network table 1010, which has a plurality of first JSC DNN structures in the First Device JSC DNN structure column for selection and use by the first device. As can be seen in master neural network table 1000, each of the first plurality of JSC DNN structures correspond to a pair of JSC DNN structures including a first JSC DNN structure for use by the first device and a second JSC DNN structure for use by the second device. Each pair of JSC DNN structures map to a JSC DNN identifier. The first neural network table 1010 includes, for each first device JSC DNN structure, a communication channel condition that the JSC DNN structure is trained for and a JSC system type the JSC DNN structures is associated with. After selecting the JSC DNN structure based on the channel condition and JSC system type, the first device during establishment of the JSC DNN operations in the communication session sends one or more control messages specifying the JSC DNN identifier of the selected pair of JSC DNN structures to the second device. The second device uses the received JSC DNN identifier with the second neural network table 1020 of FIG.10c to identify the JSC DNN structure for use at the second device for receiving JSC DNN communications from the first device. In this case, the second device is supporting the first device in performing radar sensing. [00197] In general, although the first, second and third neural network tables 1010, 1020, and 1030 are provided for and used by the first, second and third devices, respectively, such neural network tables may be stored in any configuration or combination as is appropriate. For example, the master neural network table 1000 may be stored centrally incorporating the information from each of the first, second and third neural network tables 1010, 1020, and 1030 as described and the corresponding elements accessed by the first, second and/or third devices accordingly. Conversely, each of the neural network tables 1000, 1010, 1020, and/or 1030 may be stored in a distributed manner across a number of physical devices and accessible by the first, second and/or third devices as the application demands. In other examples, the master neural network table 1000 or first neural network table 1010 is stored at the first device, the second neural network table 1020 is stored at the second device, and the third neural network table 1030 is stored at the third device. In other examples, the master neural network table 1000 is stored at the second device. In other examples, the master neural network table 1000 is stored at a network device or entity of the communication system where the first and/or second devices have restricted access to various columns/rows of the master neural network table 1000 depending on authorization and/or permissions granted. [00198] FIG.10e illustrates a flow diagram of an example JSC DNN structure selection process 1050 for a first device during establishment of JSC DNN operations with a second device. In this case, the first device accesses either the master neural network table 1000 of FIG.10a or the first neural network table 1010 of FIG.10b and the second device accesses the second neural network table 1020 of FIG.10c. For example, the first neural network table 1010 includes, for each pair of JSC DNN structures with a JSC DNN identifier, an associated communication channel condition and an associated JSC system type as illustrated in the JSC DNN identifier column, First Device JSC DNN structure column, a Communications Channel Condition/ Characteristic column, and a JSC system type column of FIG.10b. In another example, JSC DNN structure selection process 1050 may be performed by the second device with the second neural network table 1020 including, for each pair of JSC DNN structures with a JSC DNN identifier, an associated communication channel condition and an associated JSC system type as illustrated in the JSC DNN identifier column, Second Device JSC DNN structure column, a Communications Channel Condition/ Characteristic column, and a JSC system type column of FIG.10c. The JSC DNN structure selection process 1050 for the first device includes the steps of: [00199] In step 1051, identifying a communication channel condition of the communication channel between the first device and second device. [00200] In step 1052, selecting a JSC system type for use in generating radar sensing information. The JSC system type includes at least one of: bi-static, multi- static or mono-static JSC system types (or radar arrangements) as described with reference to FIGs.1, 2 and/or 3a to 3d and/or as herein described. [00201] In step 1053, selecting the JSC DNN identifier from the master neural network table 1000 or first neural network table 1010 based on the identified communication channel condition and selected JSC system type. In an example, the first device also selects the JSC DNN identifier based on capabilities or processing capabilities of the second device as described with reference to FIGs 10b. As an option, JSC DNN structure selection process 1050 may be performed by the second device (e.g., when performing uplink JSC), which may select in step 1053 the JSC DNN identifier from the second neural network table 1020. [00202] The JSC DNN structure selection process 1050, after step 1053, proceeds to perform JSC DNN process 800 or process 922/942 for establishing JSC DNN operations in a communication session between the first device and second device. [00203] As an option, when the second device establishes JSC DNN operations with the first device, the second device performs the JSC DNN structure selection process 1050, where the second neural network table 1020 is used in place of the first neural network table 1010 in steps 1051 to 1053. [00204] FIG.11 illustrates a flow diagram of an example joint training process 1100 for training a first and second JSC DNN structure (a pair of JSC DNN structures) for use, after training, by a first and second device, respectively. The joint training process 1100 stores the corresponding trained first and second JSC DNN structures of the trained pair of JSC DNN structure in, for example, the master neural network table 1000, first neural network table 1010 and/or second neural network table 1020 as described with reference to FIGs.10a to 10e. The first and second JSC DNN structures of each pair of JSC DNN structures includes one or more transmitting and/or receiving DNN models that are jointly trained for a particular JSC system type (e.g., bi-static JSC, multi-static JSC, and mono-static JSC), JSC performance requirements including one or more communication and/or radar sensing channel characteristics. Each of the transmitting and/or receiving DNN models include any type of ML model algorithm/architecture. The joint training process 1100 jointly trains the transmitting and/or receiving DNN models of the first and second JSC DNN structures using supervised or unsupervised learning. The ML model algorithm/architecture can be based on or include, by way of example only but is not limited to, one or more of: a neural network, a fully connected neural network, a convolutional neural network, a long short-term memory (LSTM) neural network, and a transformer neural network, and/or any other suitable DNN architecture, combinations thereof, modifications thereto, as herein described, and/or as the application demands. For example, supervised joint training of the first and second JSC DNN structures of the pair of JSC DNN structures for the first and second devices uses gradient back-propagation based techniques for updating the weights / parameters of the corresponding pair of JSC DNN structures using, for example, a joint radar and communication loss function. It is assumed that the DNN model architecture and the DNN model arrangements as described with reference to FIGs.3a to 3d for the first and second JSC DNN structures of the pair of JSC DNN structures for the first and second devices have been determined. The joint training process 1100 includes the following steps. [00205] In step 1160, retrieving a training dataset such as a batch of training data including a plurality of training data instances, each training data instance including known sensing data (e.g., known range, doppler, velocity, acceleration, positions, and/or any other known radar measurement data) associated with one or more objects in an environment including a communication / radar channel. The communication / radar channel is associated with selected JSC performance requirements such as, for example, a particular communication and/or radar channel condition / characteristic (e.g., one or more of NLOS communications, LOS communications, multi-path interference, multiple access interference, and narrowband interference, weather or atmospheric conditions affecting the communication/radar channel, different target sizes and types, other radar and/or communication signal interference as described with reference to FIGs.1 to 9b). The selected JSC performance requirements can further include communication performance requirements (e.g., communication performance requirements of a particular communication standard (e.g., 5G, 6G etc.)) for the communication channel/link between the first and second device, radar performance requirements for the first device in detecting objects and the like. [00206] In step 1162, processing, for each training data instance in the batch, a set of input communications data with the first JSC DNN structure, where the set of input communications data may be randomly generated and/or included in the retrieved batch of training data. Each training data instance may represent a time slot. The first JSC DNN structure connects with the second JSC DNN structure via a communications channel model (or a communications channel). The first JSC DNN structure processes the set of input communications data to generate an output JSC signal for the time slot, which is RF processed and passed through the communications channel model (e.g., simulating RF processing such as ADC and frequency up conversion, simulating transmission over the communications channel with the one or more objects based on the known sensing data for the training data instance and selected JSC performance requirements and/or channel conditions etc.) as a JSC signal waveform to the second JSC DNN structure along with reflections of the JSC signal waveform from the one or more objects based on the known sensing data for the training data instance. The output JSC signal is a digital representation of the set of input communications data transformed by the first JSC DNN structure to include a radar signal and/or radar signal characteristics for JSC. The JSC signal waveform is an RF or analog representation of the output JSC signal. The communications channel model includes one or more objects arranged as represented in the retrieved training data instance. The joint training process 1100 configures the first JSC DNN structure for generating the output JSC signal given the set of input communications data, and transmitting the output JSC signal representing the set of input communications data and radar signal/characteristics as a JSC signal waveform via the communications channel model (or communication channel) for receiving at the second JSC DNN structure. At the same time, the training process 110 configures the first JSC DNN structure for generating radar sensing information or data (e.g., range, doppler, velocity, etc.) for one or more objects. The communications channel model models a communications channel based on the communication channel condition/characteristics and the corresponding one or more objects. [00207] The second device receives the JSC signal waveform from the communication channel model (e.g., simulating RF processing such as DAC and frequency down conversion to baseband etc.) as JSC signals, which are input to the second JSC DNN structure for generating a set of reconstructed communications data corresponding to the set of input communications data and/or, depending on the JSC system type the pair of JSC DNN structures is being trained for (e.g. bi-static or multi- static), radar sensing feedback (RFB) signal waveforms based on RFB information associated with the one or more objects and reflections of the transmitted JSC signal waveform received over the communications channel model at the second device. The bi-static and multi-static JSC systems use the RFB signal waveforms as described with reference to FIGs.3a, 3b and 3d and/or as herein described. For the bi-static and multi-static JSC system types, the second JSC DNN structure provides the RFB information (e.g., doppler, velocity, position / location etc.) and the second device sends as a RFB signal waveform via the communication channel model (e.g., simulating RF processing such as ADC and frequency up conversion, simulating transmission over the communications channel etc.) to the first device, which processes the RFB signal waveform (e.g., simulating RF processing such as DAC and frequency down conversion to baseband etc.) to generate first radar sensing signal(s) representing the RFB signal waveform and RFB information therein, which is input to the first JSC DNN structure for generating radar sensing information (e.g., doppler, velocity, position / location etc.) for the one or more objects that are detected. For multi-static JSC system type, the communication channel model also processes the JSC signal waveform for the time slot and provides the first JSC DNN structure with further radar sensing signals derived from reflections of the JSC signal waveform in relation to the one or more objects being modelled in the communications channel model and received at the first device. These further radar sensing signals are input along with the first radar sensing signal(s) to the first JSC DNN structure for generating radar sensing information for the one or more objects that are detected. For mono-static JSC system type, the communication channel model processes the JSC signal waveform for the time slot and provides the first JSC DNN structure with radar sensing signals using only the reflections of the JSC signal waveform in relation to the one or more objects being modelled in the communications channel model and received at the first device. These radar sensing signals are processed by the first JSC DNN structure to generate radar sensing information for the one or more objects detected. [00208] The radar sensing information (e.g., radar object range, doppler, velocity etc.) for the time slot (i.e. for each training data instance) in relation to the one or more objects detected represents one or more from the group of: a range estimate for each of the objects that are detected, doppler estimate for each of the objects that are detected, a velocity estimate for each of the objects that are detected, a position or location estimate for each of the objects that are detected, a delay spread for each of the objects that are detected, a doppler spread for each of the objects that are detected, an average delay for each of the objects that are detected, an angular estimate for each of the objects that are detected, an azimuth and/or elevation estimate for each of the objects that are detected, or other radar measurement and the like in relation to the one or more objects for the time slot (i.e. for the training data instance). Step 1162 is performed for each of the training data instances in the batch of the training dataset. [00209] Generating the radar sensing information (or radar measurement) is performed by the first JSC DNN structure, where the DNN models of the first JSC DNN structure learn what type of radar measurement (e.g., correlation) is the optimal given the JSC signal waveform output and reflected from said one or more objects. That is, the DNN architecture of the first JSC DNN structure converges during training to the optimal scheme and/or JSC signal waveform for performing JSC for a particular selected JSC system type and set of JSC requirements. The DNN models of the second JSC DNN structure learn how to separate out the received radar/reflections from the data signal of the transmitted JSC signal waveform and generate reconstructed communication data corresponding to the input communication data incorporated in the JSC signal waveform. Constraints may be placed on the communication data signal and radar signal within the JSC signal waveform in that they are separated out to use different resources (e.g., time/frequency/space/code etc.). For example, DNN model training of the first and second JSC DNN structures can converge to a solution where radar sensing might use different resource (time/frequency/space/code). For example, a JSC signal waveform may result during training or with constraints such that the input communication data and radar signal of the JSC signal waveform occupies different communication resources (e.g., time/frequency/ space/code). In such a case, the first device can filter out, from the received JSC signal waveforms, the transmitted input communication data so that only the radar-reflections of the JSC signal waveform are included in the radar sensing signals processed by the DNN models of the first JSC DNN structure. For multiple second devices (e.g., multiple UEs) in communication with the first device (e.g., a BS), the first device can support multiple second devices using different communication resources (e.g., different time slots or frequencies/RBs/ bandwidth parts (BWPs)). Alternatively or additionally, the joint training of the first and second JSC DNN structures may converge to generate output JSC signal that, when transmitted, forms an optimal JSC signal waveform with characteristics incorporating the input communication data and a radar signal/radar signal characteristics for JSC, but constrained within the JSC performance requirements (e.g., communication and/or radar performance requirements and/or performance requirements of one or more communication standards (e.g., 4G, 5G, and/or 6G and the like)). [00210] In step 1164, after all training data instances in the batch have been processed, calculating a joint loss function using performance metrics associated with the radar sensing information generated by the first JSC DNN structure and associated with the reconstructed set of transmitted communications data for the batch generated by the second JSC DNN structure. For example, the joint loss function is a combination of performance metrics associated with the radar sensing information generated by the first JSC DNN structure of the first device and the performance metrics associated with the reconstructed communications data generated by the second JSC DNN structure of the second device for the batch. For example, the performance metric associated with the generated radar sensing information is a radar sensing error rate (or sensing error) derived from the radar sensing information generated by the first JSC DNN structure and corresponding known radar sensor data of the corresponding training data instances in the batch. For example, the radar sensing error rate for the batch is calculated from a mean squared error of the radar sensing information and known radar sensing data for the corresponding training data instances of the batch. Other error metrics such as, for example, the square error, least squared error, and/or any other error metric that is suitable as the application demands. The performance metric associated with the reconstructed communications data is a block or bit error rate (or block error, bit error, signal-to-interference plus noise ratio (SINR) or other communication performance metric) derived from the reconstructed communications data generated by the second JSC DNN structure and the corresponding set of input communication data associated with the training data instances for the batch. A joint radar and communication loss function uses the combined performance metrics (e.g., the radar sensing error rate and block or bit error rate) in updating the weights of the DNN model arrangements within the first and second JSC DNN structures. For example, the joint radar and communication loss function is a weighted sum of the radar sensing error rate (or sensing error) and block error rate (or block error). For example, gradient back-propagation based techniques using the joint radar and communication loss function can be used to update the weights of the DNN models of the first and second DNN structures. [00211] In step 1166, determining whether the JSC DNN structures of the first and second devices are validly trained. For example, the joint radar and communication loss function is used to determine whether the joint loss is below a predetermined joint loss threshold, and if so, then the JSC DNN structures may be considered validly trained, otherwise perform further training of the JSC DNN structures. If the JSC DNN structures of the first and second devices are validly trained (e.g., ‘Y’), then proceed to step 1170, otherwise (e.g., ‘N’) proceed to step 1168 for updating the model parameters and/or weights. [00212] In step 1168, updating model parameters and/or weights of the DNN models in the first and second JSC DNN structures of the pair of JSC DNN structures of the first and second devices based on the joint radar and communication loss function. For example, in supervised joint training of the first and second JSC DNN structures of the pair of JSC DNN structures for the first and second devices, the weights of the corresponding DNN models of the first and second JSC DNN structures are updated using gradient back-propagation based techniques using the joint radar and communication loss function. After updating, proceed to step 1160 for retrieving another batch of training data instances. The first and second JSC DNN structures are jointly trained such that the first JSC DNN structure is trained to process or transform the input communication data into a digital output JSC signal with radar signal capabilities or radar signal characteristics whilst incorporating the input communication data such that when the output JSC signal is transmitted (e.g. ADC and RF processed and frequency up-converted) as a JSC signal waveform, the JSC signal waveform is an optimal waveform and/or a suitable waveform that is capable, for the channel conditions (e.g., based on selected JSC requirements), for use in performing both: a) radar sensing using the reflections of the JSC signal waveform from the one or more objects, and/or RFB signal waveform from the second JSC DNN structure; and b) communications with second device in which the second JSC DNN structure is trained and optimised to generate reconstructed communication data corresponding to the input communication data incorporated into the transmitted JSC signal waveform. [00213] In step 1170, after training, the joint training process 1100 stores the pair of trained JSC DNN structures (e.g., sends to the master neural network table 1000 of FIG.10a) selection and use in the methods, JSC DNN operations, processes, systems and/or apparatus as described with reference to FIGs.1 to 10e. For example, the trained first and second JSC DNN structures form a pair of trained JSC DNN structures for first and second devices, and, along with the channel / radar conditions and/or characteristics, and JSC system type that the pair of trained JSC DNN structures have been trained for, are assigned a JSC identifier (or index) and stored in master neural network table 1000 as described with reference to FIG.10a. For example, the pair of trained JSC DNN structures of the first and second device are assigned a JSC identifier (or index) and stored in a row of the master neural network table 1000. A network device or the first device populates the first neural network table 1010 with the trained first JSC DNN structure, along with channel conditions and JSC system type and JSC identifier as described with reference to FIGs.10a and 10b. The network device or the second device populates the second neural network table 1020 with the trained second JSC DNN structure, along with channel conditions and JSC system type and JSC identifier as described with reference to FIGs.10a and 10c. [00214] In the above example, as described with reference to FIGS.1 to 3d and FIGs.10a to 10e, the first JSC DNN structure includes a transmitting JSC DNN model and a receiving radar sensing DNN model. After training and storing in the master neural network table 1000 of FIG.10a, the transmitting JSC DNN model is configurable for processing the input communication data for the time slot and generating an output JSC signal for transmission as a JSC signal waveform in the time slot. The output JSC signal and the JSC signal waveform represent the input communication data for the time slot and a radar signal for the time slot. The receiving radar sensing DNN model is configurable for processing the received one or more radar sensing signals and generating predicted radar sensing information (e.g., range, doppler, velocity, position, acceleration, and/or any other parameter) for the one or more objects. [00215] As described with reference to, for example, FIGs.3a to 3d, the transmitting JSC DNN model of the first JSC DNN structure outputs a feed forward signal to the receiving radar sensing DNN (Rx R-DNN) model of the first JSC DNN structure for each time slot. The first device configures the receiving radar sensing DNN model for processing the feed forward signal for the time slot and the received one or more radar sensing signals for the time slot and generates radar sensing information for the one or more objects. In an example, the feed forward signal includes one or more from the group of: input communication data for the time slot; one or more hidden layer outputs of the transmitting JSC DNN model for the time slot; one or more hidden layer inputs of the transmitting JSC DNN model for the time slot; and the output JSC signal for the time slot. [00216] In further examples, as described with reference to FIGs.2, 3a, 3b and 3d, and 7a to 7b, the second JSC DNN structure of the second device includes a receiving communication and radar (CR) DNN (Rx CR-DNN) model and a transmitting radar feedback (RFB) DNN (Tx RFB-DNN) model. The second device configures the Rx CR-DNN model for processing data representative of one or more received JSC signal waveforms for the time slot and generating reconstructed communication data associated with the input communication data for the time slot and generating radar sensing feedback (RFB) information for the time slot. The second device configures the Tx RFB DNN model for processing the generated RFB information and generating a RFB signal for transmission as a RFB signal waveform over the communication channel to the first device as described, for example, with reference to FIGs.2, 3a, 3b and 3d and 7a to 7b. [00217] The joint training process 1100 may be applied to any of the JSC DNN structures 103 and 114, 203 and 214, and 303 and 314 described with reference to FIGs. 1 to 10e, combinations thereof, modifications thereto, as herein described and/or as the application demands. In an example, the joint training process 1100 uses a simulated communication channel model that randomly generates the input communication data for each training data instance in each batch of training data instances. In another example the joint training process 1100 uses online real communication channels in which the input communication data is real communication data for transmission over the real communication channel from first device to the second device, where the radar sensing information associated with the objects is known or the objects being sensed are known objects with known range, doppler, velocity, acceleration, position and/or other known radar measurements and the like. Known techniques for estimating the objects range, doppler, velocity and other known radar measurements for training the first JSC DNN structure of the first device and the second JSC DNN structure of the second device. In another example, the joint training process 1100 uses federated learning techniques and/or any other real-time machine learning techniques for training the first and second JSC DNN structures in real-time on first and second devices and the like. [00218] As discussed with reference to FIGs.1 to 11, the JSC signal waveform includes both radar sensing signal and input communication data for transmission together. A JSC signal waveform that incorporates input communication data and radar signal characteristics / or a radar signal is learnt during joint training of the DNN models in the first and second JSC DNN structures. The DNN models learn or converge to an optimal JSC signal waveform given a particular JSC system type and set of JSC performance requirements, and/or other constraints, by using the feedback of a joint cost function or joint loss function that the first and second JSC DNN structures optimize. Th joint cost or loss function can be the sum of radar sensing error (e.g., doppler error, range error, velocity error, etc.) and input communication data error (e.g., block error rate (BLER)). The radar sensing error is based on an error metric such as, for example, the mean square error of the radar sensing error (e.g., range error, doppler error, etc.) Gradient back-propagation or other DNN update technique is applied to update the weights/parameters of the DNN models of the first and second JSC DNN structures with the aim to minimise the joint cost function / joint loss function. Once converged, the result is a trained first JSC DNN structure that generates an output JSC signal and subsequent transmitted JSC signal waveform that is optimal for JSC for the JSC system type and set of JSC performance requirements used when training the first and second JSC DNN structures, where the first JSC DNN structure generates radar sensing information based on reflections of the JSC signal waveform from one or more objects, and a trained second JSC DNN structure that extracts and generates reconstructed communication data from the JSC signal waveform, and/or a RFB feedback waveform for assisting the first JSC DNN structure generate radar sensing information. [00219] FIG.12 shows a non-transitory media 1200 (or computer program product) according to some embodiments. The non-transitory media 1200 may include a computer-readable storage medium 1202 (or computer-readable medium) and/or input/output mechanism 1204 for enabling a computing system to access said computer-readable storage medium 1202. Although in this example the non-transitory media is a universal serial bus (USB) stick, this is by way of example only and it is not so limited, the skilled person would appreciate the non-transitory media 1200 may be any other type of computer readable media or medium such as, for example, a compact disc, a digital video disc, a USB stick, a blue ray disk, flash drive etc. and/or any other computer readable media as the application demands. The non-transitory media 1200 stores computer program code or instructions that, when executed by a processor of an apparatus, causes the apparatus to perform one or more of the methods, operations, processes of any preceding process, for example, as disclosed in relation to the flow diagrams and schematic diagrams of FIGs.1 to 11 and related features thereof. [00220] Implementations of the methods or processes described herein may be realized as in digital electronic circuitry, integrated circuitry, specially designed ASICs (application specific integrated circuits), computer hardware, firmware, software, and/or combinations thereof. These may include computer program products (such as software stored on e.g., magnetic discs, optical disks, memory, Programmable Logic Devices) comprising computer readable instructions stored thereon that, when executed by a processor, causes the processor to perform one or more of the methods described herein. [00221] Any system feature as described herein may also be provided as a method or process feature, and vice versa. As used herein, means plus function features may be expressed alternatively in terms of their corresponding structure. In particular, method aspects may be applied to system aspects, and vice versa. Furthermore, any, some and/or all features in one aspect can be applied to any, some and/or all features in any other aspect, in any appropriate combination. It should also be appreciated that particular combinations of the various features described and defined in any aspects of the invention can be implemented and/or supplied and/or used independently. Although several embodiments have been shown and described, it would be appreciated by those skilled in the art that changes may be made in these embodiments without departing from the principles of this disclosure, the scope of which is defined in the claims and their equivalents.

Claims

Claims 1. A method performed by a first device (101), the method comprising: establishing (802) joint sensing and communication, JSC, deep neural network, DNN, operations for one or more time slots of a communication session with a second device (112); for each time slot of the one or more time slots, the method further comprising: retrieving (804) input communication data (102) for transmission to the second device in said each time slot; processing (806) the input communication data (102) with a JSC DNN structure (103) of the first device (101) to generate an output JSC signal representing the input communication data and a radar signal; transmitting (808), to the second device (112), the output JSC signal as a JSC signal waveform (107) in said each time slot over a communication channel (108); receiving (810) one or more radar sensing signals (117, 110a-110n) for said each time slot based on reflections (110a-110n, 111a-111n) of the transmitted JSC signal waveform (107) from one or more objects (109a-109n); processing (812) the one or more received radar sensing signals by the JSC DNN structure (103) of the first device (101) to generate radar sensing information (105) of the one or more objects (109a-109n) in the communication channel; and sending (814) the radar sensing information (105) of one or more objects (109a-109n) for said each time slot to one or more upper protocol layers of a protocol stack of the first device.

2. The method of claim 1, wherein the establishing (802) JSC DNN operations comprises: sending (802a) one or more control messages to the second device (112) indicating a selected JSC DNN structure for use in each time slot; and configuring (802b) the JSC DNN structure (103) of the first device (101) with the selected JSC DNN structure for use in processing the input communication data (102) and receiving the one or more radar sensing signals (117, 110a-110n).

3. The method of claim 2, further comprising: selecting, by the first device, a pair of JSC DNN structures from a neural network table (1010) comprising a plurality of pairs of JSC DNN structures, each pair of JSC DNN structures including a first JSC DNN structure for use by the first device and a second JSC DNN structure for use by the second device, and each pair of JSC DNN structures identified by a JSC DNN identifier; wherein the one or more control messages includes the JSC DNN identifier of the selected pair of JSC DNN structures; and wherein the JSC DNN structure of the first device comprises the first JSC DNN structure of the selected pair of JSC DNN structures.

4. The method of claim 3, wherein the neural network table (1010) further includes, for each pair of JSC DNN structures, an associated communication channel condition and an associated JSC system type, the method further comprising: identifying (1051) a communication channel condition of the communication channel (108) between the first device (101) and the second device (112); selecting (1052) a JSC system type for use in generating radar sensing information; and selecting (1053) the JSC DNN identifier from the neural network table (1010) based on the identified communication channel condition and selected JSC system type.

5. The method of claim 4, wherein the communication channel condition comprises one or more from a group of: a non-line of sight communication channel condition; a line of sight communication channel condition; weather or atmospheric conditions; radar channel conditions; target characteristic; channel throughput; channel frequency or frequency band; channel bandwidth; channel delay spread; channel doppler spread; channel angular spread; or any other type of condition affecting the communication channel between the first device (101) and second device (112) for JSC.

6. The method of any of claims 4 or 5, wherein the JSC system type comprises at least one of: a mono-static JSC system type; a bi-static JSC system type; or a multi-static JSC system type.

7. The method of any preceding claim, wherein the receiving (810) one or more radar sensing signals and the processing (812) the one or more received radar sensing signals further comprising: receiving (810) one or more reflected JSC signal waveforms (110a-110n) for the time slot due to reflections of the transmitted JSC signal waveform (107) from the one or more objects (109a-109n); radio frequency, RF, to base band processing the received one or more reflected JSC signal waveforms (110a-110n) for the time slot into one or more radar sensing signals for the time slot; and the processing (812) further comprising processing the one or more radar sensing signals for the time slot, using the JSC DNN structure (103) of the first device (101) for generating radar sensing information (105) for the time slot corresponding to the one or more objects (109a-109n).

8. The method of any preceding claim, wherein the receiving (810) one or more radar sensing signals and the processing (812) the one or more received radar sensing signals further comprising: receiving (810), from the second device (112), a radar sensing feedback signal waveform (117) generated using reflections (111a-111n) of the transmitted JSC signal waveform (107) from the one or more objects (109a-109n); radio frequency, RF, to base band processing the received radar sensing feedback signal waveform (117) for the time slot into one or more further radar sensing signals for the time slot; and the processing (812) further comprising processing the one or more further radar sensing signals for the time slot using the JSC DNN structure (103) of the first device (101) for generating radar sensing information (105) for the time slot corresponding to the one or more objects (109a-109n).

9. The method of any preceding claim, the method further comprising jointly training (1100) the JSC DNN structure (103) of the first device (101) and a JSC DNN structure (114) of the second device (112) over a communication channel (108) with one or more objects (109a-109n) using a training dataset comprising known radar sensor data associated with the one or more objects (109a-109n) and corresponding sets of input communication data, wherein weights of the JSC DNN structures (103, 114) of the first and second devices (101, 112) are updated based on a joint loss function comprising a combination of performance metrics associated with the radar sensing information generated by the JSC DNN structure (103) of the first device (101) and reconstructed communications data generated by the JSC DNN structure (114) of the second device (112).

10. The method of claim 9, wherein a performance metric associated with the generated radar sensing information is a radar sensing error rate derived from the generated radar sensing information of the JSC DNN structure (103) of the first device (101) and corresponding known radar sensor data of the training dataset, and a performance metric associated with the reconstructed communications data is a block or bit error rate derived from the reconstructed communications data generated of the JSC DNN structure (114) of the second device (112) and a corresponding set of input communication data associated with the training dataset.

11. The method of claims 9 or 10, wherein the JSC DNN structure (303) of the first device (101) comprises a transmitting JSC DNN (303a) and a receiving radar sensing DNN (303b), wherein the transmitting JSC DNN (303a) is configured for processing the input communication data (302) for the time slot and generating an output JSC signal for transmission in the time slot, and the receiving radar sensing DNN (303b) is configured for processing the one or more radar sensing signals and generating predicted radar sensing information (305) for the one or more objects (309a-309n).

12. The method of claim 11, wherein the transmitting JSC DNN (303a) is configured to output a feed forward signal (303c) to the receiving radar sensing DNN (303b) for the time slot, wherein the receiving radar sensing DNN (303b) is configured for processing the feed forward signal (303c) for the time slot and the one or more radar sensing signals for the time slot and generate radar sensing information (305) for the one or more objects (309a-309n).

13. The method of claim 12, wherein the feed forward signal (303c) comprises one or more of: input communication data for the time slot; one or more hidden layer outputs of the transmitting JSC DNN (303a); or the output JSC signal for the time slot.

14. The method of any one of claims 1 to 13, wherein processing the output JSC signal of the JSC DNN structure (103) of the first device (101) for transmission using a set of resource blocks and/or frequencies of the time slot, and the method further comprising: sending a control message (802a) to the second device (112) indicating the set of resource blocks and/or frequencies of the time slot for the transmission of the output JSC signal; and the transmitting (808), to the second device (112), the output JSC signal as a JSC signal waveform (107) in the time slot further comprising transmitting the JSC signal waveform (107) in the set of resource blocks and/or frequencies.

15. The method of any one of claims 1 to 14, wherein the transmitting (808), to the second device (112), the output JSC signal as a JSC signal waveform (107) in said each time slot over a communication channel (108) further comprising: radio frequency, RF, processing the output JSC signal generated by the JSC DNN structure (103) of the first device (101) for transmission as a JSC signal waveform in said each time slot, the JSC signal waveform representing the input communication data and the radar signal; and the transmitting (808), to the second device (112), the JSC signal waveform (107) in said each time slot over the communication channel (108) with one or more objects (109a-109n).

16. The method of any one of claims 1 to 15, wherein the received radar sensing signals for each time slot comprise one or more component radar signals or radar signal characteristics associated with the reflections (110a-110n; 111a-111n) of the transmitted JSC signal waveform (107) for the time slot from the one or more objects (109a-109n).

17. The method of any one of claims 1 to 16, wherein the establishing (802) JSC DNN operations for the one or more time slots comprising transmitting a control message including a JSC DNN identifier and the one or more time slots as a downlink control indicator, DCI, message.

18. The method of claim 17, wherein the transmitting the control message comprises transmitting (802a) each control message as a Radio Resource Control, RRC, message.

19. The method of any one of claims 17 or 18, wherein the transmitting the control message comprises transmitting (802a) the control message including one or more of: a JSC DNN identifier; a first set of one or more time slots used for JSC DNN operations; a first set of resource blocks and/or frequencies for use in transmitting the JSC signal waveform (107) for each time slot of the first set of time slots; a second set of one or more time slots for use by the second device in transmitting a radar sensing feedback signal waveform (117); or a second set of resource blocks and/or frequencies for use by the second device in transmitting a radar sensing feedback signal waveform (117) for each time slot of the second set of time slots.

20. The method of any preceding claim, wherein the generating of the radar sensing information comprises generating the radar sensing information including one or more of: a range estimate of each associated object; a doppler estimate of each associated object; a velocity estimate of each associated object; a location or position estimate of each associated object; a delay spread; a doppler spread; an average delay; an angular estimate of each associated object; an azimuth and/or elevation estimate of each associated object; or any other suitable radar sensing information or parameter estimated for each associated object.

21. A method performed by a second device (112), the method comprising: establishing (922) joint sensing and communication, JSC, deep neural network, DNN, operations for one or more time slots of a communication session with a first device (101); for each time slot of the one or more time slots, the method further comprising: receiving (924) a JSC signal waveform (107) transmitted from the first device (101) in said each time slot over a communication channel, the JSC signal waveform representing input communication data and a radar signal; receiving (926) one or more further JSC signal waveforms for said each time slot based on reflections (111a-111n) of the transmitted JSC signal waveform (107) from one or more objects in the communication channel; processing (928-930) the received JSC signal waveform and further JSC signal waveforms (107, 111a-111n) with a JSC DNN structure (114) at the second device (112) to generate reconstructed communication data corresponding to the input communication data and a radar sensing feedback signal associated with the one or more objects; transmitting (932) the radar sensing feedback signal as a radar sensing feedback signal waveform (117) to the first device (101) for use in generating radar sensing information for the one or more objects; and sending (934) the reconstructed communication data to a data sink of the second device.

22. The method of claim 21, wherein the establishing (422) JSC DNN operations for the one or more time slots of the communication session with the first device (101) comprises: receiving (922a) one or more control messages from the first device (101) indicating a selected type of JSC DNN structure for use by the second device (112) in processing the one or more received JSC signal waveforms (107, 111a-111n); and configuring (423, 923) the JSC DNN structure (114) of the second device (112) based on the selected type of JSC DNN structure for use in the one or more time slots.

23. The method of claim 22, further comprising a neural network table (1020) storing data representative of a plurality of JSC DNN structures, each JSC DNN structure including a specific JSC DNN identifier, wherein: the receiving (922a) the one or more control messages from the first device (101), wherein the one or more control messages from the first device (101) including a JSC DNN identifier of the selected type of JSC DNN structure; and configuring (923) the JSC DNN structure (114) of the second device (112) using the neural network table (1020) and the received JSC DNN identifier for use in the one or more time slots.

24. The method of claim 23, wherein the neural network table (1020) further includes, for each JSC DNN structure, a communication channel condition that the JSC DNN structure is trained for and a JSC system type each of the JSC DNN structures is associated with, the method further comprising: selecting the JSC DNN structure using the received JSC DNN identifier from the neural network table (1020); and configuring the JSC DNN structure of the second device (112) to the selected JSC DNN structure for the one or more time slots.

25. The method of any one of claims 21 to 24, the method further making up: jointly training (1100) the JSC DNN structure (114) of the second device (112) and a JSC DNN structure (103) of the first device (101) over a communication channel (108) with one or more objects (109a-109n) using a training dataset comprising known radar sensor data associated with the one or more objects (109a-109n) and corresponding sets of input communication data, wherein weights of the JSC DNN structures (103, 114) of the first and second devices (101, 112) are updated based on a joint loss function comprising a combination of performance metrics associated with radar sensing information generated by the JSC DNN structure (103) of the first device (101) and reconstructed communications data generated by the JSC DNN structure (114) of the second device (112).

26. The method of claim 25, wherein a performance metric associated with the generated radar sensing information is a radar sensing error rate derived from the radar sensing information generated by the JSC DNN structure (103) of the first device (101) and corresponding known radar sensor data of the training dataset, and a performance metric associated with the reconstructed communications data is a block or bit error rate derived from the reconstructed communications data generated by the JSC DNN structure (114) of the second device (112) and corresponding input communication data associated with the training dataset.

27. The method of any one of claims 21 to 26, wherein the JSC DNN structure (114) of the second device (312) includes a receiving communication and radar, CR, DNN (314a) and a transmitting radar feedback, RFB, DNN (314b), wherein: the receiving CR DNN (314a) is configured for processing one or more received JSC signal waveforms (311a-311n) for the time slot and generating reconstructed communication data (315) associated with the input communication data (302) for the time slot and radar sensing feedback information (314d) for the time slot; and the transmitting RFB DNN (314b) configured for processing the generated radar sensing feedback information (314d) and generating radar sensing feedback signal for transmission as a radar sensing feedback signal waveform (317) to the first device (301).

28. The method of claim 27, further comprising: conventionally transmitting the generated radar sensing feedback information to the first device (301) in one or more further time slots.

29. The method of any one of claims 21 to 28, wherein the JSC DNN structure (114) of the second device (312) includes a receiving communication DNN (314c), the receiving communication DNN (314c) configured for processing one or more received JSC signal waveforms (311a-311n) for the time slot and generating reconstructed communication data (315) corresponding to the input communication data (302) for the time slot.

30. A method performed by a second device (112), the method comprising: establishing (922) joint sensing and communication, JSC, deep neural network, DNN, operation for one or more time slots of a communication session with a first device (101); for each time slot of the one or more time slots, the method further comprising: receiving (924) a JSC signal waveform transmitted from the first device (101) in said each time slot over a communication channel, the JSC signal waveform representing input communication data and a radar signal; receiving (926) one or more further JSC signal waveforms for said each time slot based on reflections of the transmitted JSC signal waveform from one or more objects in the communication channel; processing (942) the received JSC signal waveform and further JSC signal waveforms with a receiving communication DNN structure (314c) at the second device (112), the receiving communication DNN structure (314c) configured for generating reconstructed communication data corresponding to the input communication data transmitted from the first device (101); and sending (944) the reconstructed communication data to one or more upper layer protocols of a protocol stack of the second device.

31. A method performed by a third device (343), the method comprising: establishing joint sensing and communication, JSC, deep neural network, DNN, operations for one or more time slots of a communication session with a first device (101); for each time slot of the one or more time slots, the method further comprising: receiving a JSC signal waveform (307) transmitted from the first device (101) to a second device (112) in said each time slot over a communication channel (308), the JSC signal waveform (307) represents input communication data for the second device (112) and a radar signal; receiving one or more further JSC signal waveforms (340a-340n) for said each time slot based on reflections of the transmitted JSC signal waveform (307) from one or more objects (309a-309n) in the communication channel; processing the received JSC signal waveform (307) and further JSC signal waveforms (340a-340n) with a radar sensing DNN structure (343) at the third device (341), the radar sensing DNN structure (343) configured for generating radar sensing feedback information for the one or more objects (309a-309n); and transmitting a radar sensing feedback signal waveform (345) comprising the generated radar sensing feedback information of the one or more objects (309a-309n) to the first device (101) for use by the first device (101) in generating radar sensing information of the one or more objects (309a-309n) for the time slot.

32. The method of any preceding claim, wherein the first device is a base station and the second device is a user equipment.

33. The method of any of claims 1 to 31, wherein the first device is a user equipment and the second device is a base station.

34. A computer program product comprising computer readable instructions that, when executed by a computer, cause the computer to perform the method according to any preceding claim.

35. A first device comprising one or more processors and a memory, the memory storing computer readable instructions that, when executed by the one or more processors, cause the first device to perform the method according to any one of claims 1 to 20.

36. A second device comprising one or more processors and a memory, the memory storing computer readable instructions that, when executed by the one or more processors, cause the second device to perform the method according to any one of claims 21 to 30.

37. A third device comprising one or more processors and a memory, the memory storing computer readable instructions that, when executed by the one or more processors, cause the third device to perform the method of claim 31.

38. An apparatus comprising: one or more antennas; one or more processors; and a memory; and wherein the one or more processors are connected to the memory and the one or more antennas, and the memory further storing computer readable instructions that, when executed by the one or more processors, cause the apparatus to perform the method according to any one of claims 1 to 33.

39. A communication system comprising: a first device configured according to claim 35; a second device configured according to claim 36; and wherein the first device and second device establish JSC DNN operations for one or more time slots of a communication session.

40. The communication system of claim 39 further comprising a third device configured according to claim 37, wherein the first device and third device establish JSC DNN operations for the one or more time slots of the communication session.

41. A computer-readable medium comprising instructions stored thereon, which when executed, causes one or more processors to perform the method according to any one of claims 1 to 33.