Aerial traffic channel feedback

Abstract

The present subject matter describes a method for positioning a Low Altitude Platform Station (LAPS) based drone-cells for supporting communication in a 3GPP network. The method comprises receiving from a UE at least one feedback parameter pertaining to a current network-traffic flow and mobility in respect of the UE. The UE is selected for availing an aerial network through a drone base station (DBS)-cell based on service-requirement of the UE determined from the feedback parameter. Further, a position for at least one DBS-cell is determined with respect to the at-least one selected UE based on said at least one parameter to serve the selected UE. An aerial-communication link is established between the selected UE and the DBS cell by deploying one or more Low Altitude Platform Station as a drone base station (DBS) in accordance with said determined-position of DBS-cell to thereby augment a network connectivity of the selected UE.

Description

DESC:FIELD OF THE INVENTIONThe present disclosure relates to aerial traffic channel feedback and in particular, relates to aerial-cell positioning in 3GPP networks.BACKGROUNDWith the improvements in wireless technology over the last decade, mobile devices are finding a place in our day-to-day lives in ways beyond imagination. Initially designed to support voice and data, with the advent of social, economic and immersive use cases, the expectation of coverage and capacity has increased many folds. New Radio (NR) designed by Third (3rd) Generation Partnership Project (3GPP) is addressing the requirements laid down by International Mobile Telecommunication system (IMT) 2020. Though the broad-level requirements address the reliability, latency, and peak throughput, it does not necessarily address network availability and per user throughput needs. Coverage and throughput needs are deployment related issues, and an operator is expected to render it. Finding a standardized solution to the above will be important requirements for Sixth Generation (6G) cellular networks. As a solution, self-optimizing networks (SON) concepts allow for dynamic optimizations of cellular-networks and adopt provisioning of network elements on the go. While user density for a location may be dynamic, 5G NR or 4G macro (terrestrial) cell-densities for the location largely remains static. Radio Access Network (RAN) deployment of evolved NodeB (eNB) or Next Generation NodeB (gNB) with Relay nodes (RN), or Radio Remote Head (RRH) or Distributed Antennae System (DAS) has posed a teething problem as it requires a substantial investment from the cellular operator and a lot of time for installation. Finding a standardized solution to the above is at-least important requirements for next generation, say Sixth Generation (6G) cellular networksTo overcome the above challenges, deploying a drone or an Unmanned Aerial Vehicle (UAV) based cells can be a great-enabler for dynamic scalability of coverage and capacity needs on the field, as it takes very less time to install from above ideas. There are several scenarios where drone cells can be used and are more apt when the need for network scalability is temporal in nature due to an event, and dependent on the user density especially in urban scenarios, for example, during peak traffic or time bound events, such as sports and business conferences. In fact, aerial-communication can even enable a business model for an on-demand network infrastructure to be monetized by service providers of event management. In conventional techniques, a plethora of use-cases and business models are analysed for deployment of drone-based aerial communication.In other conventional techniques, drone-based cellular base station for aerial communication has also been analysed, and some of the challenges have been discussed related to the channel model, system performance, Line of Sight (LOS) probability, and deployment models. The available literature discusses the physical characteristics of the link between User Equipment (UE) and drone like the Air to Ground (A2G) link. Most recently, 3GPP has published a Technical Report (TR) on Non Terrestrial Network (NTN) use cases and deployment models, for aerial communication, which is considering High Altitude Platform Station (HAPS) and satellite stations, which are in Low Earth Orbit (LEO), Medium Earth Orbit (MEO) or Geostationary Earth Orbit (GEO). Though the above aerial platforms provide for the use cases, the cost of deployment is expected to be very high. A Low Altitude Platform Station (LAPS) seems to be an ideal platform for opportunistic deployments for enhancing capacity and improving coverage in the network. In some techniques, an optimal altitude LAPS based cell was discussed with coverage optimality, though it does not discuss the per-user throughput needs to be served by LAPS. In other techniques, positioning for LAPS based drone cells is dealt with for coverage and sum-rate needs of UE and considers Signal to Noise and Interference Ratio (SINR) as the most important factor for the same. The major limitations of the available literature are that it addresses the theoretical-maximization of coverage or data rate. However, the critical component of the traffic and mobility considerations for the served UEs is not analyzed which may lead to less-utilization of the deployed aerial network. In some conventional techniques, different models for data traffic prediction are presented. Further, in other techniques, a Recurrent Neural Network (RNN) based traffic prediction model is presented which uses the traffic information gathered from the Physical Downlink Control Channel (PDCCH), and provides improved accuracy with respect to the Feed Forward Neural Network (FFNN) or to the classic Auto Regressive Integrated Moving Average (ARIMA) model. However, the models are not presented for each UE separately as it addresses the traffic prediction at a cell-level in the network. .SUMMARYThis summary is provided to introduce a selection of concepts, in a simplified format, that are further described in the detailed description of the invention. This summary is neither intended to identify key or essential inventive concepts of the invention and nor is it intended for determining the scope of the invention.From the state of the art technology, at-least the following major-limitations have been observed while deploying the LAPS based drone cells: 1. The positioning of the drone cell is based on the Signal to Noise and Interference Ratio (SINR) which only considers long term fading due to path loss and topology. The traffic and mobility considerations for the served UEs are not analyzed which may lead to lower utilization of the deployed aerial cell. 2. UE selection is not treated in the available literature. 3. The data traffic prediction model applied in the network are not applied for each UE.Meeting the requirements of coverage and per user throughput requires proper terrestrial network planning and deployment by the operator. The user density for any location due to an event is a dynamic phenomenon and it poses a challenge for network planning, as the terrestrial cell density for a location remains static. Deploying a drone or an Unmanned Aerial Vehicle (UAV), based cell for above challenge can be an enabler for Beyond 5G (B5G) systems by providing a dynamic scalability of coverage and capacity demands. Drone-based aerial communication has multiple advantages and it poses several challenges, like deployment optimization for coverage, resource allocation, interference, and energy due to altitude and flying time considerations. In the present disclosure, a Low Altitude Platform Station (LAPS) based drone-cells has been provided for opportunistically augmenting the network (e.g. the terrestrial network) and address the resource allocation challenge specifically in the aerial network. The present subject matter describes a method for positioning a Low Altitude Platform Station (LAPS) based drone-cells for supporting communication in a 3GPP network. The method comprises receiving from a UE at least one feedback parameter pertaining to a current network-traffic flow and mobility in respect of the UE. The UE is selected for availing an aerial network through a drone base station (DBS) -cell based on service-requirement of the UE determined from the feedback parameter. Further, a position for at least one DBS-cell is determined with respect to the at-least one selected UE based on said at least one parameter to serve the selected UE. An aerial-communication link is established between the selected UE and the DBS cell by deploying one or more Low Altitude Platform Station as a drone base station (DBS) in accordance with said determined position of DBS-cell to thereby augment a network connectivity of the selected UE.The state of the art literature attempts to improve the sum throughput of the aerial network, without considering the traffic distribution and user feedback. On the other hand, through the present subject matter’s analyzed augmented deployment with the availability of mobility feature in drone cells, a zone based drone service model is proposed that is based on the feedback called Feedback for Aerial Cell Trajectory (FACT) from a User terminal (UE) to the network. The feedback has parameters like predicted traffic (UL / DL), traffic preference, channel quality and mobility which are used for altitude and location adjustment of a drone cell through which the aerial network: (a) selects the users to serve, (b) controls the observed channel quality by the users and (c) schedules the resources optimally for them..The present subject matter’s augmented deployment as a solution is at least suited for dynamic capacity enhancement where while the terrestrial network provides coverage and a LAPS / drone based aerial network augments and provides for uplink (UL) and downlink (DL) resource allocation in an opportunistic manner. The mobility of a drone cell gives it an opportunity to select the UEs to serve, the duration to serve them and through its change in position, control the channel quality measurement for both UL and DL for UE. The present subject matter addresses a problem of optimally positioning and dynamically repositioning of a drone cell, such that it facilitates finding the optimal set of UEs to serve, and then improve the UL and DL resource allocation, thereby maximizing per-user throughput for each served UE. In an implementation, the Feedback for Aerial Cell Trajectory (FACT) comprises of: (a) Predicted Buffer Status from the UE in UL; (b) Channel Quality Measurement; (c) Predicted Buffer Status for the UE in DL; (d) Traffic preference by UE; (e) Probability of staying in the same zone for the next Aerial Scheduling Period (ASP). A zone is a region within a coverage area of the terrestrial cell, which has the UEs, which are chosen (based on FACT) by the terrestrial network, to be served by a drone cell, so that resource allocation from the drone cell is maximized in that zone for those chosen UEs. The zone can be temporally served by the aerial drone cell for one or more ‘Aerial Scheduling Period’ (ASP). During the ASP, a drone is statically located at a position to serve a chosen zone, and parameters for the common channel are modified to suit that position. After the ASP, based on the new FACT, a drone-cell can be moved to a new position to serve a new zone. Through extensive simulations derived from analytical modeling, the results reveal that using the above proposed deployment and procedures in accordance with the present subject matter, the bandwidth utilization for the deployed drone cell improves in comparison to a drone cell deployment discussed in the state of the art.To further clarify advantages and features of the present invention, a more particular description of the invention will be rendered by reference to specific embodiments thereof, which is illustrated in the appended drawings. It is appreciated that these drawings depict only typical embodiments of the invention and are therefore not to be considered limiting of its scope. The invention will be described and explained with additional specificity and detail with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGSThese and other features, aspects, and advantages of the present invention will become better understood when the following detailed description is read with reference to the accompanying drawings in which like characters represent like parts throughout the drawings, wherein:Figure 1 illustrates a block diagram depicting a drone cell having backhaul with terrestrial cell, in accordance with an embodiment of the present subject matter;Figure 2 illustrates positioning of drone cells affecting a served coverage area, according to an embodiment of the present disclosure;Figure 3 illustrates coverage on ground in a 2-Dimensional plane for different drone cell positions in relation to the territorial coverage, according to an embodiment of the present disclosure;Figure 4 illustrates a proposed message exchange between a User Equipment (UE) and a network, according to an embodiment of the present disclosure;Figure 5 illustrates a block diagram depicting an LSTM cell, according to an embodiment of the present disclosure; andFigure 6 illustrates a proposed LSTM model for UL / DL buffer prediction, according to an embodiment of the present disclosure. Further, skilled artisans will appreciate that elements in the drawings are illustrated for simplicity and may not have been necessarily been drawn to scale. For example, the flow charts illustrate the method in terms of the most prominent steps involved to help to improve understanding of aspects of the present invention. Furthermore, in terms of the construction of the device, one or more components of the device may have been represented in the drawings by conventional symbols, and the drawings may show only those specific details that are pertinent to understanding the embodiments of the present invention so as not to obscure the drawings with details that will be readily apparent to those of ordinary skill in the art having benefit of the description herein. DETAILED DESCRIPTION OF FIGURESFor the purpose of promoting an understanding of the principles of the invention, reference will now be made to the embodiment illustrated in the drawings and specific language will be used to describe the same. It will nevertheless be understood that no limitation of the scope of the invention is thereby intended, such alterations and further modifications in the illustrated system, and such further applications of the principles of the invention as illustrated therein being contemplated as would normally occur to one skilled in the art to which the invention relates.It will be understood by those skilled in the art that the foregoing general description and the following detailed description are explanatory of the invention and are not intended to be restrictive thereof. Reference throughout this specification to “an aspect”, “another aspect” or similar language means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the present invention. Thus, appearances of the phrase “in an embodiment”, “in another embodiment” and similar language throughout this specification may, but do not necessarily, all refer to the same embodiment.The terms "comprises", "comprising", or any other variations thereof, are intended to cover a non-exclusive inclusion, such that a process or method that comprises a list of steps does not include only those steps but may include other steps not expressly listed or inherent to such process or method. Similarly, one or more devices or sub-systems or elements or structures or components proceeded by "comprises... a" does not, without more constraints, preclude the existence of other devices or other sub-systems or other elements or other structures or other components or additional devices or additional sub-systems or additional elements or additional structures or additional components.Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skilled in the art to which this invention belongs. The system, methods, and examples provided herein are illustrative only and not intended to be limiting.Figure 1 illustrates a block diagram depicting a drone cell having backhaul with terrestrial cell. In the proposed mechanism, as shown in Figure 1, a LAPS based drone cell is deployed when the serving terrestrial cell is unable to support the required quality of service (QoS) needs for a UE or a set of UEs, due to degraded channel environment, poor coverage or increased interference or congestion due to capacity limitation emanating from the surge in number of users in the terrestrial cell. Thus, when there is a need for dynamically increasing physical channel resources or meeting the targeted QoS for a set of UEs a drone cell deployment augments the terrestrial cell. In an implementation, the Dual Connectivity (DC) or Carrier Aggregation (CA) based communication is based on frequency-carriers forming a part of a plurality of telecommunication-links defined as one or more of:a) a first link between UE and the drone base station cell defined by front-haul;b) a second-link between UE and terrestrial cell defined by front-haul; c) a third link between the DBS cell and the terrestrial cell defined by backhaul.The UEs, which are served by the drone cell, can be in Dual Connectivity (DC) or Carrier Aggregation (CA) operation mode, where resources are allocated from both the terrestrial cell as well as the drone cell. This deployment also alleviates the issue of interference as the carrier frequencies of the terrestrial cell (say f1) and the drone cell (say f2) are different. The backhaul link between the drone cell and the terrestrial cell is wireless and uses either a Point to Point (P2P) dedicated link (say f3) or an Integrated Access Backhaul (IAB) which shares the same carrier as the A2G link between UE and drone cell (f2). In the proposed deployment and analysis, the backhaul link is a P2P link on an independent carrier (f3). The terrestrial cell continues to serve the UEs, which are not served by LAPS. Accordingly, the frequency-carriers forming a part of the front-haul, back-haul, and carrier-aggregation and / or dual-connectivity based aerial communication link are either identical or differentThe determination of position of DBS cell comprises determining an amount of inter-carrier interference in respect of one or more of:a) in the front-haul between the UE and terrestrial base station;b) in the backhaul between the one or more prior-existing DBS and the TBS;c) among a plurality of prior-existing DBS forming a D2D (device to device) cluster on air;d) among one or more UEs at ground constituting a side-link; ande) any other inter-carrier interference.In an implementation, the aerial communication link as established may be further modified through removing the terrestrial base station acting as a master base station in respect of the carrier-aggregation and dual-connectivity based aerial communication link. Thereafter, the carrier-aggregation and dual-connectivity may be restricted in respect of frequency carriers configured for communication among one or more DBS and the selected UEs. In such a scenario, the plurality of DBS comprises master DBS and a secondary DBS. Moreover, in an example, the UE may also include a flying-object such as a drone that may interact with the terrestrial base station and / or the applicable drone base station.Further, each of the UE, drone and base station may be a network node in accordance with the 3GPP standards and comprises a transceiver and processor.Figure 2 illustrates positioning of drone-cells affecting a served coverage area, while Figure 3 illustrates coverage on ground in a 2-Dimensional plane for different drone cell positions in relation to the territorial coverage, according to an embodiment of the present disclosure. As per Figure 2, a drone-cell hovers over a geographical area, and the coverage provided by the drone-cell can be assumed as a two dimensional (2D) circular area or coverage zone at ground as captured in Figure 2, with radius R, dependent on the altitude of deployment. As shown, when the drone cell hovers from one position to the other, the underlying served area also moves along. The change in coverage area is shown when a drone cell moves in three-dimensional (3D) Cartesian coordinate system, between coordinates on X-axis within the range x1 and x2, Y-axis within the range y1 and y2 and Z-axis within the range h1 and h2 is also captured in the figure. Figure 3 illustrates coverage on ground in a 2-Dimensional plane for different drone cell positions in relation to the territorial coverage, according to an embodiment of the present disclosure. In Figure 3, the coverage of both terrestrial cell as well as the aerial cell is depicted. As shown the 2D coverage area for a terrestrial cell is static. However, the aerial cell can offer different 2D coverages depending on its position in the 3D plane.In an embodiment, the scheduling criteria takes cognizance of the supported coverage at each possible drone position, the UE location in the coverage area of the terrestrial network, for example, by known procedures like Reference Sequence Time Difference (RSTD) for Observed Time Difference of Arrival (OTDOA) or other such analogous-mechanism. Through UE feedback, the QoS needs of each UE, predicted UL and DL traffic from and to each UE and its reported channel quality are determined. Based on this UE feedback, a procedure is undertaken to select a target-set of UEs to serve, and then to move the drone-cell around to be able to efficiently serve those UEs of interest. The present subject matter is at-least based on the concept of ‘zone’ (shown in Fig. 3), which is a region within a coverage area of the terrestrial cell where all the UEs of interest reside. In other words, coverage-zone corresponds to a terrestrial location having a substantial number of selected UEs. Thus, a drone is positioned in in 3D space in such a way that the ‘zone’ is within its coverage area, and then the resource allocation from the drone cell is controlled to only serve those UEs of interest, for a period called as Aerial Scheduling Period (ASP). The ASP spans several seconds, which is a design parameter for the aerial network, dependent on the tradeoff of hovering time for a drone and resource utilization of the drone cell. The feedback from UE to the terrestrial network, mentioned as FACT, is a quantized information element comprising of the parameters, as captured in Table I.Table 1FACT PARAMETERSParameter DescriptionPredicted Buffer in UL This quantity is the data packet predicted to be sent from the UE for the next ASP in the UL using a trained RNN model, at the UE and learned based on traffic patterns at the application level.Predicted Buffer in DL This quantity is the data packet predicted to be sent from the UE for the next ASP in the DL using a trained RNN model, at the UE and learned based on traffic patterns at the application level.Prioritized Traffic type Though the QoS Class Identifier (QCI) is prioritized by design in cellular communication, this parameter describes the user preference based on the output of the RNN predictive model for a particular kind of uplink or downlink traffic.Probability of staying in the same zone for next ASP The probability that UE remains in the same location is very important for a drone cells resource utilization, so this parameter describes the probability as estimated by the UE that it remains in the same location or zone as beforeChannel Quality Measurement This pertains to the Channel Quality Index (CQI) measurement of the terrestrial cell, based on the reference signals measurements and averaging / filtering applied at the UEFigure 4 illustrates a proposed message exchange between a User Equipment (UE) and a network for the FACT based Zone-based Drone Positioning and Trajectory, according to an embodiment of the present disclosure. At the onset, at Step-100 a data-session is established between several UEs with the terrestrial network. At Step-101, when the terrestrial network detects that some of the UEs (or a pre-existing drone-base station (DBS)) in the coverage area are not being served well matching their QoS / QoE needs, or other such criteria it sends a request to those UEs to share FACT. Overall, it may be diagnosed that UEs are experiencing capacity-limitations within the current terrestrial or aerial-cell. The UEs may be located terrestrially or aerially. At Step-102, the UEs responds back with feedback-information or FACT, which is used by the network to select the UEs, which needs to be served with an augmented aerial cell by either Carrier Aggregation (CA) or Dual Connectivity (DC). The feedback parameter pertains to the current network-traffic flow and mobility and are determined by UEs at least based on a neural-network based model. In an example, the feedback parameters have been defined at-least through earlier presented Table 1.At Step-104, the UEs are selected for availing an aerial network through at least one drone base station (DBS)-cell at least based on service-requirement of the UE determined from the at-least one parameter. In an example, the selection of UEs comprises calculating a weighted average based on said one or more parameter for the UE at least based on a neural-network model executed by a terrestrial base station (TBS) or a pre-existing drone base station (DBS). The one or more UEs are finally shortlisted, for the next ASP, based on at least one of: a high buffer expectation, a traffic direction, and a high probability of being stationary for getting served by the DBS based cell.At Step-106, the network indicates the drone cells to reposition itself to serve the UEs of interest. As soon as the drone cell gets to the optimal position, it shares the confirmation with the terrestrial network at Step-106. Such determination of position of DBS cell comprises determining: a location of the drone in 3D space; and a 2D coverage zone of the drone at ground corresponding to said location. In addition, duration of the Aerial Scheduling Period (ASP) with respect to the drone cell is computed based on the mobility and traffic-pattern of UEs.In an implementation, from the selected UEs communicating with the drone during the ASP, the feedback parameter may be periodically re-received. The position of the DBS cell may be re-determined for the selected UEs and also to further select other prospective UEs for a subsequent ASP. The duration of the ASP may be optionally recomputed as a part of determination of said subsequent-ASP. At Step-108, an aerial communication link is established between the selected UE and the DBS cell by deploying one or more Low Altitude Platform Station as a drone base station (DBS) in accordance with said determined position of DBS-cell to thereby augment a network connectivity of the selected UE. The aerial-communication link is established between the selected UEs and the drone cell, and the data traffic is augmented through the aerial link. UE applies either CA or DC procedures to combine the UL and DL traffic as defined by 3GPP NR Protocol in accordance with a Medium Access Control (MAC) Protocol. The zone served by the drone cell can be changed after an ASP and during the ASP the UEs of interest are scheduled to match their required Buffer Status Reports (BSR).The terrestrial network can alternatively also predict the buffer status at UL and DL for each UE instead of the UE reporting it. In some deployments the terrestrial network based on the UE capability can dynamically control where the buffer status prediction happens, i.e., either at the network or at the UE.In an implementation, the coverage-zone corresponds to a terrestrial location having a substantial number of selected UEs. The location of the drone in the 3D space corresponds to the 2D coverage zone having a maximum number of selected UEs, a maximum-weighted average of a buffer-requirement for a UE.At-least an object of the present subject matter is to find optimized-coordinates for positioning of a drone from amongst the set of possible positions, for a set of selected UEs. The possible positions for a drone are from the set {L_1,L_2,….L_R }, where each? L?_i=(x_i,y_i,h_i ), is in the 3D Cartesian coordinate system, and the theoretical sum rate for each L_(i )is above a threshold, given by (1). For each position, the corresponding coverage Zone is defined in {Z_1,Z_2,….Z_R }, as shown in Fig.2 and Fig. 3, which like depicted can be assumed as a 2D circular area with a given radius, which increases in dimension as L_i changes along the Z-axis with (? h?_i>h_j). An example procedure (Procedure 1) may be presented to describe the modeling for each parameter in FACT, with a combination of closed-form equations and a deep learning model.In the proposed architecture, UEs may be assumed to be uniformly distributed in the whole coverage area of the terrestrial cell, where the network knows the location of all the UEs. A set of possible drone cell positions are assumed to be given to a system model where the achievable (theoretical) sum rate is above a threshold (Threshold_(sum_TP )) with a constraint applied on the threshold. ?_(i=1)^(N_F)¦?_(k=1)^K¦?s_k^i log_2??(1+(H_k^i p_k^i) / |(|r-r_k |)|^2 )= Threshold_(sum_TP ) ? ? where, r=(x,y,h) describes the drone-cell position and r_k= (x_k,y_k,0) describes the position of UE k, s_k^i defines if a subcarrier, i is allocated to a UE k, H_k^i defines the channel characteristics for a UE k on a subcarrier i, and p_k^i defines the power allocation on the same. The quantity N_F defines the system bandwidth in the number of subcarriers, and K defines the total number of UEs. The position of the drone cell given by (x,y,h) is derived for achieving the theoretical sum rate.Each UE maintains a data set for data traffic generated for past several days and uses it to predict the required UL and DL buffer for a period spanning the next ASP. UE reports FACT parameters to the terrestrial cell. Each of the reported parameters is modeled in the subsequent subsection. The knowledge of FACT parameters is used as a weighted measure at the network to choose a position (L_i) for the drone cell, which serves the zone (Z_i), to maximize the number of UEs, the drone utilization, and thereby improving the per-user throughput. The same is depicted in Procedure 1 as illustrated later.Figure 5 illustrates a block diagram 500 depicting an LSTM cell, according to an embodiment of the present disclosure. LSTM cells manage two state vectors viz. short-term and long-term, and for performance reasons they are kept separate by default. LSTM cell representation is captured in Figure 5. Figure 5 illustrates a block diagram depicting the LSTM cell, according to an embodiment of the present disclosure. LSTM cell looks exactly like a normal artificial neuron, except that its output is split in two vectors: s(t) as the short-term state and l(t) as the long-term state. LSTM nodes are designed such that the network over which it is operating should learn what to memorize, erase, and fetch from the long-term state. As the previous time step long-term state l(t-1) traverses the network, it first goes through a forget gate erasing some memories then it adds some new information by applying addition operation on the information selected by an input gate. Current time step, long-term state result l(t) is sent straight out of the cell without further modifications. Therefore, the design of cell is such that at each time step, there is addition and deletion in the memory present in long-term state. Post addition operation, the current time step long-term state is copied and passed through the hyperbolic tangent (tanh) transformer and the result is further modified by the output gate operations. This operation produces the current time step short-term state s(t). The same is the cell’s output for this time step, also represented by y(t).The internal design of LSTM cell has a core consisting of four independent fully connected layers, each works on their own weight matrix and biases. The current time step input vector x(t) and previous time step short-term state s(t-1) are shared as weighted transformed input to all the four layers. Each layer serves a different purpose. The first layer output f(t) serves the purpose of forget gate where based on element wise multiplication; it erases the information present in long-term state. The second and third layer output g(t) and i(t) respectively work together and serve the purpose of input gate, where the purpose of i(t) is to decide which part of g(t) should get added to the long-term state l(t). The fourth layer output o(t) serves the purpose of output gate; it operates on the hyperbolic tangent transformed version of the current time step long-term state by applying element-wise multiplication with o(t) to generate the current time step short-term state s(t) and cell output y(t).In nutshell, a LSTM cell optimally tries to solve the vanishing gradient problem and memory loss ness in large deep neural networks and learn to save an important input, store it in the long-term state, learn to preserve it for long time, and learn to extract it whenever it is required. Computation for the cell’s long-term state l(t), its short-term state s(t), and its output at each time step for a single instance is given belowf(t)= logistic(W_xf.x(t)+ W_sf.s(t-1)+b_f)i(t)= logistic(W_xi.x(t)+ W_si.s(t-1)+b_i)o(t)= logistic(W_xo.x(t)+ W_so.s(t-1)+b_o )g(t)= tanh(W_xg.x(t)+ W_sg.s(t-1)+b_g )l(t)=l(t-1)?f(t)+g(t)?i(t)s(t)=y(t)= o(t)?tanh(c(t)) (2)Where, W_xf, W_xi, W_xo, W_xg are weight matrix which operates over current time step input sequence x(t) and W_sf, W_si, W_so, W_sg are weight matrix which operates over previous time step short-term state s(t-1). Figure 6 illustrates a proposed LSTM model for UL / DL buffer prediction followed by positioning of the drone cells in accordance with steps 104 and 106, according to an embodiment of the present disclosure. In Fig. 6, the proposed model for LSTM is captured. UL & DL buffer prediction for next ASP is performed using the proposed model. For the one-step ahead ASP, the predicted UL / DL Buffer for a UE are expressed in terms of trained model hypothesis h_LSTM (x).UL_size=Predicted(UL)=h_LSTM (input UL buffers) (3)DL_size=Predicted(DL)=h_LSTM (input DL buffers) (4)For each UE, the preference for a traffic direction, is expressed using normalized version of sigmoid function with predicted buffers i.e. using (2) & (3) as input:s(UL)=1 / (1+ e^(-Predicted(UL)) )s(DL)=1 / (1+ e^(-Predicted(DL)) )Pref(UL)=(s(UL)) / (s(UL) + s(DL) ) (5)Pref(DL)=(s(DL)) / (s(UL) + s(DL) ) (6) Where, Pref(UL) and Pref(DL) normalizes s(UL) and s(DL) respectively to find the preference between UL and DL requirement. It is intuitive that data buffer requirements can be biased towards UL if the user is trying to upload data and biased towards DL if the user is trying to download, and there can be a scenario when both are equally required when the user is simultaneously trying to upload and download. A sigmoid activation may be proposed in that regard.Where, Pref_UL+ Pref_DL=1It can be observed from (5) and (6) that their sum should be equal to 1.For the given ASP, the probability of a UE to stay in the same geographical area is defined using Poisson’s Probability model, the probability mass function (pmf) is represented below.P(UE static for 'k' ASP)=(e^(-?)* ?^k) / k! = P_mobility (7)Where, ? is the average number of ASPs that a UE stays static, and k is the number of ASP’s a UE stays static.The channel quality measurement for the terrestrial cell, by a UE, is modelled as,CQI (UE) = Reported_CQI (8)Where, Reported_CQI is the CQI, as measured based on reference symbol power as defined in 5G NR system Further, the weighted average of the buffer requirement for a UE is given as:?FACT?_weighted = (((UL_size*Pref_UL )+(DL_size*Pref_DL ))*P_mobility )*((Max_CQI-Reported_CQI) / (Scale_CQI )) (9) Where, Max_CQI is defined as the maximum value of the CQI, and Scale_CQI is a scaling factor (for normalization) used in the model. The insight of (9) is based on the rationale that for the next ASP, a UE with high buffer expectation and a high probability of being static should be considered to be served by the drone cell if it is experiencing poor channel condition in the terrestrial cell. The best L_(i )using Procedure 1 maximizes the sum of ?FACT?_weighted for all m_i UEs served in the coverage of the drone for the chosen ASP. The formulation of the optimization problem to find the best L_(i )is given as follows:maximize (?_(j=1)^(j=m_i)¦??FACT?_weighted?_j ) (10)There are some additional constraints (C1,C2,C3) for the proposed model, as below:C1: ?_(j=1)^(j=m_i)¦?(U?L_size?_j+DL_(size_j ) )=Total System BW? (11)C2: Pref_(UL_j )+ Pref_(DL_j )=1, ?j ?m (12)C3: m_i>UE_thesh (13)Where Total System BW is defined as possible number traffic in bytes that the terrestrial cell can serve during an ASP. The above expression (10) is solved, as per expansions below. For each L_i which can serve m_i UEs, the weighted average of the buffer requirement for all the UEs is given by? FACT?_weighted [L_i].?FACT?_weighted [L_i ]= ?_(j=1)^(j=m_i)¦?(((UL_(size_j )*Pref_(UL_j ) )+(DL_(size_j )*Pref_(DL_j ) ))*?P_mobility?_j )*((Max_(CQI_j )-Reported_(CQI_j )) / (Scale_(CQI_j ) )) ? (14)By substituting (3) (4), (5), (6), and (7) in (14), we get (15) which denotes a weighted-average in accordance with steps 104 and 106. ?FACT?_weighted [L_i ]= ?_(j=1)^(j=m_i)¦?(((h_LSTM (UL_j )*((1 / (1+ e^(-h_LSTM (UL_j)) )) / (1 / (1+ e^(-h_LSTM (UL_j ) ) )+1 / (1+ e^(-h_LSTM (DL_j ) ) ))))+(h_LSTM (DL_j )*((1 / (1+ e^(-h_LSTM (DL_j ) ) )) / (1 / (1+ e^(-h_LSTM (UL_j ) ) )+1 / (1+ e^(-h_LSTM (DL_j)) )))))*(e^(-?_j )* ?_j^(k_j )) / (k_j !))*((Max_(CQI_j )-Reported_(CQI_j )) / (Scale_(CQI_j ) )) ? (15) By identifying the maximum value of ?FACT?_weighted [L_i] which can serve? m?_i> UE_thesh, will yield the result, which is also explained in below Procedure 1.Procedure 1: Zone-based Drone Positioning and Trajectory Procedure1: Input:2: L= {L_1,L_2,….L_R }; Drone cell locations, where sum_TP> Threshold_(sum_TP )3: Z= {Z_1,Z_2,….Z_R } ; Set of zones each corresponds to a location in L4: U= {?UE?_1,?UE?_2,….?UE?_n }; Set of connected UEs the terrestrial cell.5: Output: For each ASP, an? L?_i.6: Begin ASP_POSITION7: For each UE UE_i in U8: Get FACT parameter from each UE9: Compute FACT_weighted10: End for11: For each L_i in L12: Initialize L_i FACT_weighted=0, L_i NumUE=013: For each UE_i in the coverage area of Z_i14: Increment L_i FACT_weighted by FACT_weighted 15: if FACT_weighted ?0, Increment L_i NumUE by 116: End if17: End for18: End for19: Search L_i in L which has max (L_i FACT_weighted) and max?(L?_i NumUE)20: End ASP_POSITIONIn accordance with the present subject matter, drone-based cells for aerial communication render an operator a good solution to dynamically scale coverage and achievable capacity, essential for 6G. In accordance with the present subject matter, presented and analyzed is the augmented aerial deployment, where the terrestrial cell provides for both coverage and capacity to the served UEs, while the LAPS-based aerial cell is deployed to improve the capacity that the network can offer dynamically in a CA or DC scenario using an independent non-interfering carrier.The present subject matter proposed feedback (FACT) based positioning and trajectory of drone cells. In our proposed design, the feedback for traffic prediction, mobility, and CQI are considered from the under-served UEs and then an optimal drone position is suggested which maximizes the resource allocation for the selected UEs. In example, using the predicted data traffic along with the analytical model and system simulation, the present subject matter is able to demonstrate a gain of over 44 % in resource utilization, and over 27% in user throughput, when compared with a state of the art drone cell deploymentWhile specific language has been used to describe the present subject matter, any limitations arising on account thereto, are not intended. As would be apparent to a person in the art, various working modifications may be made to the method in order to implement the inventive concept as taught herein. The drawings and the foregoing description give examples of embodiments. Those skilled in the art will appreciate that one or more of the described elements may well be combined into a single functional element. Alternatively, certain elements may be split into multiple functional elements. Elements from one embodiment may be added to another embodiment.

Claims

,CLAIMS:We Claim:

1. A method for positioning of a Low Altitude Platform Station (LAPS) based drone cells forsupporting communication in a 3GPP network, said method comprising:receiving (step 102) from a UE at least one feedback parameter pertaining to a currentnetwork-traffic flow and mobility in respect of the UE;selecting (step 104) said UE for availing an aerial network through at least one dronebase station (DBS)-cell based on service-requirement of the UE determined from the at-least oneparameter;determining (step 106) a position for at least one DBS-cell with respect to the at-least oneselected UE based on said at least one parameter to serve the selected UE; andestablishing (step 108) an aerial communication link between the selected UE and theDBS cell by deploying one or more LAPS as a drone base station (DBS) in accordance withsaid determined position of DBS-cell to thereby augment a network connectivity of the selectedUE.

2. The method as claimed in claim 1, further comprising:querying by a terrestrial base-station (TBS) or a pre-existing drone-base station (DBS)requirements from one or more UE based on diagnosing a deficient QoE, a deficient QoS,capacity-limitations within the current terrestrial or aerial-cell, wherein said UEs are locatedterrestrially or aerially ;receiving a feedback by the TBS from said one or more UEs as said at least onefeedback parameter pertaining to the current network-traffic flow and mobility, said parametersdetermined by UEs at least based on a neural-network based model and defined by one or moreof:a Predicted Buffer in uplink (UL)a Predicted Buffer in downlink (DL)a Prioritized Traffic typea Probability of staying in the same zone for a next aerial communicationscheduling period (ASP) with respect to the drone;21a Channel Quality Measurement (CQI); anda Minimum Guaranteed service based on QoS.

3. The method as claimed in claim 2, wherein said selecting of the UEs comprises:calculating a weighted-average based on said one or more parameter for the UE atleast based on a neural-network model executed by the TBS or the pre-existing DBS; andselecting the UE, for the next ASP, based on at least one of:a high buffer expectation;a traffic direction;a high probability of being stationary for getting served by theDBS based cell.

4. The method as claimed in claim 1, wherein said determination of position of DBS cellcomprises determining:a location of the drone in 3D space; anda 2D coverage zone of the drone at ground corresponding to said location;computation of a duration of the Aerial Scheduling Period (ASP) with respect tothe drone cell based on the mobility and traffic-pattern of UEs.

5. The method as claimed in claim 4, wherein said determination of position of DBS cellcomprises further determining an amount of inter-carrier interference in respect of one or moreof :a) in the front-haul between the UE and TBS;b) in the backhaul between the one or more prior-existing DBS and the TBS;c) among a plurality of prior-existing DBS forming a D2D cluster on air;d) among one or more UEs at ground constituting a side-link; ande) any other inter-carrier interference.

226. The method as claimed in claim 4, wherein said coverage-zone corresponds to a terrestriallocation having a substantial number of selected UEs.

7. The method as claimed in claim 4, wherein said location of the drone in the 3D spacecorresponds to:the 2D coverage zone having a maximum number of selected UEs;a maximum-weighted average of a buffer-requirement for a UE.

8. The method as claimed in claim 1, wherein said establishing of an aerial communication linkachieving an aerial communication link between the selected UEs and the DBS cell;applying, by the UE, at least one of carrier aggregation (CA) and dual connectivity (DC)procedures to combine UL and DL traffic in accordance with a Medium Access Control (MAC)Protocol.periodically re-receiving the selected at-least one feedback parameter from the UEscommunicating with the drone during the ASP;re-determining the position of the DBS cell for the selected UEs and one or moreadditional UEs for a subsequent ASP; andoptionally re-computing the ASP as a part of determination of said subsequent-ASP .

9. The method as claimed in claim 8, wherein the Dual Connectivity (DC) or CarrierAggregation (CA) based communication is based on frequency-carriers forming a part of aplurality of telecommunication-links defined as one or more of:a) a first link between UE and the DBS cell defined by front-haul;b) a second-link between UE and terrestrial cell defined by front-haul;c) a third link between the DBS cell and the terrestrial cell defined by backhaul.

10. The method as claimed in claim 8, wherein said frequency-carriers forming part of the fronthaul and backhaul are identical or different.

11. The method as claimed in claim 8, wherein said frequency-carriers forming a part of the23a) front-haul;b) back-haul; andc) carrier-aggregation and / or dual-connectivity based aerial communication linkare either identical or different.

12. The method as claimed in claim 1, further comprising modifying the aerial communicationlink through at-least one of:removing the terrestrial base station acting as a master base station in respect of thecarrier-aggregation and dual-connectivity based aerial communication link; andrestricting the carrier-aggregation and dual-connectivity in respect of frequency carriersconfigured for communication among one or more DBS and the selected UEs, wherein theplurality of DBS comprises master DBS and a secondary DBS.

13. A method for selecting preferred mobile stations for coverage in a 3GPP network, saidreceiving (step 102) from one or more UEs at least one feedback parameter pertaining toa Predicted Buffer in uplink (UL) and downlink (DL)a Minimum Guaranteed service based on QoS;calculating (step 104) a weighted-average based on said one or more parameter for eachof the plurality of UEs to determine at least one of a buffer expectation and a traffic-directionwith respect to each UE; andselecting (step 104) a set of UEs out of a plurality of UEs for availing an augmentednetwork connectivity based on at least one of:said buffer-expectation being above a threshold;24the traffic direction; anda number of UEs capable of being covered in a same zone of a cell.

14. A method for determining an Aerial Cell Trajectory in a 3GPP network, said methodreceiving (step 102) at least one parameter pertaining to a current network-traffic flowand mobility from a plurality of UEs, said at-least one parameter defined by at-least one of:selecting (step 104) a set of one or more UEs from said plurality of UE for availing anaugmented terrestrial network through one or more drone base station (DBS) -cell at least basedon one or more service-requirements of the UE determined from the at-least one feedbackparameter; anddetermining (step 106, 108) a position for at least one DBS-cell in the 3D space withrespect to the at least one selected set of UEs, said position defined by one or more of:a 2D coverage zone on ground having the maximum number of UEs;a maximum-weighted average of a buffer-requirement for each UE;an aerial scheduling period ( ASP ) with respect to the drone-cell determined inreal time based on the mobility and traffic-pattern of UEs.

15. A network node (gNB) for positioning of a Low Altitude Platform Station (LAPS) baseddrone-cells for supporting communication in a 3GPP network, said network-node comprising:a transceiver for receiving from a UE at least one feedback parameter pertaining to acurrent network-traffic flow and mobility in respect of the UE;a processor configured for :25drone base station (DBS)-cell based on service-requirement of the UE determined fromthe at-least one parameter;determining (step 106) a position for at least one DBS-cell with respect to the atleastone selected UE based on said at least one parameter to serve the selected UE; andthe DBS cell by deploying one or more Low Altitude Platform Station as a drone basestation (DBS) in accordance with said determined position of DBS-cell to therebyaugment a network connectivity of the selected UE.

16. A network node (gNB) node for selecting preferred mobile stations for coverage in a 3GPPnetwork, said network node (gNB) comprising:a transceiver for receiving (step 102) from one or more UEs at least one feedbackparameter pertaining to at-least one of:calculating (step 102) a weighted average based on said one or more parameterfor each of the plurality of UEs to determine at least one of a buffer expectation and atraffic-direction with respect to each UE; andaugmented network connectivity based on at least one of:2617. A network node (gNB) for determining an Aerial Cell Trajectory in a 3GPP network, saida transceiver for receiving (step 102) at least one parameter pertaining to a currentnetwork-traffic flow and mobility from a plurality of UEs, said at-least one parameter defined bya processor configured for:availing an augmented terrestrial network through one or more drone base station (DBS)-cell at least based on one or more service-requirements of the UE determined from theat-least one feedback parameter;determining (step 106) a position for at least one DBS-cell in the 3D space with