DESC:The following specification particularly describes and ascertains the nature of this invention and the manner in which it is to be performed:-

TECHNICAL FIELD
The embodiments herein generally relate to the field of multiantenna systems and more particularly to Beam-forming (BF) training in the multiantenna systems.

BACKGROUND
Development in communication technology has increased usage of wireless communication devices, effectively raising data traffic exponentially. Currently, multiple antenna systems are being used that provide increased data rates by means of spatial multiplexing techniques, decreased error and improved signal-to-noise ratios (SNRs)/ signal-to-interference-plus noise ratios (SINRs) by means of beamforming techniques. To handle huge volume of data traffic in future an underutilized spectrum in the millimeter-wave (mmWave) frequency bands (unlicensed 60 GHz band and the lower interference licensed 3GHz to 80 GHz band) has been identified for next generation (5G) communication solution. The mmWave frequencies facilitate the use of a large number of antenna elements in a compact form factor to synthesize highly directional beams corresponding to large array gains. Hence mmWave based communication system is basically implemented as a multiantenna system. Multiantenna systems including the mmWave based communication systems utilize appropriate beam forming scheme to focus a transmitted and/or received signal in a desired direction in order to overcome the unfavorable path loss at mmWave frequency bands.
Thus, estimating an optimal beam-pair between a transmitter and a receiver for communication (before data transmission) is an important process and referred as Beam-forming (BF) training stage. Typically, due to its flexibility and superior performance, digital beamforming is generally preferred in multi-antenna systems. However, for the mmWave based multiantenna system with very large antenna arrays, digital beamforming architecture becomes formidably complex and consumes substantial amount of power due to multiple Radio Frequency (RF) chains and Analog to Digital Converters (ADCs)/Digital to Analog Converters (DACs). Therefore, analog beamforming is considered as an alternative solution for the mmWave systems since it employs array processing at the RF level using only a set of analog phase shifters, limiting the number of RF chains and ADCs/DACs to one. The optimal beam-pair denotes directions of a transmit (Tx) beam and receive (Rx) beam directions which yield the best link quality, for example, good received Signal to Noise Ratio (SNR). The BF training plays a critical role in the performance of the beamforming operation since beamforming weights (spatial pre-coders/filters) are determined by the estimation of directions. However, in mm-Wave Multiple Input, Multiple Output (MIMO) communication systems where a large number of antennas are used, BF training becomes highly challenging due to multiple reasons. Firstly, the optimal beam pairs for communication are quite finer due to the high directionality, thereby making the BF training procedure more complex and burdensome. Secondly, owing to a very large path loss, it is difficult to estimate the channel state information prior to beamforming, as there is no information on the optimal direction.
In state of the art systems, this stage adds a significant burden on the communication in terms of time required for the BF training and computational complexity involved. A current solution employs spread spectrum which involves amplitude and phase components in the analog domain which can render complexity during implementation. Further, the current solution uses PN sequences which have variations in the amplitude leading to complexity and power inefficiency. Another existing solution uses a bidirectional protocol between the transmitter and the receiver using multiple exchanges of packets to determine the beam-former weights. However, the existing solution consumes more time to determine the beam-former weights and makes no usage of spreading sequences across beams.

OBJECT OF INVENTION
The principal object of the embodiments herein is to provide methods and multi antenna systems for Beam forming (BF) training, wherein the BF training is based on beam coding techniques utilizing a set of beam coding sequences.
Another object of the invention is to provide methods and multiantenna systems, wherein the BF training based on the beam coding techniques can be performed at a receiver or a transmitter of a communication entity in the multiantenna system.

SUMMARY
In view of the foregoing, an embodiment herein provides a method for Beam-forming (BF) training in a multiantenna system. The method comprises performing, by a first communication entity in the multiantenna system, the BF training at one of a first receiver and a first transmitter of the first communication entity. The BF training is based on beam coding technique utilizing a set of beam coding sequences with a unit weightage to a plurality of antenna weights corresponding to a plurality of antennas of the first communication entity. Further, the method comprises determining, by the first communication entity, an optimal Transmitter-Receiver (Tx-Rx) beam pair between the first communication entity and a second communication entity of the multiantenna system based on the BF training performed using one of a Direction of Arrival (DoA) at the first receiver and a Direction of Departure (DoD) at the first transmitter.
Embodiments further disclose a communication entity for Beam-forming (BF) training in a multiantenna system. The communication entity comprises a BF training unit further comprising a first transmitter and a first receiver. The BF training unit is configured to perform the BF training at one of the first receiver and the first transmitter. The BF training is based on beam coding technique utilizing a set of beam coding sequences with a unit weightage to a plurality of antenna weights corresponding to a plurality of antennas of the communication entity. Further, the BF training unit is configured to determine an optimal Transmitter-Receiver (Tx-Rx) beam pair between the first communication entity and a second communication entity of the multiantenna system based on the BF training performed using one of a Direction of Arrival (DoA) at the first receiver and a Direction of Departure (DoD) at the first transmitter.
These and other aspects of the embodiments herein will be better appreciated and understood when considered in conjunction with the following description and the accompanying drawings. It should be understood, however, that the following descriptions, while indicating embodiments and numerous specific details thereof, are given by way of illustration and not of limitation. Many changes and modifications may be made within the scope of the embodiments herein without departing from the spirit thereof, and the embodiments herein include all such modifications.

BRIEF DESCRIPTION OF FIGURES
The embodiments of this invention are illustrated in the accompanying drawings, throughout which like reference letters indicate corresponding parts in the various figures. The embodiments herein will be better understood from the following description with reference to the drawings, in which:
FIG. 1a and 1b illustrate an example multiantenna system for Beam-forming (BF) training, according to embodiments as disclosed herein;
FIG. 1c illustrates BF frame transmissions and a BF frame structure for the BF training, according to embodiments as disclosed herein
FIG. 2 illustrates a plurality of components of a communication entity for the BF training, according to embodiments as disclosed herein;
FIG. 3 is a flow diagram illustrating a method for the BF training in the multiantenna system, according to embodiments as disclosed herein;
FIG. 4 is a flow diagram illustrating a method for the BF training at a receiver of a communication entity in the mutliantenna system, according to embodiments as disclosed herein;
FIG. 5 illustrates implementation of the BF training performed at the receiver of the communication entity, according to embodiments as disclosed herein;
FIG. 6 is a flow diagram illustrating a method for the BF training at a transmitter of the communication entity, according to embodiments as disclosed herein; and
FIG. 7 illustrates implementation of the BF training performed at the transmitter of the communication entity, according to embodiments as disclosed herein.

DETAILED DESCRIPTION
The embodiments herein and the various features and advantageous details thereof are explained more fully with reference to the non-limiting embodiments that are illustrated in the accompanying drawings and detailed in the following description. Descriptions of well-known components and processing techniques are omitted so as to not unnecessarily obscure the embodiments herein. The examples used herein are intended merely to facilitate an understanding of ways in which the embodiments herein may be practiced and to further enable those of skill in the art to practice the embodiments herein. Accordingly, the examples should not be construed as limiting the scope of the embodiments herein.
The embodiments herein achieve methods and multi antenna systems for Beam-forming (BF) training, wherein the BF training is based on beam coding techniques utilizing a set of beam coding sequences. The BF training is a bidirectional training procedure for identifying an optimal Transmitter (Tx)-Receiver (Rx) beam pair between two communication entities in the multiantenna system, for example, a first communication entity and a second communication entity. During the BF training one of the communication entities transmits training beam coding sequences in BF frames (packets) along a plurality of pre-defined Tx beams while the other communication entity receives the BF frames on a plurality of Rx beams.
In an embodiment, the BF training between the first communication entity and the second communication entity can be performed at a receiver (first receiver) of the first communication entity for the beam coding sequences transmitted in the BF frames by a transmitter (second transmitter) of the second communication entity. The beam coding sequences used by the second transmitter are known apriori to the first receiver. The BF training at the receiver of the communication entity is based on the estimation of Direction of Arrival (DoA) to identify the optimal Tx-Rx pair.
In an embodiment, the BF training between the first communication entity and the second communication entity can be performed at a transmitter (first transmitter) of the first communication entity by transmitting the beam coding sequences in the BF frames towards a receiver (second receiver) of the second communication entity. The first transmitter apriori knows that direction of reception of the second receiver is tuned to k-th beam among the plurality of Rx beams of the second receiver. The BF training when performed at the transmitter is based on the estimation of Direction of Departure (DoD) to identify the optimal Tx-Rx pair.
The terms first and second used in conjunction with the communication entity, the receiver and the transmitter are merely for distinguishing communication entities or components of the communication entities from one another.
In both the embodiments, beam coding technique used is based on orthogonal codes containing beam coding sequences with the Constant Amplitude Zero Auto-Correlation (CAZAC) that make is possible to set antenna weights corresponding to a plurality of antennas (Tx antennas and Rx antennas) of the communication entity to have unit magnitude. Thus, unit magnitude requirement enables the antenna weights to be implemented using only analog phase shifters. Therefore, the BF training can be implemented using a single Radio Frequency (RF)-chain and a set of analog phase shifters, effectively reducing system complexity of the multiantenna BF training. Further, the BF training proposed involves flat power variations and hence only single value of Automatic Gain Control (AGC) needs to be communicated to the receiving communication entity through an AGC field of the BF frame. This reduces inclusion of plurality of AGC values in the AGC field of each BF frame, effectively reducing size of the transmitted BF frame (packet).
Thus, utilization of beam coding sequences for the BF training proposed provides faster, low-complexity implementation solution making it useful for any multiantenna systems such as next generation of mmWave systems in Wireless Local Area Network LAN, 5G cellular communications and the like.
In an embodiment, the communication entity participating in the BF training can be a User Equipment such as a mobile phone, a tablet, a palmtop, a Base Station (BS) and the like with beam forming capability.
Referring now to the drawings, and more particularly to FIGS. 1 through 7, where similar reference characters denote corresponding features consistently throughout the figures, there are shown embodiments.
FIG. 1a and 1b illustrate an example multiantenna system 100 for the BF training, according to embodiments as disclosed herein. In the multiantenna system 100 of FIG. 1a and FIG. 1b, a transmitter of a communication entity (a first communication entity 102 or a second communication entity 104) is equipped with a linear array of Nt transmit antennas with Nt beams, and a receiver is equipped with an array of Nr receive antennas with Nb receive beams (wherein, assumption is Nr = Nb for simplicity of understanding). In an embodiment herein, it is assumed that the antenna arrays have Uniform Linear Array (ULA) configurations, where the antenna elements are uniformly spaced. Embodiments herein further assume critically spaced arrays (arrays with half-wavelength antenna spacing), where d=?/2. Even though the description specifically mentions ULAs, the BF training proposed can be extended to planar arrays as well.
The FIG. 1a illustrates the multiantenna system 100 with the first communication entity 102 and the second communication entity 104 involved in process of identifying the optimal Tx-Rx beam pair through the BF training, when the BF training is performed at the first receiver of the first communication entity 102. For the BF training at the first receiver, the second communication entity 104 can be configured to begin transmitting BF frames from the second transmitter. The second transmitter performs beam-sweeping by steering the second transmitter through Nt beams positions. Further, in each Tx beam among the Nt Tx beams, the second transmitter transmits a BF frame with each BF frame (BF frame 1, ….BF frame l,..... BF frame Nt) comprising beam coding sequence for the Nb Rx beams, wherein the beam coding sequences used by the second transmitter are known apriori to the first receiver of the first communication entity. With the beam coding sequences known apriori, the first receiver can be configured to receive the transmitted BF frames on each Rx beam among the Nb Rx beams of the first receiver. The received signal (BF frames), which are in element space are transformed to a beam space by beam transformation. Further, a beam encoding is performed on the received signal in the beam space and then analyzed for estimating the DoA using a correlator with a pre-defined correlator Metrix. Once all the received Nt BF frames are analyzed, the TxRx beam pair that maximizes the correlator Metrix is identified as the optimal TxRx beam pair (here Tx1Rx2 106) for communication between the first communication entity 102 and the second communication entity 104.
The BF training procedure at the first receiver is explained in conjunction with FIG. 4 and FIG. 5
The FIG. 1b illustrates the multiantenna system 100 with the first communication entity 102 and the second communication entity 104 involved in process of identifying the optimal Tx-Rx beam pair through BF training, where the BF training is performed at the first transmitter of the first communication entity 102. During the BF training, the first transmitter can be configured to transmit a signal (plurality of BF frames) through a corresponding plurality of Tx beams (Tx1, Tx2, Tx3 and so on towards the second receiver of the second communication entity 104. The first transmitter apriori knows that direction of reception of the second receiver of the second communication entity 104 is tuned to k-th beam among the plurality of Rx beams of the second receiver. The signal (BF frames to be transmitted) are transformed into the beam space providing spatial differentiation by beam encoding, Further, signal is converted from beam space to the element space through inverse beam transformation before transmitting the signal ( BF frames) through the Nt antenna arrays over Nt Tx beams at the first transmitter. Further, the second receiver of the second communication entity 104 on receiving the BF frames can be configured to determine a correlation Metrix, in a baseband domain of the second receiver, for the beam coding sequences received from the first transmitter. From the correlation Metrix the DoD is estimated and the optimal Tx-Rx pair (here, Tx2Rx1 108) among the plurality of Tx beams and the plurality of Rx beams is identified such that the optimal Tx-Rx pair maximizes the correlation Metrix.
FIG. 1c illustrates the BF frame transmissions and a BF frame structure for the BF training, according to embodiments as disclosed herein. The figure depicts BF frame transmissions 112 comprising BF frame1, ….BF frame l,..... BF frame Nt transmitted by the second transmitter of the second communication entity 104 over corresponding Tx beams 110 comprising Tx1, Tx2…., Txl,….TxNt. Each BF frame consists of a Preamble field, Channel estimation (CE) field, the AGC (automatic gain control) field, followed by a BF code sequence field 114 with a set of beam coding sequence comprising Nb number of codes. Each code among the Nb codes is associated with every Rx beam of the Nb Rx beams of the first receiver’s beam space. As depicted in figure the beam coding sequences in the BF code sequence field include codes BF TRN code 1, BF TRN code 2 to BR TRN code Nb. As can be seen there is only one AGC field in each BF frame due to flat power variations across all antennas that effectively reduces BF frame size as only one AGC value need to be transmitted.
FIG. 2 illustrates a plurality of components of the communication entity (first communication entity 102) for the BF training, according to embodiments as disclosed herein. Referring to figure 2, the first communication entity 102 is illustrated in accordance with an embodiment of the present subject matter. In an embodiment, the communication entity 102 may include a processor 202, an input/output (I/O) interface 204 and a memory module 206. The I/O interface 204 enables the communication entity 102 to communicate with other entities of the multi antenna system 100 such as communication entity 104 for uplink or downlink using beam forming through a transmitter 210 and a receiver 212 of the BF training unit 208. The communication entity may also utilize other communication technologies such as cellular networks, Wi-Fi networks, and the like. Further, the receiver 208 also comprises a correlator unit 214 configured to estimate the DoA to identify the optimal TxRx pair, when the BF training is implemented at the receiver 212 (first receiver). However, when the BF training is implemented at the transmitter 210 (first transmitter) a correlator unit at the second communication entity 104 estimates the DoD of the first transmitter (transmitter 210) to identify the optimal pair Tx-Rx pair. The transmitter 210 and the receiver 212 are configured to perform functions as explained in conjunction with FIG. 1 and FIG. 3, through FIG.7 and not repeated for brevity.
The FIG. 2 is described in conjunction with the first communication entity 102, however, it can be understood that same modules and components are present in the other communication entities, such as communication entity 104, of the multiantenna system 100 that participate in BF training for communication through beam forming communication technology.
FIG. 3 is a flow diagram illustrating a method for the BF training in the mutliantenna system, according to embodiments as disclosed herein. At step 302, the method 300 includes allowing the BF training unit 208 of the first communication entity 102 to perform the BF training the first receiver or the first transmitter of the first communication entity 102. The steps for BF training at the first receiver are explained in conjunction with FIG. 4 and FIG. 5 and the steps for BF training at the first transmitter are explained in conjunction with FIG. 6 and FIG 7. The BF training is based on beam coding technique utilizing the set of beam coding sequences. The method 300 enables the first receiver to assign unit weightage the plurality of antennas of the first communication entity. The plurality of antenna weights with weightage of unit magnitude are generated from plurality of analog phase shifters in the first communication entity, thus reducing the system complexity of the communication entities in the multiantenna system 100. At step 304, the method 300 allows the BF training unit 208 to determine the optimal Tx-Rx beam pair between the first communication entity and a second communication entity of the multiantenna system 100 based on the BF training performed using the DoA at the first receiver or the DoD at the first transmitter.
The various actions in method 300 may be performed in the order presented, in a different order or simultaneously. Further, in some embodiments, some actions listed in FIG. 3 may be omitted.
FIG. 4 is a flow diagram illustrating a method for the BF training at the receiver of the communication entity, according to embodiments as disclosed herein. At step 402, the method 400 includes allowing the BF training unit 208 of the first communication entity 102 to receive the plurality of BF frames on the plurality of Rx beams (Nb) of the receiver 212 for every BF frame transmitted on each Tx beam among the plurality of Tx beams (Nt) of the second transmitter of the second communication entity 104. The transmitted BF frame comprises the preamble field, the AGC field and the BF code sequence field with the set of beam coding sequences (Nb beam coding sequences). The beam coding sequences used by the second transmitter are known apriori to the first receiver of the first communication entity. The beam coding technique utilized comprises beam coding step involving the beam space transformation and beam encoding.
At step 404, the method 400 includes allowing the BF training unit 208 to perform the beam space transformation on the received set of beam coding sequences in the plurality of BF frames. The beam space transformation is performed in the Radio Frequency (RF) stage of the first receiver to convert the received set of beam coding sequences from the element space to the beam space for every BF frame transmitted by each Tx beam.
At step 406, the method 400 includes allowing the BF training unit 208 to perform the beam encoding on the received set of beam coding sequences in the beam space for every BF frame transmitted by each Tx beam. The beam encoding is performed in the Radio Frequency (RF) stage of the first receiver, wherein the beam encoding is a linear operation generating a resultant signal for every received set of beam coding sequences. The resultant signal is a sum the received set of beam coding sequences in the beam space after applying a corresponding antenna weight from the plurality of antenna weights.
At step 408, the method 400 includes allowing the BF training unit 208 to determine correlation, using the correlator unit 214 in the baseband domain, for the encoded beam coding sequences to estimate the DoA for the identifying the optimal Tx-Rx pair. The optimal Tx-RX pair maximizes the pre-defined correlation Metrix. The method 400 is explained with implementation details and mathematical analysis in conjunction with FIG. 5.
The various actions in method 400 may be performed in the order presented, in a different order or simultaneously. Further, in some embodiments, some actions listed in FIG. 4 may be omitted.
FIG. 5 illustrates implementation of the BF training performed at the receiver of the communication entity, according to embodiments as disclosed herein. The figure depicts the implementation details along with the mathematical model and analysis of the first receiver unit of the first communication unit 102.
Let x^((e)) (t) be the N_t ×1 dimensional vector transmitted from the transmitter array of the second transmitter at epoch . The N_r ×1 dimensional received signal vector at the receiver antenna array of receiver 214 ( first receiver) is given by equation 1 below:
y^((e)) (t)=Hx^((e)) (t)+z^((e)) (t) (1)
Where, z^((e)) (t) is N_r ×1 dimensional vector of spatially and temporally uncorrelated AWGN noise process withz^((e)) (t)~ N(0,s^2 I_(N_r )) . Matrix is a N_r×N_t channel matrix representing the coupling and channel gains between the transmitter and receiver antenna arrays. The representation of the received signal in equation 1 is referred to as the element space 502 representation while representation of the received signal in beam space is called the beamspace 508 representation. To derive the beam space representation, let e_N (O) be the steering vector of an N-element ULA ( second transmitter) in the directional cosine (O?cos?(?). For the critically spaced antennas case e_N (O) is given by
e_N (O)=1/vN [1,exp?(-jpO),….,exp?(-jp(N-1)O) ]^T.
Let D_N represent an N×N unitary DFT matrix whose elements are given by
[D_N ]_(k,l)=1/vN exp?(-j 2pkl/N)
Thus, D_N=[e_N (0),...,e_N (2l/N),...,e_N ((2(N-1))/N) ]
For a critically spaced ULA with N antennas, it can be shown that the columns of the DFT matrix D_N form an orthogonal basis. Let? D?_(N_t ) and D_(N_r )be the DFT matrices associated with the N_t- dimensional transmit beamspace and the N_r-dimensional receiver beamspace respectively. Then, an equivalent representation of the channel matrix in the beamspace (angular domain) is obtained as
H_b=D_(N_r)^† HD_(N_t )
Matrix H_b is a N_r×N_t matrix, with elements h_(i,j)^((b) )=[H_b ]_(i,j) . It can be observed that H_b is a 2D DFT of and thus a completely equivalent representation of .
Let x^((b) ) (t)=D_(N_t ) x^((e) ) (t) and y^((b) ) (t)=D_(N_r ) y^((e) ) (t), where x^((b) ) (t) and y^((b) ) (t) respectively, are the transmitted and the received beamspace signal vectors. Thus, the received signal of equation 1 is represented in beamspace as in equation 2 below:
y^((b) ) (t)=H_b x^((b) ) (t)+z^((b) ) (t) (2)
Where z^((b) ) (t) is an Additive white Gaussian noise (AWGN) vector in the beamspace. Observe that, due to white Gaussian nature of the element-space noise vector z^((e)) (t) and the unitary transformation, z^((b)) (t)~ N(0,s^2 I_(N_r )).
Without the loss of generality, embodiments herein assume the dimension of receiver beamspace N_B=N_r. Let C?{c_0,…, c_(N_B-1)} be the orthogonal code used for the beam coding, where c_k=[c_k (0),…, c_k (N_s-1) ],k=0,...,N_B the sequence is assigned to receive beam? B?_k. N_S is the length of the beam-codes or beam coding sequences. Due to orthogonality, c_i^† c_j=d_(i,j), where d_(i,j) is the Dirac delta function. For the sake of simplicity, embodiments herein assume that the training symbols and the beam codes have the same length. Let Y_e?[y^((e) ) (0)...,y^((e) ) (N_s-1) ] and Y_b?[y^((b) ) (0)...,y^((b) ) (N_s-1) ] be the sequence of training symbols in an element-space 502 and a beamspace 506 respectively, and y^((e) ) (t) and y^((b) ) (t) are snap-shot, at time t, of element space 502 and beamspace 506 vectors, respectively.
Conceptually, beam coding is made of two phases the beamspace transformation 504 and beam encoding 508. In the beamspace transformation phase, the received element-space vector sequence,Y_e, is transformed into the corresponding beamspace vector sequence Y_b. That is given by equation 3 below:
y^((b) ) (t)=D_(N_r ) y^((e) ) (t),t=0,...,N_S-1 (3)
In the beam encoding phase 508, the received vector sequence in the beamspace, Y_b, is encoded using orthogonal codes. For encoding of signals in the beamspace, the sequence r_k=[r_k (0),…, r_k (N_s-1) ], is obtained for each , which is generated by chip-wise multiplication of c_k , with the corresponding beamspace signal sequence y_k^((b) ). In other words,
?r_k=c_k?y?_k^((b) ),r_k (t)=c_k (t) y_k^((b) ) (t)
Where, denotes the Hadamard product (element-wise product) operation. Next, the sequences {r_k }_(k=0)^(N_S-1)k=0 are added chip-wise to generate the beam coded sequence, u=[u(0),…, u(N_s-1) ]. That is as in equation 4a,
u=?_(k=0)^(N_B-1)¦??c_k?y?_k^((b) ),? u(t)=?_(t=0)^(N_s-1)¦?c_k (t)y_k^((b) ) (t)? (4a)

Implementation of beam coding: Next, embodiments herein show that the beam coding is performed by the linear operation on the element-space vector as in equation 4b.
u(t)=?_(n=0)^(N_B-1)¦??_n (t) ? y_n^((e) ) (t),t=0,...,N_B-1 (4b)

Where, ?_n (t) are the antenna weights.
Thus, noting that y_k^((b) ) (t)=[D_(N_B ) y^((e) ) (t) ]_k, and from (4a),
u(t)=?_(k=0)^(N_B-1)¦?c_k (t)(?_(n=0)^(N_B-1)¦?1/v(N_B ) e?^(-j2pk/N_B n) y_n^((e) ) (t) ) ?
=?_(n=0)^(N_B-1)¦(1/v(N_B ) ?_(k=0)^(N_B-1)¦?c_k (t) e^(-j2pk/N_B n) ?) y_n^((e) ) (t)
=?_(n=0)^(N_B-1)¦??_n (t) ? y_n^((e) ) (t) (5)
It can be observe that
?(t)=[?_0 (t)...,?_(N_B-1) (t) ]^T=D_(N_B ) c(t).
BF training by receiver beam coding: Let the transmitter beamspace be divided into N_T orthogonal beams/sectors. Let receiver beamspace be divided into N_B beams/sectors. Without the loss of generality, embodiments herein assume that N_B=N_r. The following steps are involved in the BF training procedure:
- Transmission of BF training sequence from the second transmitter and received at element space 502 of the first receiver
- Beam-encoding at the first receiver 508
- DoA Estimation by beam decoding at the first receiver through ADC 512, the correlator unit 214 Transmission of BF training sequence: During one BF training frame, the second transmitter transmits along only one given beam/sector. During the BF training duration, the second transmitter transmits a BF training frame, which contains a training sequence (beam coding sequence) of N_S symbols. This sequence is known apriori to the first receiver. For illustration, embodiments herein assume that the second transmitter transmits through its first beam/sector. Thus, only the sequence x_1^((b) )is non-zero and x_j^((b) )=0,j?1. The transmitted vector of training symbols is given by
?x^((e) ) (t)=D?_(N_t ) [x_1^((b) ) (t),0,...,0]^T
Beam-encoding at the receiver: The receiver receives the sequence of N_B training symbols simultaneously from all N_B beams, and performs beam encoding 508 as depicted. As shown in the transmission of BF training sequence, this operation is performed by the linear combing operation as in equation 4b above and repeated below:
u(t)=?_(n=0)^(N_B-1)¦??_n (t) ? y_n^((e) ) (t),t=0,...,N_B-1 (4b)
Where ?_n (t) is the weight of the n-th antenna for the t-th symbol.
DoA Estimation by beam-decoding at the receiver: The beam coded signal u (t) from the antenna array output is sampled and passed through an Analog-to-Digital converter (ADC) 510 for digital processing. Let the baseband version of beam coded received sequence u(t) is given by the vector, u=[u(0),…, u(N_B-1) ]. The DoA estimation block perform correlation of the received signal using the correlator unit 214 with all the beam codes and computes the metric. That is, as given in equation
??_k=c?_k^† u,k=0,…, N_B-1, (6)
Where ?_k is the correlation metric of the received signal in the direction of B_k. The absolute value of the correlation metric, |?_k | essentially captures the energy of the received signal in the beam B_k. Indeed, it can be seen from (4) that
??_k=c?_k^† (?_(j=0)^(N_B-1)¦?c_j?y?_j^((b) ) ), (7)
?=c?_k^† (?_(j=0)^(N_B-1)¦?c_j?? (h_(1,j)^((b) ) x_1^((b) )+z_j^((b) ) ) ), (8)
Where, the lase equality follows from equation (2). Since the training sequence x_1^((b) )1 is known to the first receiver apriori, without loss of generality, it can be assumed that x_1^((b) )=[1,...,1]. Thus,
?_k=h_(1,k)^((b) )+(z_k ) ~ , (9)
Where, (z_k ) ~~ N(0,s^2).
Thus, ?_k can be used as a statistic of the channel in the direction of B_k. Thus, the candidate DoA 512 can be estimated by applying a threshold based energy detection:
B_DoA={B_k:|?_k |>G}
Uniform amplitude property of antenna weights: In the following, it can be proved that under certain circumstances, the antenna weights have uniform magnitude, i.e. |?_n (t) |=1 ?n,t. This property ensures that the antenna weights can be implemented only with the analog phase-shifters. The objective is to choose the orthogonal code [c_0,…, c_(N_B-1) ] such that c_k^† c_j=d_(k,j) ?k,j?{0,...,N_B-1 }, with the condition that |?_n (t) |=1 ?n. To this end, embodiments herein propose codes based on Zadoff-Chu sequences which have the CAZAC property.
Theorem 1: Let the length of the orthogonal sequence, N_S=N_B. Further, let orthogonal code C=[c_0,…, c_(N_B-1) ] be chosen such that the chips
c_k (t)="exp" (jp (k-1)^2/N_B +jp/4),?k,t=0,...,N_B-1.
Then, the antenna weights have unity amplitude. Specifically, the antenna weights are given by the phasors as in equation 10 below:
?_n (t)="exp" (jp (n(n+2t))/N_B ) (10)
Proof: First, it can be noted that C is generated such that c_k’s are obtained by the cyclic-shift of c_0=[c_0 (0),…, c_0 (N_B-1) ] with shifts. Then the proof follows from the CAZAC property of c_0.
FIG. 6 is a flow diagram illustrating a method for the BF training at the transmitter of the communication entity, according to embodiments as disclosed herein. At step 602, the method 600 includes allowing the BF training unit 208 to perform the beam encoding, in a Radio Frequency (RF) stage of the first transmitter, on the time domain signal for the BF training to generate the set of beam coding sequences in the beam space. The beam coding sequences are generated by applying corresponding antenna weight from the plurality of antenna weights (Nt), wherein the beam encoding is a linear operation.
At step 604, the method 600 includes allowing the BF training unit 208 to performing the inverse beam space transformation, in the RF stage of the first transmitter, on the set of beam coding sequences. This generates plurality of BF frames in the element space from the beam space.
At step 606, the method 600 includes allowing the BF training unit 208 to transmit the plurality of BF frames in the element space through corresponding plurality of Tx beams (Nt) towards the second receiver of the second communication entity. The first transmitter apriori knows that direction of reception of the second receiver of the second communication entity 104 is tuned to k-th beam among the plurality of Rx beams of the second receiver.
At step 606, the method 600 includes allowing the BF training unit of the second communication entity 104 to determine correlation, by a correlator unit in the baseband domain of the second receiver, for the beam coding sequences received from the first transmitter to estimate the DoD. Once the DOD is estimated for every transmitted BF frame, the second receiver identifies the optimal Tx-Rx pair that maximizes the correlation Metrix.
The various actions in method 600 may be performed in the order presented, in a different order or simultaneously. Further, in some embodiments, some actions listed in FIG. 6 may be omitted. The method 600 is explained with implementation details and mathematical analysis in conjunction with FIG. 7
FIG. 7 illustrates implementation of the BF training performed at the transmitter of the communication entity, according to embodiments as disclosed herein. The description of the mathematical system model for the first transmitter 210 of the first communication entity 102 is same as explained for the first receiver in FIG. 5 through equation 1 and equation 2. The implementation of entire beam coding procedure at the first transmitter of the first communication entity 102 can be simplified to a linear combining operation in the element space 712. Without the loss of generality, it is assumed the dimension of second receiver beamspace N_B=N_t. Let C?{c_0,…, c_(N_B-1)} be the orthogonal code used for the beam coding, where c_k=[c_k (0),…, c_k (N_s-1) ]^T,k=0,...,N_B-1 is the sequence assigned to receive beam B_k. N_S is the length of the beam codes. Due to orthogonality, have c_k^† c_j=d_(k,j), where d_(k,j) is the Dirac delta function. The beam coding is made of two phases: beam encoding 706 and inverse beamspace transformation 710. In beam encoding, the training sequence v received from the base band stage 702 through the DAC 704 are encoded by beam codes or beam sequence codes at the beam encoding stage 706. In particular, the signal component in beam B_k is encoded using the sequence,c_k. This is achieved by using the code waveform.
g_Ck= ?_(l=0)^(N_s-1)¦?c_k [l] g_(T_c ) (t-?lT?_c ) ?
Where,

g_(T_c )= {¦(1/T_c @0 otherwise)¦
It can be observed that,
g_Ck (t)=c_k (l),t?[lT_c,(l+1) T_c ).
The signal in beamspace, ,t?[0,(N_S-1) T_c )] is given by
x_k^((b) ) (t)=g_Ck (t)v(t) (11)
=v(t) ?_(l=0)^(N_s-1)¦?c_k [l] g_(T_c ) (t-?lT?_c ) ? (12)
In the inverse beamspace transformation phase, the beamspace signal vector is converted into element space. That is,
x^((b) ) (t)=D_(N_t ) x^((b) ) (t),t=0,...,N_S-1 (13)
Implementation aspect of the beam coding. From equation 12 and equation 13, it is clear
?x^((e) ) (t)=v(t)D?_(N_t ) ?_(l=0)^(N_s-1)¦?c_k [l] g_(T_c ) (t-?lT?_c ) ? (14)
For t?[lT_c,(l+1) T_c ), the piece-wise representation of x^((e) ) (t) is given by
?x^((e) ) (t)=v(t)D?_(N_t ) c[l],t?[lT_c,(l+1) T_c ).
Therefore, for the signal at the output of n-th antenna (element space signal),
x_n^((e) ) (t)=v(t)(1/v(N_B ) ?_(k=0)^(N_B-1)¦?c_k [l] e^(j2pn/N_B k) ?) (15)
By setting ?_n [l]=1/v(N_B ) ?_(k=0)^(N_B-1)¦?c_k [l]e^(j2pn/N_B k) ?, it can be obtained as
x_n^((e) ) (t)= ?_n [l]v(t),t?[lT_c,(l+1) T_c ) (16)
It can be observed that ?_n [l] is the antenna weight of n-th antenna during the time interval [lT_c,(l+1) T_c].
DoD estimation at the receiver: Here, it is assumed that the direction of reception of the receiver is tuned to k-th beam. Note that when N_r=1 , the reception degenerates to omnidirectional reception. In order to determine the DoD, channel conditions for Tx-beam B_m , are estimated by calculating the statistic
??_m=c?_m^† y,m=0,…, N_B-1, (17)
Where, y=[y_k^((b) ) (0),…, y_k^((b) ) (N_S-1) ]^T is the vector of received signal samples. Substituting equation (12) in equation (2), it is observed that

y^((b) ) (t)=H_b ?_(l=0)^(N_s-1)¦?c_k [l] g_(T_c ) (t-?lT?_c ) ?+z^((b) ) (t) (18)

The method examines the signal received in k-th beam, which is given by
y_k^((b) ) (t)=h_k^(b)T (t)?_(l=0)^(N_s-1)¦?c_k [l] g_(T_c ) (t-?lT?_c ) ?+z_k^((b) ) (t) (19)

Obtaining the sampled representation for t=?lT?_c, by a ADC 714 as given by
y_k^((b) ) [l]=h_k^(b)T (t)c[l]+z_k^((b) ) (t),l=0,…, N_S-1 (20)

Therefore,
y=[¦(?_(j=0)^(N_B-1)¦?h_(k,j )^((b) ) c_j [0] ?@.@.@.@?_(j=0)^(N_B-1)¦?h_(k,j )^((b) ) c_j [l] ?@.@.@.@?_(j=0)^(N_B-1)¦?h_(k,j )^((b) ) c_j [N_S-1] ?)]+z_k^((b) ) (21)

Substituting for y in (17), it can be obtained as
?_m=?_(n=0)^(N_S-1)¦(?_(k=0)^(N_B-1)¦?h_(k,j )^((b) ) c_j [l] c_m^* [l] ?) +c_m^* [l] z_k [l]

=?_(j=0)^(N_B-1)¦?h_(k,j )^((b) ) ?_(l=0)^(N_S-1)¦?c_j [l] c_m^* [l] ??+z ~_k
=?_(j=0)^(N_B-1)¦?h_(k,j )^((b) ) d_(j,m) ?+z ~_k
=h_(k,m )^((b) )+z ~_k (22)

Where, z ~_k~ N(0,s^2). Thus, ?_m can be used as a statistic to estimate the link quality corresponding to the Tx beam direction B_m. For example, the set of beams that correspond to the DoD (considering the channel with multipaths) can be obtained by a simple thresholding B={B_m:|?_m |^2=G} 718 at a correlator 716 of the second receiver, where is an appropriately chosen threshold.
The unity amplitude property of antenna weights for the first transmitter can be explained with the same proof provided with support of equation 10 for the unity amplitude property of antenna weights for the first receiver in FIG. 5.
The embodiments disclosed herein can be implemented through at least one software program running on at least one hardware device and performing network management functions to control the network elements. The network elements shown in FIG. 1 through FIG. 7 include blocks which can be at least one of a hardware device, or a combination of hardware device and software module.
The foregoing description of the specific embodiments will so fully reveal the general nature of the embodiments herein that others can, by applying current knowledge, readily modify and/or adapt for various applications such specific embodiments without departing from the generic concept, and, therefore, such adaptations and modifications should and are intended to be comprehended within the meaning and range of equivalents of the disclosed embodiments. It is to be understood that the phraseology or terminology employed herein is for the purpose of description and not of limitation. Therefore, while the embodiments herein have been described in terms of preferred embodiments, those skilled in the art will recognize that the embodiments herein can be practiced with modification within the spirit and scope of the embodiments as described herein.
,CLAIMS:STATEMENT OF CLAIMS
We claim:

1. A method for Beam-forming (BF) training in a multiantenna system, the method comprising:
performing, by a first communication entity in the multiantenna system, the BF training at one of a first receiver and a first transmitter of the first communication entity, wherein the BF training is based on beam coding technique utilizing a set of beam coding sequences with a unit weightage to a plurality of antenna weights corresponding to a plurality of antennas of the first communication entity; and
determining, by the first communication entity, an optimal Transmitter-Receiver (Tx-Rx) beam pair between the first communication entity and a second communication entity of the multiantenna system based on the BF training performed using one of a Direction of Arrival (DoA) at the first receiver and a Direction of Departure (DoD) at the first transmitter.
2. The method as claimed in claim 1, wherein the plurality of antenna weights with weightage of unit magnitude are generated from a plurality of analog phase shifters in the first communication entity.
3. The method as claimed in claim 1, wherein performing the BF training at the first receiver comprises:
receiving a plurality of BF frames on a plurality of Rx beams for every BF frame transmitted on each Tx beam among a plurality of Tx beams of a second transmitter of the second communication entity, wherein the transmitted BF frame comprises a preamble field, an Automatic Gain Control field and a BF code sequence field with the set of beam coding sequences;
performing a beam space transformation , in a Radio Frequency (RF) stage of the first receiver, on the received set of beam coding sequences in the plurality of BF frames to convert the received set of beam coding sequences from an element space to a beam space for every BF frame transmitted by each Tx beam;
performing a beam encoding, in the RF domain of the first receiver, on the received set of beam coding sequences in the beam space for every BF frame transmitted by each Tx beam, wherein the beam encoding is a linear operation generating a resultant signal for every received set of beam coding sequences by summing the received set of beam coding sequences in the beam space after applying a corresponding antenna weight from the plurality of antenna weights; and
determining a correlation, in a baseband domain, among the encoded beam coding sequences to estimate the DoA for the identifying the optimal Tx-Rx pair among the plurality of Tx beams and the plurality of Rx beams, wherein the optimal Tx-RX pair maximizes a pre-defined correlation Metrix.
4. The method as claimed in claim 1, wherein performing the BF training at the first transmitter comprises:
performing a beam encoding, in a Radio Frequency (RF) stage of the first transmitter, on a time domain signal for the BF training to generate a set of beam coding sequences in a beam space by applying a corresponding antenna weight from the plurality of antenna weights, wherein the beam encoding is a linear operation;
performing an inverse beam space transformation, in the RF stage of the first transmitter, on the set of beam coding sequences to generate a plurality of BF frames in an element space from the beam space; and
transmitting, the plurality of BF frames in the element space through a corresponding plurality of Tx beams towards a second receiver of the second communication entity.
5. The method as claimed in claim 4, wherein the method comprises, determining a correlation Metrix, in a baseband domain of the second receiver, for the beam coding sequences received from the first transmitter to estimate the DoD for the identifying the optimal Tx-Rx pair among the plurality of Tx beams and the plurality of Rx beams, wherein the optimal Tx-RX pair maximizes the correlation Metrix.
6. A communication entity for Beam-forming (BF) training in a multiantenna system, the communication entity comprises a BF training unit comprising configured to
performing the BF training at one of a first receiver and a first transmitter in the BF training unit, wherein the BF training is based on beam coding technique utilizing a set of beam coding sequences with a unit weightage to a plurality of antenna weights corresponding to a plurality of antennas of the communication entity; and
determine an optimal Transmitter-Receiver (Tx-Rx) beam pair between the communication entity and a second communication entity of the multiantenna system based on the BF training performed using one of a Direction of Arrival (DoA) at the first receiver and a Direction of Departure (DoD) at the first transmitter.
7. The communication entity as claimed in claim 6, wherein the plurality of antenna weights with weightage of unit magnitude are generated from a plurality of analog phase shifters in the communication entity.
8. The communication entity as claimed in claim 6, wherein the BF training unit is configured to perform the BF training at the first receiver by:
receiving a plurality of BF frames on a plurality of Rx beams for every BF frame transmitted on each Tx beam among a plurality of Tx beams of a second transmitter of the second communication entity, wherein the transmitted BF frame comprises a preamble field, an Automatic Gain Control field and a BF code sequence field with the set of beam coding sequences;
performing a beam space transformation , in a Radio Frequency (RF) stage of the first receiver, on the received set of beam coding sequences in the plurality of BF frames to convert the received set of beam coding sequences from an element space to a beam space for every BF frame transmitted by each Tx beam;
performing a beam encoding, in the RF domain of the first receiver, on the received set of beam coding sequences in the beam space for every BF frame transmitted by each Tx beam, wherein the beam encoding is a linear operation generating a resultant signal for every received set of beam coding sequences by summing the received set of beam coding sequences in the beam space after applying a corresponding antenna weight from the plurality of antenna weights; and
determining a correlation, in a baseband domain, among the encoded beam coding sequences to estimate the DoA for the identifying the optimal Tx-Rx pair among the plurality of Tx beams and the plurality of Rx beams, wherein the optimal Tx-RX pair maximizes a pre-defined correlation Metrix.
9. The communication entity as claimed in claim 6, wherein the BF training unit is configured to perform the BF training at first transmitter by:
performing a beam encoding, in a Radio Frequency (RF) stage of the first transmitter, on a time domain signal for the BF training to generate a set of beam coding sequences in a beam space by applying a corresponding antenna weight from the plurality of antenna weights, wherein the beam encoding is a linear operation;
performing an inverse beam space transformation, in the RF stage of the first transmitter, on the set of beam coding sequences to generate a plurality of BF frames in an element space from the beam space; and
transmitting, the plurality of BF frames in the element space through a corresponding plurality of Tx beams towards a second receiver of the second communication entity.
10. The communication entity as claimed in claim 9, wherein the second receiver is configured to determine a correlation Metrix, in a baseband domain of the second receiver, for the beam coding sequences received from the first transmitter to estimate the DoD for the identifying the optimal Tx-Rx pair among the plurality of Tx beams and the plurality of Rx beams, wherein the optimal Tx-RX pair maximizes the correlation Metrix.
Dated this 21st of September 2016
Signature:
Name of the Signatory: Dr. Kalyan Chakravarthy