A shape index modulation transmitter for a multi-mode orbital angular momentum (OAM) system

The shape index modulation transmitter in OAM systems addresses the challenge of activating high-reliability OAM modes by processing binary sequences to select optimal OAM mode combinations, resulting in improved bit error rates and spectral efficiency.

WO2024237853A1PCT designated stage expired Publication Date: 2024-11-21NANYANG TECH UNIV
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
PCT/SG2024/050290
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2023-05-15
Filing Date
2024-05-03
Publication Date
2024-11-21

AI Technical Summary

Technical Problem

Existing multi-mode orbital angular momentum (OAM) systems face challenges in activating high-reliability OAM modes during transmission due to random activation of OAM modes and mode combinations, leading to undesirable decoding thresholds and poor bit error rates (BERs, despite efforts to enhance spectral efficiency and mitigate inter-mode interference.

Method used

A shape index modulation transmitter is developed, which processes binary coded-bit sequences into sub-sequences to increase the probability of bits comprising a predetermined binary value, and selects optimal OAM mode combinations for transmission using a shaping encoder module and OAM mode selector module, ensuring that index symbols primarily consist of this binary value for improved reliability.

Benefits of technology

This approach significantly enhances the reliability of OAM mode combinations, leading to better bit error rates and spectral efficiency by optimizing the activation of OAM modes, thus improving the overall performance of OAM systems in line-of-sight scenarios.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure SG2024050290_21112024_PF_FP_ABST
    Figure SG2024050290_21112024_PF_FP_ABST
Patent Text Reader

Abstract

This document describes a shape index modulation transmitter for a multi-mode orbital angular momentum (OAM) system that increases the probability that OAM modes with high reliability are activated during a transmission process of the OAM system.
Need to check novelty before this filing date? Find Prior Art

Description

A SHAPE INDEX MODULATION TRANSMITTER FOR A MULTI-MODE ORBITAL ANGULAR MOMENTUM (OAM) SYSTEMCROSS-REFERENCE TO RELATED APPLICATION

[0001] This application claims the benefit of priority to Singapore patent application no. 10202301348Y which was filed on 15 May 2023, the contents of which are hereby incorporated by reference in its entirety for all purposes.TECHNICAL FIELD

[0002] This application relates to a shape index modulation transmitter for a multi-mode orbital angular momentum (OAM) system that increases the probability that OAM modes with high reliability are activated during a transmission process of the OAM system.BACKGROUND

[0003] Multiple-input multiple-output (MIMO) technology is widely recognized as an effective solution for enhancing data transmission reliability and spectral efficiency in wireless communication systems. However, traditional MIMO architectures necessitate complex radio frequency (RF) chains for each antenna element, and this leads to a substantial increase in hardware costs and power consumption. Additionally, MIMO systems require intricate digital signal processing techniques, such as the transmitting and receiving of precoding and equalization algorithms to mitigate inter-channel interference (ICI). A significant challenge for MIMO technology is its limited capacity for further enhancement without the introduction of additional resources, and its suboptimal performance in line-of-sight (LoS) scenarios which are characterized by high channel correlation and low rank. LoS channels are prevalent in various communication contexts, including LoS-dominant intelligent reflecting surfaces (IRS), unmanned aerial vehicles (UAV), satellites, and backhaul networks. To address these limitations while preserving the inherent benefits of MIMO, considerable research efforts have been directed towards the integration of orbital angular momentum (OAM) with mode division multiplexing (MDM) transmission.

[0004] OAM is a phenomenon resulting from the phase variation of electromagnetic (EM) waves with respect to the azimuthal angle around the propagation axis. In EM waves which have OAM, the wavefront takes on a helical structure, represented by the exponential function exp(-jξφ), where j is the imaginary unit, ξ, is the OAM mode number, and φ is the azimuthalangle. OAM-assisted mode division multiplexing (MDM) systems utilize EM waves with distinct OAM modes, which exhibit relatively low correlation. The multiplexing and demultiplexing of OAM modes are achieved through RF analog processing technologies, such as phase shift networks, which help to mitigate ICI and reduce computational complexity. Due to these advantageous properties, extensive research efforts have been devoted to developing and optimizing methods for generating OAM EM waves, including the use of spiral phase plates, helicoidal parabolic antennas, meta-surfaces, and uniform circular arrays (UCAs). Among these, UCAs are particularly promising due to their lower hardware costs and simpler implementation. Furthermore, numerous studies have demonstrated the viability of UCA-based OAM in both theoretical and practical contexts.

[0005] Uniform Circular Array-based OAM systems have garnered significant attention in a variety of wireless applications, encompassing multiple access communications and line-of- sight communications. OAM systems often encounter significant inter-mode interference due to misalignments between transmitting and receiving UCAs. To mitigate this issue, those skilled in the art have focused their efforts on a beam steering approach tailored for transceivers, as well as an OAM receiver that combines beam steering with amplitude detection. Moreover, optimization strategies for power allocation and precoding have been developed to eliminate inter-mode interference. To enhance spectral efficiency, an index modulation (IM) technique has been incorporated into the OAM framework, resulting in OAM- IM systems. In parallel with the development of index modulation techniques, those skilled in the art have proposed a coded modulation solution that is able to achieve high reliability-and- throughput requirements of modem communications.

[0006] Despite efforts of those skilled in the art to improve the performance of existing OAM-IM systems, transceivers in existing systems still ignore the difference in reliabilities between different OAM modes and / or mode combinations and as a result, these various OAM modes and / or mode combinations are activated randomly during the transmission of data, and this results in undesirable decoding thresholds and poor bit error rates (BERs).SUMMARY

[0007] Tn one aspect, the present application discloses a shape index modulation transmitter for a multi-mode orbital angular momentum (OAM) system. The system comprises a processing unit and a non-transitory media readable by the processing unit, the media storinginstructions that when executed by the processing unit causes the processing unit to: divide, using a shaping encoder module, a binary coded-bit sequence received from a bit-splitter module into a K-number of sub-sequences, whereby K is an integer that is determined based on a total number of OAM modes and a number of active OAM modes. The processing unit then proceeds to process, using a shaping function provided within the shaping encoder module, the sub-sequences to increase a probability that bits in the sub-sequences comprise a predetermined binary value and then formulates, using the shaping encoder module, index symbols based on the processed sub-sequences, whereby a first bit of each of the index symbols is obtained from a first processed sub-sequence. The processing unit then causes the shaping encoder module to provide the index symbols to an OAM mode selector module for selecting corresponding OAM mode combinations for transmission by the transmitter, whereby optimal OAM mode combinations of the transmitter are associated with index symbols that comprise primarily of the predetermined binary value.

[0008] In another aspect, the present application discloses a method for shaping index modulation in a multi-mode orbital angular momentum (OAM) system whereby the system comprises a shape index modulation transmitter having a shaping encoder module communicatively coupled to an OAM model selector module. The disclosed method then comprises the steps of dividing, using the shaping encoder module, a binary coded-bit sequence received from a bit-splitter module into a K-number of sub-sequences, whereby K is an integer that is determined based on a total number of OAM modes and a number of active OAM modes; processing, using a shaping function provided within the shaping encoder module, the sub- sequences to increase a probability that bits in the sub-sequences comprise a predetermined binary value; formulating, using the shaping encoder module, index symbols based on the processed sub-sequences, whereby a first bit of each of the index symbols is obtained from a first processed sub-sequence; and providing, using the shaping encoder module, the index symbols to an OAM mode selector module for selecting corresponding OAM mode combinations for transmission by the transmitter, whereby optimal OAM mode combinations of the transmitter are associated with index symbols that comprise primarily of the predetermined binary value.BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Various embodiments of the present disclosure are described below with reference to the following drawings:Figure 1 illustrates a block diagram of a protograph low-density parity-check (PLDPC)-coded orbital angular momentum (OAM)-index modulation (IM) system over line of sight (LoS) channels in accordance with embodiments of the present disclosure;Figure 2a illustrates a block diagram representative of a transmitting side of a shape index modulation transceiver for a multi-mode OAM system in accordance with embodiments of the present disclosure;Figure 2b illustrates a block diagram representative of a receiving side of a shape index modulation transceiver for a multi-mode OAM system in accordance with embodiments of the present disclosure;Figure 3 illustrates a block diagram of a processing system for performing embodiments of the present disclosure;Figure 4 illustrates a flowchart of a process for shaping index modulation in a multi-mode OAM system in accordance with embodiments of the present disclosure;Figure 5 illustrates a flowchart of a process for processing the sub-sequences using a shaping function in accordance with embodiments of the present disclosure;Figure 6a illustrates histograms of the log-likelihood ratio (LLR) values liobtained as output from a channel detector with coded bits bi= 0 and bi= 1 under SNR = 80 dB, and transformed LLR values , and a regular-(3, 6) PLDPC code with5 and is consideredin OAM-TM systems withover LoS channels:Figure 6b illustrates histograms of the log-likelihood ratio (LLR) values lias obtained as output from a channel detector with coded bits bL= 0 and bi= 1 under SNR = 80 dB, and transformed LLR values and a regular-(3, 6) PLDPC code withand is consideredin OAM-TM systems withover LoS channels:Figure 7a illustrates histograms of the log-likelihood ratio (LLR) values liobtained as output from a channel detector with coded bitsunder SNR = 82 dB, and transformed LLR values, and a regular-(3, 6) PLDPC code with is consideredin OAM-IM systems with over LoS channels: θ = 0;Figure 7b illustrates histograms of the log-likelihood ratio (LLR) valuesas obtained as output from a channel detector with coded bits bi= 0 and bi= 1 under SNR = 82 dB, and transformedLLR valuesand a regular-(3, 6) PLDPC code with = 0.5 and = 2400 is consideredin OAM-IM systems with NA= 2 over LoS channels: θ = 0.8;Figure 8 illustrates a flow diagram of a CBCT-PEXIT algorithm used for analyzing and designing PLDPC-coded OAM-IM systems over LoS channels in accordance with embodiments of the present disclosure;Figure 9 illustrates protographs for a family of AR4JA codes with a wide range of rate from 'A to (e+1) / (e+2);Figure 10a illustrates protographs for the family of IP1 PLDPC codes with a wide range of rate from 1 / 2 to (e + 1 ) / (e + 2) in accordance with embodiments of the present disclosure;Figure 10b illustrates protographs for the family of IP2 PLDPC codes with a wide range of rate from 1 / 2 to (e + 1 ) / (e + 2) in accordance with embodiments of the present disclosure;Figure 1 la illustrates channel gains of a 4 × 4 OAM-IM system without inter-mode interference (i.e., θ = 0);Figure 11b illustrates channel gains of a 4 × 4 OAM-IM system with inter-mode interference (i.e., θ = 0.8);Figure 12 illustrates extrinsic MIs output from detector with and without CBCT in a 4 × 4 OAM-IM system with θ = 0.8, which correspond to the coded bits transmitted by 4 different modes whereby the regular-(3, 6) PLDPC code and Gray-labeled QPSK modulation are used; Figure 13 illustrates constellation-constrained capacities of the natural index and Mi-based index schemes where Gray-labeled QPSK and 8PSK constellations are considered;Figure 14a illustrates BER curves of two MIMO LDPC codes (i.e., Code I and Code II) and the regular-(3, 6) code, the AR4JA code, and the IUP, IP1, and IP2 PLDPC codes in 4 × 4 OAM-IM systems with NA= 1 and θ = 0.8 whereby a natural index scheme is considered with QPSK modulation;Figure 14b illustrates BER curves of two MIMO LDPC codes (i.e., Code I and Code II) and the regular-(3, 6) code, the AR4JA code, and the IUP, IP1, and IP2 PLDPC codes in 4 × 4 OAM-IM systems with NA= 1 and θ = 0.8 whereby a natural index scheme is considered with 8PSK modulation;Figure 15a illustrates BER curves of two MIMO LDPC codes (i.e., Code I and Code II) and the regular-(3, 6) code, the AR4JA code, and the IUP, IP1, and IP2 PLDPC codes in 4 × 4 OAM-IM systems with NA= 2 and θ = 0.8 whereby a natural index scheme is considered with QPSK modulation;Figure 15b illustrates BER curves of two MIMO LDPC codes (i.e., Code I and Code II) and the regular-(3, 6) code, the AR4JA code, and the IUP, IP1, and IP2 PLDPC codes in 4 × 4 OAM-IM systems with NA= 2 and θ = 0.8 whereby a natural index scheme is considered with 8PSK modulation;Figure 16 illustrates BER curves of a 5GLDPC code, three irregular LDPC codes (i.e., irregular LDPC codes with maximum VN degrees dv = 4, 5, and 50), (3, 6) SC-LDPC code, and the IUP, IP1, and IP2 PLDPC codes in 4 × 4 OAM-IM systems with NA= 2 and θ = 0.8 when a QPSK modulation and natural index scheme are considered;Figure 17 illustrates BER curves of the high-rate AR4JA code, and proposed IP1 and IP2 PLDPC codes in 4 × 4 OAM-IM system with NA= 2 and θ = 0.8 when a natural index scheme is considered;Figure 18a illustrates BER curves of the regular-(3, 6), the AR4JA code, and proposed IUP, IP1, and IP2 PLDPC codes in 4 × 4 OAM-IM systems with NA= 2 and θ = 0.8 when both natural index and proposed shaped index schemes are considered with QPSK modulation;Figure 18b illustrates BER curves of the regular-(3, 6), the AR4JA code, and proposed IUP, IP1, and IP2 PLDPC codes in 4 × 4 OAM-IM systems with NA= 2 and θ = 0.8 when both natural index and proposed shaped index schemes are considered with 8PSK modulation;Figure 19a illustrates BER curves of the proposed improved PLDPC codes in 4×4 OAM-IM systems with NA= 2, θ = 0.8, and two different index schemes (i.e., ED based index and proposed shaped index schemes) when both QPSK and 8PSK modulations are considered with the IUP PLDPC code;Figure 19b illustrates BER curves of the proposed improved PLDPC codes in 4×4 OAM-IM systems with NA= 2, θ = 0.8, and two different index schemes (i.e., ED based index and proposed shaped index schemes) when both QPSK and 8PSK modulations are considered with the IP1 PLDPC code;Figure 19c illustrates BER curves of the proposed improved PLDPC codes in 4×4 OAM-IM systems with NA= 2, θ = 0.8, and two different index schemes (i.e., ED based index and proposed shaped index schemes) when both QPSK and 8PSK modulations are considered with the IP2 PLDPC code;Figure 20 illustrates BER curves of the (3, 6) PLDPC code in the 4×4 MIMO-VLC systems with NA=1 and 2, and four different index schemes (i.e., the CONSM scheme, ED scheme and their shaped counterparts);Figure 21a illustrates BER curves of (3, 6) PLDPC code in multi-cell network systems with two different index schemes (i.e., CONSM and shaped CONSM schemes) when the value of multi-user interference is 0 dB;Figure 21b illustrates BER curves of (3, 6) PLDPC code in multi -cell network systems with two different index schemes (i.e., CONSM and shaped CONSM schemes) when the value of multi-user interference is 6 dB; andFigure 21c illustrates BER curves of (3, 6) PLDPC code in multi-cell network systems with two different index schemes (i.e., CONSM and shaped CONSM schemes) when the value of multi-user interference is 9 db.DETAILED DESCRIPTION

[0010] The following detailed description is made with reference to the accompanying drawings, showing details and embodiments of the present disclosure for the purposes of illustration. Features that are described in the context of an embodiment may correspondingly be applicable to the same or similar features in the other embodiments, even if not explicitly described in these other embodiments. Additions and / or combinations and / or alternatives as described for a feature in the context of an embodiment may correspondingly be applicable to the same or similar feature in the other embodiments

[0011] In the context of various embodiments, the articles “a”, “an” and “the” as used with regard to a feature or element include a reference to one or more of the features or elements.

[0012] In the context of various embodiments, the term “about” or “approximately” as applied to a numeric value encompasses the exact value and a reasonable variance as generally understood in the relevant technical field, e g., within 10% of the specified value.

[0013] As used herein, the term “and / or” includes any and all combinations of one or more of the associated listed items.

[0014] As used herein, “comprising” means including, but not limited to, whatever follows the word “comprising”. Thus, use of the term “comprising” indicates that the listed elements are required or mandatory, but that other elements are optional and may or may not be present.

[0015] As used herein, “consisting of’ means including, and limited to, whatever follows the phrase “consisting of’. Thus, use of the phrase “consisting of’ indicates that the listed elements are required or mandatory, and that no other elements may be present.

[0016] As used herein, the term “distinct binary sequence” or “distinct bits” refers to a sequence of binary digits or bits that are unique and are not repeated within a given set of binary numbers.

[0017] Further, one skilled in the art will recognize that certain functional units in this description have been labelled as modules throughout the specification. The person skilled in the art will also recognize that a module may be implemented as circuits, logic chips or any sort of discrete component. Still further, one skilled in the art will also recognize that a module may be implemented in software which may then be executed by a variety of processor architectures. In embodiments of the disclosure, a module may also comprise computer instructions or executable code that may instruct a computer processor to carry out a sequence of events based on instructions received The choice of the implementation of the modules is left as a design choice to a person skilled in the art and does not limit the scope of the claimed subject matter in any way.

[0018] A block diagram of a protograph low-density parity-check (PLDPC)-coded orbital angular momentum (OAM)-index modulation (IM) system over line of sight (LoS) channels is illustrated in Figure 1. At the transmitter side of PLDPC-coded OAM-IM system 100, an information bit sequenceof lengthis first passed to PLDPC encoder 102. PLDPC encoder 102 converts information bit sequence nSinto a binary coded-bit sequence nEof length Then, nEis permuted by bit-level random interleaver 104 to yield a bit-interleaved sequence Subsequently, a bit-splitting operation is performed onby bit-splitter module 106, and this bit-interleaved sequence is divided into two subsequences, i.e., of lengthandof lengthwherewhich are then provided to OAM-based index module 110 and modulator 108 respectively.

[0019] OAM-based index module 110 is configured to have an OAM mode set Φ of size N and an active number of OAM modes between 1 and N In embodimentsof the disclosure, the number of effective OAM mode combinations used for the index isdefined aswhereis a floor operation. During each signal transmission slot, bits are extracted from to form a vector bIndand are sent to O AM-based indexmodule 110 for selecting the corresponding 0AM mode combinationthat is he used by the transmitter in system 100. In particular,

[0020] As for modulator 108, module 108 is configured to utilize modulation schemes such as, but not limited to, the M-ary phase shift keying (PSK) or quadrature amplitude modulation (QAM) constellation setDuring each signal transmission slot,bits are extracted from to form a vector bModand sent to modulator 108 for determining a modulated symbol sequence where and Mis defined as the sizeof the modulation scheme. The outputs from O AM-based index module 110 and modulator 108, i are respectively passed to 0AM multiplexing module 112 which in turnis communicatively coupled to power allocation module 113. Power allocation module 113 is configured to determine the optimal distribution of power between different 0AM modes in order to guarantee that all possible active 0AM modes can share the same energy intensity. Based on these received values, 0AM multiplexing module 112 is then configured to compute the spectral efficiency, 77, (i.e., bits per channel use (bpcu)) of a PLDPC-coded OAM-IM system as:

[0021] In order to transmit an 0AM signal with the mode number0AM multiplexing module 112 is then configured to communicate its corresponding modulated symbolto all array elements in transmitting UCA 1 14 and this is implemented with a successive phase shift between array elements. As a result, the signal on the array elementmay be defined as:where is defined as a power allocation factor for the signal with a mode number

[0022] Based on the above, a composite signal, (i.e., an OAM signal) emitted by theth array element may be defined as a linear superposition of the signals with different active modes and as such may be defined as:

[0023] For simplicity, the transmitted OAM signal vector may be denoted as s = WPx where swhere the superscriptdenotes a transpose operation; W ∈ is defined as a unitary discrete Fourier transform (DFT) matrix with elements is defined as a power allocation matrix, whilethe total transmission power equals P and comprises acomplex data vector containing NAmodulated symbols andzero entries where the indices ofmodulated symbols within x are determined by the entries (i.e., mode numbers) within

[0024] At the receiver side of system 100, receiving UCA 116 is configured to have the same number of array elements as transmitting UCA 114. When both UCAs are aligned with each other, the received OAM signal vector y =may be defined as:where H ∈is the channel response matrix and characterizes the feature of an LoS transmission environment between both UCAs; n ∈is an additive white Gaussian noise (AWGN) vector, and each entry is a complex Gaussian variable with zero mean and variance The channel response,(i.e., channel gain) between the n -th transmitting (OAM) mode and the p -th receiving (OAM) mode may then be computed as:where is the antenna gain, is the wavelength, and dpn, is the distance between the n-th andp-th array elements. If both UCAs are aligned perfectly, may be computed as:where D is the distance between both the centers of transmitting UCA 114 and receiving UCA 116, and denotes the radius of the transmitting (receiving) UCA.

[0025] The OAM signal vector y as obtained by receiving UCA 116 will then be provided to demultiplexing module 118, where it is then processed by an inverse DFT (IDFT) operation (i.e., an IDFT matrix, which yields a demultiplexing-aided OAM signal vectorthat may be defined as:

[0026] In a scenario of perfect alignment between both UCAs,may be defined as a circulant matrix and its A eigenvalues formulate a diagonal matrix Λ

[0027] In view of equation (8), equation (7) above may then be simplified to:

[0028] As the radiation energies of different active OAM modes are different due to the characteristics of a Bessel function, the power allocation factor in P isused to ensure that all possible active OAM modes can share the same energy intensity. In particular, the power allocation factor may be defined as:

[0029] Unlike ideal LoS transmission environments, both UCAs are usually misaligned in actual applications and as a result, inter-mode interference will occur. Under the assumption that the plane of the receiving UCA is tilted towards that of the transmitting UCA with an angle) the distance between two array elements for such two UCAs (i.e.,in equation (6)) may be computed as:whereand

[0030] When all received OAM signals are obtained using a demultiplexing operation, the corresponding extrinsic log-likelihood ratio (LLR) sequencecan be obtained from detector 120. Based on a specific deinterleaving procedure, deinterleaver 122 may transforminto and this information may then be passed to decoder 124, which will use this information to produce the decoded information-bit sequenceIn an embodiment of the disclosure, detector120 and decoder 124 may each be implemented by the maximum-likelihood (ML) and belief- propagation (BP) algorithms, respectively. The full channel state information (CSI) is assumed for the detection because a well-designed channel estimator may be adopted by the receiver to perfectly acquire the CSI in practical applications. The transmitted OAM signal vectorin each transmission slot may then be determined by the coded-bit vectorwhere denotes a vector consisting of coded bits that are used for the index, whiledenotes a vector composed of N sub-vectorswhere each sub-vector consists of M coded bits used for modulation. Underthe assumption that all effective OAM mode combinations constitute a setthe soft outputs of detector 120, i.e., the LLRs for the and codedbits, respectively, in b may be computed as:wheredenotes a subset of and it is mapped bywith thecoded bit being b Also, denotes the z-th constellation subset mapped bywith the j coded bit being denotes theconstellation subset mapped by. . , m) coded bit being b ∈0, 1 }, where denotes the norm, and log(.) denotes thenatural logarithm operation. From the above, it is understood that the LLRs of the bits usedfor index can be estimated even if they are not transmitted. The SNR may be defined aswhich equals , whereandare the average energy per bit and noise powerspectral density respectively. Furthermore, it should be noted that the inter-mode interference has been considered in the procedure of computing the LLR (see equations (12) and (13)).

[0031] A block diagram representative of a transmitting side of a shape index modulation transceiver for a multi-mode OAM system in accordance with embodiments of the present disclosure is disclosed in Figure 2a. As shown, shape index modulation transmitter 200 includes many of the same modules as the transmitter section of PLDPC-coded OAM-IM system 100 as previously described in Figure 1. Additionally, two new modules are incorporated into transmitter 200: (i) shaping encoder module 202, which is connected to the output of bit-splitter 106; and (ii) an interleaver-2 module 204, which is positioned between an output of the shaping encoder module 202 and an input of OAM-based index module 110. In embodiments of the disclosure, the output of interleaver-2 module 204 is concatenated with OAM-based index module 110 and is used to process the output bits from shaping encoder module 202. Further, it should be noted that the output bits from interleaver-2 module 204 are used to activate the transmitting modes / antennas / beams of transmitter 200 and that these index bits are not transmitted with the modulation signals.

[0032] In embodiments of the disclosure, optimal OAM mode combinations are precomputed for shape index modulation transmitter 200. In further embodiments of the disclosure, the optimal OAM mode combinations of the transmitter may comprise four optimal mode combinations that were selected by channel estimation of a 4 x 4 orbital angular momentum-index modulation system that has two active OAM modes. In another embodiment of the disclosure, the optimal OAM mode combinations of the transmitter may comprise optimal mode combinations that were selected using a mutual information (Ml)-based index scheme. The detailed steps to compute these optimal OAM mode combinations will be described in greater detail in the later sections in this description. Furthermore, these precomputed mode combinations are configured such that each of these optimal OAM mode combinations are associated with index symbols that comprise primarily of a predetermined binary value - which may comprise ‘0’ bits or ‘ 1’ bits.

[0033] Embodiment 1 of the Transmitter

[0034] In embodiments of the disclosure, when the optimal OAM mode combinations of the transmitter comprises four optimal mode combinations that were selected by channel estimation of a 4 x 4 orbital angular momentum-index modulation system having two active OAM modes (i.e., N = 4 OAM modes and NA= 2 active OAM modes), it was found that the index modulation (IM) scheme may comprise ofmode combinations The reliabilities of these Cemode combinations may then be estimated withthe aid of a standard channel estimation process. The detailed workings of the standard channel estimation process are omitted for brevity in this description as it is well known to one skilled in the art.

[0035] As an example, with the aid of the channel estimation process, it is assumed that four (4) mode combinations which provide the optimum system performance for 4 X 4 OAM-IM systems with NA= 2 were obtained as: {(m1, m2), (m1, m3), (m1,m4), (m2, m3)} - and was arranged in descending order or reliability from the left to the right. For this example, it was determined that the optimal OAM mode combinations are associated with index symbols that comprise primarily of ‘0’ bits, i.e., the predetermined binary value is ‘0’.40036] Specifically, it was set that the smaller the numberis, the higher the reliability of that can be achievedEach of the C index symbols, of which eachcomprises of a binary sequence composed of index bits is then defined to have anone-to-one correspondence between theindex symbols andmode combinations. Moreover, theindex symbol corresponding towas set to be a binary representation of the decimal numberAn exemplary representation of this is shown in Table 1.

[0037] Table 1 shows the shaped Natural Index Modulation scheme for a 4 x 4 OAM system with NA= 2 and θ = 0.8. It is useful to note that for a conventional Natural Index Modulation scheme (not shown), the occurrence probabilities of all OAM mode combinations are usually set to be the same However, for the shaped Natural indexscheme shown in Table 1 above, as it was determined that the OAM mode combinationsand have better reliabilities as compared to the other two mode combinations, the OAM mode combinations Φ1and Φ2were set to have higher occurrence probability than other mode combinations

[0038] Based on the above process, for themode combinations with the highest reliabilitiesit was determined that their corresponding index symbols have one thing in common, which is that their first coded bits are = 0. In view of the above,it can be seen that a significant system performance gain may be achieved if more index symbols with nexist in the index procedure (i.e., the mode combinations have a highoccurrence probability

[0039] With reference to Figure 2a, in operation, it can be seen that an interleaved coded- bit sequenceis first divided into two sub-sequences and by bit splitter 106. Sub-sequence is then provided to shaping encoder 202 having the parameterswhile is provided to modulator 108 to be used in the generation of modulation symbols. Shaping encoder 202 is configured to first divide into sub-sequences (i.e.,whereby the length of, while the length of the remaining (k — 1) sub-sequences is defined to be ns(ks< ns) Sub-sequence is then processed by ashaping function and converted to a new sub-sequence of length ns. The detailed (ns, ks)shaping function is illustrated in Algorithm 1 below. Both and ns(ns> ks) arenon-negative integers, and they satisfy the condition where i is anon-negative integer (i.e., i — 1,2, ...) and nelis the length of a coded-bit subsequence used for OAM mode selection. It is worth noting that the larger the valueis, the better the system performance can be achieved, but at the expense of spectral efficiency. Therefore, the values and can be determined based on the actual communication requirements. It should be noted that the aim of the shaping encoder is to increase the occurrence of the predetermined binary value, i.e., bit “0”, within (i.e., increase the probability of a bit being 0). Finally,shaping encoder 204 formulates each index symbol by extracting k bits from k different sub- sequences of lengthin a sequential order. As a result, the bits within will determine the bit Hi within each index symbol. It should be noted that ks(ns) denotes the number of bits sent to (output from) the shaping encoder.

[0040] In summary, Algorithm 1 outlines a shaping principle for a sequence which isdivided into sub-sequences, with the first sub-sequence having a lengthof ksand the remaining (k — 1) sub-sequences each having a length of ns(ks< ns). A codebook ensemble is created, containingdifferent binary sequences, each formulated by ksbinary bits. The z-th binary sequence inrepresents the binary version of the decimal number where z ranges from 1 to

[0041] The shaping process for involves mapping the i-th binary sequence withinto a length-binary sequence. If i =1, the sequence is mapped to an all-zero sequence of length ns. For values of i > 1, the sequence is mapped to a lengthbinary sequence having a Hamming distancewhich should be as small as possible while ensuring that the current i- th sequence is distinct from all previous sequences. As a result, a codebook ensemble isformed, comprising oflength binary sequences, with a one-to-one correspondencebetween the sequences in and Based on the mappings between these two codebooks,the sub-sequence is then converted to a new sub-sequence of lengthFinally, eachindex symbol is formed by extractingbits fromdifferent sub-sequences of length in asequential order.

[0042] As it was determined that the optimal OAM mode combinations are associated with index symbols that comprise primarily of ‘0’ bits, the formed index symbols may then be used to increase the probability of desirable transmitting modes / antennas / beams being activated for transmitting the signals, which is beneficial to avoid signal fading issues during transmission of the signals. More specifically, the proposed technique considers choosing the optimal transmission path for signals in free space based on previously obtained extrinsic channel feedback.

[0043] As the technique described above utilizes a Multiple-Input Multiple-Output (MIMO) framework, this enables the transmitter to be used in MIMO Line-of-Sight (LoS) scenarios,Visible Light Communication (VLC) scenarios, and multi cell scenarios due to the high throughput that may be achieved by transmitter 200. In particular, the throughput may be defined as:

[0044] Embodiment 2 of the Transmitter

[0045] In another embodiment of the disclosure, when the optimal OAM mode combinations of the transmitter comprises optimal mode combinations that were selected using a mutual information (Ml)-based theoretical-analysis method, anumber of index symbols may be defined such that each symbol comprises a binary coded-bit sequence composed of coded bits where b, is thecoded bit). Moreover,the binary coded-bit sequence of theindex symbol is a binary representation of the decimal number A one-to-one correspondence is then defined between theindex symboland the mode combinationwhere the firstindex symbolscorrespond to themode combinations withhighest reliabilities Also, in this embodiment, the firstindex symbolswe set to have one thing in common: which is that their first coded bits areAn example of such a relationship is illustrated in Table 2 below.

[0046] Table 2 above shows optimal mode combinations that were selected using a mutual information (Ml)-based theoretical-analysis method. As can be seen, if more index symbols with bi = 0 exist in the index procedure (i.e., the mode combinations have a high occurrence probability , where j = 1, 2, . . . , Be / 2), more modulated symbols can be transmitted by theOAM modes with relatively high reliabilities. In other words, the OAM-1M system can achieve better performance.

[0047] Table 3 above shows a shaped MI-based Index Modulation scheme for a 4 x 4 O AM- IM system with NA= 2,= 0.8, and a shaping encoder (4,2) where the reliabilities of these mode combinations were estimated using the mutual information (Ml)-based theoretical- analysis method and these mode combinations were ranked accordingly with OAM mode combinationsandhaving better reliabilities as compared to the other two mode combinations. As such, the OAM mode combinationsandwere set to have higher occurrence probability than other mode combinations and

[0048] With reference to Figure 2a, in operation, it can be seen that an interleaved coded- bit sequence is first divided into two sub-sequencesandby bit splitter 106. Sub-sequence is then provided to shaping encoder 202 having the parameterswhile is provided to modulator 108 to be used in the generation of modulation symbols. Shaping encoder 202 is configured to first divide sub-sequenceswhereby the length of while the length of the remaining (q —1) sub-sequences is defined to be where j is defined as a scaling factor and maycomprise any integer = 1, 2, .... For this embodiment, the detailed shaping functionsperformed by shaping encoder 202 is set out in Algorithm 2 below. The function of shaping encoder 202 is to increase the number of bit 0 (i.e., the predetermined binary value) within (i.e., to increase the probability that’s the bits are ‘0’). Finally, each index symbol is formulated by extracting q bits from q sub-sequences of length in a sequential manner. Asthe bits withindetermine the bit within each index symbol, more index symbols with will exist upon completion of the shaping function described above.

[0049] In summary, Algorithm 2 outlines a shaping procedure for creating two codebook ensembles and each containing binary sequences, with one-to-one mappingbetween the binary sequences ofand those of The parameters for Algorithm 2 areand which are utilized to determine the lengths of the binary sequences in the respective codebooks.

[0050] The algorithm proceeds as follows: For each index i ranging from 1 to a binarysequence is generated by converting the decimal number i-1 to a binary sequence of length Then, a corresponding binary sequence of lengthis created with the smallest possibleweight ensuring that is distinct from all previously generated sequences for valuesof i > j. The weightis defined as the which counts the number of non-zeroelements in the sequence. Finally, each sequence in codebook is mapped to a uniquesequence inestablishing the one-to-one correspondence between the two codebooks.

[0051] In further embodiments, when the parameterswere set to be (4, 3) and (5, 3), it was found that the occurrence probabilities of bit ‘0’ in the index symbols was 0.687 and 0.755 respectively.

[0052] A block diagram representative of a receiving side of a shape index modulation transceiver for a multi-mode OAM system in accordance with embodiments of the present disclosure is illustrated in Figure 2b. As shown, shape index modulation receiver 210 includes many of the same modules as the receiver section of PLDPC-coded OAM-IM system 100 as previously described in Figure 1. Additionally, three new modules are incorporated into receiver 210: LLR splitter module 212, deinterleaver-2 module 214 and shaping decoder module 216.

[0053] LLR splitter module 212 is configured to divide the LLR sequence output from detector 120 into two distinct sub-sequences, denoted asand . Specifically,1 corresponds to the data stream, which is the output from the interleaver-2 module 204 in the transmitter; whileis linked to the data stream, which is generated from the bit splitter 106 and is utilized for modulation purposes.

[0054] Deinterleaver-2 module 214, which is matched to interleaver-2 module 204 as used in the transmitter, is designed to permute (i.e., deinterleave) the LLR sequenceand subsequently output a new LLR sequence. This new sequenceis directly related to the data stream nE3in the transmitter, ensuring a coherent data flow between the transmitter and receiver components.

[0055] As for shaping decoder 216, this module is tasked with executing a shaping decoding procedure on the LLR sequence LE1output from the deinterleaver-2 module 214. This shaping decoding process can be regarded as the inverse of the shaping encoding procedure carried out in the transmitter. As a result of this decoding, a new LLR sequenceis obtained, which corresponds to the original data streamin the transmitter, thereby completing the decoding process and ensuring data integrity.

[0056] In accordance with embodiments of the present disclosure, a block diagram representative of components of processing system 300 that may be provided within shaping encoder module 202, OAM-based index module 110, interleaver-2 module 204 and / or any of the modules shown in Figures 2a and 2b to carry out the digital signal processing functions in accordance with embodiments of the disclosure. One skilled in the art will recognize that the exact configuration of each processing system provided within these modules may be differentand the exact configuration of processing system 300 may vary and the arrangement illustrated in Figure 3 is provided by way of example only.

[0057] In embodiments of the disclosure, processing system 300 may comprise controller301 and user interface 302. User interface 302 is arranged to enable manual interactions between a user and the computing module as required and for this purpose includes the input / output components required for the user to enter instructions to provide updates to each of these modules. A person skilled in the art will recognize that components of user interface302 may vary from embodiment to embodiment but will typically include one or more of display 340, keyboard 335 and optical device 336.

[0058] Controller 301 is in data communication with user interface 302 via bus 315 and includes memory 320, processing unit or processor 305 mounted on a circuit board that processes instructions and data for performing the method of this embodiment, an operating system 306, an input / output (I / O) interface 330 for communicating with user interface 302 and a communications interface, in this embodiment in the form of a network card 350. Network card 350 may, for example, be utilized to send data from these modules via a wired or wireless network to other processing devices or to receive data via the wired or wireless network. Wireless networks that may be utilized by network card 350 include, but are not limited to, Wireless-Fidelity (Wi-Fi), Bluetooth, Near Field Communication (NFC), cellular networks, satellite networks, telecommunication networks, Wide Area Networks (WAN) and etc.

[0059] Memory 320 and operating system 306 are in data communication with processor 305 via bus 310. The memory components include both volatile and non-volatile memory and more than one of each type of memory, including Random Access Memory (RAM) 323, Read Only Memory (ROM) 325 and a mass storage device 345, the last comprising one or more solid-state drives (SSDs). One skilled in the art will recognize that the memory components described above comprise non-transitory computer-readable media and shall be taken to comprise all computer-readable media except for a transitory, propagating signal. Typically, the instructions are stored as program code in the memory components but can also be hardwired. Memory 320 may include a kernel and / or programming modules such as a software application that may be stored in either volatile or non-volatile memory.

[0060] Herein the term “processor” or “processing unit” is used to refer generically to any device or component that can process such instructions and may include: a microprocessor, a processing unit, a microcontroller, a programmable logic device or other computational device. That is, processor 305 may be provided by any suitable logic circuitry for receiving inputs, processing them in accordance with instructions stored in memory and generating outputs (for example to the memory components or on display 340). In this embodiment, processor 305 may be a single core or multi -core processor with memory addressable space. In one example, processor 305 may be multi-core, comprising — for example — an 8 core CPU. In another example, it could be a cluster of CPU cores operating in parallel to accelerate computations.

[0061] A flowchart which sets out the process for shaping index modulation in a multimode OAM system in accordance with embodiments of the present disclosure is illustrated in Figure 4. In embodiments of the disclosure, process 400 as illustrated in Figure 4 may be performed by shaping encoder module 202 and / or OAM-based index module 110.

[0062] Process 400 begins at step 402, with process 400 dividing a binary coded-bit sequence received from a bit-splitter module into a K-number of sub-sequences, whereby K is an integer that is determined based on a total number of OAM modes and a number of active OAM modes. Process 400 then proceeds to step 404 where it processes, using a shaping function, the sub-sequences to increase the probability that bits in the sub-sequences comprise a predetermined binary value.

[0063] At step 406, process 400 then formulates index symbols based on the processed subsequences, whereby a first bit of each of the index symbols is obtained from a first processed sub-sequence. At the final step 408, process 400 then provides the index symbols to an OAM mode selector module for selecting corresponding OAM mode combinations for transmission by the transmitter, whereby optimal OAM mode combinations of the transmitter are associated with index symbols that comprise primarily of the predetermined binary value.

[0064] A flowchart which illustrates a process for processing the sub-sequences using a shaping function in accordance with embodiments of the present disclosure is shown in Figure 5. In embodiments of the disclosure, process 500 as illustrated in Figure 5 may be performed by shaping encoder module 202 and / or OAM-based index module 110.

[0065] Process 500 begins at step 502 by generating a codebook comprising an initial set of distinct binary sequences that are mapped to a final set of distinct binary sequences that comprise primarily of the predetermined binary value, whereby a length of each binary sequence in the initial set of distinct binary sequences is equal to the first length and a length of each binary sequence in the final set of distinct binary sequences is equal to the second length. Process 500 then proceeds to convert the first sub-sequence to a shaped first sub- sequence based on the mappings in the codebook at step 504.

[0066] In embodiments of the disclosure, the step of formulating the index symbols based on the processed sub-sequences at step 406 may comprise the steps of, for the formulation of each index symbol, sequentially extracting, using the shaping encoder module, a predefined number of distinct bits from each of the processed sub-sequences and concatenate the extracted bits to form the index symbol.

[0067] In embodiments of the disclosure, the step of selecting the optimal mode combinations using a Mi-based theoretical-analysis method comprises the steps of performing one or more rounds of mode-combination selection, using a Ml-based theoretical -analysis module that is communicatively coupled to the OAM mode selector, each round comprising the steps of: generating a modulated symbol sequence for a distinct signal-to-noise ratio (SNR); sequentially dividing the modulated symbol sequence into a N number of sub-sequences, whereby the modulated symbols within a k-th sub-sequence are to be transmitted by a k-th mode, where k is defined as a number from a set { 1 , 2, ..., N] and N is defined as the total number of different OAM modes; computing, using an MI calculator, extrinsic MI values for each mode based on the sub-sequence associated with the mode; and determining mode combinations to be used as the optimal mode combinations based on mode combinations that have the largest MI values.

[0068] In embodiments of the disclosure, the MI calculator may be configured to compute extrinsic MI values based on a customized Box-Cox transformation (CBCT)-based protograph extrinsic information transfer (PEXIT) algorithm and the binary coded-bit sequence may comprise a bit sequence coded based on a protograph low-density parity-check (PLDPC) code,an improved unpunctured (IUP) PLDPC code, or one of two improved punctured (IP1 and IP2) PLDPC codes.

[0069] Customized Box-Cox Transformation

[0070] In certain transmission environments, it is useful to consider channel characteristics so that coding schemes that approach channel capacity may be developed. To achieve this, Monte-Carlo simulations were performed and the extrinsic LLR values of coded bits output from the channel detector (as shown in Figure 1) are analyzed. The simulation parameters for the simulated OAM-IM systems are set out in Table 4 below.

[0071] For this simulation, a regular-(3, 6) PLDPC code with a code rate Rp= 0.5 and a codeword lengthwas adopted, while the modulation scheme comprised a quadrature PSK (QPSK) modulation scheme. Although the conditional distribution analysis for LLRs obtained from the detector does not require channel coding, a regular-(3, 6) PLDPC code was used to provide a more practical analysis for PLDPC-coded OAM-IM systems, was denotedas the coded bit in a frame (codeword) and 4 was denoted as the corresponding LLR value obtained from the channel detector.

[0072] After the LLRs were generated from the simulations, the range of LLR values are sequentially divided into several uniformed intervals, each with a length of = 0.1, and theoccurrence frequencies of these LLRs were computed for different intervals. These occurrence frequencies are then scaled by a factor of lto generate the LLR histograms andwhich correspond to the Probability Density Functions (PDFs) of an appropriatelyquantized version of LLRs for the values of and respectively.

[0073] Further, symmetric Gaussian distributions andwere considered for comparison, where u is the mean, 2u is the variance, anddenotes the average extrinsic LLR values for. With reference to Figure6a, when an OAM-IM with NA= 1 and a SNR = 80 dB is considered for 9 = 0 (i.e., there is perfect alignment between the transmitting and receiving UCAs), it can be seen that the extrinsic LLR histogramsand(i.e., the scattered points) are consistent with the symmetric Gaussian distributionsplot 602 (i.e., solid lines), respectively.

[0074] When two UCAs are misaligned with each other the extrinsic LLRhistogramsmay not exhibit such PDF characteristics. An example when an OAM-IM with NA= 1 and a SNR = 80 dB is considered for 9 = 0.8 is illustrated in Figure 6b, whereand) tend to follow skewed distributions (i.e., the scattered points), but not the symmetric Gaussian distributions- plot 606 (i.e., solid lines), respectively.

[0075] A similar phenomenon is illustrated in Figures 7a and 7b when an OAM-IM with NA= 2 and a SNR = 82 dB is considered for 9 = 0 (Figure 7a) and for 9 = 0.8 (Figure 7b). In particular, it can be seen from Figure 7a that the extrinsic LLR histogramsand) (i.e., the scattered points) are consistent with the symmetric Gaussian distributions - plot 702 (i.e., solid lines), respectively. However, itcan be seen from Figure 7b that the extrinsic LLR histogramsthe scattered points) are not consistent with the symmetric Gaussian distributions- plot 708 and- plot 706 (i.e., solid lines), respectively.

[0076] It should be noted that these numerical results shown in Figures 6a and 6b, Figures 7a and 7b are all calculated based on extrinsic LLRs obtained from simulations. Further, discrete point representations were adopted in these plots as compared to the traditional histograms in order to make the observations clearer. Thus, it can be seen that the conventional theoretical-analysis tools used for the density evolution analysis of LDPC codes (i.e., extrinsic information transfer (EXIT) chart and protograph extrinsic information transfer (PEXIT) algorithm) and coding schemes tailored for channels with symmetric Gaussian characteristics cannot be directly applied to OAM-IM systems with inter-mode interference.

[0077] From a mathematical perspective, given the original data has a non-Gaussian distribution (especially skewed distribution), a parametric transformation technique, named Box-Cox transformation (BCT) may be utilized to convert this data into a new data set (i.e., transformed data) that is able to approximate a Gaussian distribution. The BCT may be defined as:where denotes the original data, andis the transformation parameter which is used to make the transformed dataapproximate a Gaussian distribution.

[0078] However, in certain scenarios, some LLR values(i.e., original data) obtained from the channel detector may comprise a negative value, especially for the set of coded bits(see the scattered points in Figures 6a and 6b, and Figures 7a and 7b). As a result, the BCT technique as disclosed in equation 15 may not be used. Instead, a variant referred to as the Customized Box-Cox transformation (CBCT) may be utilized and the CBCT is defined as:where denotes an absolute value operator. The transformed data can then be attained through the use of equation (16), whereby histograms associated with these transformed data may approximate symmetric Gaussian distributions. To verify this, the histograms of transformed extrinsic LLR values whenwithwithare plotted in Figure 6b (i.e., the dashed line and the line with circle in Figure 6b). Based on these plots, it can be seen that histograms associated with these transformed extrinsic LLR values are able to match with the symmetric Gaussian distributions having the values ofand uA similar occurrence may be observed in Figure 7b, where the histograms of transformed extrinsic LLR values are attained usingfor bothand 1, and the mean values of the corresponding symmetric Gaussian distributions are

[0079] In order to determine the optimalto be used in the CBCT procedure, a Kullback- Leibler divergence (KLD) analysis is utilized together with a wave-crest-based constraint.When these two approaches are combined, this ensures that a small KLDexists between the transformed distribution and its corresponding Gaussian distribution while ensuring that such two distributions have similar wave crests. For other angles (e.g.,= 1.0), the original LLRs obtained from the channel detector of OAM-IM systems may also follow a skewed distribution. As such, the CBCT technique may be utilized to attain the corresponding transformed LLRs, whose histograms approximate a symmetric Gaussian distribution. It is useful to note that the plots illustrated in Figures 6 and 7 may be made smoother by increasing the number of samples (i.e., the number of frames).

[0080] As observed from Figures 6b and 7b, the proposed CBCF technique primarily shapes the long left / right tails of the skewed distributions, which make them approach the symmetric Gaussian distributions. In particular, as can be seen from equation (16), both the original and transformed LLRs have the same positive / negative sign. This implies that when a coded bit has a high probability of being decoded as 0 / 1 in the original model (i.e., the original LLR has a positive / negative sign), it is also likely to be decoded as 0 / 1 in the transformation model (i.e., a CBCT-aided OAM-IM system over an LoS channel that utilizes the CBCT technique) and vice versa. Therefore, the transformation model does not deteriorate any accuracy of the BP decoding, but only simplifies the EXIT analysis and code design. For the above reason, the transformation model is able to capture the performance of the PLDPC codes in the original model. Furthermore, through the use of the CBCT, a feasible low-complexity theoretical- analysis tool may be devised to guide the design and optimization of PLDPC codes and index schemes for improving the performance of OAM-IM systems over LoS channels. However, the CBCT itself cannot directly improve the performance of the system since it is not used in the practical data transmission process.

[0081] Protograph-Based Low-Density Parity-Check (PLDPC) Codes

[0082] PLDPC codes comprise codes which are generated by expanding a small Tanner graph, referred to as protograph. Specifically, a protograph can be represented byis a variable-node (is a check-nodeis an edge set anddenotes the edge connecting theCN to the The code rate of a protographisA protograph can be expressed as a base matrixwhose rowsand columns correspond to the CNs and VNs, respectively, wheredenotes the number of edges connecting to Given a protograph (i.e., base matrix), a lifting operation may beperformed on it and the corresponding derived graph (i.e., parity-check matrix) may be attained, which in turn may be used to generate the PLDPC code In particular, unlike the conventional LDPC codes, multiple edges connecting a CN and a VN (i.e., parallel edges) are permitted in a protograph, but they must be eliminated in a lifting process. A lengthPLDPC code can be generated from a protograph with is a liftingfactor. In embodiments of this disclosure, the lifting operation is implemented using a modified progressive-edge-growth (PEG) algorithm.

[0083] A CBCT-PEXIT Algorithm

[0084] A Protograph Extrinsic Information Transfer (PEXIT) algorithm is an extension of the Extrinsic Information Transfer (EXIT) analysis used specifically for protograph-based Low-Density Parity-Check (LPDC) codes. Protograph-based LDPC codes are a class of structured LDPC codes that are designed based on a small bipartite graph, called a protograph, which is then lifted to create a larger LDPC code. The PEXIT algorithm is an asymptotic- performance analysis tool, which can trace the mutual information (MI) evolution of a PLDPC code to predict its decoding threshold. The decoding threshold is regarded as the minimum SNR that allows the PLDPC code to achieve error-free transmission, and thus can be used to evaluate the convergence performance of PLDPC-coded OAMIM systems. For the PEXIT algorithm, the Ml evolution process includes two different parts. The first part is the calculation of MI output from the channel detector, while the second part is the MI update between the VN decoder and CN decoder. Note that the computational complexity of the PEXIT algorithm is mainly dominated by the first part.

[0085] To be specific, for a given channel environment, the extrinsicobtained from the detector can be calculated as:where i s a converted binary symbol and is a coded bit. Underthe assumption that the conditional PDFin equation (17) is approximated by a symmetric Gaussian distribution the conditional PDF may be defined as:As a result, equation (17) may be simplified as:where IE [•] is the expectation function and / is the LLR corresponding to

[0086] For existing OAM-IM systems (i.e., the original model), equation (17) has to be utilized to calculate the extrinsic MIas the original extrinsic LLRs tend to follow a skewed distribution. Based on equation (17), the discrete numerical values have to be computed, the corresponding conditional PDFs must be estimated, while this involves an integration operation. In particular, for a non-Gaussian distribution (e g., a skewed distribution), it is challenging to obtain a close-form solution for the integration, as such the numerical method (i.e., histogram) is usually adopted to calculate the extrinsic MIand this leads to a highly complex calculation.

[0087] This issue may be addressed through the use of the CBCT, in which the transformed extrinsic LLRs have a close correlation with a symmetric Gaussian distribution. A transformed extrinsic MImay then be obtained for the CBCT-aided OAM-IM systems (i.e., the transformation model) based on equation (19), which involves much simpler steps as compared to equation (17) as it only involves discrete numerical calculations and an expectation operation. In view of the above, a CBCT-PEXIT algorithm is defined below.

[0088] Figure 8 illustrates a principle of this CBCT-PEXIT algorithm, in which the mutual information (MI) is defined as follows.• represents the transformed extrinsic MI of as obtained from the MIcalculator;.and indicate the a-priori Mis sent fromto Vj andtorespectively;• and indicate the extrinsic Mis sent from to andtorespectively;•indicates the a-posteriori MI of Vj.

[0089] In particular, and are equal to and ,respectively, in each inner iteration between the VN decoder and the CN decoder. Tmis set as the maximum number of inner iterations and the CBCT-PEXIT algorithm is defined as follows.

[0090] During an initialization step, an initial SNR is set, and a Monte Carlo simulation is implemented. The receiving UCA is then configured to obtain a sequence of channel outputs 801 . This sequence 801 will be demultiplexed to beand then processed by channel detector 802.

[0091] During the step of computing the transformed extrinsic Mis output from MI calculator 806, channel detector 802 outputs a sequence of extrinsic LLRs toCBCT module 804. Through the use of equation (16), these extrinsic LLRs are converted to a new sequence . Based on the transformed extrinsic LLRs and theircorresponding coded bits, the transformed extrinsic MI of whereis calculated as:where{+1, -1} and are the A-th converted binary symbol andcoded bit, respectively. Subsequently, is sent to VN decoder 808 and used as initialWhen theVNis punctured in a protograph,In embodiments of the disclosure, MI calculator 806 may comprise a module that is configured to calculate the mutual information between the input and output of a channel in order to assess the channel capacity and the efficiency of the channel’s transmission. MI calculators typicallyapply a formula for mutual information, as known to one skilled in the art, to probability distributions of the random variables and then compute the entropy of the individual variables. The output produced by the MI calculator comprises the mutual information value, which is a non-negative number measured in bits (for binary variables).

[0092] During the step of determining the extrinsic Mis obtained from VN decoder 808, i.e., the initial Mis and the a-prioriare exploited. The extrinsic Misare updated and considered as The detailed process can be foundin the publication by Z. Yang et. al. titled "‘Analysis and optimization of tail-biting spatially coupled protograph LDPC codes for BICM-ID systems, ” in the IICI CI I 1r' ans. Veh. Technol., vol. 69, no. 1, al pp. 390- -404, Jan. 2020,' and in the publication by G. Liva andM. Chiani titled “Protograph LDPC codes design based on EXI T analysis, ” in the Proc. IEEE Glob. Commun. Conf. (GLOBECOM), Nov. 2007, atpp. 3250 3254.

[0093] During the step of determining the extrinsic Mis output from CN decoder 810, given the a-priori the extrinsic MIscan be updated if andconsidered as in next inner iteration. The detailed process of this step is also set outin the two publications mentioned above.

[0094] During the step of computing the a-posteriori MIs for VNs, based on the initial Mis an d the MIsthe a-posleriori(where j = 1, 2, . . . , nv)can be computed. More detailed calculations concerning can be found in the twopublications mentioned above.

[0095] At the final step, the CBCT-PEXIT algorithm is terminated, and the current SNR is used as a decoding threshold if = 1 for all VNs. Otherwise, i) this algorithm will restartwith a larger SNR ifis reached; ii) return to Step 3) and continue to trace MI convergence if is not reached.

[0096] A traditional EXIT chart is not applicable to PLDPC codes because the protographs allow the presence of degree-1 VNs and punctured VNs. Also, different protographs may have the same degree distribution. In the traditional PEXIT algorithm, the initial MI calculation is based on the assumptions of infinite-length codeword, binary modulation, and an AWGNchannel. For the OAM-IM systems with finite-length codeword, high-order modulation, and non-Gaussian channel, the initial MI calculation (i.e., equation (20)) is different from that in the PEXIT algorithm.

[0097] In summary, for the CBCT-PEXIT algorithm, the CBCT is an essential link between the channel detector and VN decoder, which allows equation (19) to be used when the transformed extrinsic Mis for OAM-IM systems with inter-mode interference are calculated. The analysis based on the Kullback-Leibler Divergence (KLD) has revealed that a short codeword length does not provide sufficient extrinsic LLRs for accurate approximation. Consequently, the conditional histogram of the transformed extrinsic LLRs obtained through CBCT cannot accurately approximate a symmetric Gaussian distribution. Additionally, the standard PEXIT analysis assumes an infinite-length codeword to calculate the decoding threshold. Therefore, to make an accurate prediction of the convergence performance limit of the PLDPC code in this algorithm, it is recommended to set a large codeword length9000). While the transformed extrinsic LLRs exhibit symmetric and Gaussian characteristics, it is important to note that they are the output of a transformation model (i.e., the CBCT-aided OAM-IM system with inter-mode interference) rather than Additive White Gaussian Noise (AWGN) channels

[0098] Proposed PLDPC Codes IUP PLDPC / IPX PLDPC / LP2 PLDPC codes

[0099] A PLDPC code that performs well in a specific transmission environment may not necessarily be a promising candidate for other transmission environments due to the difference in channel characteristics. Moreover, even if the channel’s characteristics are given, a PLDPC code is usually unable to achieve desirable performance in both low and high-SNR regions simultaneously. For AWGN channels, the accumulate-repeat-by-4-jagged-accumulate (AR4JA) code is viewed as a competitive PLDPC code since it can achieve a good performance tradeoff between the waterfall and error-floor regions.

[0100] Figure 9 illustrates protographs for a family of AR4JA codes with a wide range of rates fromwheredenotes an arbitrarily large positive value; the dark circles, blank circle, and circles with a plus sign represent the transmitted VNs, punctured VN, and CNs, respectively. The rate AR4JA code is constructed by repeatedly addingdegree-4 VNs that connect the two highest-degree CNs into a rate-1 / 2 AR4JA code. In OAM- IM systems over LoS channels, the transmitted signals will suffer from inter-mode interference and their extrinsic LLRs output from channel detector may follow a non-Gaussian distribution. Consequently, the PLDPC-code design for OAM-IMsystems has to be addressed

[0101] In the following section, a CBCT-PEXIT-aided computer-search method is employed to construct two types of rate-1 / 2 improved PLDPC codes tailored for the proposed transformation model. These two types of rate-1 / 2 improved PLDPC codes are also the promising alternatives for use in original OAM-IM systems (i.e., without using CBCT)

[0102] Firstly, a small-size all-zero-base matrix Bo is defined The searchspace then equals if an exhaustive computer-search method is used tofind the resultant base matrixthat corresponds to a PLDPC code having an optimal decoding threshold. In particular,is the maximum integer element inWith an aim of reducing the search space and lowering the encoding / decoding complexity, it is assumed that within satisfies the condition thatMeanwhile, two fundamental characteristics of a good PLDPC code should betaken into account which is that: (1) an appropriate proportion of degree-1 and degree-2 VNs should be added into the protograph to ensure that the resultant PLDPC code attains a relatively low decoding threshold; and (2) a CN should be used to connect a degree-1 unpunctured VN and a highest-degree punctured VN to formulate a precoding structure that can further improve the threshold performance for a punctured PLDPC code.

[0103] Based on the above, the design for two different types of improved PLDPC codes are described below when simple constraints are imposed onto formulate an initial base matrix B with the same size, and when the correspondingis attained with the optimal decoding threshold using the CBCT-PEXIT-aided computer-search method depicted in Algorithm 3 below.Algorithm 3 CBCT-PEXIT-aided computer-search methodInput: Binit, constraints related toand a pre-set decoding threshold (e.g., T- 100);1 : for i = 1 , 2, ..., n do2: Generate a tentative base matri based onvia a constraint-based enumeration;3: Calculate the decoding threshold ofusing the CBCT-PEXIT algorithm;4: if5.6: end if7: end for

[0104] Improved unpunctured PLDPC codes:

[0105] For a base matrix, the column swapping operation does not change its decoding threshold. Under the assumption that an initial 4 x 8 base matrixwith a degree-1 VN and a degree-2 VN is given, the corresponding may then be defined as:

[0106] When three different constrains (i.e., the CN-degree, VN-degree, and element constrains) are considered:with the aid of a computer-search-basedprocedure, a rate-1 / 2 improved unpunctured PLDPC code, called IUP PLDPC code may be obtained and its base matrix may be defined by:

[0107] Improved punctured PLDPC codes:

[0108] When a 3 x 5 base matrixis initialized and added with a degree-1 VN and a degree-2 VN, and under the assumption that the second column of the initial base matrixcorresponds to a punctured VN, the correspondingmay be defined as:

[0109] When three different constrains (i.e., the CN-degree, VN-degree, and element constrains) are considered: i)By further adjusting thenumber of degree-2 VNsand using the computer-search method, two different rate-1 / 2 improved punctured PLDPC codes denoted as IP1 and IP2 PLDPC codes are obtained. Their base matricesare given by:[001 10] In particular, the TP1 PLDPC code was found to achieve a desirable tradeoff between the waterfall and error-floor performance; while the IP2 PLDPC code was found to have the optimal waterfall performance but suffers from error-floor behavior in high-SNR region. For most wireless transmission environments, high-rate PLDPC codes are preferred over low-rate counterparts, which inspires the use of a simple rate-extension scheme to construct high-rate IP1 and IP2 PLDPC codes As it has been determined that when VNs with degrees no less than 3 are appended to a protograph, it results in PLDPC code that is able to achieve good waterfall performance without deterioration in the error-floor performance, the same VN pattern composed of two different degree-3 VNs is thus repeatedly added into the protographs of rate- 1 / 2 1P1 and 1P2 PLDPC codes. Under the assumption that the number of newly added VN patterns is e > 1, the code rate of an extended 1P1 / 1P2 PLDPC code is.... The protographs corresponding to such two families of improved punctured PLDPC codes are presented in Figures 10a in relation to the family of IP1 PLDPC codes and Figure 10b in relation to the family of IP2 PLDPC codes, while their base matrices are expressed as:

[0111] It is noted that better high-rate PLDPC codes may be constructed if the CBCT- PEXIT-aided computer-search method is used to search the large-size base matrices without considering any constraints. However, to reduce the search space (i.e., lower the PLDPC-code construction complexity), a rate-1 / 2 punctured PLDPC code is first constructed with a small size base matrix based on some given constraints, and then the same VN pattern is repeatedly introduced into the protograph of this code to formulate the high-rate counterparts.

[0112] Also, as OAM systems are mostly used for LoS backhaul links between base stations, the misaligned condition usually remains constant for a long time. As a result, the channel coefficient, which is closely related to the 0AM system setting, also remains constant during each transmission period. The OAM-based LoS channel then can be viewed as a wireless channel with inter-mode interference. Hence, the natural step would be to design PLDPC codes for such a particular channel configuration since the codes designed for conventional fading MIMO channels may not be able to preserve their performance advantage for OAM applications.

[0113] Mutual Information (MI)-based Index Scheme[001 14] For an OAM-IM system with N array elements in both transmitting and receiving UCAs andactive modes, the number of effective OAM mode combinations,may be defined asFor example, when N andare respectively set to be 4 and 2, the number of effective OAM mode combinations,equals 4 Channel gains of a 4 x 4 OAM- IM system without inter-mode interference are illustrated in Figure 1 la; and withinter-mode interferenceare illustrated in Figure 11b. As shown, it can be seenthat in the case with inter-mode interference (see Figure 1 lb), the signals emitted by different modes will suffer from different channel gains. In the following section, a method for determining an optimal number of effective OAM mode combinations, Bewill be disclosed.

[0115] From the perspective of Monte-Carlo simulations, an N x N OAM-IM system can be viewed as N different 1 x N OAM-IM systems. Moreover, the CBCT-PEX1T algorithm described above can be used to evaluate the reliability of signals transmitted by N different OAM modes. Specifically, for a given SNR, a sufficiently large, modulated symbol sequence is first generated, and then divided into N subsequences in a sequential order The modulated symbols within the A-th sub-sequence will be transmitted by the A-th mode, where k = 1, 2, . . . , N. Subsequently, the previously described CBCT-PEXIT algorithm is utilized to calculate N different extrinsic Mis as obtained from the MI calculator. By repeatedly executing the above procedure for different SNRs, a N number of different extrinsic MI convergence curves corresponding to N different modes may be obtained. Based on the extrinsic MI analysis, Bemode combinations having the largest Mis may then be determined, and these mode combinations may then be arranged in a descending order such that the reliability of themode combination is higher than that of the mode combination, where k = 1, 2, . . .

[0116] As an example, the extrinsic Ml curves in a 4 x 4 OAM-IM system when 6 = 0.8, and where the regular-(3, 6) PLDPC code and Gray-labeled QPSK modulation are used is illustrated in Figure 12. In this figure, the extrinsic Mis obtained from the detector using CBCT are analyzed, and it can be seen that these values correspond to coded bits transmitted by 4 different modes (see dashed lines). The extrinsic Mis obtained from the detector using conventional PEXTT algorithms (see solid lines) are also analyzed, and it can be seen that their trends are consistent with the dashed lines. This implies that if a code is able to achieve a good convergence performance in the transformation model, it can also achieve a good convergence performance in the original model. In this embodiment of the disclosure, is defined as thereliability of the Gth mode. Hence, the greater the extrinsic MI of coded bits transmitted by the mode, the higher the correspondingwill be. As shown in Figure 12, 4 different modes satisfy this condition,Under the assumption that the number of active modes is N , this implies that there will be 4 different mode combinations, i.e.,with largest MIs, and this hereinafter is defined as an MI-based index scheme. Table 5 shows theMl-based index scheme, the natural index scheme, and the mapping relationship between them and coded bits.

[0117] To verify the superiority of the mode-combination selection over the natural index scheme, the constellation-constrained capacities are computed for the natural index and ML based index schemes and the results are plotted in Figure 13. For both QPSK and 8PSK modulations, it can be seen that the Mi-based index scheme has larger capacities than the natural index scheme over a wide range of SNRs (i.e., 65 ~ 85 dB). In particular, it should be noted that the performance of the Mi-based index scheme may be further improved using a shaped Index Modulation scheme (as previously described in Embodiment 2 of the transmitter).

[0118] Decoding-Threshold Analysis

[0119] In order to demonstrate the superiorities of the improved PLDPC codes (i.e., IUP, IP1, and IP2 PLDPC codes) and the shaped index scheme, the convergence performance of these codes and scheme are assessed for 4 x 4 OAM-IM systems with 0 = 0.8 over LoS channels. Additionally, two different PLDPC codes (i.e., regular-(3, 6) and AR4JA PLDPC codes) and two different index schemes (i.e., natural and ED-based index schemes) are considered as benchmarks for the proposed code construction and shaping index design, respectively.

[0120] Table 6 below illustrates the decoding thresholds of the above five PLDPC codes with QPSK modulation and the natural index are calculated by exploiting the above described CBCT-PEXIT algorithm.

[0121] When NA= 1, the proposed TUP PLDPC code achieves a smaller decoding threshold over regular-(3, 6) PLDPC code. The proposed 1P1 and 1P2 PLDPC codes also exhibit smaller decoding thresholds over AR4JA PLDPC code. Furthermore, a similar phenomenon can be observed in the case of NA= 2. This observation demonstrates that the proposed improved PLDPC codes are promising alternatives for use in OAM-IM systems.

[0122] The decoding thresholds of these five PLDPC codes in the case of using original LLRs (i.e., without CBCT) are analyzed in Table 7 below.

[0123] When NA= 1 , one may notice that the proposed TUP PLDPC code and IP1 / IP2 PLDPC code significantly outperforms the regular-(3, 6) PLDPC code and AR4JA PLDPC code for both QPSK and 8PSK modulations, respectively. The convergence performance trends of these PLDPC codes are also consistent with those in Table 6 above, which further implies the effectiveness of the CBCT-PEXIT algorithm.

[0124] In Table 8, the decoding thresholds of these PLDPC codes are compared with three different index schemes for NA= 2.

[0125] Specifically, for a given QPSK / 8PSK modulation, the PLDPC codes with the proposed shaped index scheme are capable of achieving better decoding thresholds than those with existing index schemes (i.e., non-shaped index schemes) Therefore, the performance of PLDPC-coded OAM-IM systems can be significantly enhanced with the use of shaping index described in the previous sections.

[0126] Simulation Results

[0127] In these simulations, the simulated BERs of two known LDPC codes (i.e., Code I and Code II) designed for MIMO channels, the regular-(3, 6), AR4JA, and the IUP, IP1, and IP2 PLDPC codes in 4 x 4 OAM systems over LoS channels are compared. In particular, QPSK and 8PSK modulations as well as three different index schemes (i.e., natural index, ED-based index, and shaped index schemes) are considered. Unless otherwise stated, it is assumed that the parameters are 0 = 0.8, and N= 1 and 2. Furthermore, all PLDPC codes are configured to have the same code rate a transmitted codeword length 2400 and the maximumnumber of BP decoding iterations is set to be equal to 100.

[0128] A. PLDPC Codes with (Natural Index Scheme)

[0129] With reference to Figure 14a, when NA= 1 and QPSK modulation are considered, it is shown from the plots that the IUP, IP1, and IP2 PLDPC codes (as described in the sections above) significantly outperform the other four benchmark codes. Specifically, at a BER of 10-5, the proposed IUP PLDPC code achieves performance gains around 0.3, 0.3, and 0.1 dB over the regular-(3, 6) PLDPC code, Code I, and Code II, respectively. Also, the proposed IP1 and IP2 PLDPC codes achieve performance gains around 0.2 and 0.25 dB over the AR4JA PLDPC code, respectively.

[0130] A similar observation can be seen in Figure 14b, when NA= 1 and 8PSK modulation are used, the proposed IP2 PLDPC code exhibits the best waterfall performance among these seven different codes but suffers from an error-floor behavior in the high-SNR region. This is because the IP2 PLDPC code includes a high proportion of degree-2 VNs (i.e., 40% VNs in each codeword), which forms a girth-4 cycle (see equation (24)). By increasing additional VNs with degree at least 3 in the protograph of the IP2 PLDPC code, its error-floor performance can be improved significantly. It is useful to note that the large value of in simulationsis caused by the system parameter settings (i.e., P, which results in a small channel gain For practical OAM applications, there exist some OAM antennas which can generate a high-gain OAM vortex beam (e g., the antenna gain 100), and then one onlyneeds a relatively small SNR to lower the BER.

[0131] As shown in 15a and 15b, when = 2 is considered, it can be observed that the performance of the IUP, IP1, and IP2 PLDPC codes are still better than that of other four benchmark codes for both QPSK (see Figure 15a) and 8PSK (see Figure 15b) modulations. Thus, it may be concluded that for different modulations and numbers of active OAM modes the proposed improved PLDPC codes are promising alternatives for use in OAM-1Msystems over LoS channels.

[0132] It is useful to note at this stage that both the TUP and TP1 PLDPC codes have a low proportion of degree-2 VNs and no girth-4 cycle. For this reason, these codes are able to avoid the error floor and achieve better performance in the high SNR region with respect to the IP2 PLDPC code.

[0133] With reference to equations (22) and (24), it can be observed that the base matrix of the IUP PLDPC code has (i) higher edge density (i.e., the average degree for one VN is 3.625) and (ii) higher proportion of parallel edges (i.e., the average number of parallel edges for one VN is 1. 125) than the base matrix of the IP1 PLDPC code, whose average degree and average number of parallel edges for one VN are 2.8 and 0.8, respectively. Then, the IUP PLDPC code is more likely to have short cycles in the resultant parity-check matrix and thus perform worse in the high SNR region, compared with the IP1 PLDPC code.

[0134] Figure 16 further compares the performance of the 5G LDPC code, three irregular LDPC codes (i.e , irregular LDPC codes with maximum VN degrees dv= 4, 5, and 50), (3, 6) spatially coupled (SC)-LDPC code, and the proposed improved PLDPC codes. It is obvious that the proposed improved PLDPC codes are the best-performing codes, especially in the high- SNR region. The proposed IUP, IP 1, and IP2 PLDPC codes exhibit gaps of about 0.59, 0.62, and 0.40 dB, respectively, to the Shannon limit (i.e., 79.71 dB). The gaps of the three proposed PLDPC codes to the Shannon limit are very desirable for the code design over OAM-based LoS channels.

[0135] B. PLDPC Codes with (Natural Index Scheme)

[0136] Figure 17 illustrates the BER results of the AR4JA, IP1, and IP2 PLDPC codes with code ratesBased on this, it was found that the IP1 and IP2 PLDPC codes are superior to the AR4JA PLDPC code. Specifically, the IP1 and IP2 PLDPC codes respectively possess additional performance gains of about 0.2 and 0.2 ~ 0.3 dB over the AR4JA PLDPC code in the whole range of code rates. Additionally, since the degree of newly added VNs is greater than 2, the high-rate IP2 PLDPC codes have no error-floor behavior even when the BER equals 10-6. Based on the above, it can be seen that the IUP, IP1, and IP2 PLDPC codes with rate-compatible structures are good candidates for use in OAM-IM systems with a high-transmission-rate demand.

[0137] C. PLDPC Codes with Shaped Index Schemes

[0138] Figures 18a and 18b compare the BER performance of the regular-(3, 6) PLDPC code, AR4JA PLDPC code, and three improved PLDPC codes with the natural index and proposed shaped index schemes in the case ofIt can be seen that all PLDPC codes are capable of achieving significant performance gains with the aid of proposed shaped index scheme. Specifically, for QPSK modulation (see Figure 18a), the regular-(3, 6), AR4JA, and proposed IUP, IP1, and IP2 PLDPC codes achieve performance gains of about 2.2, 2.2, 2.3, 2.4, and 2.5 dB, respectively, at a BER of 2 x 10-5by using the proposed shaped index scheme. On the other hand, this technique suffers from a spectral-efficiency loss of only 0.5 bpcu (i .e , 8.33% of the original bpcu). Additionally, a similar phenomenon can be noticed in Figure 18b where 8PSK modulation is considered, but the spectral-efficiency loss is only 6.25% of the original bpcu.

[0139] Figures 19a, 19b and 19c investigate the performance of three improved PLDPC codes with the ED-based index and proposed shaped index schemes. Specifically, for IUP PLDPC code (see Figure 19a) the shaped index scheme achieves a significant performance gain over the ED-based index scheme. Also, a similar phenomenon can be observed for IP1 and IP2 PLDPC codes in Figures 19b & 19c, respectively. In view of the above, it can be said that the proposed shaped index scheme is a good alternative for conventional PLDPC codes (e g., regular-(3, 6) and AR4JA PLDPC codes), and thus can be treated as an appealing technique for boosting system performance with a slight spectral -efficiency loss.

[0140] Simulated Implementation in a multiple-input multiple-output (MIMO) communications for 5G

[0141] Figure 20 illustrates the performance of the (3, 6) PLDPC code in 4 x 4 MIMO visible light communication (VLC) systems with NA= 1, 2 , and four different index schemes, including a conventional spatial modulation (CONSM) scheme, an ED scheme and their shaped counterparts. From this plot, it can be seen that the MIMO-VLC systems can achieve significant performance gains with the aid of the shaped index scheme (i.e., shaped ED (proposed) and shaped CONSM (proposed))

[0142] Simulated Implementation in multi-cell network systems

[0143] Figures 21a, 21b and 21c illustrate the performance of (3, 6) PLDPC code in multicell network systems with two different index schemes, including conventional spatial modulation (CONSM) scheme and its shaped counterpart. It can be seen from these plots that cell-edge users achieve significant performance gains with the aid of the shaped index scheme. Specifically, the larger the multi-user interference (i.e., superimposition of user interferences in a cell) is, the larger the performance gain can be obtained.

[0144] Numerous other changes, substitutions, variations, and modifications may be ascertained by the skilled in the art and it is intended that the present application encompass all such changes, substitutions, variations, and modifications as falling within the scope of the appended claims.

Claims

CLAIMS1. A shape index modulation transmiter for a multi-mode orbital angular momentum (OAM) system comprising: a processing unit; and a non-transitory media readable by the processing unit, the media storing instructions that when executed by the processing unit causes the processing unit to: divide, using a shaping encoder module, a binary coded-bit sequence received from a bit-splitter module into aof sub-sequences, whereby K is an integer that is determined based on a total number of OAM modes and a number of active OAM modes; process, using a shaping function provided within the shaping encoder module, the subsequences to increase a probability that bits in the sub-sequences comprise a predetermined binary value; formulate, using the shaping encoder module, index symbols based on the processed sub-sequences, whereby a first bit of each of the index symbols is obtained from a first processed sub-sequence; and provide, using the shaping encoder module, the index symbols to an OAM mode selector module for selecting corresponding OAM mode combinations for transmission by the transmitter, whereby optimal OAM mode combinations of the transmitter are associated with index symbols that comprise primarily of the predetermined binary value.

2. The transmiter according to claim 1 whereby a length of the first sub-sequence of the series of sub-sequences comprises a first length and lengths of remaining sub-sequences of the series of sub-sequences each comprise a second length, wherein the first length differs from the second length.

3. The transmitter according to claim 2 whereby the instructions to process the series of subsequences using the shaping function comprises instructions for directing the processing unit to: generate, using the shaping encoder module, a codebook comprising an initial set of distinct binary sequences that are mapped to a final set of distinct binary sequences that comprise primarily of the predetermined binary value, whereby a length of each binary sequence in the initial set of distinct binary sequences is equal to the first length and alength of each binary sequence in the final set of distinct binary sequences is equal to the second length; and convert, using the shaping encoder module, the first sub-sequence to a shaped first sub- sequence based on the mappings in the codebook.

4. The transmitter according to any one of claims 1 to 3 whereby the instructions to formulate the index symbols based on the processed sub-sequences comprises instructions for directing the processing unit to: for the formulation of each index symbol, sequentially extract, using the shaping encoder module, a predefined number of distinct bits from each of the processed sub-sequences and concatenate the extracted bits to form the index symbol.

5. The transmitter according to claim 4 wherein the predefined number of distinct bits extracted from each of the processed sub-sequences is defined as the number K of the subsequences.

6. The transmitter according to any one of claims 1 to 5, whereby the optimal OAM mode optimal mode combinations that wereselected by channel estimation of aorbital angular momentum-index modulation system having NAactive OAM modes where N is defined as a number of array elements and NAis defined as a number of active modes, whereby7. The transmitter according to any one of claims 1 to 5, whereby the optimal OAM mode combinations of the transmitter comprise optimal mode combinations that were selected using a mutual information (Ml)-based index scheme.

8. The transmitter according to claim 7, whereby the selection of the optimal mode combinations using the Mi-based index scheme comprises instructions for directing the processing unit to: perform one or more rounds of mode-combination selection, using a MI-based index scheme module that is communicatively coupled to the OAM mode selector, each round comprising: generating a modulated symbol sequence for a distinct signal -to-noise ratio (SNR);sequentially dividing the modulated symbol sequence into a N number of sub- sequences, whereby the modulated symbols within a k-th sub-sequence are to be transmitted by a k-th mode, where k is defined as a number from a set { 1, 2, ..., N} and N is defined as the total number of different OAM modes; computing, using an MI calculator, extrinsic MI values for each mode based on the sub-sequence associated with the mode; and determining mode combinations to be used as the optimal mode combinations based on mode combinations that have the largest MI values.

9. The transmitter according to claim 8, whereby the MI calculator computes extrinsic MI values based on a customized Box-Cox transformation (CBCT)-based protograph extrinsic information transfer (PEXIT) algorithm.

10. The transmitter according to any one of claims 1-9 wherein the binary coded-bit sequence comprises a bit sequence coded based on protograph low-density parity-check (PLDPC) codes, an improved unpunctured (IUP) PLDPC code, or one of two improved punctured PLDPC codes.

11. A method for shaping index modulation in a multi-mode orbital angular momentum (OAM) system comprising a shape index modulation transmitter having a shaping encoder module communicatively coupled to an OAM model selector module, the method comprising the steps of: dividing, using the shaping encoder module, a binary coded-bit sequence received from a bit-splitter module into aof sub-sequences, whereby K is an integer that is determined based on a total number of OAM modes and a number of active OAM modes; processing, using a shaping function provided within the shaping encoder module, the sub- sequences to increase a probability that bits in the sub-sequences comprise a predetermined binary value; formulating, using the shaping encoder module, index symbols based on the processed sub- sequences, whereby a first bit of each of the index symbols is obtained from a first processed sub-sequence, and providing, using the shaping encoder module, the index symbols to an OAM mode selector module for selecting corresponding OAM mode combinations for transmission by thetransmitter, whereby optimal OAM mode combinations of the transmitter are associated with index symbols that comprise primarily of the predetermined binary value.

12. The method according to claim 11 whereby a length of the first sub-sequence of the series of sub-sequences comprises a first length and lengths of remaining sub-sequences of the series of sub-sequences each comprise a second length, wherein the first length differs from the second length.

13. The method according to claim 12 whereby the step of processing the series of sub- sequences using the shaping function comprises: generating, using the shaping encoder module, a codebook comprising an initial set of distinct binary sequences that are mapped to a final set of distinct binary sequences that comprise primarily of the predetermined binary value, whereby a length of each binary sequence in the initial set of distinct binary sequences is equal to the first length and a length of each binary sequence in the final set of distinct binary sequences is equal to the second length; and converting, using the shaping encoder module, the first sub-sequence to a shaped first sub- sequence based on the mappings in the codebook.

14. The method according to any one of claims 11 to 13 whereby the step of formulating the index symbols based on the processed sub-sequences comprises: for the formulation of each index symbol, sequentially extracting, using the shaping encoder module, a predefined number of distinct bits from each of the processed sub-sequences and concatenate the extracted bits to form the index symbol.

15. The method according to claim 14 wherein the predefined number of distinct bits extracted from each of the processed sub-sequences is defined as the number K of the sub-sequences.

16. The method according to any one of claims 11 to 15, whereby the optimal OAM mode combinations of the transmitter compriseoptimal mode combinations that were selected by channel estimation of a orbital angular momentum-index modulationsystem having active OAM modes where N is defined as a number of array elementsand is defined as a number of active modes, whereby17. The method according to any one of claims 11 to 15, whereby the optimal OAM mode combinations of the transmitter comprise optimal mode combinations that were selected using a mutual information (Ml)-based index scheme.

18. The method according to claim 17, whereby the step of selecting the optimal mode combinations using the Mi-based index scheme comprises: performing one or more rounds of mode-combination selection, using a Ml-based index scheme module that is communicatively coupled to the OAM mode selector, each round comprising the steps of: generating a modulated symbol sequence for a distinct signal -to-noise ratio (SNR); sequentially dividing the modulated symbol sequence into a N number of sub- sequences, whereby the modulated symbols within a k-th sub-sequence are to be transmitted by a k-th mode, where k is defined as a number from a set {1, 2, N) and N is defined as the total number of different OAM modes; computing, using an MI calculator, extrinsic MI values for each mode based on the sub-sequence associated with the mode; and determining mode combinations to be used as the optimal mode combinations based on mode combinations that have the largest MI values.

19. The method according to claim 18, whereby the MI calculator computes extrinsic MI values based on a customized Box-Cox transformation (CBCT)-based protograph extrinsic information transfer (PEXIT) algorithm.

20. The method according to any one of claims 11-19 wherein the binary coded-bit sequence comprises a bit sequence coded based on protograph low-density parity-check (PLDPC) codes, an improved unpunctured (TUP) PLDPC code, or one of two improved punctured PLDPC codes.

Citation Information

Patent Citations

  • Method for transmitting and receiving signal in optical wireless communication system, and transmission terminal and reception terminal therefor

    US20220321213A1

  • Information transmission by circle selection and detection in coaxial multi-circle orbital angular momentum (OAM) communication system

    WO2022141425A1

  • Mode division duplex for orbital angular momentum communications

    WO2022178705A1