ABSTRACT In this paper, we present a modulation scheme called generalized symbol sequence modulation (GSSM), which incorporates multiple symbol sequence modulation mappers. In GSSM, additional information is conveyed through symbol sequences from different symbol constellations to increase data rates. In order to convey information, the sequence of symbols is used in three time slots; therefore, a single antenna can convey at least four bits. Unlike spatial modulation (SM), GSSM conveys information over fading and non‐fading channels. The performance of the GSSM system has been validated through bit error rate (BER) simulations and theoretical framework based on various receiving antennas and fading channels. Furthermore, in terms of secrecy rate, it is superior to conventional single input, multiple output systems (SIMO), space shift keying (SSK), and symbol sequence modulation (SSM). Also, GSSM has shown a reduction in receiver computational complexity when compared to SM at the same bit rate. BER results are presented for non‐fading, fading, correlated, and non‐correlated scenarios, under GSSM.

Abstract This paper discusses a new mapping scheme known as face coded modulation (FCM) system. In FCM, peak energy symbols are mapped onto an innermost ring according to the eight sockets in the human face, that is, brain, mouth, nostrils, eyes and ears. For example, FCM is formed when the constellation diagram from M ‐ary quadrature amplitude modulation ( M QAM) system is modified to reduce peak‐to‐average power ratio (PAPR) by relocating the four corner symbols of the M QAM, with peak energy, to the innermost ring in a way that forms the figure of a cross. Unlike APSK, FCM mapping introduces non‐uniform sequence of symbols on the ring, face width factor and multiple modulator circuits that can be used to lower power requirements for high power amplifiers (HPA) as used in M QAM transmission systems. Symbol error rate (SER) for FCM is calculated and the results compared with M QAM and M PSK. It is shown that at equal energy efficiency, FCM scheme has a better response to errors than both M PSK and M QAM and a better energy efficiency due to lowered PAPR than M QAM. Moreover, the simulation results exhibit a tight match for the proposed analytical framework when assessed under Additive White Gaussian Noise (AWGN) channel.

Abstract This paper discusses a new symbol modulation scheme known as polarized amplitude and phase shift keying, (‐APSK) modulation with four rings in its basic form. The new scheme maps power‐polarized symbol pairs on its constellation in order to increase the number of data symbols. ‐APSK exploits a symbol mapper based on a conjugate power splitting algorithm. Product modulation is applied, where a voltage signal of a specified amplitude and phase is multiplied by another current signal of a specified amplitude and phase, thus forming polymorphic signals in the product constellation. At the receiver, an isolated detection is performed, where the voltage signal is detected independently of the received current signal. A selection combining scheme is then used. The results depict unity peak‐to‐average power ratio and low average symbol energy, which is desirable for Green communications in 5G networks. It presents lower bit error rates when compared with the state‐of‐the art ‐ary quadrature amplitude modulation, with a signal‐to‐noise ratio gap of at least . The proposed analytical framework closely matches the simulations for bit error rates under .

Abstract This paper discusses layered baud‐space modulation system under arbitrary spatial correlation both at the transmitter and the receiver. Layered baud‐space modulation provides better reliability in spatial modulation systems as well as flexibility in data rates compared to multiple‐active space modulation. Several transmissions occur asynchronously within the time slot. As a result, fractional sampling is employed at the receiver to detect the signals. For this scheme, it is more practical to analyse spatial correlation both at the transmitter and the receiver, since temporal correlation is a likely occurrence. However, asynchronous transmission will lead to higher probability of packet drops when synchronization errors are present. Lower bounds of bit error rates are derived from theoretical criteria based on mutual information and sample covariance of the joint spatial and modulation space. Analytical results for bit error probability are tightly verified from simulations. The channel is assumed to be of frequency‐flat Rayleigh fading form. The tests are carried out under different antenna spacing, amplitude and phase modulation schemes and various number of transmit and receive antenna configurations.

Abstract 5G and future generation networks require physical layer security (PLS) to reinforce the security at the higher communication layers. A novel technique of securely transmitting data in a wireless network is presented, by ensuring its security and reliability on the physical layer. Information is encoded in two domains: the spatial domain, in terms of the sequence of transmitting antennas and in the signal domain, in terms of the symbol that the antennas transmit. Hence the name antenna sequence spatial modulation (ASSM). The case of three antennas is presented, where each antenna transmits the same symbol of the two possible symbols, of the bipolar phase shift keying (BPSK) modulation, in one of the three time slots. Both the transmitter and the receiver have a mapper table relating the antenna transmitting sequence and the symbol transmitted to a code word. Using the information theoretic approach, the performance analysis of the system is presented through the bit error rate (BER) and the secrecy capacity metrics on Rayleigh fading channel. The comparison of simulation and analytical results are presented. It is inferred that the ASSM has a higher secrecy rate than the conventional spatial modulation (SM), albeit at a lower spectral efficiency.

A simple optimal error analysis of multiple‐branch equal gain combining (EGC) and selection combining (SC) in the presence of arbitrarily correlated Rayleigh fading channels is presented. It is shown that the error rates over correlated branches for diversity combining schemes can be analysed at the output of an eigenfilter splitter, which is placed at the output of the diversity combiner. Since the outputs of the splitter are uncorrelated, the error analysis of maximal ratio combining (MRC) technique is applied to EGC and SC. This analysis is possible since error performance of MRC with correlated branches is optimal and invariant to eigenfilter or eigenvector decorrelation and the decorrelated outputs can simply be analysed similar to independent branches. The authors exploit extensive simulations for average bit error probability, which show tight bounds with the proposed analytical approach. Channel correlations for different practical antenna spacing of 0.15, 0.2 and 0.5 of the signal carrier wavelength, λ are analysed.

In this study, symbol error probability of transmitted modulation signals for a single‐input–multiple‐output system is used to derive overall average bit error probability (ABEP) of multiple active‐spatial modulation (MA‐SM) system as a function of the number density of bits in error. Specifically, a theoretic method of determining the number density of bits in error in the joint detector of MA‐SM is proposed. Sample variance (SV) criterion uses the sample covariance of detector noise data, while mutual information (MI) criterion uses relative entropy derived from the Kullback–Liebler distance to compute the number density of bits in error. The choice for the better criterion is formulated as a model selection problem between the SV and the MI criteria. Analytical results for ABEP are tightly verified from simulations, which are carried out under arbitrarily correlated flat Rayleigh fading channels. Various data rates, antenna spacing and number of transmit and receive antenna configurations are tested, where it is also shown that SM is a special case of MA‐SM. Moreover, a basis spectral efficiency (BSE) formula for MA‐SM systems is proposed, which can be used to select a preferred antenna configuration without the need for simulations.

A scheme is proposed, called polar coded spatial modulation (PCSM), to provide both detection/decoding of the transmitted symbol and the transmit antenna index in multiple‐input–multiple‐output (MIMO) antenna systems. Based on the free polar set and the frozen polar set in polar codes, the proposed PCSM scheme maps the transmitted information data in the free polar set and the transmit antenna index bits in the frozen polar set whose positions are known at the receiver. This has the advantage that the bits that identify the transmit antenna are not transmitted to the receiver and do not need to be detected as is the case in conventional spatial modulation (SM). Thus, PCSM presents a spectrally efficient and a power‐efficient system. For the simulations, MIMO channels with transmission over independent and identically distributed flat fading characteristics are exploited. Firstly, capacity results for coded SM which is a requirement for the construction of the codes are presented. Consequently, the bit error rate performances of the M‐ary quadrature amplitude modulation SM are analysed for various MIMO architectures after which the desirable MIMO architecture is proposed for the PCSM. In addition, the upper and lower bounds on probability of frame error for the transmitted symbol information, under successive cancellation decoding are derived and the PCSM scheme suits the bounds for the suiting architecture.