Abstract The ever-increasing demand for massive machine-type communication requires multiple communication channels to coexist simultaneously. Beyond the classical orthogonal multiple access (OMA), a promising emerging scheme is non-orthogonal multiple access (NOMA), which offers better bandwidth utilization. An important measure of efficiency is the achievable capacity region for the communicating channels, which defines the theoretical limits of any implementation. In the literature, this capacity region is usually presented only for two users. In this paper, we present a systematic method to determine the convex hull of the capacity region for any number of users in uplink NOMA communication. We also extend the discussion to cases where one or more users experience fading. We provide examples of capacity regions for some indicative cases and present the necessary formulas to apply our method in a general setting.

The wireless communication system used by an Unmanned Aerial Vehicle has special requirements. Two separate data streams are used, one for telemetry and control of the vehicle, and one for payload. The first stream is of relatively low data rate, but must be very robust. The second stream requires high data rate, but a best effort approach is acceptable. Real-time remote control usage mandates low response time (turn around latency) of the system, which demands small frames and frequent direction changes over the half duplex channel. These two different requirements must be accounted for in the design of the modulation system. In this paper our solution based on the joint application of a chirp spread spectrum modulation with a correlator based demodulator for the telemetry and control data stream and a high data rate orthogonal frequency division multiplexing (OFDM) modulation for the payload is presented. The OFDM communication occupies the same bandwidth as the spread spectrum modulation, but - unlike the Chirp signal - utilizes its full bandwidth. By using a correlation detector for all timing recovery tasks and keeping the same timing for the OFDM part of the frame, all control loops normally needed for this task (symbol synchronization PLL, COSTAS loop) can be eliminated. This results in the whole signal processing path being a finite impulse response system which aids us in a highly pipelineable design implementation and makes it possible to create a high speed implementation with a low-end field programmable gate array (FPGA) circuit.

This paper presents a new unified analytical technique for accurate interference analysis of noncoherent M-ary frequency shift keying (MFSK) wireless communication systems in the presence of additive white Gaussian noise (AWGN) and multiple arbitrary interfering signals. New exact bit error rate expressions are derived for nonorthogonal noncoherent binary FSK in the presence of an arbitrary number of interfering signals having arbitrary spectral shapes at arbitrary frequency offsets. Furthermore, tight upper bounds for the symbol and bit error rates are given for noncoherent MFSK with arbitrary tone spacings. These results are extended to include MFSK signals experiencing Rayleigh, Nakagami-m or Rician fading. Some specific interference scenarios are analyzed in details, including jammers with a given Doppler-spread, direct sequence spread spectrum and orthogonal frequency division multiplexing signals, and a Poisson field of arbitrary interferers.

This paper presents a new analytical performance analysis of slow frequency-hopped synchronous spread-spectrum multiple-access (FHSS-MA) networks employing nonorthogonal binary frequency-shift keying (BFSK) and having arbitrary spacing between the hopping carriers. Taking into account the crosstalk due to the nonorthogonality of both hopping carriers and BFSK tones, and based on an accurate interference analysis, we derive new expressions for the average probability of error of a reference signal subjected to a number of similar signals in the additive white Gaussian noise Rayleigh-faded channel. These are used to investigate the effect of carrier and tone spacing on the spectral efficiency of an FHSS-MA network. It is shown that the spectral efficiency depends primarily on the shape of the employed pulses, in addition to both carrier and tone spacing. Numerical results show that rectangular pulses give higher spectral efficiency as compared with the half-sine pulses. Furthermore, orthogonal tone spacing is shown to be more efficient in the case of rectangular pulses, whereas half-sine pulses become more efficient with the nonorthogonal signaling.

This paper presents new accurate packet error rate analysis of ALOHA type networks that employ BPSK or QPSK modulation in a Rayleigh fading environment. Relying on a precise interference model, we derive new accurate expressions for the packet success probabilities that account for bit to bit error dependence, and are valid for unequal power levels. The accuracy of the new expressions is validated by simulation. These results are used to investigate the throughput performance of Aloha type networks in different local area environments

Abstract Multiuser/multisubchannel detection schemes based on a recurrent neural network (RNN) structure are a promising alternative to conventional detection schemes, both in terms of performance and complexity. This paper gives an overview of different approaches that make use of the results of RNN research for multiuser/multisubchannel detection. The detectors' results are compared with the results of conventional detection schemes in simulations of the uplink of a DS‐CDMA packet transmission system. Copyright © 2002 John Wiley & Sons, Ltd.

We consider the performance of infrastructure wireless LANs based on frequency hopping transmission techniques. The service area is partitioned into overlapping cells, and each cell is assigned a distinct signature sequence that is shared by all terminals inside the cell. A sequence of request-to-send/clear-to-send (RTS/CTS) handshaking procedures precedes the transmission of data frames. Though handshake procedures prevent primary interference, the inter-cell interference lasts and may limit the performance. We develop a unified analysis framework for modeling the inter-cell interference and evaluating the performance of FHSS WLANs in the case of a finite number of users, fixed frame lengths and asynchronous frame transmissions.

This paper is concerned with the performance analysis of frequency-hopped packet radio networks with random signal levels. The authors assume that a hit from an interfering packet necessitates a symbol error if and only if it brings on enough energy that exceeds the energy received from the wanted signal. The interdependence between symbol errors of an arbitrary packet is taken into consideration through the joint probability generating function of the so-called effective multiple access interference. Slotted networks, with both random and deterministic hopping patterns, are considered in the case of both synchronous and asynchronous hopping. A general closed-form expression is given for packet capture probability, in the case of Reed-Solomon error only decoding. After introducing a general description method, the following examples are worked out in detail: (i) networks with random spatial distribution of stations (a model for mobile packet radio networks), (ii) networks operating in slow fading channels, and (iii) networks with different power levels which are chosen randomly according to either discrete or continuous probability distribution (created captures).>