Performance of nonorthogonal FSK for the Internet of Things

IEEE 802.15.4 is a popular standard for low-rate wireless communications used for the Internet of Things (IoT). This IEEE standard contains a variety of physical layers (such as SUN, TVWS, RCC LMR, and LECIM) that make use of different modulations, such as nonorthogonal frequency-shift keying (FSK) and position-based FSK. This paper proposes simple closed-form expressions to evaluate the bit-error rate (BER) performance of nonorthogonal FSK schemes used for the IoT. Both coherent and noncoherent detection cases are investigated in additive white Gaussian noise (AWGN) and frequency-nonselective fading channels. In case of AWGN channels, the proposed BER expressions are obtained by selecting and modifying two terms of a union bound on the error probability. In case of fading channels, the proposed BER expressions are obtained by integration of the conditional BER over the fading statistics. Simulation results confirm the good accuracy of the proposed analytical BER expressions for different channel models (such as AWGN, Nakagami-m, Rice, lognormal) and for different detection techniques (such as noncoherent, coherent, and coherent with a residual phase error). Experimental results with wireless transmission and reception confirm the accuracy of the proposed approach. In addition, this paper also presents a BER expression for coherent detection of position-based FSK. Simulation results demonstrate that the presented BER expression for position-based FSK is more accurate than the BER expressions proposed in the literature.