Electromagnetic Emission-Aware Schedulers for the Uplink of OFDM Wireless Communication Systems

Abstract

Abstract-The popularity and convergence of wireless communications have resulted in continuous network upgrades in order to support the increasing demand for bandwidth. However, given that wireless communication systems operate on radiofrequency waves, the health effects of electromagnetic emission from these systems are increasingly becoming a concern due to the ubiquity of mobile communication devices. In order to address these concerns, we propose two schemes (offline and online) for minimizing the EM emission of users in the uplink of OFDM systems, while maintaining an acceptable quality of service. We formulate our offline EM reduction scheme as a convex optimization problem and solve it through water-filling. This is based on the assumption that the long-term channel state information of all the users is known. Given that, in practice, long-term channel state information of all the users cannot always be available, we propose our online EM emission reduction scheme, which is based on minimizing the instantaneous transmit energy per bit of each user. Simulation results show that both our proposed schemes significantly minimize the EM emission when compared to the benchmark classic greedy spectral efficiency based scheme and an energy efficiency based scheme. Furthermore, our offline scheme proves to be very robust against channel prediction errors.

Electromagnetic Emission-Aware Scheduling for the Uplink of Multicell OFDM Wireless Systems

“…We further assume that all the users transmit the same target number of bits, i.e., B m k = B ∀k, m, while the fast fading and path-loss follow the model used in [14]. We compare our scheme against a state of the art uplink multicell (MC) SE maximization scheme of [16] and two single cell (SC) schemes, i.e., our EM emission reduction scheme of [14] and a classic SE maximization scheme [11], where each sector uses different frequency bands i.e., the number of subcarriers is split amongst the M sectors, to avoid interference.…”

“…In order to ensure that both (8a) and (8b) are met, the transmission duration could be varied depending on the target number of bits. Indeed, we have shown in [14] that EM emission reduces as the transmission duration increases. This is because more subcarriers become available to the users and, as a result, a lower transmission power is needed to transmit the target number of bits.…”

“…• Simulation results demonstrate that our proposed scheme significantly outperforms the single cell EM emission minimization scheme of [14] (i.e., without intercell interference) by up to 70%, and by over 3 and 4 orders of magnitude when compared to the multicell SE maximization scheme of [16] and the single cell SE maximization scheme of [11], respectively. Furthermore, the results show that the uplink EM emission in the network is proportional to the number of users and their QoS requirements.…”

“…The subcarrier ordering is done in ascending order of the worst Γ j,n (t) on each subcarrier, as performed in [14]. Starting from the first column, the user with the best Γ j,n (t) on that subcarrier is allocated to it and then we move to the next column until S subcarriers are allocated to each user, where S = N T /K and .…”

“…In the single cell scenario, subcarrier and power allocation in the uplink of OFDM systems have been very well investigated but mainly for SE improvement [10], [11] and, recently, EE [12], [13]. In the context of EM emission reduction, in [14], we proposed a scheduler for the uplink of OFDM systems to minimize EM emissions and maintain QoS, based on the assumption that the channel state information (CSI) of all the users can be predicted over a certain transmission window (i.e., spanning several time slots). Our scheme takes into account the signaling power of all the users in the network and data transmission is confined to happen within the transmission window.…”