In this paper, we consider the problem of admission control in 5G networks where enhanced mobile broadband (eMBB) users and ultra-reliable low-latency communication (URLLC) users are coexisting. URLLC users require low latency and high reliability while eMBB users require high data rates. Thus, it is essential to control the admission of eMBB users while giving priority to all URLLC users in a network where both types of users are coexisting. Our aim is to maximize the number of admitted eMBB users to the system with a guaranteed data rate while allocating resources to all URLLC users. We formulated this as an l_0 minimization problem. Since it is an NP-hard problem we have used approximation methods and sequential convex programming to obtain a suboptimal solution. Numerically we have shown that the proposed algorithm achieves near-optimal performance. Our algorithm is able to maximize the number of admitted eMBB users with an optimal allocation of resources while giving priority to all URLLC users.
The 5th generation (5G) mobile communication systems are expected to revolutionize everything seen so far in wireless systems. The requirements for 5G vary by application but will include data rates ranging from very low sensor data to very high video content delivery, stringent low latency requirements, low energy consumption, and high reliability. All of these technological requirements are expected to be achieved while keeping the same or lower cost than today's technologies. The application scenarios range from usual broadband mobile to machine-to-machine communications, real-time highly reliable control with low latency and low data rate sensor networks with large number of nodes, to mention a few.
Interference prediction and resource allocation are critical challenges in mission-critical applications where stringent latency and reliability constraints must be met. This paper proposes a novel Gaussian process regression (GPR)-based framework for predictive interference management and resource allocation in future 6G networks. Firstly, the received interference power is modeled as a Gaussian process, enabling both the prediction of future interference values and their corresponding estimation of uncertainty bounds. Differently from conventional machine learning methods that extract patterns from a given set of data without any prior belief, a Gaussian process assigns probability distributions to different functions that possibly represent the data set which can be further updates using Bayes' rule as more data points are observed. For instance, unlike deep neural networks, the GPR model requires only a few sample points to update its prior beliefs in real-time. Furthermore, we propose a proactive resource allocation scheme that dynamically adjusts resources according to predicted interference. The performance of the proposed approach is evaluated against two benchmarks prediction schemes, a moving average-based estimator and the ideal genie-aided estimator. The GPR-based method outperforms the moving average-based estimator and achieves near-optimal performance, closely matching the genie-aided benchmark.
We study the sensitivity of stable rates to imperfect sensing in cognitive radio systems comprised of a set of source-destination pairs having different priorities. The adopted cognitive access protocol allows the secondary user not only to exploit the idle slots of the primary user but also to transmit along with the primary user with some probability. This is aimed at achieving full utilization of the shared channel with capture. The abolition of strong primacy, however, requires the secondary user to properly regulate its multi-access probability in order not to impede the primary user's stability guarantee at any stabilizable input demand. To this end, the stability region of the system is characterized which describes the theoretical limit on rates that can be pushed into the system while maintaining the queues stable. Interestingly, we found that even with non-zero sensing error rates, there exists a condition for which we can achieve identical stability region that is achieved with perfect sensing, i.e., the stability is insensitive to the sensing errors. This happens when relatively strong capture effect is present. For the case when the stability is sensitive to the sensing errors, we precisely quantify the loss due to the imperfect sensing in terms of the size of the stability region.
The wide-band code-division multiple-access (WCDMA) concept FMA2 developed in the European Future Radio Wide-Band Multiple-Access Systems (FRAMES) project supports variable-data-rate transmission. The data rate can be altered by changing either the spreading factor or by using multiple spreading codes in parallel. In order to facilitate flexible changing of the data rate, variable-length orthogonal Walsh codes are used for channel separation, followed by the scrambling code. Optional short scrambling codes can be used if interference suppression is used. The downlink performance of the FMA2 system using both the conventional RAKE and LMMSE-RAKE receivers is studied. According to the numerical results, the performance of the conventional RAKE receivers is significantly degraded at the highest data rates, whereas the LMMSE-RAKE performs well in those cases also.
In this paper, using the so-called parallel space-time Markov chain (PSTMC) framework we analyze the IEEE 802.11 Distributed Coordination Function (DCF) frame service time, jitter, and queuing delay in a single-hop non-saturated wireless network. PSTMC framework provides the possibility of simultaneous modeling of backoff and post-backoff procedures, in addition to the transmission queue status of a non-saturated 802.11 station. To the best of our knowledge, the presented contribution is the first analysis of service time, i.e., access delay and retransmission delay, plus queuing delay at the same time, when the precise modeling of binary exponential backoff (BEB) scheme in medium access control (MAC) layer is the main issue of concern. The model is validated by extensive simulations, showing its remarkable level of accuracy.
End-to-end learning of a communications system using the deep learning-based autoencoder concept has drawn interest in recent research due to its simplicity, flexibility and its potential of adapting to complex channel models and practical system imperfections. In this paper, we have compared the bit error rate (BER) performance of autoencoder based systems and conventional channel coded systems with convolutional coding (CC), in order to understand the potential of deep learning-based systems as alternatives to conventional systems. From the simulations, autoencoder implementation was observed to have a better BER in 0-5 dB $E_{b}/N_{0}$ range than its equivalent half-rate convolutional coded BPSK with hard decision decoding, and to have only less than 1 dB gap at a BER of $10^{-5}$. Furthermore, we have also proposed a novel low complexity autoencoder architecture to implement end-to-end learning of coded systems in which we have shown better BER performance than the baseline implementation. The newly proposed low complexity autoencoder was capable of achieving a better BER performance than half-rate 16-QAM with hard decision decoding over the full 0-10 dB $E_{b}/N_{0}$ range and a better BER performance than the soft decision decoding in 0-4 dB $E_{b}/N_{0}$ range.
Coming cellular systems are envisioned to open up to new services with stringent reliability and energy efficiency requirements. In this paper we focus on the joint power control and rate allocation problem in Single-Input Multiple-Output (SIMO) wireless systems with Rayleigh fading and stringent reliability constraints. We propose an allocation scheme that maximizes the energy efficiency of the system while making use only of average statistics of the signal and interference, and the number of antennas $M$ that are available at the receiver side. We show the superiority of the Maximum Ratio Combining (MRC) scheme over Selection Combining (SC) in terms of energy efficiency, and prove that the gap between the optimum allocated resources converges to $(M!)^(1/(2M))$ as the reliability constraint becomes more stringent. Meanwhile, in most cases MRC was also shown to be more energy efficient than Switch and Stay Combining (SSC) scheme, although this does not hold only when operating with extremely large $M$, extremely high/small average signal/interference power and/or highly power consuming receiving circuitry. Numerical results show the feasibility of the ultra-reliable operation when $M$ increases, while greater the fixed power consumption and/or drain efficiency of the transmit amplifier is, the greater the optimum transmit power and rate.
