The use of a reliable and accurate localization system is essential to ensure the smooth integration of visually impaired people into society and making their daily lives easier. Many of the proposed systems are based on the use of Machine Learning (ML) where increasingly complex models are introduced. However, this is not always necessary given that comparable performance can be achieved with simpler models by better understanding and adjusting key model parameters that impact model performance. In this paper, we present an indoor localization framework that applies a simple 1D Convolutional Neural Network (CNN) on a real collected LoRa dataset. We concentrate our work on studying and analyzing the impact of essential parameters and data criterion on the efficiency and performance of the system. We mainly study the impact of the LoRa frequency, the impact of the distribution of training points and the impact of the quality of collected fingerprints. Several conclusions have been drawn showing the correlation between localization performance and the specific features of propagation and model training.

In this paper, we propose a new PHY layer trans-mission scheme for the uplink of a Single Input Multiple Output (SIMO) massive Machine Type Communication (mMTC). Our scheme is based on a Chirp Spread Spectrum (CSS) modulation at the transmitter side and a joint Maximum-Ratio Combiner (MRC) and Successive Interference Cancellation (SIC) decoder. One of the key characteristics of CSS signals is the linear variation of the frequency carrier over time, with symbol-dependent cutoff points. In the frequency domain, this temporal behavior leads to a peak power with magnitude and carrier-frequency position that are symbol-dependent. Indeed, adjacent carriers to the peak-carrier also hold a substantial amount of power. In a multi-user context, the non-homogeneous power-peak distribution across the different symbols biases the in-terpretation of the spectrum at the receiver side. Indeed, in a noisy environment, the high-power on the neighboring carriers of the peak-one induces erroneous decision. These two issues are further exacerbated by the SIC decoder as errors propagate from one device to another, creating a bottleneck. To address these challenges, we propose to (i) balance the power at the peak-carriers regardless of the symbol; (ii) refine the position of the peak-power index using a trellis-coded modulation (TCM). Our results show that the power balance combined with the TCM refinement of the peak-power index enhances the error performance rate of the SIC decoder.

In this paper, we provide a software-defined radio implementation for our transmission scheme proposed in [1], [2] to secure the wireless transmission on a wiretap channel with a passive eavesdropper. We show that a message modulated with our BICM-IPM scheme can be correctly decoded at the legitimate receiver. However, the decoding of this message at the eavesdropper results in a near random bit stream.

Low Power Wide Area Networks (LPWANs) enable cost-effective, low power consumption, and long-range IoT applications but often operate under low Signal-to-Noise Ratio (SNR) conditions. Conventional deep learning models for signal recognition struggle to generalize in such dynamic environments. We propose a lightweight framework combining ensemble learning, a Mixture of Experts (MoE) with uncertainty-aware soft gating, and meta-adaptation using the Almost No Inner Loop (ANIL) method. Expert models trained at distinct lowSNR levels share frozen base weights, while only the classifier head is adapted during few-shot learning for rapid specialization. The uncertainty-aware gating mechanism produces sparse, temperature-controlled expert weights, enhancing robustness and decision reliability. Experiments demonstrate improved accuracy and generalization across both seen and unseen SNR levels, highlighting the framework's effectiveness for real-world wireless signal technology recognition in spectrum monitoring systems.

In this paper, we propose a noisy and dynamic-index partitioned modulation (IPM) to secure a quadrature amplitude modulation (QAM) constellation transmission over a non-degraded wiretap channel. Unlike approaches that rely on multi-antenna systems or tight amplitude-phase tracking, IPM partitions QAM into multiple disjoint subsets that are separately indexed by a dynamic key, known at the transmitter (Alice) and the legitimate receiver (Bob), but not at the eavesdropper (Eve). The proposed IPM maps the information bits into multiple sequences, each one lying in a different partition of the QAM constellation space. This mapping is performed through a cross-bit labeling that we define to increase the confusion at the eavesdropper, while minimizing the bit error rate (BER) at the legitimate receiver. As the eavesdropper is not aware of the dynamic index pointing to the different partitions, the IPM creates larger Voronoï detection zones around Bob’s symbol compared to Eve. To induce further confusion at Eve, we inject random uniform noise into IPM symbols. The noise varies in a domain that is fully included in Bob’s larger detection zone, but it exceeds Eve’s detection zone. The performances of the noisy IPM scheme in terms of error rate and secrecy rate are analytically and numerically evaluated. Our results show that the IPM scheme creates on the Eavesdropper’s link an error floor, independently of its signal-to-noise ratio (SNR). However, the IPM scheme preserves the legitimate link, for which the BER decreases as the SNR increases. Indeed, the secrecy rate remains positive for all SNR values and achieves an asymptotical constant plateau value.

The increase in the number of IoT devices invading both private and professional spaces opens the way for different cyberattacks. Facing such threats requires new approaches capable of reading the context of a given situation and acting accordingly to protect the users’ data and applications. Logical-based approaches can be harnessed in this case because of the semantic dimension they model and implement. In this paper, we make use of an existential rule-based knowledge base to model an IoT environment and detect anomalies as inconsistencies inside the obtained logical system. Accordingly, we first introduce an algorithm for knowledge base rewriting into a context-based knowledge base. Then, we detail our contextualized derivation algorithm for inconsistency detection in such a logical system.

In this paper, we consider a Low Power Wide Area Network (LPWAN) operating in a licensed-exempt band. The LoRa network provides long-range, wide-area communications for a large amount of objects with limited power consumption. In terms of link budget, nodes that are far from the collector suffer collisions caused by nodes that are close to the collector during the data transmissions. Chirp Spread Spectrum (CSS) modulation is adopted by assigning different Spreading Factors (SF) to active sensors to help reduce destructive collisions in LoRa network. In order to improve the energy efficiency and communication reliability, we propose an application of multi-agent Q-learning algorithm in the dynamic allocation of power and SF to the active nodes for uplink communications in LoRa. The main objective of this paper is to reduce power consumption of the uplink transmissions and to improve the network reliability.

In this paper, we consider a latency tolerant 2× nrfree space optical multiple-input multiple-output (FSO-MIMO) communication with a time diversity order of two. For this configuration, we propose an extended golden light code (E-GLC) construction that encodes data over two independent optical MIMO channel realizations. The main objective of this code construction is to jointly extract the spatial and the temporal diversity while keeping a multiplexing gain of two symbols per channel use. The proposed coding scheme is compliant with the optical intensity modulation and direct detection constraints that restrict the laser outputs to positive magnitude values with peak and average constraints. The error rate performance evaluations show that the E-GLC outperforms the other known replication-based schemes.

In this paper, we propose a space time code construction with a symbol rate of two for free space optical multiple input multiple output (FSO-MIMO) communication systems. Unlike the radio-frequency space time coding design that deals with complex symbols holding phase and magnitude information, the phase detection in wireless optical communications remains very challenging. This restricts optical MIMO codes to carry positive magnitude information belonging to a finite range. To meet these requirements, we propose to design the FSO-MIMO codes considering first the whole real space R, followed by a bipolar to unipolar conversion to fit in the intensity margin. We consider the MIMO configurations with two or four transmitters (lasers). We base our shape-preserving construction on quadratic extension fields that linearly combine the real symbols to form a full diversity space time code with a symbol rate equal to two. For each configuration, we show how to optimize the choice of the algebraic field numbers to jointly optimize the non-vanishing code determinant and the bipolar to unipolar attenuation factor.

In this paper, we propose a space time code construction for the 2 × 2 optical Multiple Input Multiple Output (MIMO) communication systems. The code construction, denoted Golden-Light Code, is compliant with the optical limited range constraints and the requirements of intensity modulation and direct detection. We base our construction on a two-parametric dependent full-rate 2 × 2 MIMO coding scheme with real components that conserves the real symmetrical constellation shape. Then, we show how to optimize the choice of these two parameters to jointly optimize the non-vanishing code determinant, and the bipolar to unipolar attenuation factor.

Dimensioning the LTE maritime Mobile Ad-hoc Networks relies essentially on the accuracy of the channel propagation path-loss model. The maritime ITU-R path-loss is computed based on empirical field strength measures that statistically capture, depending on the weather condition, the behavior of the wireless channel over a certain percentage of time. In this paper, we provide an analytical method to linearly adjust the required number of radio resource blocks (RBs) computed in [1] in function of the channel fluctuations. Based on a reference path-loss model, we compute the variation in the number of RBs required to achieve the same target resource outage probability as a function of the path-loss exponent and attenuation factor variations.

This work gives an analytical model for the radio dimensioning of uplink Long Term Evolution (LTE) systems. It relies on the computation of an upper bound of the outage probability, the outage being defined by an insufficient number of Resource Blocks (RBs) for serving all the users in the network. The number of RBs, that has to be dedicated to each communication in order to ensure its Quality of Service (QoS), depends on the transmitting users random positions and their radio conditions. Throughout this paper, this number of RBs is assimilated as a marked Poisson point process. The upper bound of the outage probability is computed for different service classes according to the users throughput requirements, for the fair and the opportunistic RB allocation algorithms. Once validated by Monte-Carlo simulations, it estimates the necessary cells bandwidth. Finally, we apply these results to LTE standard bandwidths and give the associated resulting outage probabilities. Our work shows that the analytical upper bound of the outage probability helps the radio system planners to decrease the networks outage probability to a desired threshold at the price of a slight and controlled radio over dimensioning.

In this paper, we propose a transceiver structure for the multi-user MIMO uplink channel of the Long Term Evolution (LTE) networks. This transceiver structure is based on the use of an adequate encoder at the transmitter side, to take advantage of the multiple transmit antenna diversity, and a combination of a joint zero forcing (ZF) and a maximum likelihood (ML) decoder at the receiver side. Based on this transceiver design, we show how to extend an opportunistic multiuser spatial multiplexing allocation to the whole LTE bandwidth with respect to Single Carrier Frequency Division Multiple Access (SC-FDMA) constraints. The evaluated rate performances of the proposed transceiver in a scheduling uplink LTE context outperform the ones of other well known decoders.

This work proposes two new Resource Block (RB) allocation algorithms for the LTE uplink. They take into account the RB adjacency constraint and update the allocation metric. Two different heuristics are proposed: an Opportunistic and Efficient RB allocation Algorithm (OEA) and a Quality of Service based Opportunistic and Efficient RB allocation Algorithm (QoS based OEA). Both algorithms seek to maximize the aggregate throughput and avoid RB wastage unlike other algorithms in the literature. The complexity of the proposed algorithms are also computed analytically and compared to other well known heuristics.

The design of perfect space time codes constructed from cyclic division algebra (CDA) on the quasi-static MIMO channel has received lots of attention in industry over the last few years. However, the recent standards that use multiple antennas terminals such as IEEE 802.11n, IEEE 802.16e or LTE are based on more realistic assumptions involving the use of outer codes and multi-taps channels. In this paper, the question regarding the utility of using a high dimensional modulation schemes such as space time codes in a standard context is addressed. For this, we evaluate the performance of the perfect space time codes in term of an upper-bound on the pairwise error probability, and we compare it to the one of simple spatial division multiplexing (SDM) schemes. Finally, simulation results considered in the context of IEEE 802.11n show that the 2 × 2 MIMO perfect code, i.e. the golden code performs relatively similarly as the SDM scheme in the presence of a good outer code.

In this paper, we study the feasibility of the interference alignment (IA) scheme for the selective fading interference channel with time-frequency selective links. Unlike the frequency selective channel case, we show here that, for the scalar time-frequency (TF) selective underspread channels, there is no need to scale the bandwidth as the number of users. However, a constraint on the delay-Doppler spread of the channel itself is required. A physical interpretation of this condition in the time-frequency domain is provided and is shown to be easily satisfied in practical system. Our results are based on the polynomial channel decomposition that we derive for the TF channel with large signal duration and bandwidth.

In this paper, we extend the non-vanishing determinant (NVD) criterion to the selective fading channel case. A new family of split NVD parallel codes, that satisfies this design criterion, is proposed to achieve the optimal diversity multiplexing tradeoff (DMT) derived by Coronel and Bölcskei, IEEE ISIT, 2007. This result is of significant interest not only in its own right, but also because it settles a long-standing debate in the literature related to the optimal DMT of selective fading channels.

In a broadcast channel with random packet arrival and transmission queues, the stability of the system is achieved by maximizing a weighted sum rate capacity with suitable weights that depend on the queue size. The weighted sum rate capacity using Dirty Paper Coding (DPC) and Zero Forcing (ZF) is asymptotically equivalent to the weighted sum capacity over parallel single-channels. In this paper, we study the Diversity Multiplexing Tradeoff (DMT) of the fading broadcast channel under a fixed weighted sum rate capacity constraint. The DMT of both identical and different parallel weighted MISO channels is first derived. Finally, we deduce the DMT of a broadcast channel using DPC and ZF precoders.

In this paper, the selective time division multiple access (S-TDMA) strategy is studied in the downlink channel. This strategy consists in transmitting data to the user with the largest capacity. The diversity and multiplexing gains that can be achieved by this sub-optimal strategy are evaluated and then compared to the optimal gains over the broadcast channel. Codes construction is then proposed to achieve the diversity multiplexing tradeoff (DMT) of the S-TDMA. These codes are extended to the scenario of cooperating asynchronous broadcasting base stations where new codes that are suitable for this scenario are proposed and analyzed.

In this paper, we consider the isotropic fading broadcast channel. This channel refers to the case when no directional information is available at the transmitter side, and was studied by Jafar et al in. It was shown that the isotropic vector broadcast channel (BC-V) can be reduced to an equivalent scalar broadcast channel (BC-S). It is well known from [2]http://www.pimrc2008.org/ that BC-S is degraded in the same order as the channel magnitude. This implies that the optimal strategy that maximizes the sum capacity for BC-S and BCV consists on allocating the whole power to the strongest user. Based on these results, we derive in this paper the diversity multiplexing tradeoff (DMT) of the isotropic fading BC-S and BC-V, and we propose optimal schemes that achieve these DMT.