This study addresses the challenge of robust human pose recognition in complex indoor environments by proposing a system that generalizes to unseen user locations using only Angle of Arrival (AoA) data from Bluetooth Low Energy (BLE) 5.1 devices. A comparative evaluation was performed, assessing a Random Forest model alongside three neural network architectures— CNN, LSTM, and hybrid models—for identifying four physical states: standing, sitting, lying down, and walking. Data were collected in a real indoor environment using smart Personal Protective Equipment (PPE) equipped with BLE transmitters placed on a helmet, upper and lower shirt regions, and a boot. The preprocessing pipeline included filtering, interpolation, feature extraction, and application of the Fast Fourier Transform (FFT). Models were evaluated under two scenarios: (i) known positions (used in training) and (ii) unknown positions (never seen before). The best overall performance on unknown positions was achieved by theAttention-based BiLSTM–CNN,reaching an accuracy ofA= 94.89% and a macro averaged F1-score ofF1= 95.59%, demonstrating strong generalization capabilities. In the scenario with known positions, the highest performance was obtained by theRandom Forest, withA= 99.95% andF1= 99.95%, followed by theDeep CNN,withA= 97.27% andF1= 97.28%, closely followed by theAttention-based BiLSTM–CNN,which also showed robust classification accuracy across all posture classes. The analysis confirmed that using four wearable devices led to the highest performance, and sensor combinations significantly affected accuracy. The study also provides a publicly available dataset to support reproducibility and benchmarking.

We study the cross-layer energy efficiency of cooperative MIMO networks. At the physical layer, the transmit power determines the outage probability and the number of contending nodes per unit area. At the medium access control (MAC) layer, accessing the channel takes additional time and energy, which impacts the throughput and the energy efficiency. Then, we propose a new cooperative MAC protocol based on IEEE 802.11 in order to coordinate cooperation and to improve the energy efficiency. Moreover, we focus on slow fading scenarios and we analyze the performance of antenna selection (AS), singular value decomposition (SVD) diversity beamforming, and spatial multiplexing (SM) transmission schemes. Results show that SVD and SM have smaller delay and, consequently, larger throughput; however, AS is more energy efficient.

In this paper we study the energy efficiency of multiple antennas (MIMO) networks considering both the physical (PHY) and the medium access control (MAC) layers. At the PHY layer, the transmit power determines the communication rate and the area containing nodes that contend for channel access. In turn, accessing the channel takes additional time and energy at the MAC layer, which impacts the throughput and the energy efficiency. Moreover, we focus in slow fading scenarios, so that we remodel former cross-layer models in order to be representative of quasi-static MIMO networks, in which we employ the e-capacity as the system metric. In addition, we analyze the impact of antenna selection (AS) and singular value decomposition (SVD) beamforming MIMO diversity transmission schemes in terms of throughput and energy efficiency. Results show that SVD has smaller delay and higher throughput; however, this scheme is significantly outperformed in terms of energy efficiency by AS.

This paper proposes an iterative power allocation algorithm that maximizes the energy efficiency of a multiple-relay cooperative multiple antenna (MIMO) network. Energy efficiency is defined as the achievable spectral efficiency normalized by the total power consumption, which also accounts for the RF circuitry consumption, modeled in a wireless sensor network. Moreover, relay selection is performed using the distributed opportunistic selection protocol, while we derive closed-form expressions for outage probability and energy efficiency for a single antenna (SISO) and two MIMO scenarios: 1) based on antenna selection (AS) and 2) based on single value decomposition (SVD) beamforming. In order to evaluate the performance of the proposed algorithm, we compare it to an exhaustive search optimal power allocation method. Results show that the proposed algorithm performs very close to an exhaustive search approach and that, although MIMO SVD has the best outage performance, AS has the best energy efficiency for low and moderate source-destination distances and spectral efficiencies.